MEMORANDUM OPINION ON MOTION FOR PRELIMINARY INJUNCTION

HAMILTON, District Judge.

The plaintiffs challenge the constitutionality of a new Indiana statute regulating abortions and seek a preliminary injunction against its enforcement. The law, known as Public Law 187, requires in almost all cases that at least 18 hours before an abortion can be performed, a woman must be given certain medical information and information concerning alternatives to abortion. Public Law 187 is similar to a Pennsylvania law upheld by the Supreme Court in Planned Parenthood of Southeastern Pennsylvania v. Casey, 505 U.S. 833, 112 S.Ct. 2791, 120 L.Ed.2d 674 (1992). Plaintiffs contend that differences between the Indiana law and the Pennsylvania law together with recently available evidence about the actual effects of such laws show that Public Law 187 violates *1440 the “undue burden” test adopted by the Supreme Court in Casey.

The court’s decision on the motion for preliminary injunction is not the court’s last word on this matter. It is instead a decision based on incomplete evidence in the face of threatened irreparable harm. As explained in detail below, the court concludes that plaintiffs have established a reasonable likelihood of prevailing on the merits of their challenges to Public Law 187. They have shown that Indiana’s requirement that some information be provided “in the presence” of the woman at least 18 hours before the abortion is likely to impose an undue burden on a woman’s right to choose whether to continue or terminate a pregnancy. Unlike the plaintiffs in Casey, the plaintiffs here have come forward with evidence showing that the burdens of the law are likely to prevent abortions for approximately 11 to 14 percent of women who would otherwise choose to have them. Plaintiffs have also shown that the “medical emergency” exception of Public Law 187 probably fails to meet constitutional standards. Under principles of Indiana law, and based on the law’s legislative history, it appears that the mandatory disclosure and waiting period requirements apply even when compliance would cause severe but temporary physical problems for a woman, or when compliance would cause severe psychological harm. (However, as defendants have requested, the court will certify the state law issues of statutory interpretation for decision by the Supreme Court of Indiana before this court makes a final decision on the constitutional issues in this case.) Plaintiffs have also shown that enforcement of the law while this lawsuit is pending would cause irreparable harm to a significant number of women. On the other side of the scale, defendants have not shown that temporarily preserving the status quo is actually likely to cause substantial irreparable harm. The court will therefore enjoin enforcement of Public Law 187 pending a trial on the merits.

I. The Indiana Statute

Since long before 1995, Indiana has required physicians performing abortions to obtain the informed consent of their patients. Abortions in Indiana are criminal unless “the woman submitting to the abortion has filed her consent with her physician.” Ind.Code § 16-34-2-l(l)(B). Informed consent generally requires that the patient be told the general nature of her condition, the proposed treatment or procedure, the expected outcome, the material risks, and the reasonable alternatives to the treatment or procedure. See Ind.Code § 27-12-12-2 (informed consent for purposes of medical malpractice action).

Indiana Public Law 187-1995 (referred to here as “Public Law 187”) adds special mandatory disclosure and waiting period provisions for informed consent for abortions. The law requires in almost all eases that certain medical information and information about alternatives to abortion be provided to a woman orally at least 18 hours before she may have an abortion. Some of the medical information must be provided “in the presence of the pregnant woman.” The law was to have gone into effect on September 1, 1995, but was enjoined from operation by this court’s temporary restraining order.

The central provisions of Public Law 187 state:

An abortion shall not be performed except with the voluntary and informed consent of the pregnant woman upon whom the abortion is to be performed. Except in the case of a medical emergency, consent to an abortion is voluntary and informed only if the following conditions are met:

(1) At least eighteen (18) hours before the abortion and in the presence of the pregnant woman, the physician who is to perform the abortion, the referring physician or a physician assistant (as defined in IC 25-27.5-2-10), an advanced practice nurse (as defined in IC 25-23-l-l(b)), or a midwife (as defined in IC 27-12-2-19) to whom the responsibility has been delegated by the physician who is to perform the abortion or the referring physician has orally informed the pregnant woman of the following:

(A) The name of the physician performing the abortion.

*1441 (B) The nature of the proposed procedure or treatment.

(C) The risks of and alternatives to the procedure or treatment.

(D) The probable gestational age of the fetus, including an offer to provide:

(i) a picture of a fetus;

(ii) the dimensions of a fetus; and

(in) relevant information on the potential survival of an unborn fetus;

at this stage of development.

(E) The medical risks associated with carrying the fetus to term.

(2) At least eighteen (18) hours before the abortion, the pregnant woman will be orally informed of the following:

(A) That medical assistance benefits may be available for prenatal care, childbirth, and neonatal care from the county office of family and children.

(B) That the father of the unborn fetus is legally required to assist in the support of the child. In the case of rape, the information required under this clause may be omitted.

(C) That adoption alternatives are available and that adoptive parents may legally pay the costs of prenatal care, childbirth, and neonatal care.

(3) The pregnant woman certifies in writing, before the abortion is performed, that the information required by subdivisions (1) and (2) has been provided.

Ind.Code § 16-34-2-1.1. Anyone who “knowingly” or “intentionally” performs an abortion in violation of these requirements is subject to criminal penalties. Ind.Code §§ 16-34-2-1, 16-34-2-7.

Public Law 187 contains an exception to the disclosure and waiting period requirements when a woman faces a “medical emergency.” As defined in Public Law 187, a “medical emergency”

means a condition that, on the basis of the attending physician’s good faith clinical judgment, complicates the medical condition of a pregnant woman so that it necessitates the immediate termination of her pregnancy to avert her death or for which a delay would create serious risk of substantial and irreversible impairment of a major bodily function.

Ind.Code § 16-18-2-223.5. A physician who believes there is a medical emergency must provide information concerning that medical judgment to the woman:

When a medical emergency compels the performance of an abortion, the physician who will perform the abortion shall inform the woman, before the abortion if possible, of the medical indications supporting the physician’s judgment that an abortion is necessary to avert:

(1) the woman’s death; or

(2) a substantial and irreversible impairment of a major bodily function.

Ind.Code § 16-34-2-1.2.

II. The Parties and Their Claims

The plaintiffs are seven reproductive health care facilities that provide a range of services related to pregnancy and women’s health, including abortions up to 12 weeks of gestation; and Ulrich G. Klopfer, a licensed physician in Indiana who performs abortions. Plaintiffs filed their complaint on August 24, 1995, asserting that Public Law 187 will impose undue burdens on women’s constitutional right to choose to have an abortion. 1 Defendants are a class of all prosecuting attorneys in the State of Indiana, with Scott C. Newman of Marion County as representative of the class; and John C. Bailey, in his official capacity as Commissioner of the Indiana Department of Health.

On August 25, 1995, asserting that the enforcement of Public Law 187 would cause immediate and irreparable harm, plaintiffs filed a verified motion for a preliminary injunction and expedited hearing. Because Public Law 187 was scheduled to take effect *1442 September 1, 1995, the court treated the motion as also requesting a temporary restraining order with notice. Defendants filed a brief in opposition. After a hearing on August 30, 1995, the court entered a temporary restraining order enjoining enforcement of the challenged law for ten days. The court also issued an order certifying the defendant class. The court deferred ruling on plaintiffs’ motion for a preliminary injunction pending a further evidentiary hearing. The parties agreed to extend the temporary restraining order, and the court heard evidence on plaintiffs’ motion for a preliminary injunction on October 11-13, 1995. The parties have submitted their post-hearing briefs and some additional evidence of the law’s legislative history.

Plaintiffs contend that the new mandatory disclosure and 18-hour waiting period requirements will impose an undue burden on a woman’s right to choose whether to continue or terminate a pregnancy. Recognizing that the Supreme Court upheld a similar law against similar challenges in Casey, plaintiffs focus most of their attack on the likely effects of the requirement that some information be provided “in the presence of the pregnant woman” at least 18 hours before the abortion. This would require most women to make two trips to a clinic in order to get an abortion. Plaintiffs contend this requirement has no legitimate purpose and would effectively prevent some women from exercising their right to choose. Plaintiffs assert that the women who would be most affected by these burdens include women who, for a variety of reasons, would have difficulty explaining absences from husbands, partners, families, jobs, or schools; women who would have trouble paying for abortions; and women who would be especially vulnerable to harassment from anti-abortion protesters. They offer evidence that was not available in Casey concerning the actual effects of such laws. Plaintiffs also contend that the medical emergency exception for the mandatory disclosure and waiting period requirements is written too narrowly to meet constitutional requirements to protect the woman’s health.

Defendants contend that plaintiffs cannot succeed on the merits of these claims because the Supreme Court upheld a very similar Pennsylvania law in Casey. They argue that plaintiffs have not provided any persuasive evidence of undue burdens beyond the evidence that was held insufficient in Casey. They also argue that, despite its narrow wording, the Indiana medical emergency exception should be given the same kind of saving construction that the Third Circuit gave the Pennsylvania statute in Casey. 2

III. Preliminary Injunction Standard

Before the court may enter a preliminary injunction, the moving party must demonstrate “(1) some likelihood of succeeding on the merits, and (2) that it has ‘no adequate remedy at law1 and will suffer ‘irreparable harm’ if preliminary relief is denied.” Abbott Labs. v. Mead Johnson & Co., 971 F.2d 6, 11 (7th Cir.1992). If those elements are not satisfied, then the inquiry is over and the preliminary injunction must be denied. If those elements are met, the court then proceeds to consider “(3) the irreparable harm the non-moving party will suffer if preliminary relief is granted, balancing that harm against the irreparable harm to the moving party if relief is denied; and (4) the public interest, meaning the consequences of granting or denying the injunction to non-parties.” Abbott Labs., 971 F.2d at 11-12. Accord, e.g., Vencor, Inc. v. Webb, 33 F.3d 840, 845 (7th Cir.1994); Lawson Products, Inc. v. Avnet, Inc., 782 F.2d 1429, 1433 (7th Cir.1986).

*1443 Preliminary injunctions are designed to “minimize the hardship to the parties pending the ultimate resolution of the lawsuit,” Roth v. Lutheran General Hospital, 57 F.3d 1446, 1453 (7th Cir.1995), and “to preserve the court’s power to render a meaningful decision after a trial on the merits.” 11A Wright, Miller & Kane, Federal Practice & Procedure: Civil 2d § 2947, at 121. Some courts have described the purpose of a preliminary injunction as the preservation of the status quo. 11A Wright, Miller & Kane § 2948, at 133. E.g., E.E.O.C. v. City of Janesville, 630 F.2d 1254, 1259 (7th Cir.1980). See also Roland v. Air Line Employees Ass’n Int'l, 753 F.2d 1385, 1392 n. 9 (7th Cir.1985) (describing how preserving status quo can mean either mandating or prohibiting action in order to prevent irreparable injury).

A preliminary injunction is not a court’s last word on the question of injunctive relief, since the court is not called upon at that time to render a final decision on the merits. Therefore, at the preliminary injunction stage, the moving party does not need to prove her case beyond dispute. If the balance of harms is lopsided enough, she may not even need to prove she is more likely to prevail than not. There is no precise standard for the quantum of evidence required to obtain a preliminary injunction. Rather, the strength of the evidence and the potential harms involved are factors balanced on what the Seventh Circuit has called a discretionary sliding scale:

The [district] court, sitting as would a chancellor in equity, ... “weighs” all four factors in deciding whether to grant the injunction, seeking at all times to “minimize the costs of being mistaken.” We call this process the “sliding scale” approach: the more likely it is the plaintiff will succeed on the merits, the less the balance of irreparable harms need weigh towards its side; the less likely it is the plaintiff will succeed, the more the balance need weigh towards its side. This weighing process, as noted, also takes into consideration the consequences to the public interest of granting or denying preliminary relief. While we have at times framed the sliding scale approach in mathematical terms, it is more properly characterized as subjective and intuitive, one which permits district courts to “weigh the competing considerations and mold appropriate relief.”

Abbott Labs., 971 F.2d at 12 (citations and footnote omitted). Accord, Vencor, 33 F.3d at 845; Storck USA, L.P. v. Farley Candy Co., 14 F.3d 311, 314 (7th Cir.1994). See also 11A Wright, Miller & Kane § 2948.3, at 189-97. In fact, the Seventh Circuit has said that “[i]f the harm to the plaintiff from denial of the preliminary injunction would be very great and the harm to the defendant from granting it very small, then the injunction should be granted even if the defendant has a better chance of prevailing on the merits than the plaintiff, provided the plaintiffs chances are better than negligible; and vice versa.” Omega Satellite Products Co. v. City of Indianapolis, 694 F.2d 119, 123 (7th Cir.1982). Accord, Roland Machinery Co. v. Dresser Industries, Inc., 749 F.2d 380, 387-88 (7th Cir.1984). This discretionary sliding scale reflects the facts that only a limited amount of information may be available as evidence at the preliminary injunction stage, and that weighing risks obviously involves a fair amount of uncertainty. See generally Lawson Products, 782 F.2d at 1432-36.

This standard is critical here. Some evidence has not been fully developed and other evidence has not been fully subjected to scrutiny by the opposing side. The evidence presented thus far tends to show that portions of the new law will impose undue burdens on women, but that evidence is not complete and leaves room for some doubt. Defendants have not come forward at this stage with evidence showing that women are actually unlikely to experience such burdens. Thus, at this point, plaintiffs appear more likely to succeed than not. The evidence as to irreparable harm points decisively in favor of plaintiffs. The court will therefore grant the plaintiffs’ motion for preliminary injunction to prevent irreparable harm and to preserve the status quo until the case may be heard on the merits.

IV. Likelihood of Success on the Merits

The principles that govern the merits of plaintiffs’ challenges to the Indiana law are *1444 set forth in Casey, which must be reviewed in some detail. Those standards must then be applied to the evidence presented here concerning Indiana’s mandatory disclosure and waiting period provisions, and finally to the “medical emergency” exception for those requirements.

A. Roe, Casey, and the Controlling “Undue Burden” Standard

A fundamental right of “personal privacy” is one substantive aspect of the liberty protected by the Due Process Clause of the Fourteenth Amendment. E.g., Carey v. Population Services Int’l, 431 U.S. 678, 684, 97 S.Ct. 2010, 2015, 52 L.Ed.2d 675 (1977). In Roe v. Wade, 410 U.S. 113, 93 S.Ct. 705, 35 L.Ed.2d 147 (1973), the Supreme Court applied constitutional protection to a woman’s privacy interest in deciding whether to continue or terminate a pregnancy. At the same time, the Court acknowledged that states may regulate abortions to protect the life and health of the pregnant woman and, in some circumstances, may regulate to protect the fetus because of the state’s “interest in protecting the potentiality of human life.” Roe, 410 U.S. at 162, 93 S.Ct. at 731. The Roe Court balanced these interests in a trimester framework. The Court held, in summary, that during the first trimester of a nine-month pregnancy, the abortion decision must be left to the woman, and its effectuation must be left to the medical judgment of the physician. The state could regulate abortions during the second trimester but only to promote its interest in the health of the mother. The state could regulate or even prohibit abortions after the point of viability (approximately the beginning of the third trimester) except where necessary to preserve the life or health of the mother. Id. 410 U.S. at 164-65, 93 S.Ct. at 732-33.

The Supreme Court substantially modified this analytic framework in Planned Parenthood of Southeastern Pennsylvania v. Casey, 505 U.S. 833, 112 S.Ct. 2791, 120 L.Ed.2d 674 (1992). Casey dealt with a Pennsylvania statute very similar to the Indiana statute at issue here. It established a new controlling framework for analyzing constitutional challenges to state regulations of abortion. Accordingly, this court must review both the analysis and the specific holdings of Casey in some detail. The joint opinion signed by Justices O’Connor, Kennedy, and Souter, and joined in part by Justices Blackmun and Stevens, provided the narrowest grounds for the judgment of the Court on all questions. That opinion therefore states the controlling holdings of the Court. Planned Parenthood of Southeastern Pennsylvania v. Casey, — U.S. -, - n. 2, 114 S.Ct. 909, 910 n. 2, 127 L.Ed.2d 352 (1994) (Souter, J., in chambers) (joint opinion in Casey is controlling); Planned Parenthood, Sioux Falls Clinic v. Miller, 63 F.3d 1452, 1456 n. 7 (8th Cir.1995) (same). See Marks v. United States, 430 U.S. 188, 193, 97 S.Ct. 990, 993, 51 L.Ed.2d 260 (1977) (when no single rationale explaining the result enjoys the assent of five Justices, the holding of the Court may be viewed as that position taken by the Justices who concurred in the judgment on the narrowest grounds).

