Sandra J. Hartke v. Dr. William McKelway Sandra J. Hartke v. Dr. William McKelway

Opinion for the Court filed by Senior Circuit Judge McGOWAN.

These cross-appeals arise out of the failure of an operation to sterilize the plaintiff, resulting in the birth of a healthy baby girl. The primary issue for our consideration in this diversity case is whether, under District of Columbia law, the plaintiff would be allowed to recover from the defendant doctor some portion of the expenses of raising the child to majority. We are also called upon to decide whether, in the light of certain language in our opinion in Henderson v. Milobsky, 595 F.2d 654, 657-58 (D.C.Cir.1978), patients in informed-consent cases must testify that they would not have undergone the procedure had they known of all the material risks involved. Finally, we must decide whether the risk of subsequent pregnancy in this ease could reasonably have been considered material to a decision whether to undergo the treatment.

After a jury verdict for plaintiff on all claims, the District Court disallowed the award of childrearing expenses because it found the evidence clear that plaintiff had sought to be sterilized for therapeutic, not economic, reasons, and because she prized the child she bore. Hartke v. McKelway, 526 F.Supp. 97, 105 (D.D.C.1981). The court also held that in haec verba testimony of causation is not required to get the issue of informed consent to the jury as long as there is otherwise sufficient evidence from which the jury could infer that a patient would have declined the procedure had he or she known of the risks. Id. at 103. *1547 Finding such evidence, the court upheld the jury’s award of damages for plaintiff’s medical expenses, pain, suffering, and mental anguish resulting from the pregnancy and childbirth. Id. The court also held that a jury could have found that a reasonable person in plaintiff’s somewhat unusual position would likely have attached significance to the undisclosed risk of subsequent pregnancy here (one to three out of one thoúsand). Id. at 102-03. We affirm.

I

In the winter of 1978, Sandra J. Hartke, the plaintiff in this action, discovered she was pregnant and elected to have an abortion. For reasons that are crucial to the disposition of this appeal, and that are discussed at length below, she also sought to have herself sterilized. Her usual doctor recommended a hysterectomy, the complete removal of the uterus. Hartke, then 33, thought this a rather drastic procedure, so she approached the defendant, Dr. William McKelway, for a second opinion. Dr. MeKelway recommended a procedure known as laparoscopic tubal cauterization, which involves blocking the Fallopian tubes by burning them with instruments inserted through one or two small incisions in the abdomen. Hartke consented to the procedure, and on March 14, 1978, an abortion and tubal cauterization were performed on her. Dr. McKelway subsequently examined Hartke and termed the operation successful.

There was testimony from which the jury could infer that prior to the operation Dr. McKelway failed to disclose to Hartke that there was a risk of recanalization — where a Fallopian tube spontaneously reopens — of one to three out of one thousand. Hartke and her boyfriend, with whom she had lived for four years and whom she later married, also testified that the boyfriend offered to undergo a vasectomy if there was any risk of subsequent pregnancy, but that McKelway told them that the procedure was “a 100 percent sure operation,” and that Hartke would not have to worry about becoming pregnant again. Record Excerpts (R.E.) at 33; accord id. at 34.

Despite the surgery, Hartke again became pregnant in September 1979. After an examination confirmed that the pregnancy was normal — she had had a tubal or ectopic pregnancy in 1968 — she elected to carry it to term and in June 1980 gave birth by Caesarean section to a baby girl. At the same time, Hartke had herself resterilized by a tubal ligation, which involves actually cutting the Fallopian tubes. The record suggests that this method of sterilization involves about the same risk of subsequent pregnancy as cauterization. Transcript of July 23, 1981, at 145, 148 (testimony of Dr. Falk); see also id. at 148 (risk when ligation performed at time of delivery is greater). At the time of trial, one year after the delivery, she had not resumed sexual relations with her husband. Transcript of July 24, 1981, at 304 (testimony of Weddle (plaintiff’s husband)); see id. at 296 (testimony of Mrs. Hartke (plaintiff’s mother)).

Invoking the District Court’s diversity jurisdiction, Hartke brought suit against McKelway alleging negligence in the performance of the cauterization procedure, failure to obtain informed consent, and breach of warranty. At the conclusion of the plaintiff’s evidence, the District Court granted MeKelway a directed verdict as to the breach of warranty claim; Hartke does not now complain of this ruling. The jury returned a special verdict, finding that McKelway negligently failed to cauterize Hartke’s Fallopian tubes and that he failed to inform her of a material risk of the procedure. It awarded Hartke $10,000 in medical expenses, $100,000 for pain, suffering, and mental anguish, and $200,000 for the “[ajnticipated costs of raising this child until age 18 less any benefit [Hartke] received or in the future will receive by reason of the love, joy, happiness, etc. she experienced in raising a healthy, happy child.” R.E. at 1-2; 526 F.Supp. at 106 (copy of special verdict form used).

