CUMMINGS, Chief Judge.

This case involves a suit brought under 42 U.S.C. § 1983 by plaintiff Thaddeus J. Malak, M.D. (“plaintiff”) against numerous defendants. Plaintiff appeals from the district court’s grant of summary judgment in favor of all the defendants. For the reasons set out below, we reverse.

The ensuing analysis will be clearer by classifying the defendants into two groups. One group is the “public group,” consisting of the Porter Memorial Hospital (“PMH”), its Board of Trustees, Arthur S. Malasto (“Malasto”), and Sharon Simon (“Simon”). The second group is the “private group,” consisting of Associated Physicians, Inc. (“API”), Martin J. O’Neill, M.D. (“O’Neill”), and Clark McClure, M.D. (“McClure”). PMH is a public hospital. Malasto is the head administrator of PMH, and functions as its chief executive officer. 1 Simon is the head nurse in PMH’s emergency room. API is an Indiana corporation which contracted with PMH to operate PMH’s emergency room and staff it with physicians. O’Neill is the president of API and McClure is its secretary. O’Neill and McClure are chairmen of the emergency department at PMH and medical directors of emergency services at PMH.

Plaintiff was associated with PMH and was employed by API for five years as an emergency-room staff physician. Plaintiff was given notice of his termination by API in April 1983, and PMH subsequently revoked his staff privileges at the hospital. Plaintiff claims that the private group of defendants conspired with the public group of defendants to terminate him because he publicly criticized the emergency-room conditions at PMH, the conspirators thereby violating both the First and Fourteenth Amendments of the United States Constitution. Plaintiff also alleged a second count of liability against only the public group of defendants. Defendants countered that plaintiff’s employment contract with API provided for termination without cause, *279 and that API alone simply exercised its right to terminate him.

The defendants filed various motions under Rule 12(b) of the Federal Rules of Civil Procedure. On August 16, 1983, O’Neill filed a motion under both Rule 12(b)(1) and Rule 12(b)(6) which argued that there was no subject matter jurisdiction under either 28 U.S.C. § 1331 or 28 U.S.C. § 1343(3) because there was no state action, and that there was no constitutional claim against O’Neill. On August 18, 1983, PMH and its Board of Trustees also filed a motion to dismiss under both Rule 12(b)(1) and Rule 12(b)(6). On that same date the district court held a hearing on plaintiff’s request for a preliminary injunction and denied it. At that time, the court heard argument as well on the issues presented in the motions to dismiss.. On February 10, 1984, O’Neill, now joined by API and McClure, filed a Rule 12(b)(1) motion which argued that state action is a prerequisite to jurisdiction, and that there was no state action in the instant ease.

On March 5, 1985, the district court granted summary judgment in favor of all defendants. Because the court had heard arguments from various parties and would look to facts outside the pleadings, the court treated the February 10,1984, motion to dismiss as a motion for summary judgment. In a terse sentence devoid of any elaboration, the district court stated that plaintiff failed “to show that any acts undertaken by the defendants constituted ‘state action’ ” (Plaintiff’s App. A-2).

Before reaching the issue of whether the conduct of any of the defendants constituted state action, some procedural knots created by the pleadings must be untangled. The district court treated the motion to dismiss as a motion for summary judgment, since the court considered matters outside the pleadings, and cited Fed.R. Civ.P. 12(c) as authority for doing so. This citation is in error; Fed.R.Cir.P. 12(c) allows the conversion of a motion for judgment on the pleadings into a motion for summary judgment, and no defendant ever moved for judgment on the pleadings. Instead, the relevant authority is Fed.R. Civ.P. 12(b), which allows the conversion of a Rule 12(b)(6) motion into a motion for summary judgment when the court considers matters outside the pleadings. The difficulty is that only O’Neill and PMH and its Board of Trustees made 12(b)(6) motions: API and McClure made only a 12(b)(1) motion, while Malasto and Simon did not make any type of 12(b) motion.