The Casey joint opinion first reaffirmed both a woman’s right to choose to have an abortion before viability and the state’s interests in maternal health and in the fetus. 505 U.S. at 845-46, 112 S.Ct. at 2804. It reaffirmed the point of viability as a crucial chronological line in balancing those interests, but it did not legally define the medical viability point as the start of the third trimester. Id. 505 U.S. at 845-46, 868-72, 112 S.Ct. at 2804, 2816-18. It also rejected the Roe distinction between the first and second trimester, concluding that the state’s interest in the potential life of the fetus exists throughout pregnancy and that Roe’s trimester framework undervalues that interest. Id. 505 U.S. at 872, 876, 112 S.Ct. at 2818, 2820. The joint opinion stated:

Though the woman has a right to choose to terminate or continue her pregnancy before viability, it does not at all follow that the State is prohibited from taking steps to ensure that this choice is thoughtful and informed. Even in the earliest stages of pregnancy, the State may enact rules and regulations designed to encourage her to know that there are philosophic and social arguments of great weight that can be brought to bear in favor of continuing the *1445 pregnancy to full term and that there are procedures and institutions to allow adoption of unwanted children as well as a certain degree of state assistance if the mother chooses to raise the child herself. “‘[T]he Constitution does not forbid a State or city, pursuant to democratic processes, from expressing a preference for normal childbirth.’”

Id. 505 U.S. at 872, 112 S.Ct. at 2818 (citation omitted).

Reaffirming the central holding in Roe that before viability a woman has a constitutional right of privacy to decide for herself whether to terminate or continue the pregnancy, Casey held that the right must be protected from “undue burdens” imposed by state regulation. Id. 505 U.S. at 876, 112 S.Ct. at 2820. The joint opinion said: “An undue burden exists, and therefore a provision of law is invalid, if its purpose or effect is to place a substantial obstacle in the path of a woman seeking an abortion before the fetus attains viability.” Id. 505 U.S. at 878, 112 S.Ct. at 2821. Thus, a statute with the purpose of placing a substantial obstacle in the path of a woman seeking an abortion of a nonviable fetus is invalid “because the means chosen by the State to further the interest in potential life must be calculated to inform the woman’s free choice, not hinder it.” Id. 505 U.S. at 877, 112 S.Ct. at 2820. A statute that has the effect of placing a substantial obstacle in the path of a woman’s choice is also invalid because it cannot be considered a permissible means of serving the state’s legitimate ends. Id. The joint opinion stated that a particular provision is valid unless it will operate as a substantial obstacle to “a large fraction” of the women “for whom the law is a restriction” and therefore relevant. Id. 505 U.S. at 895, 112 S.Ct. at 2829, 2830.

The undue burden test is best understood from its application in Casey. The Supreme Court applied the analysis to several sections of a Pennsylvania statute before it. A majority of the Court held that requirements of mandatoiy disclosure, a 24-hour waiting period, and parental consent for minors had not been shown to impose an undue burden on women seeking pre-viability abortions. See id. 505 U.S. at 880-86, 900, 964-70, 112 S.Ct. at 2822-26, 2832, 2867-69 (Justices O’Connor, Kennedy, Souter, White, Scalia, Thomas, and the Chief Justice). A different majority of the Court struck down a requirement that a married woman notify her spouse before seeking an abortion. See id. 505 U.S. at 886-98, 920-22 & n. 8, 112 S.Ct. at 2826-31, 2843 & n. 8 (Justices O’Connor, Kennedy, Souter, Blackmun, and Stevens).

The portion of Casey striking down Pennsylvania’s spousal notice requirement illustrates the type of showing required to meet the undue burden standard in a facial challenge before a law takes effect. The Court reviewed extensive evidence and district court findings concerning domestic violence and the role that pregnancy too often plays in triggering domestic violence. 505 U.S. at 886-93, 112 S.Ct. at 2826-29. That evidence and “what common sense would suggest” persuaded the Court that women’s justifiable fears of abuse triggered by spousal notification meant the law was

likely to prevent a significant number of women from obtaining an abortion. It does not merely make abortions a little more difficult or expensive to obtain; for many women, it will impose a substantial obstacle. We must not blind ourselves to the fact that the significant number of women who fear for their safety and the safety of their children are likely to be deterred from procuring an abortion as surely as if the Commonwealth had outlawed abortions in all cases.

505 U.S. at 893-94, 112 S.Ct. at 2829. The Court concluded that the spousal notification law actually would be relevant for only about one percent of women who seek abortions: only about 20 percent of the women who obtain abortions are married, and 95 percent of these notify their husbands of their own volition. Id. The likely consequences of the law for those five percent of married women seeking abortions were sufficient to invalidate the law on a facial challenge before it took effect. 505 U.S. at 895, 112 S.Ct. at 2830.

Casey’s holding on the spousal notification law shows that the focus of the undue burden test is not on added expense or incon *1446 venience. The focus is on practical burdens so great that they would actually prevent a significant number of women from obtaining abortions they would otherwise choose to have. As defendants in this case point out, of course, Casey also held that the plaintiffs there had not proved that a mandatory disclosure and waiting period law very similar to Indiana’s Public Law 187 would impose an “undue burden.” The authors of the joint opinion reviewed detailed evidence and findings showing that the requirements were likely to make abortions more expensive and more difficult for many women to obtain. 505 U.S. at 885, 112 S.Ct. at 2825. Similar effects had persuaded the Court to strike down similar laws in Akron v. Akron Center for Reproductive Health, Inc., 462 U.S. 416, 450, 103 S.Ct. 2481, 2503, 76 L.Ed.2d 687 (1983) (Akron I), and Thornburgh v. American College of Obstetricians & Gynecologists, 476 U.S. 747, 762-63, 106 S.Ct. 2169, 2179-80, 90 L.Ed.2d 779 (1986). The joint opinion in Casey framed the “undue burden” issue as whether “the 24-hour waiting period is nonetheless invalid because in practice it is a substantial obstacle to a woman’s choice to terminate the pregnancy.” 505 U.S. at 885, 112 S.Ct. at 2825. The joint opinion described the findings of increased expense and difficulty as “troubling in some respects” but found they did not demonstrate that the waiting period constituted an undue burden. Id.

From Casey’s application of the undue burden standard to both the waiting period and the spousal notice requirement, it is clear that the authors of the joint opinion upheld the disclosure and waiting period provisions because the record did not persuade them that the law was likely to be so burdensome as actually to prevent women from having abortions they would otherwise choose to have. At the same time, Casey left open the possibility that plaintiffs in future cases could prove that the burdens of such laws would actually prevent women from having abortions. The joint opinion emphasized that its conclusion on this issue was based “on the record before us.” 505 U.S. at 886, 112 S.Ct. at 2826. Justice Blackmun disagreed with the joint opinion’s conclusion but said he was “confident that in the future evidence will be produced to show that ‘in a large fraction of the cases in which [these regulations are] relevant, [they] will operate as a substantial obstacle to a woman’s choice to undergo an abortion.’” 505 U.S. at 925, 112 S.Ct. at 2845 (quoting joint opinion). The dissenting justices criticized the “undue burden” test in large part because the joint opinion left the door open for a future showing that similar disclosure and waiting period provisions would in fact impose an undue burden. 505 U.S. at 988-91, 112 S.Ct. at 2879-80 (Scalia, J., joined by Rehnquist, C.J., White and Thomas, JJ.).

Two authors of the joint opinion have written more recently that the lower courts should develop a thorough record on the actual or likely effects of the particular laws that are challenged, even if the laws are nearly identical to the laws considered in Casey. In Fargo Women’s Health Org. v. Schafer, — U.S. -, -, 113 S.Ct. 1668, 1669, 123 L.Ed.2d 285 (1993), the Court denied a stay and injunction pending appeal in a challenge to North Dakota’s mandatory disclosure and waiting period law. Justice O’Connor, joined by Justice Souter, filed a brief opinion stating her view that the lower courts must consider the likely practical effects of a challenged law:

[W]e made clear that a law restricting abortions constitutes an undue burden, and hence is invalid, if, “in a large fraction of the cases in which [the law] is relevant, it will operate as a substantial obstacle to a woman’s choice to undergo an abortion.” Casey, 505 U.S., at 895, 112 S.Ct. at 2830. And the joint opinion specifically examined the record developed in the district court in determining that Pennsylvania’s informed-consent provision did not create an undue burden. See id. at 885-98, 112 S.Ct. at 2825-2831 (opinion of O’CONNOR, KENNEDY, and SOUTER, JJ.). While I express no views as to whether the particular provisions at issue in this case constitute an undue burden, I believe the lower courts should have undertaken the same analysis.

— U.S. at -, 113 S.Ct. at 1669. Similarly, in Casey itself, after remand from the Supreme Court, the case returned to Justice *1447 Souter on an application to allow the district court to hear additional evidence on the likely effects of the Pennsylvania statute. Justice Souter denied that application but explained that “litigants are free to challenge similar restrictions in other jurisdictions, as well as these very provisions as applied.” Planned Parenthood of Southeastern Pennsylvania v. Casey, — U.S. -, -, 114 S.Ct. 909, 911, 127 L.Ed.2d 352 (1994) (Souter, J., in chambers).

The portions of Casey striking down the spousal notification law but upholding mandatory disclosure and a waiting period raise an important issue concerning both the timing and scope of plaintiffs’ challenge here. The Casey plaintiffs challenged the Pennsylvania laws before they took effect and sought to have them struck down as facially unconstitutional, so that the laws could not be enforced at all. The Supreme Court was persuaded that the spousal notice law was likely to pose a substantial obstacle to some significant fraction of the women it affected, and therefore held the law facially unconstitutional before it ever took effect. 505 U.S. at 893-97, 112 S.Ct. at 2829-31. That approach contrasts with a standard for facial challenges announced and applied in some earlier cases: “A facial challenge to a legislative Act is, of course, the most difficult challenge to mount successfully, since the challenger must establish that no set of circumstances exists under which the Act would be valid.” United States v. Salerno, 481 U.S. 739, 745, 107 S.Ct. 2095, 2100, 95 L.Ed.2d 697 (1987), followed in Rust v. Sullivan, 500 U.S. 173, 183, 111 S.Ct. 1759, 1767, 114 L.Ed.2d 233 (1991) (facial challenge to regulations affecting abortion and family planning counseling); Webster v. Reproductive Health Services, 492 U.S. 490, 524, 109 S.Ct. 3040, 3060, 106 L.Ed.2d 410 (1989) (O’Connor, J., concurring in part and concurring in judgment) (facial challenge to several abortion statutes).

The statement of the undue burden standard and its application to strike down the spousal notification law in Casey do not seem consistent with the Salerno standard. Since Casey, the lower courts have disagreed about how to handle facial challenges to abortion statutes. Some have held that because Casey expressly overruled other cases and did not mention Salerno, the heightened standard from Salerno still applies to abortion cases. See Barnes v. Moore, 970 F.2d 12, 14 & n. 2 (5th Cir.1992); Jane L. v. Bangerter, 809 F.Supp. 865, 871-72 (D.Utah 1992), aff'd in part and rev’d in part on other grounds, 61 F.3d 1493 (10th Cir.1995). See also Ada v. Guam Society of Obstetricians & Gynecologists, — U.S. -, -, 113 S.Ct. 633, 634, 121 L.Ed.2d 564 (1992) (Scalia, J., dissenting from denial of certiorari, joined by Rehnquist, C.J., and White, J.,) (Casey did not purport to change Salerno standard for facial challenge; law prohibiting all abortions should be upheld against facial challenge because it could be applied to some post-viability abortions). Other courts have followed the lead of Justices O’Connor and Souter, as expressed in Fargo Women’s Health Organization v. Schafer, and have concluded that Salerno does not apply to abortion restrictions. See Planned Parenthood, Sioux Falls Clinic, 63 F.3d 1452, 1456-58 (8th Cir.1995); Casey v. Planned Parenthood of Southeastern Pennsylvania, 14 F.3d 848, 863 n. 21 (3d Cir.1994) (on remand) (dicta). See also Compassion in Dying v. State of Washington, 850 F.Supp. 1454, 1462-64 (W.D.Wash.1994) (applying undue burden test from Casey rather than Salerno standard to facial challenge of statute prohibiting physician-assisted suicide), rev’d, 49 F.3d 586, 591 (9th Cir.) (application of Casey’s undue burden standard for facial challenge was “unwarranted extension of abortion jurisprudence”), rehearing en banc granted, 62 F.3d 299 (9th Cir.1995).

It is true that Casey did not expressly overrule or limit Salerno. However, the Casey Court’s actions and statements conflict with the standard stated in Salerno. Casey is very closely on point here, and more recent. In addition, Justices O’Connor and Souter have indicated clearly in the Fargo case and in Casey on remand that a law “constitutes an undue burden, and hence is invalid, if, ‘in a large fraction of the cases in which [the law] is relevant, it will operate as a substantial obstacle to a women’s choice to undergo an abortion.’” — U.S. at -, 113 S.Ct. at 1669 (quoting 505 U.S. at -, 112 S.Ct. at 2830). Therefore, like the Third *1448 and Eighth Circuits, this court believes that Casey effectively displaced Salerno’s application to abortion laws. See Planned Parenthood, Sioux Falls Clinic, 63 F.3d at 1458 (“We choose to follow what the Supreme Court actually did — rather than what it failed to say — and apply the undue-burden test.”). See also Michael C. Dorf, Facial Challenges to State and Federal Statutes, 46 Stan.L.Rev. 235, 276 (1994) (Casey joint opinion applied “substantial overbreadth” test, which conflicted with Salerno rule but was consistent with other substantive due process decisions).

What kind of proof would suffice to successfully challenge the Indiana abortion restriction under the undue burden test? The portion of Casey striking down the spousal notice law showed that the evidence of a law’s likely effects may be so strong that the “undue burden” for a significant fraction of women will be evident enough even before the law takes effect. Casey also demonstrated that the likely effects of mandatory disclosure and waiting period laws were not so likely to cause undue burdens as to persuade the Court to strike them down before they took effect. However, Casey clearly left several avenues open for such challenges. One possibility for plaintiffs is to wait until the law takes effect and then to present testimony from a number of women explaining how the law had actually prevented them from having abortions. A moment’s reflection shows it is highly unlikely that such evidence would be available even if the law were extremely burdensome. For women for whom making two trips would pose a substantial obstacle to obtaining an abortion, the same factors (plus the later birth of a child) make it unlikely that they would step forward to testify about those burdens. A more realistic alternative is to wait for the law to take effect and then to show through statistical and other circumstantial evidence that the burdens of the law have significantly reduced the rate of abortions that women would otherwise choose to have. Casey itself alluded to this possibility, but such proof is not a simple proposition. The statistical analysis must show that other variables do not explain a drop in the rate of abortions, and there must also be evidence as to the reasons the law caused the effect. Casey allows states to regulate abortions in an attempt to persuade women not to have them. Parties challenging a disclosure and waiting period law must therefore show, under Casey, that a significant drop in the incidence of abortion is the result not of any persuasive effect the law has but of the burdens it imposes on some women.

If a law does in fact impose substantial burdens on women’s right to choose, this approach — waiting for the law to take effect and for its practical effects to be reasonably clear from statistical analysis — inflicts irreparable harm on those women who are prevented from having abortions in the interim. In this case, plaintiffs have taken a different approach to avoid this prospect. They brought this facial challenge before the Indiana law could take effect. They seek to prove the likely effects of the law in Indiana by proving the actual effects of a very similar law elsewhere (Mississippi). They have also offered evidence that there is no reason to expect Indiana’s experience to be materially different from Mississippi’s, and there is no substantial evidence to the contrary.

Defendants have criticized this approach as making “a mockery of the rule of law and stare decisis ” and inviting a “jurisprudence of caprice.” Def.Br. at 15. They say this approach would “require finding not only that a rule that is perfectly legal in one state would violate the Constitution in another, but that a rule invalid in a state one day could, on the next, with a better evidentiary presentation, be found consistent with the Constitution.” Id. These criticisms echo the Casey dissent’s attack on the undue burden standard. See 505 U.S. at 988-91, 112 S.Ct. at 2879-80 (Scalia, J.). They do not show, however, that plaintiffs’ approach is inconsistent with the governing undue burden standard as articulated and applied in Casey, focusing on the likely practical effects of the law. Plaintiffs’ attempt to prove through statistical and other evidence the likely effects of the Indiana law is entirely consistent, at least in theory, with the undue burden standard as set forth and applied in Casey. Whether plaintiffs have come forward with sufficient *1449 evidence to show a reasonable likelihood of prevailing is the question to be addressed next. 3

B. Effects of Mandatory Disclosure and Waiting Period

Plaintiffs argue that the mandatory disclosure of information and the 18-hour waiting period imposed by Public Law 187 will put undue burdens on a woman’s right to choose to have an abortion. They contend the information that must be disclosed to a woman is unnecessary and possibly misleading. They assert that the 18-hour waiting period will cause longer delays in actual practice, making abortions more difficult to obtain and, in some cases, more dangerous and expensive. Plaintiffs focus most of their attack, however, on the “in the presence” provision, which requires that some of the information be provided to the woman in her presence, at least 18 hours before the abortion, effectively requiring most women to make two trips to a clinic.