The District Court granted in part MeKelway’s motions for judgment notwithstanding the verdict and for a new trial. It held that there was no evidence of medical *1548 expenses greater than $6,000 and no evidence of future medical expenses, and ordered a new trial on this issue unless Hartke agreed to remit $4,000. Id. at 104. Hartke made such a remittitur. 1 The court also ruled that a new trial of the negligence claim would be necessary because one of Hartke’s experts should not have been allowed to testify on the standard of care for laparoscopic cauterization. Id. at 101. 2 Finally, the court disallowed the award of childrearing expenses because the “weight of authority does not, and the District of Columbia courts would not, allow recovery of the costs of raising a healthy child in circumstances such as these where the plaintiff sought sterilization solely for therapeutic reasons, and prizes the child she bore.” Id. at 105. 3 Both parties appealed. 4

II

We deal first with the judgment that Dr. McKelway failed to disclose material risks to Hartke.

A. Materiality of the Risk

McKelway first argues that he had no duty to disclose the risks of pregnancy in this case since no “reasonable person in what the physician [knew] or should [have known] to be the patient’s position would be likely to attach significance to the risks in deciding whether to accept or forego the proposed treatment,” Crain v. Allison, 443 A.2d 558, 562 (D.C.App.1982). The risk of pregnancy after laparoscopic cauterization was testified to be one to three out of one thousand.

For present purposes, the crucial language in the above formulation is “what the physician [knew] or should [have known] to be the patient’s position.” The “patient’s position” must include the patient’s medical history and other factors that might make knowledge of certain risks particularly important to a certain patient, acting reasonably. Here, there were two factors that would make even a small risk of pregnancy unusually dangerous for a patient in Hartke’s position. First, Dr. McKelway knew that Hartke had a history of gynecological and pregnancy-related problems. She had contracted -peritonitis after the birth by Caesarean section of her first child in 1964, resulting in a lengthy and traumatic hospital stay. She had had *1549 an ectopic pregnancy in 1968, apparently begun while using an IUD. She had been hospitalized numerous times for minor gynecological procedures. Hartke testified she informed Dr. McKelway that other doctors had advised her she “would not make it through [another] pregnancy.” Transcript of July 21, 1981, at 8. Second, Dr. McKelway had before him conclusive evidence of the psychological effect of pregnancy on his patient. He testified that she was “extremely upset” and “very agitated” about the pregnancy. Transcript of July 24,1981, at 311. She testified that she told him she thought she was going to die from the pregnancy. R.E. at 23.

In sum, the jury could conclude that a subsequent pregnancy would be a very serious consequence for this particular patient, which would result possibly in physical and certainly in psychological trauma. This was underscored by the testimony that Hartke’s boyfriend told the doctor he would undergo a vasectomy instead if the sterilization was not sure to be successful.

Moreover, less risky paths than relying on the cauterization were open to Hartke. Had she been able to compare the relative risks of failure, she might reasonably have changed her mind and decided to undergo a hysterectomy if, as the evidence suggested, 5 the latter procedure would have reduced the risk. Perhaps more likely given her ultimate course of conduct, even if Hartke had agreed to the cauterization, she might reasonably have decided to have her boyfriend undergo a vasectomy, abstained from intercourse, or taken other precautions that would reduce the risk.

In Canterbury v. Spence, 464 F.2d 772 (D.C.Cir.), cert. denied, 409 U.S. 1064, 93 S.Ct. 560, 34 L.Ed.2d 518 (1972), this court noted in this connection, “Whenever nondisclosure of particular risk information is open to debate by reasonable-minded men, the issue is for the finder of the facts.” Id. at 788 (footnote deleted). There we held that the jury could have found a one percent risk of a very serious harm — permanent urinary incontinence and paralysis of the bowels — to be material to a decision whether to undergo an operation for back pain. Id. at 794. In Henderson v. Milobsky, 595 F.2d 654, 659 (D.C.Cir.1978), we held that a .001% risk of permanent loss of sensation in a small section of the face could not reasonably be deemed material to a decision whether to have impacted wisdom teeth removed. In this case, the undisclosed risk was a .1% to .3% chance of subsequent pregnancy. For most people, this risk would be considered very small, but this patient was in a particularly unusual position. In view of the very serious expected consequences of pregnancy for her — possibly including death — as well as the ready availability of ways to reduce the risk, we agree with the District Court that a jury could conclude that a reasonable person in what Dr. McKelway knew to be plaintiff’s position would be likely to attach significance to the risk here.