There are two responses to this dilemma, the first being a partial solution and the second a complete solution. The partial solution is that the type of 12(b)(1) motion made by API and McClure should be treated as a 12(b)(6) motion where the federal claim is not insubstantial and frivolous, which apparently is what the district court did implicitly in the instant case. API and McClure, in their 12(b)(1) motion, contend that state action is a jurisdictional prerequisite for a Section 1983 suit, and that since the conduct of the private defendants is state action only if they acted jointly with the public defendants, which they did not, plaintiff’s suit must fail for lack of jurisdiction. However, plaintiff’s basis for jurisdiction with respect to these private defendants under 28 U.S.C. § 1331 —state action via joint action with the public defendants — is also an element of plaintiff’s federal cause of action under Section 1983. In such a situation, where a challenge to the court’s jurisdiction is also a challenge to the existence of a federal cause of action, the Supreme Court has stated that jurisdiction cannot be defeated by the possibility that plaintiff may not have stated a cause of action. Bell v. Hood, 327 U.S. 678, 682, 66 S.Ct. 773, 776, 90 L.Ed. 939. Instead, the case should be dismissed for lack of jurisdiction only if the claim “clearly appears to be immaterial and made solely for the purpose of obtaining jurisdiction or where such a claim is wholly insubstantial and frivolous.” Bell, 327 *280 U.S. at 682-683, 66 S.Ct. at 776. 2 Additionally, if plaintiff meets this Bell test, the district court should take jurisdiction and handle defendants’ motion as a direct attack on the merits of plaintiff’s case. Williamson v. Tucker, 645 F.2d 404, 415 (5th Cir.1981), certiorari denied, 454 U.S. 897, 102 S.Ct. 396, 70 L.Ed.2d 212; Student Government Association of Wilberforce University v. Wilberforce University, 578 F.Supp. 935, 940 (S.D.Ohio 1983). As the Fifth Circuit explained in Williamson, this method of treating indirect attacks on the merits as 12(b)(6) motions rather than 12(b)(1) motions properly provides more protection to the plaintiff, since under 12(b)(1) the court may determine disputed facts which determine jurisdiction, whereas under 12(b)(6) the court must treat plaintiff's allegations as true, and in addition under Rule 56 must determine that no genuine issue of material fact exists. Williamson, 645 F.2d at 415. This higher level of protection for plaintiff is warranted since in this situation he is in reality facing a challenge to the merits of his claim. Id. at 415. By thus treating the 12(b)(1) motion made by the private defendants as an attack on the merits under 12(b)(6), the district court’s conversion of the motion into a motion for summary judgment is valid. See 5 Wright & Miller, Federal Practice and Procedure § 1366 (1969) (“The element that triggers the conversion is a challenge to the sufficiency of the pleader’s claim supported by extra-pleading material. It is not relevant how the defense actually is denominated.”).

The above analysis still does not help Malasto or Simon, neither of whom made a 12(b) motion of any type. However, where one defendant files a motion for summary judgment which the court grants, the district court may sua sponte enter summary judgment in favor of additional non-moving defendants if the motion raised by the first defendant is equally effective in barring the claim against the other defendants and the plaintiff had an adequate opportunity to argue in opposition to the motion. Macon v. Youngstown Sheet & Tube Co., 698 F.2d 858 (7th Cir.1983). If the district court concluded that there was no state action with respect to PMH and its Board of Trustees or the private defendants because the public defendants were not at all involved in the termination of plaintiff, then the same reasoning would justify a finding of no state action on the part of Malasto or Simon. This reasoning validates not only the district court’s grant of summary judgment with respect to Malasto and Simon, but also with respect to API and McClure, and thus provides a complete solution to both Malasto’s and Simon’s failure to file any motion as well as API’s and McClure’s failure to file a 12(b)(6) motion.

Even though the foregoing discussion demonstrates that the district court in effect had before it the requisite motions to grant summary judgment with respect to all the defendants, a second procedural difficulty remains: the district court never gave notice to the parties that it intended to convert the 12(b)(6) motions before it into motions for summary judgment, as required by Fed.R.Civ.P. 12(b). See 5 Wright & Miller, § 1366 at p. 683. We have previously held that although a district court certainly should give notice to *281 the parties when the court converts a 12(b)(6) motion into a motion for summary judgment, the failure to do so does not necessarily mandate reversal where nothing else could have been raised to alter the entry of summary judgment. Milwaukee Typo, Etc. v. Newspapers, Inc., 639 F.2d 386, 391 (7th Cir.1981), certiorari denied, 454 U.S. 838, 102 S.Ct. 144, 70 L.Ed.2d 119; Chicago-Midwest Meat Association v. City of Evanston, 589 F.2d 278, 282 (7th Cir. 1978), certiorari denied, 442 U.S. 946, 99 S.Ct. 2895, 61 L.Ed.2d 318. Although the cited cases involve a failure to give notice where the party against whom summary judgment was entered could not possibly allege any genuine issues of material fact, whereas in the instant case there are genuine issues of material fact, the reversible error in the instant case concerns the substance of the district court’s grant of summary judgment rather than its failure to give proper notice, since plaintiff raised below the relevant issues of material fact in his response to the defendants’ various motions. We thus turn to the merits of the state action issue.