To attack the mandatory disclosure and waiting period requirements, plaintiffs offered testimony from Dr. Stanley K. Henshaw, a researcher at the Alan Guttmacher Institute; from two Indiana clinic directors, Jane Stout and Marne Greening; and from a clinic director from North Dakota, Jane Bovard. Plaintiffs also rely on a number of affidavits submitted with their motion, including an affidavit from Dr. Connie L. Best, a clinical psychologist who is an expert on the subjects of domestic violence, sexual assault, and rape. Other affiants include the directors and other employees of the plaintiff clinics. 4 Plaintiffs’ evidence describes the services and counseling they provide their patients, the types of problems and concerns confronting many women who are deciding whether to continue or terminate a pregnancy, then* experience with harassment of patients by anti-abortion protesters, and some of the practical consequences of Public Law 187. The court will provide additional detail on plaintiffs’ evidence with respect to each specific issue raised.

In support of the disclosure and waiting period requirements, defendants offered evidence summarizing the information provided to the Indiana legislature when it was considering the bill. Witnesses told the legislature that “hundreds” of women who have had abortions in Indiana have claimed they were “misinformed or ill-informed about the risks, complications and alternatives,” and later regretted their decisions to have abortions. Several women told the legislature about their own experiences, their beliefs that they should have had additional information, and their later regrets about having abortions. A physician also told the legislature about the physical and emotional complications associated with abortions and the need for informed consent. The defendants’ evidence about information provided to the legislature was not admitted for the truth of the matters asserted, but for the limited purpose of showing some of the information the legislature considered. That information is relevant to the purpose prong of the Casey undue burden standard and to whether the legislation bears a rational relationship to a legitimate state purpose.

Two women testified at the court hearing concerning their own experiences, and their testimony was admitted to show the truth of the matters asserted. Marilyn Dowden testified that she had an abortion in 1978 and was not given accurate information about her options or fetal development in the early stages of pregnancy. She regrets her decision to have an abortion. She also now counsels women after abortions. She testified that the women she counsels experience feelings of guilt, depression, and low self-esteem after having abortions. Delores Brents Huddleson testified about having two abortions in approximately the mid-1970s, which she regrets, and about her experiences that led her to believe she had been pushed into having *1450 those abortions without sufficient information or consideration of her alternatives. Defendants did not come forward with any evidence tending to show that women having abortions in Indiana today would actually make different decisions (or would feel their same decisions were better-informed) if Public Law 187 were in effect.

The court will consider in turn the key provisions of the Indiana law: the mandatory disclosure itself, the 18-hour delay requirement, the requirement that some of the medical information be provided by a physician or other health care professional, and the “in the presence” requirement.

1. The Mandatory Disclosures

Indiana law has long required a patient to give her informed consent before she has an abortion. The evidence shows that plaintiffs’ current counseling practices are designed to ensure that women are both aware of their options and sure of their decision before they have abortions. Women who indicate any ambivalence about their decision receive additional counseling and, if necessary, as much time as they need to think further about their decision. The evidence presented thus far shows that the vast majority of women who schedule an appointment for an abortion and arrive at the clinic are already aware of their options and believe they have given their choice sufficient thought. There is no evidence showing that those women who are not aware of their options or who are not sure of their decision are being pushed into having abortions they do not want. There is no evidence that any patients who have had abortions at the plaintiff clinics have later complained that they had insufficient information or were pushed into having abortions without adequate time for reflection.

The testimony of Ms. Dowden and Ms. Huddleson about their own experiences many years ago paints a picture different from plaintiffs’ evidence about their current counseling practices. Both Ms. Dowden and Ms. Huddleson had understandable difficulty testifying here about exactly what they were told and what they understood many years ago. Assuming they were actually pushed into having abortions with inadequate information, that evidence does not overcome plaintiffs’ evidence about counseling practices today. Ms. Dowden testified that women she counsels are upset and have feelings of guilt and depression after abortions. It is not surprising that some women have later regrets over what can be a difficult decision for many women. Regrets and second thoughts about important and difficult decisions are not unusual, even when the decisions are made with ample information and plenty of time for thought. Without more, the evidence of some women’s later regret does not show that any of these women actually made their decisions without important information or adequate time for thought.

Most of the evidence in the record concerning the practical effects of the new law’s informational requirements comes from Jane Bovard, director of the only clinic in North Dakota where abortions are performed regularly. North Dakota has a mandatory disclosure and delay law similar to Indiana’s, except that it has been interpreted to permit the information to be given over the telephone. Ms. Bovard testified that she provides the required information to women calling for appointments, that “a lot of times they’re impatient when we give them the information,” and that when “we get to the section where we have to tell them about medical assistance and the man supporting the child, at least half our patients laugh.” Tr. 269. She estimated that in her experience, about one to two percent of women who call ask to receive the state-provided written materials on fetal development and alternatives to abortion. The vast majority of women have no questions about the information they are given. No patient has ever told Ms. Bovard that she was cancelling her plans or changing her decision because of the mandated information.

Based on all this evidence, plaintiffs contend that the informational requirements of Public Law 187 are demeaning and patronizing, and are at best useless for most women who will be required to listen to the information. Under the joint opinion in Ca sey, however, these circumstances do not show an “undue burden.” The joint opinion in Casey allows a state to express its preference for childbirth through “truthful and not *1451 misleading” information and to legislate for the minority of women who are ambivalent about their decisions. 505 U.S. at 881-85, 112 S.Ct. at 2823-25. The information required by Public Law 187 can be provided in as little as five minutes. While the information may be useless for most women who must listen to it, there is no evidence that the requirement that they listen will itself actually impose a substantial obstacle or an undue burden in the sense that the law will be so burdensome as to prevent women from having abortions they would otherwise choose to have. See Casey, 505 U.S. at 881-85, 112 S.Ct. at 2823-24 (joint opinion). 5

Plaintiffs also suggest that the required information about available medical assistance and child support may be misleading to women. Public Law 187 requires that a woman be informed only that “medical assistance benefits may be available” and that the father of the unborn fetus is “legally required to assist in the support of the child.” Ind.Code § 16-34-2-1.1(2) (emphasis added). No one claims that this information provides a complete picture of relevant facts on either of these issues. However, the law does not forbid anyone from providing additional information about either the eligibility criteria for medical assistance or the practical realities of collecting child support from fathers. If the law were to require a more complete picture on these issues, its requirements would become more burdensome and time-consuming. Under Casey, the legislature may constitutionally require that these subjects be raised with the pregnant woman without requiring additional detail in all cases. Under Casey, therefore, the plaintiffs appear unlikely to show that the information requirement itself imposes an undue burden on a woman’s right to choose whether to continue or terminate her pregnancy.

2. The 18-Hour Delay

Plaintiffs have expressed more of a concern about the practical effects of the 18-hour delay requirement. The evidence shows that abortions are ordinarily performed only one or two days a week at most Indiana clinics. The required 18-hour delay therefore will often result in actual delays of several days or perhaps even a week or two. In some cases, these delays may cause what would have been a safer and less expensive first trimester abortion to become a riskier and more expensive second trimester abortion. (After the eighth week of pregnancy, *1452 the medical risks of abortion, which are very-low, especially in the earliest stages of pregnancy, rise by about 20 percent per week.) However, the evidence also shows that most women seeking abortions in Indiana schedule their appointments in advance, usually by telephone. It is very unusual to have an abortion procedure performed the same day the woman first contacts a clinic. Accordingly, apart from the “in the presence” requirement, if the information required by Public Law 187 could be provided when a woman first makes contact with a clinic, the 18-hour delay would rarely add to delays that women already encounter as a matter of scheduling and logistics.

There is no evidence before the court showing that women would actually benefit from this waiting period mandated by law. Cf. Casey, 505 U.S. at 920, 112 S.Ct. at 2843 (Stevens, J., dissenting in part) (courts should consider actual benefits of law in applying undue burden standard). There is evidence that some women later regret their decisions. That evidence is not surprising in view of the fact that a decision to have an abortion is important and can be difficult for many women. But that evidence does not prove either that these women’s decisions would probably have been different or that they would have regretted them less if they had been subject to a mandatory waiting period. The only evidence about current practices of Indiana abortion climes indicates that the clinics already take steps to ensure that women are aware of their alternatives and have made a considered decision before they have abortions. There is no evidence that women who are ambivalent are now having abortions without taking time for further reflection.

Under Casey, the apparent lack of genuine benefit does not mean that the waiting period amounts to an undue burden. There is no evidence, beyond the types of evidence found insufficient in Casey, that the 18-hour delay in itself would prevent a substantial number of women from having abortions they would otherwise choose to have. Thus, the delay itself does not appear to be an undue burden. This conclusion is consistent with decisions of other courts upholding brief waiting periods after Casey. E.g., Planned Parenthood, Sioux Falls Clinic, Inc. v. Miller, 63 F.3d 1452, 1467 (8th Cir.1995); Fargo Women’s Health Org. v. Schafer, 18 F.3d 526, 533 (8th Cir.1994); Utah Women’s Clinic, Inc. v. Leavitt, 844 F.Supp. 1482, 1491, 1495 (D.Utah 1994). 6

3. Role of the Physician or Health Care Professional

Public Law 187 requires that some of the medical information be provided by a referring physician, the physician who is to perform the abortion, or a physician assistant, advanced practice nurse, or midwife. This provision of Indiana’s law is somewhat less restrictive than the Pennsylvania law in Casey, which required that the information be provided by a physician. The Supreme Court upheld that requirement since there was “no evidence on this record that requiring a doctor to give the information as provided by the statute would amount in practical terms to a substantial obstacle to a woman seeking an abortion.” 505 U.S. at 884, 112 S.Ct. at 2824.

Plaintiffs have raised concerns here about the scarcity of physician time and the expense and difficulty of hiring others who qualify under the Indiana statute. There are only eight physicians in Indiana who regularly perform abortions. Plaintiffs contend there is only a small number of other health care professionals who are both (a) qualified to provide medical information required by Public Law 187 and (b) willing to do so. There is also no evidence before the court showing any substantial benefit from the *1453 law’s limits on who may provide the information. See Casey, 505 U.S. at 920, 112 S.Ct. at 2843 (Stevens, J., dissenting in part) (courts should consider actual benefits in determining whether law imposes an undue burden). To the contrary, the evidence before the court shows that the counselors at the plaintiff clinics are well-trained and well-qualified to counsel patients on the required subjects, including medical information. However, the joint opinion in Casey shows that increased cost and inconvenience, apparently even for little or no actual benefit, do not establish an undue burden in the sense that the law would actually prevent women from having abortions they would choose to have. Thus, there is no evidence on this record that the Indiana law’s limitation on who may provide the medical information is likely, in practical terms, to amount to a substantial obstacle. Cf. Armstrong v. Mazurek, No. CV-95-083-GF, slip op. at 13 (D.Mont. Sept. 29, 1995) (denying preliminary injunction against enforcement of statute that requires an abortion to be performed only by a licensed physician).

4. The “In the Presence” Requirement

Plaintiffs have directed their strongest attack against the requirement in Public Law 187 that some of the information be provided “in the presence of the pregnant woman.” 7 Plaintiffs contend that the practical effect of this provision would be to require women seeking an abortion to make two trips to a clinic. For many women, making two trips would make abortions more expensive and less convenient, but they would still be able to obtain them. However, plaintiffs contend that for many other women, the practical difficulties of arranging to be away from families or partners, jobs or classes, would actually prevent them from obtaining abortions they would otherwise choose to have. Plaintiffs express special concern about women in abusive or potentially abusive relationships who feel they must conceal from their husbands or partners the fact that they are having an abortion. The affidavit of Dr. Connie L. Best provides disturbing evidence concerning the extent of domestic violence and especially the extent of violence directed at some pregnant women and their children. Dr. Best points out that it can be very difficult for women in abusive relationships to make even one undetected trip to an abortion clinic, let alone two, and that the prospect of the second trip increases the risk of discovery and further abuse. See also Casey, 505 U.S. at 887-93, 112 S.Ct. at 2826-29 (striking down spousal notification requirement based on risk of abuse resulting from spouse’s awareness of pregnancy and planned abortion). Plaintiffs have also submitted extensive evidence at the hearing and by affidavit concerning the variety of harassing tactics used at their clinics by anti-abortion protesters. Some of those tactics include photographing patients, tracing license plates, and the like. Plaintiffs assert that women required to make two trips to clinics would be vulnerable to abuse and harassment (and unwanted disclosure to abusive partners or others) after the first visit. Plaintiffs contend that for a significant number of women, these obstacles and risks would be so great that women will forego abortions they would otherwise choose to have.

Defendants point out that the Indiana statute does not actually require two trips to a clinic because the information may be provided by a referring physician. However, as a practical matter, very few women seeking abortions are referred by other physicians. Plaintiffs’ Exhibit 25 shows that over a thirteen month period, approximately four percent of one Indiana clinic’s patients were referred by an outside physician. For purposes of evaluating the law’s practical effects, the “in the presence” requirement in the law amounts to a requirement that women make two trips to a clinic.

The Supreme Court assumed that the Pennsylvania statute upheld in Casey would require two visits. The statute required that at least 24 hours before the abortion, the physician who is to perform the abortion or the referring physician must have “orally *1454 informed the woman” of certain medical information. See 505 U.S. at 902, 112 S.Ct. at 2833. The district court in Casey interpreted this language as requiring that the information be provided in the woman’s physical presence. That court found that the requirement would subject many women to harassment and hostility; would double travel time and increase costs for travel, lost wages, babysitting, and related expenses; and would be especially burdensome for women having difficulty explaining their absences, such as battered women, school age women, and working women without sick leave. 744 F.Supp. at 1351-52. The Supreme Court described these findings as “troubling in some respects,” but held that they did not demonstrate that the law would impose an undue burden. 505 U.S. at 885-86, 112 S.Ct. at 2825. As a result, Casey established that a facial challenge to a two-trip law cannot succeed without evidence beyond that presented in Casey.

Plaintiffs argue that the experience in other states with similar laws that have taken effect since Casey was decided provides sufficient evidence to establish that the Indiana law’s “in the presence” requirement would impose an undue burden. The centerpiece of plaintiffs’ proof at this stage of the case is a study done by Dr. Stanley K. Henshaw about the effects of a similar law in Mississippi. 8 According to plaintiffs, Mississippi is the only state where a law with a two-trip requirement has been in effect long enough so that data have been available to evaluate the law’s effects. The Mississippi law imposes disclosure and delay requirements similar to Indiana’s, although the delay is for 24 hours. The law does not explicitly require that the information be given in the presence of the woman, but it has been interpreted and applied that way.

Plaintiffs argue that the experience of women in Mississippi provides good evidence — in fact the best available evidence— of the likely effects of the Indiana law. Dr. Henshaw testified that while there might be some difference in degree, he believes that Indiana and Mississippi are sufficiently similar to expect similar effects from the two laws. The parties have not brought to the court’s attention any significant differences between the populations of the two states or other factors that would affect the provision of abortion services. The Mississippi study therefore deserves close scrutiny here.

The Mississippi law took effect on August 8, 1992. Dr. Henshaw reviewed statistical data on abortions provided in Mississippi (both to Mississippi residents and residents of other states), as well as data from Alabama and Tennessee, where some women from Mississippi go to obtain abortions. He did not have data available from Louisiana and Arkansas. In his published study (admitted as Plaintiffs’ Exhibit 3), Dr. Henshaw calculated the number of abortions that Mississippi residents would have been expected to obtain from August 1992 to December 1992, and he calculated the number that they actually obtained. His calculations showed the following:

[[Image here]]

*1455 The data thus shows that the number of Mississippi residents obtaining abortions in Mississippi, Alabama, and Tennessee dropped by 13 percent from the expected level calculated by Dr. Henshaw. He also calculated that if the number of Mississippi residents obtaining abortions in Louisiana increased to the same extent that the numbers for Tennessee and Alabama did, the overall drop from expected to actual would be 11 percent. 9 Both Dr. Henshaw and defendants’ expert, Dr. Steele, agreed that that number is statistically significant, meaning that the chance that the drop was merely a random fluctuation in the abortion rate is very small. Dr. Henshaw calculated that there was less than one chance in one thousand that this drop was the result of random fluctuations. Pl.Ex. 3.

