dissenting.
I have no disagreement with Judge Cow-en or Judge Sloviter concerning the general rule governing the statute of limitations for medical malpractice under the Kubrick standard — the two year statute of limitations begins to run when a reasonable person should have known of the injury and its cause. Moreover, I freely concede that, generally speaking, the jurisprudence construing the statute of limitations under the Federal Tort Claims Act (FTCA) is not very pliable. The caselaw does not admit of tolling the limitations period by reason of infancy or mental disability or a number of other individualized traits that could reasonably interfere with the two events that, as I have noted, start the statute running: discovery of the injury and discovery of its cause. I part company with my colleagues, however, in their assessment that this is another in a line of sorry cases in which we have no power to circumvent the harsh dictates of the statute of limitations under the FTCA.
I believe that the majority has overstated and misapplied the Kubrick standard. I am persuaded that this case falls within a narrow, yet recognized equitable exception to the Kubrick general rule. Under this exception, the statute of limitations under the FTCA does not accrue for a plaintiff whose ability to perceive that the government injured him was destroyed by the government’s negligent care until the plaintiff is affirmatively informed of the injury in a way he can understand, or until a guardian is appointed. The government concedes that Barren was unable to perceive his injury and that the government’s negligent treatment of Barren aggravated his condition and contributed to that inability. And yet, in extending the Kubrick rule to the facts of this case, the majority has ignored two key factors that set Barren’s case apart: 1) that this is an aggravation case where determining what a reasonable person would know is inherently very difficult; and 2) the government itself is responsible for Barren's inability to perceive his own mental deterioration.
I.
A. Aggravation
It is essential, at the threshold, to identify and discuss the precise nature of Barren’s injury. His injury was the worsening of his mental condition caused by his negligent treatment by the VA, including its failure to diagnose his condition, mispres-cription of medication, counterproductive advice to Barren and his family, and failure to admit Barren for in-hospital observation and diagnosis.
Barren’s mental illness began with his tour of duty in Korea. His mental problems became worse because, as the district court noted, “the defendant made no meaningful effort to diagnose Barren’s condition or to treat it in the proper manner.” J.A. at 51. Dr. Anghel, who was not board certified in psychiatry and was practicing for the first time without supervision, provided substandard care. Despite the pleas of Barren’s mother and sister, Barren was fed a heavy diet of tranquilizers and pep-talks, and was assured that he needed only to “grow up.” Barren’s medication was not monitored, and Dr. Anghel dismissed as “nonsense” many of Barren’s serious subjective complaints (e.g., anxiety, sleeplessness and crying fits).
Because the actual injury was the further deterioration of Barren’s condition from mentally ill but nevertheless functioning, to totally incapacitated because of mental illness, the Kubrick standard is very difficult to apply. See Augustine v. United States, 704 F.2d 1074, 1078 (9th Cir.1983) (“[w]hen a physician’s failure to diagnose, treat, or warn a patient results in the development of a more serious medical *996problem than that which previously existed, identification of both the injury and its cause may be more difficult for a patient than if affirmative conduct by a doctor inflicts a new injury”); Greenberg v. McCabe, 453 F.Supp. 765, 772 (E.D.Pa.1978), aff'd without op., 594 F.2d 854 (3d Cir.), cert. denied, 444 U.S. 840, 100 S.Ct. 78, 62 L.Ed.2d 51 (1979) (“where the injury and cause thereof are subtler and more complicated than in the normal malpractice case, it seems particularly inappropriate to determine as a matter of law what the plaintiff should have known”) (footnote omitted).
Judge Cowen makes much of the fact that Barren left the YA to seek outside help, and contends that this fact provides evidence that Barren knew or at least should have known of his injury and its cause. See maj. at 991. But the mere switching of physicians does not necessarily impute knowledge of injury and the source of that injury. See Nicolazzo v. United States, 786 F.2d 454 (1st Cir.1986) (distinguishing the facts of Kubrick and holding that plaintiffs awareness that YA physicians were not aiding him did not constitute knowledge of injury and its cause). Furthermore, in applying the Kubrick standard, Judge Cowen does not account at all for0 the subtleties of an aggravation case. The notion that changing doctors does not necessarily establish discovery of negligence seems particularly true in an aggravation case, where the plaintiff already knows that he is sick and may not have any means for differentiating negligent care from the natural progression of the disease. Therefore, even under the rigid reasonable person standard that the majority adopts, I do not think it can confidently be said that Barren should have known of his injury.