Ill

The Federal Rules of Civil Procedure dictate that a grant of summary judgment is appropriate when “there is no genuine issue as to any material fact and ... the moving party is entitled to judgment as a matter of law.” Fed.R.Civ.P. 56(c); Black v. Henry Pratt Company, 778 F.2d 1278, 1281 (7th Cir.1985). “[I]n determining whether factual issues exist, a reviewing court must view all the evidence in the light most favorable to the non-moving party.” Black at 1281, quoting Collins v. American Optometric Association, 693 F.2d 636, 639 (7th Cir.1982). Applying this standard to the instant case, there remains a factual dispute over whether the conduct of the defendants constituted state action, and consequently the district court erred in granting the defendants’ motions for summary judgment.

The main thrust of plaintiff’s case is that the private defendants acted jointly with the public defendants in terminating plaintiff. The district court did not give any specific reasons for its conclusion that none of the acts undertaken by the defendants constituted state action, but the likeliest reason was that the district court believed that the public defendants played no role whatsoever in terminating plaintiff. However, plaintiff has certainly made a sufficient showing with respect to this claim to survive a motion for summary judgment. A private defendant acts “under color of” state law for purposes of Section 1983 when he is “a willful participant in joint action with the State or its agents.” Dennis v. Sparks, 449 U.S. 24, 27, 101 S.Ct. 183, 186, 66 L.Ed.2d 185; see also Tower v. Glover, 467 U.S. 914, 104 S.Ct. 2820, 2824, 81 L.Ed.2d 758 (“an otherwise private person acts ‘under color of’ state law when engaged in a conspiracy with state officials to deprive another of federal rights”); Adickes v. Kress & Co., 398 U.S. 144, 152, 90 S.Ct. 1598, 1605, 26 L.Ed.2d 142; Lusby v. T.G. & Y. Stores, Inc., 749 F.2d 1423, 1430 (10th Cir.1984). PMH is a public hospital, 3 and Malasto’s conduct as an official of a public hospital constitutes state action. Lopez v. Vanderwater, 620 F.2d 1229, 1236 (7th Cir.1980), certiorari dismissed, 449 U.S. 1028, 101 S.Ct. 601, 66 L.Ed.2d 491 (“Action taken by a state official who is cloaked with official power and who purports to be acting under color of official right is state action and is taken under color of state law whether or not the action is in fact in excess of the authority actually delegated to the official under state law.”). Dr. Ronneau of PMH states in his affidavit that he overheard defendant McClure say that the decision to terminate plaintiff was made by both API and PMH (Plaintiff’s App.A-4). Moreover, plaintiff states that Malasio and O’Neill conspired to make it appear that plaintiff *282 was competing with API for PMH’s emergency-room contract in order to make his termination appear legitimate, and PMH Nurse Cooley’s testimony supports both that Malak and O’Neill jointly acted to terminate plaitniff and that plaintiff was being set up to appear to be a competitor of API. Whether or not the above is sufficient for plaintiff to win his claim remains to be seen, but it is certainly sufficient to survive a summary judgment motion claiming a lack of state action.

Relying on the order issued by the district court that denied plaintiff's motion for a preliminary injunction for want of state action or action under color of law (R. Item 47 p. 9), defendants contend that plaintiff’s proffered evidence was insufficient to show the kind of conspiratorial conduct shown in Dennis, and moreover that the case of Rendell-Baker v. Kohn, 457 U.S. 830, 102 S.Ct. 2764, 73 L.Ed.2d 418, is more relevant to this case. We disagree. The issue in Rendell-Baker was whether the conduct of a private school which derives its income from public sources and which is regulated by public authorities was state action, Rendell-Baker, 457 U.S. at 831, 102 S.Ct. at 2766, and the Court held that it was not. By contrast, in the instant case the conduct of a public hospital and its employees is clearly state action and the conduct of otherwise private entities that act jointly with them is also state action. The fact that Dennis involved judicial immunity does not detract at all from this conclusion. Dennis teaches that the conduct of a private defendant who conspires with a public defendant is state action even though the public defendant (in that case a judge) is immune from damages liability, Dennis, 449 U.S. at 28, 101 S.Ct. at 186, so that clearly the conduct of a private defendant remains state action when the public defendants are not immune from damage liability. Moreover, this joint action theory has roots that extend far deeper than the single case of Dennis. Tower, 104 S.Ct. at 2824; Adickes, 398 U.S. at 152, 90 S.Ct. at 1605.