Dr. Henshaw also found that the number of Mississippi women obtaining abortions earlier than 9 weeks into their pregnancies (and thus at the earliest and safest stage) dropped by 25 percent, or more than the overall drop. He found that the number of abortions after 12 weeks (the first trimester) stayed approximately the same, meaning that, with the overall drop in numbers, the proportion of abortions after 12 weeks increased from 10.3 percent to 12.1 percent.

The study was published in September 1994. At the preliminary injunction hearing, Dr. Henshaw testified about updated information on Mississippi. He had analyzed data from 1991, 1992, and 1993, which provided a broader statistical base for analysis. The results are summarized in Plaintiffs’ Exhibit 4. Using 1991 as a base year, the additional data show that after the new law took effect, the number of abortions obtained by Mississippi residents and performed in Mississippi dropped by 24.9 percent. The number of Mississippi residents traveling to Alabama and Tennessee for abortions increased by 22.3 percent. Thus, the additional data show that the total number of abortions obtained by Mississippi residents after the law took effect (whether in Mississippi, Alabama, or Tennessee) decreased by 16 percent. With an estimated adjustment for women going to Louisiana for abortions, the relevant decrease was 14 percent. The proportion of abortions for Mississippi residents before 9 weeks dropped by 14.8 percent, and the proportion of abortions for Mississippi residents after 12 weeks increased by 22.1 percent. Dr. Henshaw found that these changes were all statistically significant, meaning again that it is highly unlikely that they were the result of mere random fluctuations in abortion rates. 10

Dr. Henshaw looked for factors other than the new law that might explain these changes. The possibilities he considered were long-term trends in abortion rates, changes in abortion providers or prices in Mississippi, and changes in the population of women of child-bearing age in Mississippi. He looked at longer-term data from Mississippi and surrounding states and determined that the significant drop was not part of a longer-term decline in abortion rates. He obtained information about abortion providers and prices in Mississippi and found no change in the relevant time period that could explain the drop. He also found no change in the population of women of child-bearing age in Mississippi. After doing his analysis, he concluded that he was “95 percent convinced” that the new Mississippi law accounted for the significant changes in abortion rates and patterns that he observed. For reasons discussed below in more detail, he also concluded that this effect of the law was not the result of Mississippi women receiving the mandated information, waiting for the required period, and then changing their minds about having abortions.

Defendants agree that the Mississippi data show a statistically significant drop in abortion rates for Mississippi residents. They dispute Dr. Henshaw’s conclusions about the reasons for the drop, and they have attacked Dr. Henshaw’s study on two different levels. *1456 First, they contend that Dr. Henshaw committed methodological errors that undermine the reliability of his statistical analysis and his conclusion that the law caused the drop in the abortion rate. Second, they attack most vigorously Dr. Henshaw’s claim that the Mississippi law had these effects not because of the persuasive power of the information and waiting period, but because of the burdens it imposes on women seeking abortions.

Dr. Gregory Steele, the state epidemiologist for the Indiana State Department of Health, evaluated Dr. Henshaw’s study for defendants. 11 Dr. Steele criticized Dr. Henshaw’s methodology on several grounds. First, he argued that Dr. Henshaw did not account properly for seasonal variations in abortion rates in calculating the expected numbers that were the basis for his comparisons after the new law took effect. Dr. Henshaw developed his “expected” data by starting with the actual data for the period January 1992 to July 1992, the period just before the law took effect, and multiplying those figures by 65 percent. 12 Dr. Henshaw supported the 65 percent figure in two ways. First, he looked at actual abortion rates in Mississippi in 1990, 1991, and 1993, and found that the number from August to December of each year averaged 65.029 percent of the number from January to July of the same year. He also looked at national data from 1986 and 1988, which showed ratios of 65.456 percent and 65.220 percent, respectively.

Dr. Steele argued first that Dr. Henshaw should have “normalized” his data for the seasonal variations in a different way, using logarithms of monthly figures as a way of smoothing out the monthly variations. However, Dr. Steele had not actually performed such an analysis with logarithms and therefore could not comment on how his preferred method of seasonal adjustment or normalizing would affect the results of the study. While there are several different ways for statisticians to deal with such seasonal variations, Dr. Steele’s testimony did not persuade the court that Dr. Henshaw’s failure to use the logarithm technique that Dr. Steele advocated undermines the study or its conclusions.

Dr. Steele sharply criticized Dr. Henshaw for having used 1993 data in calculating these seasonal variations. Dr. Steele testified that use of any data from the time after the intervention being evaluated, i.e., after the new law took effect, would contaminate the results of the analysis. That analysis depends on a comparison of actual and expected events after the intervention. The court believes it understands Dr. Steele’s concern on this point but is not yet persuaded that this factor had any material influence on the results of Dr. Henshaw’s study. Dr. Henshaw used the “contaminating” 1993 data for only one purpose — calculating the average ratio of August to December abortions to the number of January to July abortions in Mississippi. The result of that calculation, 65.029 percent, was very close to the ratio developed from the national data (65.456 percent and 65.220 percent), which suggests that the 65 percent figure was reasonably accurate despite the asserted methodological error. In addition, Dr. Steele did not do comparable calculations without the 1993 data to evaluate the extent to which the 1993 data might have affected the result. 13 Finally, there is no evidence before the court suggesting that the Mississippi law would have had any impact on the seasonal variations of abortion rates in Mississippi. Thus, the court does not believe Dr. Henshaw’s conclusions are undermined by his use of the 1993 data for this limited purpose.

*1457 Dr. Steele also criticized Dr. Henshaw’s Chi-square calculations on several grounds. 14 This testimony was based largely on his analysis of a handwritten set of calculations (Def.Ex. 10) that were not explained by either side. The court is left to wonder just what the asserted problems were. In any event, Dr. Steele did not test Dr. Henshaw’s Chi-square calculations by doing his own calculations. Accordingly, the court does not give these criticisms of the Chi-square calculations any weight based on this record. 15

More relevant are Dr. Steele’s attacks on Dr. Henshaw’s conclusions drawn from the data. Dr. Steele said that Dr. Henshaw did not sufficiently consider other explanations for the changes in the Mississippi abortion rates. Dr. Steele contended the drop could have been part of a larger trend in declining abortion rates. As noted, however, Dr. Henshaw looked at Mississippi and data from other states in the region, and the data did not indicate there was any larger trend at work to explain the sharp drop in Mississippi in 1992. Dr. Steele raised several other questions about Dr. Henshaw’s study and the conclusions to be drawn from it. Dr. Steele noted that data from the Centers for Disease Control showed a 4.5 percent drop in the pregnancy rate for Mississippi teenagers from 1991 to 1992, and he suggested that could explain part of the decrease in abortions observed by Dr. Henshaw. He noted that Dr. Henshaw’s own data do not show any correlation between the drop in Mississippi abortion rates and the age, race, educational level of women, or with their distance from an abortion provider. 16 He also noted, as did Dr. Henshaw, that while the abortion rate dropped by about 11 percent for women in Mississippi, the birth rate for the same cohort of pregnancies also dropped by 1.3 percent, instead of increasing by about 1.5 percent, as would have been expected from the drop in abortions.

Neither Dr. Henshaw nor Dr. Steele was able to explain these phenomena, but there is also no real indication that they undermine Dr. Henshaw’s conclusions. Dr. Steele agreed with Dr. Henshaw that the drop in Mississippi abortion rates was statistically significant. He also agreed that the two-trip requirement in the Mississippi law could be burdensome and could have accounted for the decline. He did not agree with Dr. Henshaw that the available evidence allowed a reliable conclusion that the law caused the change, but he also did not point to any evidence that actually tends to refute Dr. Henshaw’s conclusion.

Accordingly, the court finds that plaintiffs have established a reasonable likelihood of showing that the Mississippi law caused a drop of 11 to 14 percent in the number of Mississippi women obtaining abortions, and that Public Law 187 is likely to have a comparable effect in Indiana. While the number cannot be calculated precisely, that is a significant fraction of women who would otherwise choose to have abortions. The magnitude of the effect therefore appears sufficient to satisfy the undue burden test as set forth in Casey. See 505 U.S. at 896, 112 S.Ct. at 2830.

However, plaintiffs must also show why the Mississippi law had this effect. Both the Indiana and Mississippi laws appear to have been enacted in the hope that they would *1458 reduce the number of abortions. Casey instructs that such laws and such effects may be constitutional if they reduce the number of abortions by providing women with more information and requiring them to continue considering their decisions for a brief time after they receive that information. If the persuasive effects of the law reduce the number of abortions, that result is perfectly consistent with Casey, which held that a state may impose such requirements in order to encourage childbirth over abortion, and to ensure that a woman’s decision is informed and well-considered. If, on the other hand, the law causes a drop in abortions by making it more difficult to obtain an abortion, the law may impose an “undue burden” that is unconstitutional under Casey.

Dr. Henshaw testified that in his opinion, the Mississippi law caused the decline in abortion rates because of burdensome effects rather than persuasive effects. He based that conclusion primarily on two factors. First, as reported in his published study, he asked the directors of two of the three abortion providers in Mississippi at that time whether women were changing their minds after coming for an initial visit and receiving the initial counseling. He testified: “They both responded that it was quite rare for women to change their minds and not return; that if women came for the first visit, they almost invariably returned for the procedure. They didn’t see a change in women changing their minds as a result of the — excuse me, after the law went into effect.” Tr. 28. Second, Dr. Henshaw noted the 17 percent increase in the number of Mississippi women going to other states to have abortions and the 30 percent decrease in the number of women from other states coming to Mississippi to have abortions. Those substantial changes cannot be explained by the persuasive effects of the Mississippi law. They indicate that women are going to less convenient climes in order to avoid the inconvenience of the Mississippi law.

The evidence on this issue is not limited to Dr. Henshaw’s opinion and the Mississippi data. The experience under North Dakota’s disclosure and delay law also tends to indicate that the Mississippi law’s effect is a result of burdens rather than persuasion. Jane Bovard, the director of the only clinic in North Dakota, explained that the North Dakota law permits the information to be provided by telephone. See also Fargo Women’s Health Org. v. Schafer, 18 F.3d at 530-32. When a woman calls to schedule an appointment for an abortion, the woman is told the state-mandated information. As noted above, Ms. Bovard estimated that only one or two percent of women ask for the additional state-provided written materials that must be offered, and she said that no patient has ever told her she was going to cancel her appointment because of the information provided. She has seen no decrease in the number of abortions provided. 17

Also highly relevant here is the complete absence of any evidence suggesting that similar disclosure and delay laws in any states actually have the constitutionally permissible persuasive effect on women. As discussed elsewhere in this opinion, defendants have shown that there exists information upon which a rational legislature could rely suggesting that such laws may produce some benefits. Since Casey was decided, similar laws have taken effect in several states without two-trip requirements. There is no evidence before this court indicating that these laws are persuading women not to have abortions or that the additional time for reflection is changing any women’s minds.

Plaintiffs’ evidence about the ways in which the two-trip requirement could prevent women from having abortions is also relevant here. Plaintiffs provided substantial evidence, particularly from the affidavit of Dr. Connie L. Best, about behavior of men and women in abusive relationships. She testi *1459 fled that men who subject women to violence and abuse often subject them to surveillance and use other tactics to keep them under control. She also testified that men who abuse their wives or partners often escalate the abuse when the woman becomes pregnant. Those forms of behavior make it very difficult for some abused women to make even one trip to an abortion clinic, let alone two, because the result of the two-trip requirement may be that the man learns of the woman’s plan to have an abortion. Dr. Best’s affidavit is not enough by itself to show that the two-trip requirement in Indiana will in fact impose an undue burden on a significant number of women, but it provides one reasonable explanation for how a two-trip requirement could prevent a significant number of women from exercising their right to choose. Moreover, that real possibility in many cases is one of the principal foundations for the Supreme Court’s decision in Casey to strike down the spousal notification law. 505 U.S. at 887-96, 112 S.Ct. at 2826-30. As the Supreme Court explained:

For the great many women who are victims of abuse inflicted by their husbands, or whose children are the victims of such abuse, a spousal notice requirement enables the husband to wield an effective veto over his wife’s decision. Whether the prospect of notification itself deters such women from seeking abortions, or whether the husband, through physical force or psychological pressure or economic coercion, prevents his wife from obtaining an abortion until it is too late, the notice requirement will often be tantamount to the veto found unconstitutional in [Planned Parenthood of Central Missouri v. Danforth, 428 U.S. 52, 96 S.Ct. 2831, 49 L.Ed.2d 788 (1976)]. The women most affected by this law — those who most reasonably fear the consequences of notifying their husbands that they are pregnant— are in the gravest danger.

Id. 505 U.S. at 897, 112 S.Ct. at 2831. 18

Defendants objected to Dr. Henshaw’s opinion about the law’s burdens causing the drop in the Mississippi abortion rate, and Dr. Steele criticized Dr. Henshaw for relying on the information provided by the two clinic directors in reaching any conclusions. Dr. Henshaw explained that in his studies, he and other researchers often rely on information supplied by providers of abortion services. On cross-examination, he explained why he considered the providers’ statements that very few women failed to return after receiving the state-required information sufficiently reliable to use in his study. He said that is the type of information that should be objectively verifiable. The climes should have records on whether or not women come back, their actions have economic consequences for the facilities, and such actions would cause inconvenience to the clinics such that the directors should be aware of the extent of such actions. Tr. 89-90.

Defendants contend Dr. Henshaw’s opinion on this issue fails to satisfy Daubert v. Merrell Dow Pharmaceuticals, Inc., — U.S. -, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993), in which the Supreme Court instructed district judges to act as “gatekeepers” concerning scientific testimony. In considering the admissibility of proffered scientific testimony, the court should consider several factors: (1) whether a theory or technique can be and has been tested; (2) whether the theory or technique has been subjected to peer review and publication; (3) the known or potential rate of error; and (4) the “general acceptance” of the theory. Bradley v. Brown, 42 F.3d 434, 437 (7th Cir.1994). This list of factors is not exhaustive, and some of these factors may not be readily applicable in all cases, such as those *1460 involving social science opinions. See Daubert v. Merrell Dow Pharmaceuticals, Inc., 43 F.3d 1311, 1317 & n. 4 (9th Cir.1995) (on remand) (list of factors illustrative rather than exhaustive, and factors may not be equally applicable, or applicable at all, in every case); In re Aluminum Phosphide Antitrust Litigation, 893 F.Supp. 1497, 1506 (D.Kan.1995) (each of these factors may not be relevant in determining reliability of social science expert testimony, but Daubert requires the court to act as a “gatekeeper” in determining reliability and relevance). The ultimate question is whether the scientific knowledge will assist the trier of fact. Federal Rule of Evidence 702 requires “a grounding in the methods and procedures of science,” and “more than subjective belief or unsupported speculation.” Daubert, — U.S. at -, 113 S.Ct. at 2795. Daubert does not require a scientific opinion to be beyond criticism before it becomes admissible. The Supreme Court instructed lower courts to continue to rely on “[vjigorous cross-examination, presentation of contrary evidence, and careful instruction on the burden of proof’ as the means for dealing with “shaky but admissible evidence.” — U.S. at -, 113 S.Ct. at 2798.

Upon consideration of these factors, the court concludes that Dr. Henshaw’s opinion that the Mississippi law’s effects are the result of its burdens rather than it persuasive effects is sufficiently supported to be admissible. Dr. Henshaw did not claim his opinion is the final and definitive word on the subject, nor does the court take it as such. Rather, his opinion is a preliminary scientific opinion reached on the basis of limited but reasonable and objective information. That information includes the dramatic changes in travel for abortions into and out of Mississippi, and no one suggests those changes can be explained by the law having persuasive effects. The information also includes the information from two of three clinic directors in Mississippi, and Dr. Henshaw explained why he and other researchers rely on such information. Other evidence that Dr. Henshaw did not cite also tends to corroborate his opinion. That evidence includes the data indicating, if not definitively proving, the absence of any persuasive effect from similar information requirements in North Dakota. It includes the absence of any evidence showing that any similar laws around the country actually have the type of persuasive effects that give them a constitutionally legitimate purpose under Casey. It also includes the evidence concerning behavior in abusive relationships. Similar evidence and similar concerns led the Supreme Court in Casey to strike down the spousal notice statute. And while the Supreme Court was not persuaded in Casey that a two-trip waiting period law would impose an undue burden, the Court did not have available to it evidence showing that such a law had in fact caused a significant drop in the incidence of abortion. All that information supports the view that the burdens of a two-trip requirement are not merely a figment of Dr. Henshaw’s imagination or advocacy. (Dr. Steele conceded the two-trip requirement could be burdensome, that the decrease in abortions was statistically significant, and that the decrease could have been caused by the law’s burdensome effects.)