B. The Government’s Responsibility
Second, and more important, I do not think the Kubrick standard should apply at all under the special facts of this case, where, as the majority opinion concedes, “the VA’s malpractice was a substantial factor in Barren’s inability to recognize that very malpractice.” Maj. at 991. This case arrives to us in an unusual and compelling posture: because the district court held a trial on the merits before determining the statute of limitations questions, we are not dealing with alleged negligence but rather with proven negligence by the VA, which the government does- not challenge on appeal. The district court accepted the plaintiff’s argument “that John Barren was at all times during [the relevant] period unable to comprehend that his mental condition had deteriorated while he was under the care of the V.A. medical personnel and that the deterioration was caused by the treatment received at the hands of the V.A.” J.A. at 29. There is thus no question that the poor care Barren received exacerbated his mental illness, directly contributed to his present mental incapacity, and most important, prevented him from understanding the nature or source of his injury.
Basic principles of equity dictate that the government cannot profit from its own negligence. At bottom my reasoning derives from the simple and compelling precept that the government cannot render someone unable to comprehend his injury and then impose a reasonable person discovery rule. The majority’s emphasis upon the objective-subjective distinction in Kubrick is simply misplaced in this context. It ignores the added special factor of the government’s participation in Barren’s inability to comprehend his injury and its cause. In recognizing the unique and pernicious effect of the government’s malpractice, we do no harm to the basic Kubrick discovery rule. Rather, we carve a narrow exception that prohibits the government from insulating itself from its own wrongful activity.
Barren’s special situation strikes me as significantly distinguishable from mere mental incapacity which does not, in and of itself, toll the statute of limitations under the FTCA. The added factor here of the government’s causation of that mental incapacity changes the rules of the game. I agree with Judge Bramwell in Dundon v. United States, 559 F.Supp. 469 (E.D.N.Y.1983), where he considered the govern*997ment’s argument based on a strict interpretation of the Kubrick rule and observed that
[t]o treat this case as one involving mere mental incompetency in the general sense proffered by the government is to lose sight of the decedent’s extraordinary situation. The decedent’s mental condition, allegedly caused by his treating physicians, directly prevented his understanding the nature and cause of his injuries.
Id. at 474.
Other courts have managed to circumvent the obvious injustice of the result the lead opinion and concurrence reluctantly advance. Two lines of cases support the notion that where the government itself is responsible for the plaintiff’s admitted inability to perceive the worsening of his condition, the suit will not be time barred. The first consists of coma cases, where the plaintiff’s lack of consciousness is directly attributable to the government’s negligence.1 The second involves malpractice by therapists who violate the transference relationship by engaging in sexual relations with their patients thereby causing severe mental distress and aggravating a pre-exis-tent mental illness. I discuss them in turn,
1. The Coma Cases
In Clifford by Clifford v. United States, 738 F.2d 977 (8th Cir.1984), a comatose patient represented by his guardian-father sued the United States for malpractice stemming from a drug overdose. The court held that the statute of limitations accrued when his father was appointed as guardian, and not, as the government contended, when the patient fell into the coma. Judge Arnold, speaking for the court, analogized Clifford’s case to Zeidler v. United States, 601 F.2d 527 (10th Cir.1979), where the statute of limitations did not run for someone who was lobotomized by the government until a conservator was appointed on his behalf. As Judge Arnold explained, in both cases the government had “destroyed those plaintiffs’ capacities to realize the existence and cause of their injuries.” Clifford, 738 F.2d at 979. Essentially, the court in Clifford held that as a matter of fairness the statute of limitations must be tolled lest the government “profit from its own (alleged) wrong.” Id. at 980. The court specifically held that because the patient had reached the age of majority and had not yet been declared legally incompetent, the knowledge of his family members could not be imputed to the patient himself. Furthermore, the court observed that tolling the statute of limitations in the case of a coma “does not disturb in any way [the] well-recognized rules” that ordinarily the statute of limitations is not tolled for infancy or mental incapacity and that its holding was fully consonant with Kubrick. Id. at 980.