Both groups of defendants contend that the conduct of PMH’s Malasto and Simon is not state action and cite the case of Place v. Shepherd, 446 F.2d 1239, 1244-1245 (6th Cir.1971), to support their contention. Although Place states that the termination of a registered nurse in a public hospital by the head nurse in that hospital is not state action, reasoning that nothing done by the head nurse was dependent upon or enhanced by any state power, Place, 446 F.2d at 1244-1245, such is not the law generally or in this Circuit. See our Lopez opinion, supra p. 281. The Sixth Circuit assumed that a city official is liable only if her conduct arose out of a “statute, ordinance, regulation, custom or usage of the city,” Place, 446 F.2d at 1245; yet this argument was rejected by the landmark case of Monroe v. Pape which held that a person acts “under color of” state law even though his acts were not authorized by the state or indeed were even forbidden by the state. Monroe v. Pape, 365 U.S. 167, 172, 187, 81 S.Ct. 473, 476, 484, 5 L.Ed.2d 492; see generally Wright, The Law of Federal Courts 120 (4th ed.1983). The public defendants further cite numerous additional cases to support their contention, but the issue in those cases, as in the previously cited case of Rendell-Baker, is whether a privately owned entity acts jointly with the state, serves a public function, or has a sufficiently close nexus with the state because of financial support from the state or state regulation, so that the entity’s conduct or that of its officials is state action. See, e.g., Musso v. Suriano, 586 F.2d 59 (7th Cir.1978), certiorari denied, 440 U.S. 971, 99 S.Ct. 1534, 59 L.Ed.2d 788. Those cases are simply irrelevant to the instant case since PMH is publicly owned.

PMH contends that the district court implicitly decided that the hospital and its Board of Trustees were not liable because plaintiff’s theory of liability against the hospital rested on the doctrine of respondeat superior, and this doctrine does not apply to Section 1983 liability. It is true that the doctrine of respondeat superior liability is not recognized in Section 1983 cases. Monell v. Department of Social Services of the City of New York, *283 436 U.S. 658, 98 S.Ct. 2018, 56 L.Ed.2d 611. However, PMH’s contention is in error for several reasons. First, the district court based its conclusion on the lack of state action (Plaintiffs App.A-2), whereas PMH’s respondeat superior argument and the Monell case involve the nature of the duty owed and the degree of causation required in a Section 1983 case against a governmental entity rather than the state action requirement of Section 1983. Monell, 436 U.S. at 690-695, 98 S.Ct. at 2035-2038; Sheldon H. Nahmod, Civil Rights & Civil Liberty Litigation § 6.06 (1979). PMH’s argument would also not explain the grant of summary judgment with respect to the remaining defendants.

Second, plaintiff does not allege that PMH is liable because of the doctrine of respondeat superior. As noted, Dr. Ronneau’s affidavit states that PMH itself, and not its employees, helped to terminate plaintiff. Additionally, defendant Malasto, who allegedly played an integral role in the termination of plaintiff, functions as the chief executive officer of the hospital (see supra, note 1), and the acts of a high-ranking official of a governmental entity may render the entity liable under Section 1983 where the official is a policymaker whose conduct is in effect that of the govermental entity. City of Oklahoma City v. Tuttle, - U.S. -, 105 S.Ct. 2427, 2436, 85 L.Ed.2d 791 (1985); Monell, 436 U.S. at 694, 98 S.Ct. at 2037 (governmental entity is liable under Section 1983 when a person “whose edicts or acts may fairly be said to represent official policy” causes the injury); Soderbeck v. Burnett County, Wisconsin, 752 F.2d 285, 292 (7th Cir.1985), certiorari denied, — U.S. -, 105 S.Ct. 2360, 86 L.Ed.2d 261 (county is liable for the acts of its policymakers); Reed v. Village of Shorewood, 704 F.2d 943, 953 (7th Cir.1983) (“The official acts of municipal policymakers are acts of the municipality for the purposes of section 1983 liability”); Smith v. Stoner, 594 F.Supp. 1091, 1113 (N.D.Ind. 1984); Granberg v. Ashland County, 590 F.Supp. 1005, 1010-1011 (W.D.Wis.1984). Other circuits faced with cases similar to the instant case have held that a municipal hospital is liable for the acts of its officials whose conduct establishes the policy of the hospital. Vinyard v. King, 728 F.2d 428 (10th Cir.1984); Rookard v. Health & Hospitals Corp., 710 F.2d 41, 45 (2d Cir.1983). The Rookard case is strikingly similar to the instant case: in that case the Second Circuit ruled that a former nurse at a municipal hospital stated a valid claim under Section 1983 against the municipal corporation which operates New York City’s municipal hospitals by alleging that the executive director had ordered her transferred because she spoke out about practices at the hospital. Rookard, 710 F.2d at 45.