To apply the Daubert factors: First, Dr. Henshaw’s opinion about the burdensome effects of the Mississippi law certainly can be tested and is subject to falsification. If there is one basic characteristic that marks an assertion as “scientific” from one that is not, it is that the assertion must be capable of being proved false. 19 Dr. Henshaw’s opinion could be proven incorrect by evidence, for example, that significant numbers of Mississippi residents contemplating abortions listen to the state-mandated information and then decide not to have abortions. Evidence that similar law in other states have significant persuasive effects might also tend to disprove Dr. Henshaw’s opinion. There may be many other forms of evidence that would disprove Dr. Henshaw’s opinion. But there *1461 is no evidence now before the court tending to disprove or falsify it.

Second, Dr. Henshaw’s opinion has been published by a respected institution in the field, the Alan Guttmaeher Institute, but has not yet been subjected to formal peer review. Dr. Henshaw testified that there was a desire not to delay publication. He was also aware his analysis might be used in litigation. As the Ninth Circuit explained in Daubert on remand, evidence from a scientist testifying about research that she has done independently of the litigation and that has been subjected to peer review and publication is likely to satisfy the Daubert reliability test. See 43 F.3d at 1317-19. These criteria are not indispensable, however. There may be instances, and this appears to be one, where time pressures affect either publication or peer review. See Daubert, — U.S. at -, 113 S.Ct. at 2797; 43 F.3d at 1318 n. 9. Although Dr. Henshaw conducted his research with the realization that it could be relevant to litigation, that fact does not disqualify the evidence concerning a field where litigation is constant and the Supreme Court’s opinion in Casey amounted to an invitation to conduct such studies. Healthy skepticism about such evidence is necessary, but the evidence should not be excluded because of the connection with the litigation. The court also finds credible Dr. Henshaw’s testimony that he was surprised by the results of his research on Mississippi.

Third, the known or potential rate of error factor seems to implicate considerations not directly relevant here, but again, Dr. Henshaw’s opinion is certainly falsifiable. Fourth, in terms of general acceptance, the principal challenge is to Dr. Henshaw’s reliance on what defendants call “anecdotal reports of two clinic directors.” The characterization as “anecdotal” is not accurate. Dr. Henshaw explained that researchers in his field often rely on information from health care providers, but they do not do so uncritirally. He explained why he believed the information from two of three available directors was reliable in this instance. They provided not anecdotes but the fact that they had not observed patient behavior consistent with the law having persuasive effects. Defendants can rightly complain that these observations were not precisely quantified. However, the data show a 22 percent drop in abortions actually performed in Mississippi (a total decrease of more than 700 over a five month period). It is reasonable to conclude that if the law’s persuasive effects could account for a drop of that magnitude, the directors would have noticed a significant number of patients visiting once to hear the state-mandated information and then cancel-ling or not scheduling abortions. 20

Dr. Henshaw’s preliminary opinion that the Mississippi law’s effects cannot be attributed to its persuasive effect is reasoned and based on relevant evidence; it is falsifiable (ie., capable of being disproved); and it is consistent with a great deal of other evidence available to the court. It is highly relevant to a central issue in the case, and it is sufficiently helpful to this trier of fact to consider at this preliminary stage of the case, where the court must make a decision on the best available evidence in the face of threatened irreparable harm. Dr. Steele compared the new law to a clinical drug trial, and he said that “if you’re trying a drug and you find out it’s harmful, then you should stop using it.” Tr. 351. Dr. Henshaw’s opinion and the other evidence before the court show a reasonable probability that a two-trip law imposes significant burdens and irreparable harms. Dr. Henshaw’s opinion and the other evidence discussed above therefore support the issuance of a preliminary injunction to prevent irreparable harm (ie., to stop the “clinical trial”) at least until the critical issues can be explored in more depth. Whether Dr. Henshaw’s opinion would be sufficient to support a final judgment for plaintiffs after a *1462 trial on the merits is another question. Before it must be faced, all parties and Dr. Henshaw will have ample opportunity to gather and analyze all available data. 21

Both Dr. Henshaw and Dr. Steele agreed that a better indicator of the reasons for the drop in abortion rates in Mississippi could be obtained through a study that would interview women who chose to have abortions, women considering abortions who received the state mandated information and then changed their minds, and a sample of women who chose to carry their pregnancies to term. Dr. Henshaw had initially proposed a study that would include some interviews of women obtaining abortions concerning their response to the state-provided information. Dr. Steele said he would prefer such a study interviewing all three groups of women as the basis for drawing any conclusions about the effects of the Mississippi law. Tr. 306-07, 355-57. The court does not doubt that such information could be useful in drawing conclusions with greater certainty about the effects of the Mississippi law. However, Dr. Steele also testified that he was not aware of any ongoing surveys dealing with abortion that used such methods. Such interviews would appear to require either significant invasions of privacy about an extremely sensitive subject or, if researchers relied on voluntary waivers of confidentiality by women, a serious risk of studying non-representative samples. The court has doubts about the ability to conduct such studies reliably where privacy and confidentiality are as important as they obviously are in this context.

The evidence now before the court does not allow anyone to draw definitive conclusions about the effects of the Mississippi law or the likely effects of the Indiana law. But this case is before the court on a motion for a preliminary injunction. Plaintiffs need not prove they are certain to prevail on the merits. They must show at this point a reasonable likelihood of prevailing on the merits of this claim. They can prevail ultimately by showing by a preponderance of the evidence the probable effects of the Indiana law. The evidence before the court, which goes well beyond the evidence presented in Casey on this issue, shows that the “in the presence” requirement of the Indiana law is likely to have effects in Indiana comparable to the effects of the similar law in Mississippi. That is, the number of Indiana women obtaining abortions is likely to drop by approximately 11 to 14 percent, and this effect is likely to be the result the burdens of the law rather than the persuasive effect of the required information and delay. That effect means that for a large fraction of women seeking abortions, the law is likely to operate as a substantial obstacle to a woman’s choice to undergo an abortion. Under Casey, therefore, the “in the presence” requirement of the Indiana law appears reasonably likely to impose an undue burden that is unconstitutional.

These conclusions at this stage of the case are necessarily based on incomplete information and are provisional. By the time of trial in this ease more information should be available concerning the effects of similar laws— either with or without two-trip requirements — in other states. Because of the prospect of irreparable harm, however, the court is required to make its best judgment now on the basis of the information that is available now, notwithstanding witnesses’ or the court’s desires for more complete and definitive information.

C. The “Purpose” Prong of the Undue Burden Test

Under Casey, a law may impose an undue burden on women’s right to choose *1463 if it was enacted for the purpose of placing a substantial obstacle in the path of a woman seeking an abortion. 505 U.S. at 877, 112 S.Ct. at 2820. Casey upholds as constitutional at least the general purposes of similar mandatory disclosure and waiting period laws — ensuring that a woman’s decision is based on sufficient information and has been made after time for reflection. Plaintiffs argue that even if Indiana has a legitimate purpose in imposing the mandatory disclosure and waiting period, it has no legitimate purpose in requiring that the required information concerning medical risks and fetal development be provided in person rather than over the telephone. Defendants suggest the Indiana legislature reasonably could have concluded that “sound medical practice required the assessment of risk and fetal age to be done in person, and that the woman contemplating abortion should have some time to reflect on that information in privacy____” Def.Br. at 20.

Casey holds that a state’s purpose of reducing the rate of abortions by trying to ensure that women’s decisions are well-informed and made after adequate time for reflection is a legitimate and constitutional purpose. It also holds that a state’s desire to reduce the rate of abortions by making them more difficult to obtain is not. But Casey does not explain how lower courts should go about determining the purpose of a challenged statute regulating abortions, and there are few areas of constitutional law as troublesome as determining the “purpose” of a statute. See, e.g., Michael M. v. Superior Court of Sonoma County, 450 U.S. 464, 469-70, 101 S.Ct. 1200, 1204-05, 67 L.Ed.2d 437 (1981) (plurality opinion) (“This Court has long recognized that inquiries into congressional motives or purposes are a hazardous matter, and the search for the ‘actual’ or ‘primary’ purpose of a statute is likely to be elusive.”) (citations omitted). Sometimes the test seems to turn at least in part on the actual subjective motives and intentions of lawmakers. See, e.g, Edwards v. Aguillard, 482 U.S. 578, 585-94, 107 S.Ct. 2573, 2578-83, 96 L.Ed.2d 510 (1987) (law requiring teaching of “creation science” found to have unconstitutional purpose of promoting certain religious beliefs); Wallace v. Jaffree, 472 U.S. 38, 74-77, 105 S.Ct. 2479, 2499-2501, 86 L.Ed.2d 29 (1985) (O’Connor, J., concurring in judgment) (inquiry into legislative purpose should be deferential and limited, but courts should not accept an announced purpose that is a “sham”). More often, the Supreme Court generally avoids inquiries into subjective legislative motives and focuses the analysis of legislation on its actual or likely effects and other objective evidence. See, e.g., Arlington Heights v. Metropolitan Housing Dev. Corp., 429 U.S. 252, 265-68, 97 S.Ct. 555, 563-65, 50 L.Ed.2d 450 (1977) (discussing objective evidence relevant to racially discriminatory purpose, but allowing consideration of legislative history); United States v. O’Brien, 391 U.S. 367, 383-84, 88 S.Ct. 1673, 1682-83, 20 L.Ed.2d 672 (1968) (focus is on inevitable effects of legislation; Court declined to strike down legislation that would be upheld if a legislator had made a “wiser” speech about it); Fletcher v. Peck, 10 U.S. (6 Cranch) 87, 131, 3 L.Ed. 162 (1810) (court will not strike down statute “in consequence of the impure motives which influenced certain members of the legislature which passed the law”).

Although the joint opinion in Casey does not elaborate on the “purpose” prong of the undue burden test, the opinions of other justices advocating other standards show what the joint opinion’s test is not. Justice Blackmun advocated “strict” constitutional scrutiny, upholding restrictions only if the government can demonstrate that the restriction is both necessary and narrowly tailored to serve a compelling governmental interest. 505 U.S. at 928-29, 112 S.Ct. at 2847. Under this test, the government would need to show that the “in the presence” requirement itself is actually likely to serve a compelling governmental interest. On this record, the “in the presence” requirement fails that test. The Chief Justice argued in Casey that abortion regulations should be upheld so long as the regulation is “rationally related to a legitimate state interest.” 505 U.S. at 966, 112 S.Ct. at 2867. The state’s lawyers have articulated an interest that appears to be legitimate, and the “in the presence” requirement appears to meet the minimal hurdle of being rationally related *1464 to that objective. However, the joint opinion’s “purpose” test seems to he somewhere between these two positions. It is not clear to what extent its application would depend on evidence of the legislature’s subjective motives and goals, evidence of the more objective effects of the challenged restriction, articulation of a legitimate purpose by counsel in litigation, or a combination of these factors.

This court will briefly review the evidence that might be relevant to the purpose issue. As introduced and passed by the Indiana Senate, the legislation required that the mandatory disclosure be made “orally” to the pregnant woman. The “in the presence of the pregnant woman” language was added by a floor amendment in the Indiana House. Debate on the amendment was brief. The sponsor of the amendment explained the result of the words he proposed but not his underlying reasoning for requiring the information session to be conducted in person:

[A]ll my amendment does is on page 2, line 22, after the word “abortion” it inserts “and in the presence of the pregnant woman” so it’s very clear that the information will be given to the pregnant woman in person and not by telephone. That information can be given to her by her family doctor or by the clinic. It just simply requires that it be given in person and I would appreciate your support. I might add, Mr. Speaker, there’s some confusion about the language that presently is in the bill whether the information can be given over the telephone or not and this would clear that up and make it very clear that it should be, it will be given in the presence of the woman and not by telephone or some other conveyance.

Pl.Ex. 38 at 3. Then a member spoke in opposition to amendment and raised some of the same objections that plaintiffs have raised here:

Some people have referred to the bill itself as the women — women are stupid bill. This amendment is the women are really stupid amendment and I don’t mean that as offensively to Representative Alderman. I really feel that very sincerely. Because now we’re saying that not only do we have to have this informed consent as if a woman hasn’t thought this through, we’re saying that if she calls in and talks to the physician and hears all the options and is given the option of seeing the pictures that people want her to see or the information that she has to receive, that she’s not capable of understanding that over the telephone. This is dearly an effort to make the woman have to go into the dinic, to have to go through all the protesters, and to alert those protesters that she’ll be back in 18 hours, we’ll know exactly who she is, we’ll know exactly why she’s back, it gives them the second chance. I mean if we can have informed consent, we can have the bill as it is now with the language in it that requires the informed consent, that requires her to have to call in and get that, all the information and all the graphic descriptions that you want to give, if that’s what you so choose, but we don’t have to have her go into the facility twice in order to receive that information. I would ask that you reject this amendment.

Id. at 3-4 (emphasis added). Another member spoke in opposition:

I’ve had a test, biopsy if you will, for cancer and if you can get that information on the phone when something is life threatening as cancer, I should think if you choose to, you could get any other information on the phone as far as abortion or not to have an abortion or whatever is concerned.

Id. at 4.

After the assertion that supporters of the amendment had an “obstacle” purpose in mind, the chief sponsor of the bill in the House defended the “in the presence” requirement:

Thank you, Mr. Speaker, members of the House. Real quickly, the purpose of this amendment is quite simply, somebody that’s going to make a decision that is this important, you want to make sure that you’re talking to the physician, the advanced practical nurse, the midwife or whomever is spelled out in this bill that that’s the person you’re talking to. I mean, the first time you may see this doctor, maybe on the operating table while you’re already anesthetic and how do you *1465 know this is the doctor or the person you’re supposed to be talking to. I mean, that is terrible consequence to rely on somebody you’ve never met before in your life and to take them at face value that this is the person who may be performing the operation or has the authority from the physician to talk to me over a telephone. That is very dangerous, especially when you’re talking about the symptoms and consequences of an abortion. I would think you would want to talk personally to the person who may be performing that and know that they have — know that they are the person indeed that they are.

Id. at 4 (emphasis added). The House then adopted the amendment, and the Senate eventually concurred with all House amendments.

There is no evidence before the court suggesting that telephone impersonation of health care professionals is a material risk here. Yet when supporters of the “in the presence” requirement were challenged to come up with a legitimate purpose, that was the response. The constitutionality of a law should not necessarily depend, of course, on how nimble its proponents are in a legislative debate, especially in a body trying to deal quickly with a large number of bills. See United States v. O’Brien, 391 U.S. at 384, 88 S.Ct. at 1683 (“What motivates one legislator to make a speech about a statute is not necessarily what motivates scores of others to enact it____”). Therefore courts are ordinarily willing to consider other purposes not stated in the floor debate. In this lawsuit, the defendants have suggested only one other purpose for the requirement, the possible benefit of providing some of the medical risk and fetal development information on the basis of an actual physical examination rather than information provided by the woman by telephone. There is no evidence that the legislature actually considered that purpose for the provision. There is also no evidence now before the court showing that this suggested purpose is reasonably likely to provide any genuine benefit.