Similarly, in Washington v. United States, 769 F.2d 1436 (9th Cir.1985), the court, citing Clifford, held that the cause of action accrued when the comatose plaintiff finally died, not when she entered her fourteen-yéar coma. The court reasoned that the plaintiff satisfied the rule because she was never aware of her injury or its cause. Id. at 1439. The court specifically held that the knowledge of plaintiff’s husband was irrelevant to her ability to bring suit. Id. at 1438. See also Dundon v. United States, 559 F.Supp. 469 (E.D.N.Y.1983) (tolling the statute of limitations for medical malpractice under the FTCA where government negligently sent plaintiff into a coma).2
*9982. The Transference Cases
The second line of cases involves malpractice by a mental health therapist who by engaging in sexual relations violated the transference relationship with his patient.3 In Simmons v. United States, 805 F.2d 1363 (9th Cir.1986), the Court of Appeals for the Ninth Circuit held that the statute of limitations had not run on a patient who sued her government sponsored counsellor for malpractice for wrongfully engaging her in sexual relations during the course of her treatment. The court held that the two year statute of limitations under the FTCA only began to run once she had been informed by another therapist that the source of her present injury (deep depression, attempted suicide) stemmed from her meretricious relationship with her former therapist. Id. at 1367. The court discussed the nature of the transference phenomenon at great length, explaining that given the special patient-therapist relationship, the court could not hold that the plaintiff reasonably should have known of her injuries any earlier. Id. The high degree of dependence and trust of the patient on her therapist, it was held, prevented a patient from understanding the nature of her injury or its cause.
In Greenberg v. McCabe, 453 F.Supp. 765 (E.D.Pa.1978), aff'd without op:, 594 F.2d 854 (3d Cir.), cert. denied, 444 U.S. 840, 100 S.Ct. 78, 62 L.Ed.2d 51 (1979), Judge Joseph S. Lord, III considered a similar question under Pennsylvania’s law of psychiatric malpractice. Although this case preceded Kubrick, and is not an FTCA case, it applied the same standard as the Kubrick discovery rule. The court inquired into when the plaintiff, in exercise of reasonable diligence, should have discovered her injury and its cause. 453 F.Supp. at 768. It noted the deep dependence of the plaintiff on her therapist and the assurances by the therapist that their sexual encounters were part of her therapy. In rejecting defendant’s motion for a judgment n.o.v., Judge Lord explained that the jury could have inferred that the extreme dependence of the plaintiff on her therapist and the therapist’s improper behavior impeded the plaintiffs “powers of judgment” and that she “neither knew nor by objective standards could have known under those circumstances” the nature of her injury or its cause. Id. at 771-72. The court explained: “At issue is the objective effect of the defendant’s [conduct] on discoverability by a reasonable person.... [T]he statutory period does not begin to run if ... the plaintiff’s failure of discovery, objectively determined, is brought about by the very nature of the defendant’s conduct.” Id. at 769. See also Person v. Kieffer, 634 F.Supp. 892 (E.D.Pa.1986) (evidence that *999attack by police dog may have caused psychological difficulties impeding plaintiff’s memory tolled applicable two-year Pennsylvania tort statute for suits against police officers).
The nature of the relationship and the transference phenomenon blinded the patients in Simmons and Greenberg from comprehending the true nature of their injuries and its source. These cases are very similar to the coma cases in that the negligence itself (sleeping with a patient) was insulated from the patient’s detection by the wrongdoer. Because of the special relationship, which the therapist abused, the patients did not question the wrongful activity or recognize it as the source of their injuries.