The doctrine that the acts of a policymaker can render a governmental entity liable is well supported by the reasoning of Monell. Monell teaches that there should be no liability under Section 1983 unless a . defendant caused plaintiff’s alleged injury, and a governmental entity does not cause an injury solely because an employee’s conduct caused the injury. Tuttle, 105 S.Ct. at 2433. A common example of this principle is when a citizen brings a Section 1983 suit against both a police officer who unconstitutionally injured him and the city that employs the police officer. The city will not be liable under Section 1983 merely because a police officer unconstitutionally injured a citizen; such an isolated incident by a police officer does not show that a policy of the city caused the injury. See, e.g., Strauss v. City of Chicago, 760 F.2d 765, 767 (7th Cir.1985). On the other hand, a governmental entity can act only through staff, and at a sufficiently high level the acts of officials are the acts of the governmental entity. The courts, including this one, have correctly concluded that when the conduct of a policymaker causes the alleged injury, that conduct may be fairly attributed to the governmental entity. Soderbeck, 752 F.2d at 292.

Although one district court in this Circuit has stated that the single act of a high-level policymaker cannot render a governmental entity liable under Section 1983 (Perry v. Larson, 599 F.Supp. 727, 731 (E.D.Wis.1984)), we disagree. Where the official is in a sufficiently high position *284 such that his conduct is that of the governmental entity, then even a single incident can render the entity liable under Section 1983. See Tuttle, 105 S.Ct. at 2435 (Monell “was intended to prevent the imposition of municipal liability ... where no wrong could be ascribed to municipal decisionmakers”). Moreover, Perry relied on Bell v. City of Milwaukee, 746 F.2d 1205 (7th Cir. 1984), but Bell does not conflict with our position here. Bell stated that a municipality cannot be liable because of a single incident of unconstitutional conduct of a district attorney, and that there was “no evidence linking [the conduct of the district attorney] with the policy of County officials. ” Bell at 1272 (emphasis supplied). While a district attorney may not be a policymaker for a county, a chief executive officer of a governmental hospital is surely a policymaker for that hospital.

For the above reasons, the summary judgment for defendants is reversed and the cause is remanded for further proceedings. Plaintiff should be granted further discovery in connection with his First Amendment claim if he deems it necessary.

. In its application to the State of Indiana Board of Health for a license to operate a hospital, in response to a question which asked for the identity of the chief executive officer of the hospital, PMH answered Arthur S. Malasto (Plaintiff's Exhibit 4).

. Although Bell teaches that the existence of a meritorious claim is not required to invoke jurisdiction under 28 U.S.C. § 1331, 28 U.S.C. § 1343(3) is an exception to the Bell analysis, and under the latter jurisdictional statute a court must first determine if a Section 1983 claim exists before determining if the court has jurisdiction. Mt. Healthy City Board of Education v. Doyle, 429 U.S. 274, 278-279, 97 S.Ct. 568, 571-572, 50 L.Ed.2d 471; Haldorson v. Blair, 449 F.Supp. 1025, 1027 (D.Minn.1978). This fact was often of crucial importance prior to 1980, since at that time the statute granting jurisdiction for federal questions, 28 U.S.C. § 1331, had an amount in controversy requirement of $10,000, and therefore a Section 1983 litigant with less than this amount in controversy had to assert jurisdiction under 28 U.S.C. § 1343(3). Congress, however, amended 28 U.S.C. § 1331 in 1980 so that it no longer contains this amount in controversy requirement, and thus a Section 1983 litigant such as plaintiff who meets the Bell standard articulated above can assert federal jurisdiction under 28 U.S.C. § 1331 without a prior determination that his claim is meritorious, regardless of the amount in controversy.

. In its application to the State of Indiana Board of Health for a license to operate a hospital, PMH checked "governmental" in response to the question "type of ownership” (Plaintiffs Exhibit 4).