If the purpose test under Casey requires the government to come forward with evidence that a restriction on abortions is reasonably likely to serve a useful and legitimate purpose, the “in the presence” requirement appears to fail that test on the record now before this court. Cf. Casey, 505 U.S. at 920 & n. 6, 112 S.Ct. at 2843 & n. 6 (opinion of Stevens, J.). If the purpose test focuses on evidence of the subjective motivations or intentions of the legislature and allows the court to infer an improper purpose from the absence of an expression of legitimate purpose in the legislature, the “in the presence” requirement also appears to fail that test. The only attempt to offer a legitimate purpose in the legislature — concerns about telephone impersonation of health care providers — deserves no credence. Cf. Edwards v. Aguillard, 482 U.S. at 587, 107 S.Ct. at 2579 (Court normally defers to a state’s articulated purpose, but only if purpose is “sincere and not a sham”); Weinberger v. Wiesenfeld, 420 U.S. 636, 648 n. 16, 95 S.Ct. 1225, 1233 n. 16, 43 L.Ed.2d 514 (1975) (Court need not accept assertions of legislative purpose at face value when examination of legislative scheme and its history demonstrates that the asserted purpose could not have been a goal of the legislation). 22

This court has wrestled with the opinions in Casey and has tried to identify between the rational relation test and the strict scrutiny test some middle ground that both (a) remains distinct from the “effects” prong of the undue burden test and (b) avoids at least complete reliance on evidence of subjective motivation. Cf. 505 U.S. at 920 & n. 6, 112 S.Ct. at 2843 & n. 6 (Stevens, J., *1466 dissenting in part) (undue burden standard should analyze both the severity of the regulatory burden and the legitimacy of its justification, and should require evidence that the restriction “serves a useful and legitimate purpose”). The applicable legal standard is not yet clear. At this time, however, this court need not decide plaintiffs’ likelihood of success on the “purpose” issue to decide their motion for a preliminary injunction. To obtain a preliminary injunction, plaintiffs must prove an imminent threat of irreparable harm. At least with regard to the “in the presence” requirement in this case, the plaintiffs could not prove a threat of irreparable harm without proving that the requirement is likely to impose an undue burden in its practical effects. They have done so. That proof is both necessary and sufficient to meet the first two essential elements for a preliminary injunction. Proof of an improper purpose would add nothing to the equation. The court therefore declines to characterize at this point the plaintiffs’ likelihood of succeeding on the purpose prong of the undue burden standard.

D. Medical Emergency Exception

Since Roe v. Wade and Doe v. Bolton, 410 U.S. 179, 93 S.Ct. 739, 35 L.Ed.2d 201 (1973), it has been clear that states may not regulate abortions in ways that actually cause harm to women’s health. In Roe, the Court held that, after viability, a state may regulate or prohibit abortion, except where the abortion is necessary to preserve the life or health of the mother. 410 U.S. at 163-65, 93 S.Ct. at 732. In Doe v. Bolton, the Supreme Court upheld the relevant statutory exception for a woman’s health on the ground that the statute allowed a physician to exercise a medical judgment “in the light of all factors — physical, emotional, psychological, familial, and the woman’s age — relevant to the well-being of the patient. All these factors may relate to health.” 410 U.S. at 192, 93 S.Ct. at 747. In Casey the Supreme Court reaffirmed this basic limitation on any regulation of abortion: “[T]he essential holding of Roe forbids a State from interfering with a women’s choice to undergo an abortion procedure if continuing her pregnancy would constitute a threat to her health.” 505 U.S. at -, 112 S.Ct. at 2822. Casey makes clear that any disclosure and waiting period law must contain an exception to ensure that the law will not be applied so as to harm the health of the woman. Id.

Public Law 187 excepts from its mandatory disclosure and waiting period requirements a woman who is faced with a “medical emergency.” Ind.Code § 16-34-2-1.1. “Medical emergency” is defined in Public Law 187 as

a condition that, on the basis of the attending physician’s good faith clinical judgment, complicates the medical condition of a pregnant woman so that it necessitates the immediate termination of her pregnancy to avert her death or for which a delay would create serious risk of substantial and irreversible impairment of a major bodily function.

Ind.Code § 16-18-2-223.5.

The Pennsylvania statute upheld in Casey contained almost identical language. The construction of that language by the Third Circuit, as a matter of state law, was critical to the Supreme Court’s decision. In Casey, the district court had held the language of the medical emergency exception unconstitutional because it found that the plain language of the narrow definition “may require a physician to act in ways that do not protect or restore the health of his or her patient in instances where a delay in the abortion procedure will pose a risk to the patient’s health, but an immediate abortion may not be necessary to avert her death or to prevent a serious risk of the substantial and irreversible loss of a major bodily function.” 744 F.Supp. at 1345. On appeal, the Third Circuit read the statutory language more broadly and held it was facially constitutional. 947 F.2d at 699-702. That court reasoned that the Pennsylvania legislature would favor an interpretation of the problematic language “serious risk of substantial and irreversible impairment of a major bodily function” so that compliance with the regulation “would not in any way pose a significant threat to the life or health of a woman.” Id. at 701. The Supreme Court said that if the statute were read so narrowly as to *1467 “foreclose[] the possibility of an immediate abortion despite some significant health risks,” the Court “would be required to invalidate the restrictive operation of the provision.” 505 U.S. at 880, 112 S.Ct. at 2822. However, in construing state statutes, the Supreme Court generally defers to lower federal courts unless an interpretation amounts to plain error. It found no plain error in the Third Circuit’s interpretation of the medical emergency exception as a matter of Pennsylvania law. Id.

Plaintiffs contend the medical emergency exception in Public Law 187, interpreted as a matter of Indiana law, does not pass constitutional muster. The question at this preliminary stage of the case is whether the plaintiffs have shown a reasonable likelihood of success on their claim that Indiana’s medical emergency exception does not meet the constitutional standard. The court will review first the relevant evidence and then Indiana law on statutory construction.

Much of the evidence concerning the health risks faced by pregnant women comes from the hearing testimony and affidavit of Dr. Jane Hodgson, a physician specializing in obstetrics and gynecology. 23 Dr. Hodgson’s testimony establishes that women face many increased health risks from pregnancy, including complication of a wide range of preexisting conditions such as asthma, epilepsy, diabetes, cardiac problems, chronic renal failure, myasthenia gravis, obesity, pulmonary emboli, and cancer. The pregnancy itself may pose some health risks, including hypertension and diabetes. Where a pregnancy complicates a pre-existing chronic condition such as asthma, epilepsy, or diabetes, Dr. Hodgson’s testimony shows that the complications can create medical conditions that range all across a spectrum of severity, from immediate threats to life to very mild and temporary conditions. Dr. Hodgson’s testimony shows that pregnant patients can experience a number of serious health problems such that delaying an abortion may cause severe but temporary pain, or may pose a threat of temporary but not irreversible damage.

In addition to the evidence from Dr. Hodgson about serious physical problems, Dr. Connie Best testified in her affidavit about the psychological risks for women who are pregnant and who are victims of rape, incest, or domestic violence. Her testimony indicates that such women may suffer severe psychological harm if they are required to delay an abortion or if they are required to hear some of the information required by Public Law 187.

The statute must be evaluated in light of this evidence about the kinds of medical situations in which questions about its application may arise. The language chosen by the legislature includes several limitations. The reference to “substantial and irreversible impairment” seems to exclude even the most severe and painful physical effects, so long as they are temporary. That limitation would appear to apply regardless of the extent of time, pain, and treatment that would need to be endured in healing. The reference to “major bodily function” appears to exclude consideration of a woman’s mental and emotional health. The term “major bodily function” creates by clear implication a category of “minor bodily functions” such that even substantial and irreversible impairment of them could not warrant an exception from the law. Looking solely at the language of the statute, it is difficult to see how these restrictive concepts can be reconciled with the requirement in Casey, Roe, and Doe that a medical emergency exception apply where compliance would cause “a threat to her health.” E.g., 505 U.S. at 880, 112 S.Ct. at 2822.

Of course, courts do not construe statutes in a vacuum or as an academic exercise. The defendants rely on two important principles of statutory construction to save the Indiana law. First, Indiana courts construe statutes, if reasonably possible, so as to render them constitutional. Second, where the legislature copies language from another jurisdiction *1468 where it has been construed by the courts, the legislature should be presumed to have intended that the language be given the same construction. Defendants also contend that because they are the public officials charged with enforcement of the law, the federal courts owe some deference to their construction of the law. Weighing against these considerations are the force of the statutory language actually enacted, evidence of the legislature’s actual (as opposed to its presumed) intention, and the context in which the statutory language must be applied.

After balancing these considerations, the court believes the plaintiffs have a reasonable likelihood of prevailing on the merits of their claim that the Indiana medical emergency exception is unconstitutionally narrow. The court will issue a preliminary injunction on that basis. However, as defendants point out, any decision the federal courts may make on this issue of Indiana law is provisional and subject to correction by the Indiana courts. Defendants have asked this court to certify to the Supreme Court of Indiana the interpretation of the medical emergency definition. The court will do so in order to obtain a definitive resolution of this aspect of Indiana law. Because certification does not provide an immediate answer, however, this court must resolve the question as best it can for purposes of ruling on plaintiffs’ motion for preliminary injunction. If the Supreme Court of Indiana disagrees, this court will promptly review its own decision and take appropriate action.

1. Plain Language

The dominant principle of statutory construction in Indiana law is that the plain language of the statute is the expression of the legislature’s intent that must be given effect. E.g., Indiana Dep’t of State Revenue v. Horizon Bancorp, 644 N.E.2d 870, 872 (Ind.1994) (“Nothing may be read into a statute which is not within the manifest intention of the legislature as gathered from the statute itself’; “unambiguous statute must be held to mean what it plainly expresses, and its plain and obvious meaning may not be enlarged or restricted”); Benham v. State, 637 N.E.2d 133, 136 (Ind.1994) (clear and unambiguous statutory language leaves no room for judicial construction; words and phrases must be taken in their plain, ordinary, and usual senses); Department of Public Welfare v. Couch, 605 N.E.2d 165, 167 (Ind.1992) (“words are to be given then-plain, ordinary, and usual meaning unless a contrary purpose is clearly shown by the statute itself’). It is the statutory language that is presented to the governor for signature or veto. It is the statutory language that is most readily available to those who are trying to conform their conduct to law.

Would the Indiana courts construe the language of the medical emergency exception in a way that is constitutional? Defendants argue that this language includes concepts that require some degree of construction, such as “serious” risk, “substantial” impairment, and “major” bodily function. The terms “serious,” “substantial,” and “major” all require some degree of quantitative judgment in application to specific cases. Those terms allow some latitude for the exercise of such judgments, and in that room left for those judgments, the concerns of physicians and patients may be reconciled with concerns of proponents of the legislation who fear a broad exception may invite abuse. 24

The constitutional problem lies not in these flexible terms of degree but in the definition’s other terms. The difference between “irreversible” and severe-but-temporary is not easily washed away as a matter of degree. Similarly, the gap between a “major bodily function” and mental or psychological health also cannot be bridged so easily by construction.

The issue here is whether, as a matter of Indiana law, the provision containing all those terms can be construed as the equivalent of a statute that excuses compliance with an abortion regulation whenever compliance would constitute “a threat” or “a significant threat” to the woman’s “health.” *1469 See Casey, 505 U.S. at 880, 112 S.Ct. at 2822 (using both terms). The court may construe statutory language so as to render it constitutional only to an extent that the construction is reasonable. In re Public Law No. 154-1990, 561 N.E.2d 791, 792 (Ind.1990) (court’s duty is to bring statute in harmony with constitutional requirements “if the language permits”); Grody v. State, 257 Ind. 651, 278 N.E.2d 280, 285 (1972) (court “will not consider the abstract question of whether Congress might have enacted a valid statute but instead must ask whether the statute Congress did enact will permissibly bear a construction rendering it free from constitutional defects”) (quoting Aptheker v. Secretary of State, 378 U.S. 500, 515, 84 S.Ct. 1659, 1669, 12 L.Ed.2d 992 (1964)). As examples of this reluctance expressed in Grody, see, e.g., Brady v. State, 575 N.E.2d 981, 988-89 (Ind.1991) (plain language of statute violated defendant’s right under state constitution to “meet the witnesses face to face”); In re Public Law No. 154-1990, 561 N.E.2d at 792-93 (statutory declaration of absolute immunity from attorney disciplinary rules clearly violated separation of powers). As the Supreme Court of Indiana explained:

A court cannot amend a statute or establish public policy within its judicial authority to confine legislative products to constitutional limits. However, a court in reading a statute for constitutional testing, may give it a narrowing construction to save it from nullification, where such construction does not establish a new or different policy basis and is consistent with legislative intent. State v. Kuebel (1961), 241 Ind. 268, 172 N.E.2d 45.

State v. Downey, 476 N.E.2d 121, 123 (Ind.1985). In Downey the court found that the particular language at issue (“may endanger” in a child neglect statute) did not “indicate a critical legislative choice or represent the resolution of important issues within the social problem involved.” Id. The court therefore found that a saving interpretation was permissible. See also Burris v. State, 642 N.E.2d 961, 968 (Ind.1994) (giving saving construction to death penalty statute); Price v. State, 622 N.E.2d 954, 963-64 (Ind.1993) (giving saving construction to statutory reference to “public nuisance”).

Although some of the statutory terms can be stretched to some extent, defendants have not explained how language of “irreversible” impairments can be extended to temporary but severe conditions. Nor have they explained how language referring to a “major bodily function” can be extended to psychological and mental health. At least they have not offered such explanations based on the language itself. Thus, the assertion that the courts should construe this language as constitutional appears to take the approach rejected in Grody v. State: consider whether the legislature might have enacted a valid statute on the subject, and then impose on the enacted language whatever construction is needed to save the statute. 278 N.E.2d at 285. That approach effectively requires courts to write statutes rather than to construe them, and Indiana courts ordinarily refuse invitations to do so. Cf. Jane L. v. Bangerter, 61 F.3d 1493, 1501-02 (10th Cir.1995) (reversing district court’s saving construction of ban on fetal “experimentation” as a judicial “rewriting” of statute).

In fact, in a very similar context, the Supreme Court of Indiana adhered to a remarkably restrictive construction of a statutory exception for medically necessary abortions. In Cheaney v. State, 259 Ind. 138, 285 N.E.2d 265 (1972), decided a few months before Roe, the court considered whether the medical exception to the Indiana law proscribing abortions was unconstitutionally vague. The criminal law then prohibited the performance of all abortions unless “necessary to preserve [the pregnant woman’s] life.” Id. at 266. The court held that that language was not vague but plainly meant, in light of “the context of the entire statute,” that “an abortion will be allowed only when the continuation of the pregnancy will directly and proximately result in the death of the mother.” Id. at 271. Justice DeBruler dissented from the judgment on constitutional privacy grounds. He did not suggest the majority had misread the statute, but pointed out that the statute, as worded, prohibited all abortions except in cases “where death will certainly result to the pregnant woman,” thus requiring a pregnant woman to forego an abortion and “run the risk of dying when *1470 that risk is not a certainty, but only a middle range possibility.” 285 N.E.2d at 274 (emphasis added). The majority responded: “If this law appears too strict or harsh for some, the place to change it is not in the courts but in the legislature.” Id. at 271. Subsequent developments in constitutional standards would cause the Supreme Court of Indiana to decide some of the federal constitutional issues differently, but the court’s faithfulness to the strong language of the statute is evident.

2. Legislative Intent

It is unclear whether the Supreme Court of Indiana would be willing to consider evidence beyond the language of the statute in construing it, at least as to the terms “irreversible” and “major bodily function.” See, e.g., Orr v. Turco Mfg. Co., 484 N.E.2d 1300, 1302 (Ind.App.1985) (when statute is unambiguous, legislative history should not be used to determine its meaning). Cf. O’Laughlin v. Barton, 582 N.E.2d 817, 821 (Ind.1991) (when legislation is susceptible to several different constructions, court may look at legislative journals to determine legislative intent); County Dep’t of Public Welfare v. Potthoff, 220 Ind. 574, 585, 44 N.E.2d 494, 498 (1942) (same). Assuming that the court would be willing to consider such materials, defendants base their strongest argument on the near identity between the Indiana language and the Pennsylvania language that was construed and then upheld in Casey. Defendants contend that the Indiana legislature intended to adopt the construction that had been placed on the Pennsylvania statute. Indiana courts may presume that when the Indiana legislature adopts statutory language from a sister state, the constructions placed on such language are adopted as well. In re Marriage of Hudson, 434 N.E.2d 107, 113 (Ind.App.1982); In re City Investing Co., 411 N.E.2d 420, 427 (Ind.App.1980). See also Miller v. Mayberry, 506 N.E.2d 7, 11 (Ind.1987) (legislature presumed to be aware of prevailing judicial construction of a statute in Indiana); Lewis v. Lockard, 498 N.E.2d 1024, 1026 (Ind.App.1986) (same). 25

It is not entirely clear how the Supreme Court of Indiana might resolve the tension between the plain language of a statute and the foregoing presumption if there were no other pertinent evidence on the question. As the Supreme Court of the United States has explained, however, the presumption that judicial constructions are borrowed from other jurisdictions along with their statutory language is “a presumption of legislative intention ... which varies in strength with the similarity of the language, the established character of the decisions in the jurisdiction from which the language was adopted and the presence or lack of other indicia of intention.” Carolene Products Co. v. United States, 323 U.S. 18, 26, 65 S.Ct. 1, 5, 89 L.Ed. 15 (1944). Here the material language is identical and the single decision construing the Pennsylvania language was certainly known to the Indiana legislature. However, the force of defendants’ argument is limited not only by the weight of the language actually enacted by the legislature (“substantial and irreversible impairment of a major bodily function”) but also by the “other indicia of intention” — in this case, evidence of the Indiana legislature’s actual intention when it squarely confronted this issue.