The government attempts to distinguish the transference cases. It claims that “Barren was never in a therapist-patient relationship with the YA’s psychiatrist, and, therefore, the conditions for unquestioning trust were not present.” Brief of Appellant at 21 n. 11. The government’s attempted distinction misses the point. It essentially argues that Barren’s case is distinguishable because he complains of too little psychiatric assistance, not too much. The analogy to the transference abuse cases does not derive from the fact that both cases involve therapists, but rather that both cases involve wrongdoers who are insulated from suit because the patient cannot reasonably be expected to understand the nature or the source of his or her injuries. In the case of transference abuse, the therapist shields himself from the patient’s discovery of the wrong by taking advantage of the special, dependent relationship to cover up the injury he is inflicting. The therapist assures the patient that the sexual encounter is an important part of the therapy thus insulating his harmful conduct from detection as the source of the patient’s worsened mental health. In Barren’s case, the progression of Barren’s mental illness under Dr. An-ghel’s care prevented Barren from discovering his injury.
3. Barren’s Case
I have discussed the above precedents in detail to demonstrate that my objection to the holding in this case stems from more than personal distaste for the obvious unfairness of the result, a sentiment shared by the majority. I find no principled way to distinguish Barren’s case from the cases discussed above. In both lines of cases the plaintiff is unable to detect the injury and its source through no omission of his own, but rather because of the malfeasance of the wrongdoer himself. Admittedly, a coma seems more dramatic, and perhaps is easier to understand. I must stress, however, that Barren's case is analytically no different. Clear, unchallenged evidence indicates that, because of the government’s negligence, Barren was unable to perceive the worsening of his mental condition. For the purposes of recognizing his injury, he was akin to a comatose patient (Clifford, Washington), or a patient whose therapist has violated the transference relationship (Simmons, Greenberg).
The government claims that the rule I espouse is “unworkable as a practical matter,” because it “does not indicate how long accrual should be delayed.” Appellant’s Br. at 22. I have four responses to the concern that this narrow rule might engender a case with no foreseeable statutory time limit.4 First, this is obviously a narrow rule. In adding Barren’s case to the short list of situations, like comas and transference abuse, where the patient cannot reasonably be expected to discover the injury and its cause because of the very negligence, we do not open the courthouse doors to a floodgate of stale claims. See Dundon, 559 F.Supp. 469, 475 (“[t]he exception is narrow and merely prevents ‘blameless ignorance’ from being penalized, by avoiding the anomalous result of having an arguably wronged comatose patient de*1000nied his right to press a claim by virtue of the very malpractice of which he seeks to complain”).
Second, as other courts that have considered this same question have observed, the passage of time handicaps the interests of both sides, and a plaintiff who brings suit after too long a period will have trouble adducing evidence and proving his claim. For example, the Ninth Circuit noted in Washington that “[t]he [plaintiffs], as well as the government, share this increased burden caused by the passage of time, and the [plaintiffs] should not be prevented from maintaining this action when the government was responsible for the condition which caused the delay.” 769 F.2d at 1439 (citations omitted). Similarly, Judge Arnold noted that
[t]he government’s best argument is that the rule advocated by plaintiff might leave it open to suit indefinitely. No doubt that is true, at least in theory.... Probably the real exposure of the government to liability would be slight in such cases, though. The passage of time should make it progressively more difficult for plaintiff to prove his case.
Clifford, 738 F.2d at 980.
Third, anyone conducting the balance in this case would at once note that the government would not have to wait for the claim to accrue for years on end. At the very latest, the statute of limitations would have begun to run when John Barren was deemed mentally incompetent and his sister was appointed as guardian. See Clifford, 738 F.2d at 980 (statute of limitations did not begin to run until guardian was appointed); Zeidler v. United States, 601 F.2d 527 (10th Cir.1979) (statute of limitations did not begin to run until conservator appointed). I believe that the doctrine of laches can be relied upon to balance the rights of the plaintiff against the need of the government to know when it will be sued. Laches has historically been an effective barrier to such claims not subject, for one reason or another, to a statutory time bar. Here the doctrine of laches would not appear to bar the plaintiff's claim.
Finally, and most important, even if we find that the government would be exposed to additional liability, I believe we must balance the harm to the government against the intolerable effect of the majority’s holding, which allows the government to benefit from its own negligence. See Clifford, 738 F.2d at 980 (“[W]e are persuaded that the rule contended for by the government would be still more objectionable. For under it the government would profit from its own (alleged) wrong.”).