As introduced and as originally passed by the Indiana Senate, the medical emergency exception tracked the language from the Pennsylvania statute quoted above, “serious risk of substantial and irreversible impairment of a major bodily function.” The statute upheld in Casey was plainly the model for this language. However, the Indiana House committee considering the bill amended the definition. It substituted the phrase “serious risk of substantial permanent impairment to *1471 the woman’s life or health.” This language was taken from Ind.Code § 16-34-2-1(3), which is the “life or health” exception for post-viability abortions. The bill then came to the House floor for amendments on second reading. The bill’s chief sponsor in the House appealed to his colleagues to remove the reference to “health” and to restore the Senate version of the medical emergency definition:

This deals with medical emergencies and the purpose of the bill that if there’s a medical emergency where the woman’s life is in danger, that it doesn’t seem prudent unless the doctor has some time to inform the woman, go through the consent, have her see everything that’s necessary and all that, that we realize that there is an emergency. It’s something different than someone contemplating an abortion, and this may be something that she doesn’t even want. This is a medical emergency. However, it was changed in the committee to put permanent impairment, if you look on page two, line five, and also page three, line 27 of the life or physical health of the pregnant woman. The reasons we need to change this back quite frankly is this opens it up to every abortion you can think of whether it’s an emergency or not. Complete opposite of the supporters of this bill is what they do not want here. Quite frankly, an opinion we have from the language instead of talking about preventing her death, this language will talk about her physical health instead of her imminent death because of not having an abortion.

Also, by having the word “health” at the end of those two sentences instead of the words that I propose, impairment of major bodily junction, those words could mean anything that affects a woman’s well being from including physical, sociological or family problems. So, if she has a disagreement with her husband it now becomes an emergency and they can perform an abortion without an informed consent or the items go along with it, that’s the danger of those words. If you believe in this bill, we need to take those words out and go back to the purpose of this bill, especially for a medical emergency, if the woman’s life is in danger which we all can agree with. I would ask your support of this amendment.

Pl.Ex. 38 at 1-2 (emphasis added). He later explained:

This is talking about someone’s life and again, just to point out that the federal government under Doe v. Dalton [sic] has defined “health” which is in this bill to mean anything which affects a woman’s well being including psychological, sociological or family problems and that’s certainly what we do not want, even though her intentions may be good, this completely reverses that bill. I ask again for your support for my amendment.

Pl.Ex. 38 at 2 (emphasis added). The House then voted to amend the definition to restore the original language.

The House committee’s adoption of a broader definition, the House floor debate, and the restoration of the narrower language show that the Indiana legislature did not intend for the statutory language to be construed as broadly as the Third Circuit interpreted the language in Casey or as broadly as the Constitution requires. The legislature copied the statutory language approved in Casey, but when the specific issue came to a head, supporters of the bill were asked to restore the narrower language precisely because it would not permit the broader “construction” proposed by defendants here and would have a meaning different from the meaning of “health” in Doe v. Bolton, 26 They did so. Thus, it appears that the legislature made “a critical legislative choice” on this issue that prevents a different interpretation for the purpose of saving the legislation. See State v. Downey, 476 N.E.2d at 123 (court may give narrowing construction where it is “consistent with legislative intent” *1472 and where language “does not indicate a critical legislative choice or represent the resolution of important issues within the social problem involved”).

In holding that states may “regulate, and even proscribe, abortion” during the third trimester, Roe mandated an exception to such regulations “where it is necessary, in appropriate medical judgment, for the preservation of the life or health of the mother.” 410 U.S. at 165, 93 S.Ct. at 732. Roe and Doe v. Bolton, decided the same day, “are to be read together.” 410 U.S. at 165, 93 S.Ct. at 733. Therefore, when Roe forbids state regulation when “preservation of the life or health of the mother” is necessary, the Court was referring to the broader concept of “health” found in Doe v. Bolton. Casey did nothing to alter the threshold of the required medical exception set forth in Roe. See 505 U.S. 878-80, 112 S.Ct. at 2821-22. By rejecting the broader concept of health when the specific issue was raised, the Indiana legislature provided substantial evidence about how it intended the language to be applied. Its intent seems to have been that the words of the statute mean what they say. In the face of both the statutory language and the evidence of the actual legislative intent, it appears the Indiana statute cannot be saved by the preference for an interpretation to save the statute or the presumption that the legislature intended to adopt another jurisdiction’s judicial construction.

3. The Role of “Fair Notice” in Statutory Construction

In its entry explaining the issuance of the temporary restraining order in this case, the court based its initial view of the Indiana definition of medical emergency in part on the need for physicians to have fair notice of what the law requires and prohibits. These concerns take on special force when a law applies in the context of medical emergencies, where prompt action may be required. A violation of the law can result in severe legal and professional consequences for the physician, while a misunderstanding of the law in the other direction could have avoidable adverse health consequences for the patient.

Defendants point out that the Third Circuit in Casey rejected a “void for vagueness” challenge to the definition, 947 F.2d at 702, and that the Eighth Circuit in Fargo Women’s Health Organization rejected a vagueness challenge to the virtually identical language in North Dakota’s statute, 18 F.3d at 534-35. This court is not suggesting that defendants’ proposed construction of the Indiana law would render the law unconstitutionally vague. However, the Indiana courts could consider issues of fair notice and context in determining whether the proposed construction is correct as a matter of statutory interpretation.

A proposed construction of a statute can survive constitutional “void-for-vagueness” scrutiny, yet still be so at odds with the apparent meaning of the language that the Indiana courts would reject the proposed construction. This may be such a case. As Dr. Hodgson’s testimony indicates, the language of the statute is not entirely clear to her as a physician, but it appears to exclude from the definition of a medical emergency some situations that doctors would consider to be emergencies warranting immediate abortions. The Indiana courts might consider relevant the fact that physicians would need to apply the law in the context of medical emergencies, understood broadly. This is not a situation where the parties trying to conform their conduct to law will always have an opportunity to ask for a legal opinion from counsel before they take action. The Indiana courts could consider, in interpreting the law, the need for physicians making decisions about the life and health of their patients to be apprised of what treatment they may legally prescribe. In that context, even a decision by the Supreme Court of the United States construing similar language in another state’s statute in a way not obviously consistent with its language might not be readily accessible to physicians trying to conform their conduct to law in an emergency situation.

Assuming that a physician had an opportunity to seek a legal opinion on a particular situation, even the fact that Casey is on the books does not necessarily solve the problem as a matter of Indiana statutory construction. *1473 In addition to the statutory language and legislative history discussed above, the Indiana courts might also consider relevant the fact that the same chapter of the Indiana Code already has two other medical emergency provisions. See Ind.Code § 16-34-2-1(3)(C) (permitting post-viability abortions where abortion is “necessary to prevent a substantial permanent impairment of the life or physical health of the pregnant woman”); Ind.Code § 16-34-2-4(i) (parental consent for minor’s abortion not required where “continuation of the pregnancy provides an immediate threat and grave risk to the life or health of the pregnant woman”). The Seventh Circuit has upheld the constitutionality of the second exception, construing it as allowing a physician to use his or her medical judgment, “a judgment that would be questioned only if the physician did not believe there was a health danger.” Indiana Planned Parenthood Affiliates Ass’n v. Pearson, 716 F.2d 1127, 1139 (7th Cir.1983) (emphasis added). Courts ordinarily would consider significant the legislature’s use of three different formulas in such close proximity dealing with the same type of question. See, e.g., Federal Election Comm’n v. NRA Political Victory Fund, — U.S. -, -, 115 S.Ct. 537, 541, 130 L.Ed.2d 439 (1994); United States v. American Bldg. Maintenance Indus., 422 U.S. 271, 277, 95 S.Ct. 2150, 2154-55, 45 L.Ed.2d 177 (1975). A physician who had an opportunity to ask a lawyer to research these matters might reasonably hesitate in view of these different formulas for medical emergencies in the abortion context. And the Indiana courts might reasonably consider the plight of the physician-facing severe legal and professional sanction if she errs in one direction and avoidable harm to her patient if she errs in the other— in deciding how to construe the language of the statute.

4. Application to Victims of Rape and Incest

Much of the debate in the Indiana House of Representatives concerned whether the bill would exempt victims of rape and incest from the mandatory disclosure and waiting period requirements. This debate focused on a small fraction of the women seeking abortions, where a subject already emotionally charged becomes even more so. The evidence at the hearing showed that about one percent of the abortions performed are for women who are victims of rape or incest. The legislature ultimately decided to excuse rape victims from being told that the father could be liable for child support. It rejected all other efforts to create such exceptions.

The defendants contend correctly that Casey, Roe, and the Constitution do not require an exception from the disclosure and waiting period for victims of rape and incest, as such. See Planned Parenthood, Sioux Falls Clinic v. Miller, 63 F.3d at 1467, 860 F.Supp. at 1421. Nor is it appropriate for federal judges to evaluate such statutes based on their own views of such legislative policy decisions. However, the Supreme Court noted in Casey that a physician was not required to comply with Pennsylvania’s informed consent and waiting period requirements “if he or she can reasonably demonstrate by. a preponderance of the evidence that he or she reasonably believed that furnishing the information would have resulted in a severely adverse effect on the physical or mental health of the patient.” 505 U.S. at 883-84, 112 S.Ct. at 2824 (quoting statute) (emphasis added). The Indiana statute contains no similar language. Casey did not state explicitly that such an exception for mental health is constitutionally required. However, the holdings of Casey and Roe that a state may not impose abortion regulations that would pose a significant threat to a woman’s health, and the broad view of “health” adopted in Doe v. Bolton, together indicate that there must be such an exception to the mandatory disclosure and waiting period provisions. Accord, United States v. Vuitch, 402 U.S. 62, 71-72, 91 S.Ct. 1294, 1299, 28 L.Ed.2d 601 (1971) (medical necessity exception not unconstitutionally vague; “health” included psychological as well as physical well-being). See also Casey, 505 U.S. at 882, 112 S.Ct. at 2823 (“It cannot be questioned that psychological well-being is a facet of health.”).

The affidavit of Dr. Connie L. Best concerning women who are victims of rape and incest shows that a significant proportion of victims suffer significant psychological difficulties, including post-traumatic stress syn *1474 drome or a major depressive episode. Such difficulties can be made worse by pregnancy resulting from the rape or incest. Dr. Best also testified in her affidavit concerning the effects the law may have on battered women. She stated that some of the state-mandated information could be “extremely psychologically harmful” to victims of rape, incest, and other forms of violence. Her testimony stands unrebutted on this record. There may be many eases where a physician will conclude that an individual victim of rape, incest, or other violence would not risk serious harm to her health by complying with the statute. Dr. Best does not deny that possibility. But the narrow definition of “medical emergency” in the Indiana law would require compliance with the law despite a reasonable medical judgment that compliance appears to cause serious harm to the mental health of a particular victim.

The issue is not whether the Constitution requires an exception for women who are the victims of rape, incest or domestic violence, but whether it compels an exception for the mental health of the pregnant woman, whatever event or condition may trigger the concern. 27 The evidence shows that some women, because of rape, incest, violence, fetal anomalies, or other reasons, will experience severe emotional or psychological harm if they must wait to have an abortion or be provided all the prescribed information. The absence of a medical exception that includes aspects of mental and emotional health appears likely to violate the Constitution.

5. Deference to Construction by State Officials

Defendants assert that this court is required to give considerable deference to their own interpretation of the law because they are charged with enforcing it. See Ward v. Rock Against Racism, 491 U.S. 781, 795-96, 109 S.Ct. 2746, 2755-57, 105 L.Ed.2d 661 (1989); Frisby v. Schultz, 487 U.S. 474, 483, 108 S.Ct. 2495, 2501-02, 101 L.Ed.2d 420 (1988); Kolender v. Lawson, 461 U.S. 352, 355, 103 S.Ct. 1855, 1857, 75 L.Ed.2d 903 (1983). Their proposed construction is that adopted by the Third Circuit in Casey: the exception applies where compliance would “in any way pose a significant threat to the life or health of a woman.” 947 F.2d at 701. For the reasons explained above, this court does not believe that the defendants’ proposed construction of the statute can be reconciled, as a matter of Indiana law, with the language the legislature chose to enact. The court therefore believes that plaintiffs have shown a reasonable likelihood of succeeding on the merits on this claim. However, as defendants point out correctly in their motion for certification to the Supreme Court of Indiana, any decision this court might make on this question of Indiana law is merely provisional and subject to erasure by the Supreme Court of Indiana. The court will therefore grant the motion for certification so that the Supreme Court of Indiana may resolve these issues.

V. Irreparable Harm, the Balance of Harms, and the Public Interest

The Casey joint opinion made clear that the essential holding from Roe that a woman has a constitutional right to choose whether to continue or terminate her pregnancy is still the law. Given the nature of the right and the high standard that must be satisfied to show that a law’s effects impose an “undue burden” on the right, a plaintiff who proves the likelihood of an undue burden has also shown the threat of irreparable harm. See Planned Parenthood v. Casey, — U.S. at -, 114 S.Ct. at 910 (Souter, J., in chambers) (proof of “substantial obstacle” would establish irreparable harm); Doe v. Mundy, 514 F.2d 1179, 1183 (7th Cir.1975) (violation of privacy right to choose abortion would cause irreparable harm); Doe v. Bellin *1475 Memorial Hosp., 479 F.2d 756, 759 (7th Cir.1973) (interference with right to have timely abortion would be irreparable). Here, too, there would be irreparable harm if Public Law 187 is not enjoined.

The evidence that tends to show the “in the presence” requirement of Public Law 187 imposes an undue burden on a woman’s right to choose to have an abortion supports the plaintiffs’ claim of irreparable harm. The Mississippi data support the inference that the Indiana law is likely to prevent approximately 11 to 14 percent of Indiana women who would otherwise choose to have an abortion from doing so. The information from Mississippi, North Dakota, and the plaintiff clinics in Indiana, together with the absence of evidence showing that mandatory disclosure laws actually change women’s minds, all support the inference that the likely drop in the number of women having abortions would be the result of burdensome effects of the Indiana law rather than the persuasive effects of the mandatory disclosures and waiting period. The evidence indicates that more than 15,000 abortions are performed in Indiana each year. See Pl.Ex. 5 (15,840 in 1992). Applying the rate of 11 to 14 percent to an annual total of 15,000, if the Indiana law were to go into effect with its “in the presence” requirement, the result would be that over the next year, approximately 1,650 to 2,100 women in Indiana would be unable to obtain abortions they would otherwise choose to have. That is substantial and irreparable harm. 28

On the other side of the equitable scale, defendants argue that a preliminary injunction will cause irreparable harm because there is a “small, but real group of women that clearly benefit from additional information and a reflective period.” The defendants’ evidence on this point shows that under the deferential rational relation standard of constitutional review, the Indiana legislature could rationally have found that there is such a group of women whose needs for information and time are not met by existing Indiana law requiring informed consent for medical procedures and by current practices among Indiana providers of abortion services. But the evidence offered at the preliminary injunction hearing does not show that such a group of women exists in fact.

The only testimony on this score admissible for the truth of the matters asserted came from Ms. Dowden and Ms. Huddleson. They testified about their own experiences in the mid-1970s. Their accounts of their experiences are troubling, for each believes that she was misled about relevant information and pressured into having an abortion that she now regrets. Because their experiences were so long ago, however, and because the evidence from the plaintiffs indicates that such problems occur now very rarely, if at all, their testimony does not show that current Indiana law and actual practices today are insufficient to ensure that women make informed and thoughtful decisions. See Ragsdale v. Turnock, 625 F.Supp. 1212, 1231 (N.D.Ill.1985) (in balancing harms at preliminary injunction stage, court could consider fact that other existing laws would remain in effect to protect the public), aff'd in part, vacated in part on other grounds, 841 F.2d 1358 (7th Cir.1988), appeal dismissed, Turnock v. Ragsdale, 503 U.S. 916, 112 S.Ct. 1309, 117 L.Ed.2d 510 (1992). Also, it is far from clear that compliance with the specific provisions of Public Law 187 would have caused either of these witnesses to have changed her mind under the circumstances she faced at the time.

To show irreparable harm from an injunction, defendants also rely on evidence from Jane Stout and Dr. Hodgson concerning the ambivalence that some women and teenagers show about their choice. The evidence from plaintiffs indicated, however, that their coun *1476 selors are trained and instructed to make sure that patients are aware of and have considered their options, and that if a patient seems ambivalent about her choice, additional time and counseling are provided. Plaintiffs do not require such “options counseling” where there is no indication of a need for it. The record now before this court does not show that plaintiffs’ practices are inadequate such that an injunction is likely to prevent any woman from either receiving information she actually needs or taking any time she wants or needs to make her choice.

The task of the district court facing an application for a preliminary injunction is to evaluate the merits of the case on a tentative and provisional basis and to weigh the consequences of granting the injunction against the consequences of denying the injunction. As the Seventh Circuit has explained, this task is not subject to an exact formula, but the court should try to minimize the risk and consequences of an erroneous decision. E.g., Abbott Labs., 971 F.2d at 12. For the reasons explained above, the plaintiffs have shown a reasonable likelihood of showing that the medical emergency exception (which is essential to all the provisions of Public Law 187) and the “in the presence” requirement for the mandatory disclosure and waiting period provisions are likely to impose undue burdens on women’s rights to make their own decisions about whether to continue or terminate their pregnancies. 29

If an injunction were denied, the evidence before the court at this point shows that the burdens of the “in the presence” requirement would likely prevent hundreds of women in Indiana each year from having abortions they would otherwise choose to have. Those individual women cannot be identified at this time, but the irreparable harm to them is both predictable and substantial. If an injunction is granted, there is a possibility that some women might choose to have abortions without being sufficiently aware of their alternatives and without taking sufficient time to consider their choice. The evidence now before the court does not establish, however, that this risk is either real or substantial. The court’s provisional evaluation of the merits and the evidence relevant to the issue of irreparable harm therefore weigh in favor of enjoining enforcement of Public Law 187 pending final resolution of this case. That final resolution will be reached after both sides have had a sufficient opportunity to develop and present additional evidence, and after the Supreme Court of Indiana has had an opportunity to consider the determinative issues of Indiana law.

VI. Conclusion

For the foregoing reasons, the court is issuing today a preliminary injunction enjoining enforcement of Public Law 187 pending a final judgment on the merits or further order of this court.

Notes

. In their complaint, plaintiffs also challenge an Indiana statute providing that abortions after the first trimester may be performed only "in a hospital or ambulatory outpatient surgical center.” Ind.Code. § 16-34-2-1 (2)(B). Plaintiffs have not requested a preliminary injunction against this provision, so it need not be addressed at this stage in the litigation.

. On October 13, 1995, defendants filed a motion to certify the interpretation of the medical emergency exception to the Supreme Court of Indiana. This court may certify such a question, see United Farm Bureau Mut. Ins. Co. v. Metropolitan Human Relations Comm'n, 24 F.3d 1008, 1015 n. 10 (7th Cir.1994), and the Supreme Court of Indiana in its discretion may answer the question pursuant to Ind.App.R. 15(0). By separate order today, the court grants defendants’ motion to certify the question, and will await action by the Supreme Court of Indiana before trying the merits in this case. However, the certification of the statutory issue does not affect this court’s obligation to interpret the medical emergency exception at this time to the extent required by plaintiffs’ motion for a preliminary injunction.

. This court recognizes that the Fifth Circuit upheld the Mississippi law because Casey was controlling. Barnes v. Moore, 970 F.2d 12 (5th Cir.1992). That decision was made before the law had taken effect, and thus before evidence of its actual effects was available.

. By separate order today the court has ruled on numerous objections to portions of certain affidavits.

. Justice Stevens asserted in Casey that the Court should not decide whether a restriction imposes an "undue” burden without considering the extent of any actual benefits of the restriction. 505 U.S. at 920, 112 S.Ct. at 2843. He suggested that "the future may also demonstrate that a standard that analyzes both the severity of a regulatory burden and the legitimacy of its justification will provide a fully adequate framework for the review of abortion legislation....” Id. 505 U.S. at 920 n. 6, 112 S.Ct. at 2843 n. 6. Several commentators have suggested that the joint opinion in Casey did not intend to treat "purpose” and "effects” as two entirely separate tests for its undue burden analysis. See Alan Brownstein, How Rights Are Infringed: The Role of Undue Burden Analysis in Constitutional Doctrine, 45 Hastings L.J. 867, 887-92 (1994); Sandra L. Tholen & Lisa Baird, Con Law Is As Con Law Does: A Survey of Planned Parenthood v. Casey in the State and Federal Courts, 28 Loy. L.A.L.Rev. 971 (1995) (text accompanying notes 101-04). These commentators and at least one other have suggested that "undue burden” analysis in abortion cases should balance burdens against benefits to at least some degree (although they do not suggest that any precise weighing is necessary). They argue that such an approach would be more consistent with “undue burden” analysis in other areas of constitutional law where the extent of any benefits may be considered. See Gillian E. Metzger, Unburdening the Undue Burden Standard: Orienting Casey in Constitutional Jurisprudence, 94 Colum.L.Rev. 2025, 2077-89 (1994); Brownstein, 45 Hastings L.J. at 891-92; Tholen & Baird, 28 Loy.L.AX.Rev. 971 (text accompanying notes 93-126). This court must follow the joint opinion in Casey. This court reads that opinion as calling for separate consideration of purpose and effect, and as not considering the extent of any genuine benefits from a restriction, at least so long as the restriction appears to have a legitimate purpose and will not prevent women from exercising their right to choose. That seems to be one of the key points of disagreement between the joint opinion and Justice Stevens' opinion. The subsequent history of Casey itself, however, suggests that as district courts makes their factual findings in these cases, they should keep in mind possible developments in the constitutional standards. See Planned Parenthood v. Casey, 822 F.Supp. 227 (E.D.Pa.1993) (allowing presentation of additional evidence to apply "undue burden” standard), rev’d, 14 F.3d 848 (3d Cir.1994), stay denied, - U.S. -, 114 S.Ct. 909, 127 L.Ed.2d 352 (1994) (Souter, J., in chambers).

. This form of analysis focuses on the effects of a particular restriction on abortions, considered in isolation. The joint opinion in Casey did not indicate whether or how it might be possible to prove that a state has imposed an undue burden on a woman’s right to choose through the cumulative effect of many restrictions which may each serve a legitimate purpose, but only to a marginal degree. Cf. Ohio v. Akron Center for Reproductive Health, 497 U.S. 502, 527, 110 S.Ct 2972, 2985-86, 111 L.Ed.2d 405 (1990) (Blackmun, J., dissenting); see also Walter Dellinger & Gene B. Sperling, Abortion and the Supreme Court: The Retreat from Roe v. Wade, 138 U.Pa.L.Rev. 83, 99-103 (1989).

. The information the statute requires to be delivered in person relates to fetal development and medical risk. See Ind.Code § 16-34-2-1.1(1). The statute permits information concerning child support, medical assistance, and adoption alternatives to be delivered "orally,” which could include by telephone. See Ind.Code § 16-34 — 2-1.1(1) & (2).

. Dr. Henshaw is deputy director of research at the Alan Guttmacher Institute in New York. He has a doctorate in sociology. Most of his professional work concerns quantitative statistical studies related to pregnancy, abortion, and reproductive health services.

. Dr. Henshaw appears not to have addressed the possibility that Mississippi women obtained more abortions in Arkansas, the fourth state bordering Mississippi, but defendants have not pressed the point. The court will not speculate about the matter but will limit its evaluation to the evidence actually presented.

. All of these data were significant at a level of at least 95 percent, and most to higher degrees of significance. See Pl.Ex. 4 (values of "p" ranging from less than 0.05 to less than 0.001).

. Dr. Steele has a doctorate in public health. The focus of his professional work for many years has been work with health statistics of a wide variety, including data on pregnancy, births, and abortions.

. If abortions were evenly distributed among months, the ratio would be 71.429 percent. Because abortion rates are generally slightly higher in the early months of the year, decline in late summer and fall, and increase again in November or December, some seasonal adjustment is necessary for use of partial-year data.

. In their post-hearing reply brief defendants included a table of ratios they had calculated for the years 1987-1994, but that table is not meaningful without further analysis and calculations of the sensitivity of Dr. Henshaw's conclusions to such changes.

. Chi-square calculations are mathematical tests to determine the extent to which differences between expected and observed phenomena are likely to have been the result of random chance. Such calculations provide one basis for concluding whether a difference between expected and observed phenomena is "statistically significant,” which can be quantified in several ways. See generally Federal Judicial Center, Reference Manual on Scientific Evidence 378-81 (1995).

. The court is concerned, however, about Dr. Steele's testimony that he had thus far been unable to duplicate some of Dr. Henshaw's calculations. The testimony of both witnesses at the preliminary injunction hearing had been prepared under significant time pressure, so there may not have been adequate time to iron out any apparent discrepancies in the arithmetic. If necessary, the court will take appropriate steps through the discovery process in preparing for the trial on the merits to ensure that the data and calculations of experts on each side are readily accessible to the other side.

. Dr. Henshaw’s initial study of the 1992 data showed that the drop in abortion rates appeared greater for women with less than a high school education. That correlation did not appear in the later study looking at data from 1991, 1992, and 1993. See Pl.Ex. 4.

. The North Dakota law took effect on March 17, 1994. Ms. Bovard presented data showing the number of abortions her clinic provided from January to September of 1993, 1994, and 1995. Pl.Ex. 29. The numbers are virtually identical. Defendants say that the North Dakota statistics are "crude” and that the numbers can be analyzed in different ways such that it is possible the North Dakota law caused a slight decline in the abortion rate. The North Dakota data have not yet been subjected to even the limited analysis performed on the Mississippi data. The parties will have an opportunity to do such analysis before trial.

. This portion of Casey illustrates the issue plaintiffs face. A spousal notice law like Pennsylvania’s could conceivably reduce the incidence of abortion by causing husbands and wives to talk about the decision and reach joint decisions that would be different from the wives’ initial decisions. That result would be permissible under Casey, but the Supreme Court's decision to strike down the spousal notice law shows that the Court concluded that scenario was unlikely. The evidence concerning abusive relationships appears to have played an important role in convincing the Supreme Court that the likely decrease in the incidence of abortion caused by the spousal notice law was unlikely to result from a woman changing her mind after a constructive dialogue, mandated by the state law, with her husband.

. Judges and other non-scientists often speak of “verifying” scientific claims, see, e.g., Porter v. Whitehall Laboratories, Inc., 9 F.3d 607, 615 (7th Cir.1993), but in view of the at least theoretical possibility that even the best established scientific verities could be proven false in the future, it is more accurate to refer to ''falsifiability. See Daubert, - U.S. at -, ---, 113 S.Ct. at 2795, 2796-97; Sierra Club v. Marita, 46 F.3d 606, 622 (7th Cir.1995).

. Defendants make much of Dr. Henshaw's deposition testimony about “speculation” and the limits of his opinion. What Dr. Henshaw could not explain specifically, based on the data he had, is “what was going on in [women’s] heads to cause them not to come to have abortion services as the consequence of the law.” Tr. 98-99. His opinion that the law's effects are not the result of the required counseling or persuasive effects is not "speculation,” regardless of his inability to identify precisely how and why the law’s burdens prevent women from having abortions.

. The preliminary character of this decision and the evidence upon which it must be based is illustrated by two episodes in the preliminary injunction hearing. First, through Dr. Steele, defendants offered evidence to show that the incidence of abortions in Indiana, which had not changed its law, had dropped by an amount comparable to the drop in Mississippi at the same time. By the next day it became apparent that the Indiana government data on which this testimony was based came from a year for which the state government had lost much of the relevant records and had then scrambled to reconstruct them as best it could. Defendants therefore withdrew the key exhibit. By the same token, plaintiffs asked the court to leave the hearing record open for a few days so they could have an opportunity to obtain better evidence directly from the Mississippi clinic directors. The court provided that opportunity, which passed without proffer of additional evidence and without explanation.

. The evidence here shows that the challenged legislation was supported by a number of groups opposed to abortion in all or nearly all circumstances. Plaintiffs ask the court to treat these groups' support as evidence of the law's unconstitutional purpose. The court declines to do so. The fact that supporters of a law may desire to change the law in ways that would be held unconstitutional does not poison any legislation on the general subject that they may endorse. Plaintiffs have pointed to no case taking such a step, and the Supreme Court has treated a desire or goal of reducing or even outlawing abortion as within the bounds of legitimate political debate. See Bray v. Alexandria Women's Health Clinic, 506 U.S. 263, 274, 113 S.Ct. 753, 762, 122 L.Ed.2d 34 (1993).

. Dr. Hodgson was trained and educated at the University of Minnesota Medical School and the Mayo Clinic, among other places. She has been a licensed physician for over 55 years and has been board certified in obstetrics and gynecology since 1949. She has published many papers and articles in the area of obstetrics and gynecology.

. In construing the Pennsylvania statute as constitutional in Casey, the Third Circuit focused on three specific conditions and relied on the flexibility in these same statutory terms of degree to hold that the medical emergency exception would apply to those conditions. See 947 F.2d at 700-02.

. Defendants also point out that other federal courts have adopted similar constructions of other state laws copying the Pennsylvania law. See Planned Parenthood, Sioux Falls Clinic v. Miller, 63 F.3d at 1467; Fargo Women's Health Org. v. Schafer, 18 F.3d at 533; Barnes v. Moore, 970 F.2d at 15; Utah Women's Clinic, Inc. v. Leavitt, 844 F.Supp. at 1492-94. However, the courts’ opinions in those cases do not address the specific issues raised here. They do not explain whether or how "irreversible" may apply to temporary but severe impairments, nor whether or how “major bodily function” may apply to mental health. They also did not have before them evidence of legislative history undermining the reliance on the interpretation of Casey, as discussed below.

. The transcript refers to "Doe v. Dalton," but the sponsor surely meant Doe v. Bolton. The court is unaware of a case called Doe v. Dalton, and Doe v. Bolton discusses the precise concerns the sponsor raised, an expansive definition of the word "health" in the abortion context. See 410 U.S. at 192, 93 S.Ct. at 747. See also Pl.Ex. 31 at 100305 (letter of April 9, 1995, received by bill's sponsor (and apparently the "opinion” referred to in House debate) urging use of original language because Doe v. Bolton concept of "health" was too broad).

. Similar considerations may apply in the cases of women carrying so-called anomalous fetuses. Dr. Hodgson testified that about 50 percent of women who find out that they are carrying an anomalous fetus elect to terminate the pregnancy and that many such women are quite anxious to obtain an immediate abortion. Most such pregnancies are planned pregnancies, and women are often distraught to learn that the child would be unable to survive or would be bom with very severe physical or developmental disabilities. Thus, compliance with the information and waiting period requirements of the Indiana law could produce extreme anguish. Defendants point out that some women in this situation could also benefit from additional time. However, at least in severe situations, the Constitution appears to require that such matters be left to the physician’s judgment.

. Defendants point out that there is no “Jane Roe” as a plaintiff in this case, and that harm to the plaintiff clinics from enforcement of the Indiana law would be only financial. However, because of both the privacy concerns at stake and the very brief time any one individual pregnant woman has a stake in the issue, providers of abortion services have standing to raise the privacy rights of their patients. E.g., Singleton v. Wulff, 428 U.S. 106, 117, 96 S.Ct. 2868, 2875, 49 L.Ed.2d 826 (1976) (plurality opinion). In addition, the court must make its decision on a motion for preliminary injunction in light of consequences for those who are not actually parties to the lawsuit. E.g., Duct-O-Wire Co. v. U.S. Crane, Inc., 31 F.3d 506, 509 (7th Cir.1994). Also, even financial harm to the clinics would be irreparable in light of the immunity doctrines that would block any damages action.

. The defendants have suggested that the "in the presence” requirement might be severed from the remainder of the disclosure and waiting period requirements. The apparent constitutional flaws in the medical emergency exception for the entire law call for an injunction against enforcement of Public Law 187 as a whole. If the Supreme Court of Indiana construes the medical emergency exception so as to eliminate those flaws, the issue of severability may need to be addressed at that time.