At bottom, I believe that the Kubrick rule simply cannot apply to this case. As I have stressed above, I advocate an exception to that rule not because of Barren’s mental incompetency simpliciter, but because of the government’s conceded participation in Barren’s inability to perceive his injury. Judges Cowen and Sloviter emphasize that the test cannot be subjective. For this proposition they recite black letter law, but they cite no cases that deal with the unusual circumstances of this case. Nor do they directly confront the coma or transference lines of cases. In sum, I find the Kubrick rule tragically inappropriate under the circumstances. Instead, I advocate following the wise jurisprudence of courts that have carved a narrow equitable exception to the reasonable person standard of Kubrick where the government itself has fostered the plaintiff’s incapacity to comprehend the nature and source of his injury.5
II.
Finally, in view of my conclusion above, I address the question that the majority and concurrence do not reach concerning the *1001Pennsylvania immunity statutes. See maj. at 992 n. 8. I find these to be vexing questions and I hesitate to pronounce on them here in the depth that otherwise would be justified given my dissenting posture. Nevertheless, at least to indicate that my vigorous opposition to the holding of this case is not a vain exercise, I will explain briefly why I do not believe that the government is entitled to full immunity under the Pennsylvania statutes.
The government argues that it is entitled to all immunity available to any individual or government body. For this proposition it cites General Electric Co. v. United States, 813 F.2d 1273 (4th Cir.1987), vacated on other grounds, — U.S. — , 108 S.Ct. 743, 98 L.Ed.2d 756 (1988), holding the United States immune from tort suit because a Maryland employer would be immune from tort suit under Maryland’s workers’ compensation law. Proceeding from this premise, the government claims immunity pursuant to 50 Pa.Stat. § 4603, which was effective when Dr. Anghel negligently treated Barren but was repealed in 1976.6 The statute provides that
No person and no governmental or recognized nonprofit health or welfare organization or agency shall be held civilly or criminally liable for any diagnosis, opinion, report or any thing done pursuant to the provisions of this act if he acted in good faith and not falsely, corruptly, maliciously or without reasonable cause; provided, however, that causes of action based upon gross negligence or incompetence shall not be affected by the immunities granted by this section.
Mental Health and Mental Retardation Act of 1966, § 603, 50 Pa.Stat. § 4603 (1969), repealed by 1978 P.L. 1399 No. 330, § 802.
I note that two courts in Pennsylvania have held that this provision applies only to protect admission decisions. See Saunders v. Latrobe Area Hospital, Inc., 14 Pa. D. & C. 3d 458, 461-62 (Arbitration Panel for Health Care), permission to appeal denied, 14 Pa. D. & C. 3d 466 (Pa.Commw.1980); Hanczar v. Trellis, 14 Pa. D. & C. 3d 466, 468-69 (Arbitration Panel for Health Care), permission to appeal denied, 14 Pa. D. & C. 3d 470 (Pa.Commw.1980). It seems likely to me that the Supreme Court of Pennsylvania would similarly decide that the statute applies only to admission decisions. The statute would not, therefore, insulate the negligent treatment of Dr. Anghel {e.g., improper drug dispensation, detrimental advice), though it would protect the failure to admit Barren for in-hospital psychological evaluation.
For the reasons set forth in Part I, I would affirm the district court’s holding on the statute of limitations question. For the reasons briefly set forth in this section, I would remand for the district court to decide two matters. First, I believe that the district court should consider the question whether Dr. Anghel’s treatment rose to the level of gross negligence or incompetence, as alleged in Barren’s complaint. It is clear from the language of the statute that if Dr. Anghel committed gross negligence, or was deemed incompetent, the statute would not bar recovery at all. See Freach v. Commonwealth, 471 Pa. 558, 370 A.2d 1163 (1977). Second, under my view, even if the statute applies, the district court must recalculate the damage award. I believe that under Pennsylvania law Barren is entitled to receive at least some of his award — that attributable to the negligent care he received — even if the district court ultimately must reduce the damages to immunize the government from its negligence in failing to admit Barren for observation and diagnosis.
I respectfully dissent.
. Judge Sloviter in her concurrence correctly notes that the parties have agreed (as did the district court) that these coma cases do not apply. Barren, of course, was successful below and, I presume, did not wish to challenge on appeal the district court’s reasoning, given that its decision was in his favor. However, the coma cases are before us both in the district court opinion and the government’s brief. Because they materially enhance the analysis and discussion of this very troubling question, Barren's concessions cannot foreclose their consideration.
. In her concurrence, Judge Sloviter contends that the coma cases themselves distinguished the extraordinary situation of a coma and lobotomy from some lesser level of mental incapacity. Concurring Op. at n. 2. I make two responses. First, the two cases upon which I rely most heavily, Washington and Clifford, make no such distinction. In particular, I disagree with *998Judge Sloviter's characterization of the Clifford case. Although it is true that the case takes note of Allen Clifford's "extraordinary situation,” it seems clear to me that, in quoting the language of Dundon, the Clifford case perceives the "extraordinary situation” to be the government’s causation of the plaintiffs inability to perceive his injury. The Clifford case explains:
And, while mental incompetence generally does not toll the statute of limitations,
[t]o treat this case as one involving mere mental incompetence in the general sense proffered by the government is to lose sight of [Allen’s] extraordinary situation. [His] mental condition, allegedly caused by his treating physicians, directly prevented his understanding the nature and cause of his injuries. Dundon, supra, 559 F.Supp. at 474. (Emphasis supplied).
Clifford, 738 F.2d at 980. This interpretation is bolstered by the fact that the Clifford case supplied the emphasis to this quote from Dundon, thereby focusing on the question of the government’s causation.
Second, to the extent that Zeidler and Dundon do note their extraordinary facts, the reasoning derived from those cases is nevertheless directly on point where, as here, the government concedes that the plaintiff was unable to detect his mental deterioration because of the government’s negligence.
. As the court explained in Simmons v. United States, 805 F.2d 1363 (9th Cir.1986):
Transference is the term used by psychiatrists and psychologists to denote a patient’s emotional reaction to a therapist and is "generally applied to the projection of feelings, thoughts and wishes onto the analyst, who has come to represent some person from the patient’s past.”
Id. at 1364 (quoting Stedman’s Medical Dictionary 1473 (5th Lawyers' Ed.1982)). When a therapist abuses the transference relationship and becomes sexually involved with a patient the result is as devastating psychologically as an incestuous relation. 805 F.2d at 1365.
. Additionally, I note that the very nature of the Kubrick discovery rule indicates that in some cases the government will have to wait a long time after its original tort before it is sued. Furthermore, other factors, such as continuous treatment, often result in the de facto tolling of the limitations period. See, e.g., Otto v. National Institute of Health, 815 F.2d 985 (4th Cir.1987).
. Judge Sloviter suggests that the position I advocate would "change the statute." Concurring Op. at 994. I disagree. The statute requires that "[a] tort claim against the United States shall be forever barred unless it is presented in writing to the appropriate Federal agency within two years after such claim accrues." 28 U.S.C. § 2401(b). The determination of when a medical malpractice claim accrues, however, is an inherently judicial inquiry. As discussed above, other courts of appeals have not found themselves to be powerless in circumstances analogous to Barren’s case, and for the reasons I have set forth I think we should follow their lead.
. The government also claims immunity pursuant to 50 Pa.Stat.Ann. § 7114 (Purdon Supp. 1978). However, I discuss only the first statute because I believe the second one, 50 Pa.Stat. § 7114, does not apply. Section 7114 was passed after the negligence in this case had already occurred. See 1 Pa. Const.Stat.Ann. § 1926 (Purdon Supp.1987) (effective 1972)
("No statute shall [be] construed to be retroactive unless clearly and manifestly so intended by the General Assembly.”). See Hanczar v. Trellis, 14 Pa. D. & C. 3d 466, 468 (Arbitration Panel for Health Care), permission to appeal denied, 14 Pa. D. & C. 3d 470 (Pa. Commw. 1980).