B. Proximate Cause

In order for there to be liability in tort, there must be both breach of duty— here, the failure to divulge a material risk — and proximate causation. McKelway argues that the issue of whether the failure to disclose the risk of subsequent pregnancy here proximately caused the harm should not have gone to the jury because Hartke had not testified that she would not have undergone the treatment had she known of the risks. The source of McKelway’s argument is certain language in our opinion in Henderson v. Milobsky, 595 F.2d 654 (D.C.Cir.1978). Discussing the earlier case of Haven v. Randolph, 494 F.2d 1069 (D.C.Cir.1974), the Henderson court wrote:

*1550 Haven did not, however, add anything really novel to our jurisprudence on risk-disclosure. In result, it merely reemphasized the claimant’s burden of showing that the alleged breach of duty to disclose led to the injury for which compensation is sought. In Canterbury we had held that when damages are sought for a condition attributed to a medical procedure, causation by breach of duty cannot be demonstrated simply by the claimant’s unadorned hindsight-statement that had he known of the risk he would not have consented to the procedure. Haven merely stands for the cognate proposition that when the claimant has not even made such an assertion, the issue of causation cannot possibly go to the jury.

595 F.2d at 657-58 (emphasis added, footnote deleted).

The District Court rejected McKelway’s argument, finding that Henderson should not be read to require in haec verba testimony from the plaintiff “as long as there is sufficient evidence from which the jury could infer that she would have refused.” 526 F.Supp. at 103. The court held that to require such testimony in such a ease “would only set a trap for the unwary.” Id. Citing Hartke’s husband’s offer to undergo a vasectomy, the court found that there was evidence in this case “from which the jury could find that plaintiff would have declined the procedure had she been informed of the risks.” Id.

In its ruling the District Court appears to have applied a standard of causation based on what Hartke herself would have done. This is inconsistent with the standard of causation adopted by this court in Canterbury v. Spence, 464 F.2d at 790-91. In that case the court held that the issue of causation should be resolved on an objective basis, “in terms of what a prudent person in the patient’s position would have decided if suitably informed of all perils bearing significance.” Id. at 791 (footnote deleted). The rule was based on a distrust of the patient’s hindsight testimony that he or she would have foregone the treatment. Id. at 790 (The patient’s testimony “hardly represents more than a guess, perhaps tinged by the circumstance that the uncommunicated hazard has in fact materialized.”) (footnote deleted). Such testimony was believed to be the primary, if not only, evidence on the question of causation in the usual case.

The District Court’s confusion in this case — it charged the jury under a reasonable-person standard, Transcript of July 29, 1981, at 431 — is understandable, because in both Henderson and Haven, issued after the Canterbury decision, this court appears to have applied a causation-in-fact rather than a reasonable-person standard. In Henderson, the court held that the plaintiff’s testimony that he would not have undergone the procedure had he known of the risk of temporary paresthesia could not be believed because he did in fact continue the treatment once the paresthesia had appeared. 595 F.2d at 658; see id. at 659 (“We can hypothesize no more telling evidence of what he would have done .... ”). In Haven, this court affirmed the District Court’s grant of a directed verdict for the defendant, “primarily” for the reasons stated in the District Court’s opinion, 494 F.2d at 1070. The trial court’s holding rested in part on its finding that “there was no evidence that the parents would not have given their consent” had they known of the risks involved. Haven v. Randolph, 342 F.Supp. 538, 544 (D.D.C.1972), aff’d, 494 F.2d 1069 (D.C.Cir.1974).

Since the trial court’s ruling in this case, however, the District of Columbia Court of Appeals has unequivocally adopted, albeit in dictum, the objective, prudent-person standard. Citing the Canterbury case, the court in Crain v. Allison, 443 A.2d 558, 563 n. 14 (D.C.App.1982), wrote, “[T]he test of causation is objective. The test is what would a prudent person in the patient’s position have decided if informed of all relevant factors ....” 6

*1551 Under this standard, it is no longer possible to argue that a patient’s testimony is necessary for the issue of causation to get to the jury. The entire motivation for jettisoning the subjective standard of causation was distrust of precisely this testimony. As the Canterbury court pointed out, under the objective standard “[t]he plaintiff’s testimony is relevant ... but it would not threaten to dominate the findings.” 464 F.2d at 791; accord Crain, 443 A.2d at 563 n. 14 (“Although the patient’s testimony is relevant on the issue of causation, the test of causation is objective.”); Sard v. Hardy, 281 Md. 432, 450, 379 A.2d 1014, 1025 (1977) (“Under this rule, the patient’s hindsight testimony as to what he would have hypothetically done, though relevant, is not determinative of the issue.”). While it might be helpful, the jury certainly does not need the patient’s testimony to decide what a reasonable person in that position would have done. 7

Ill

The District of Columbia courts have not offered the same kind of guidance as to the other major issue raised by this appeal: whether Hartke may recover some portion of the costs of raising to majority the child born after the failed sterilization. Moreover, as the District Court noted, the case law from other jurisdictions is almost evenly divided, some courts allowing some recovery under various formulas, others allowing no recovery whatsoever. See Hartke v. McKelway, 526 F.Supp. at 104 & nn. 2 & 3 (citing cases).

In large part, the differences appear to revolve around whether the child can be considered a kind of damage to the par *1552 ents. 8 A number of courts have ruled that as a matter of law no healthy child can ever be considered an injury to its parents, because, as one court put it, “it is a matter of universally-shared emotion and sentiment that the intangible but all-important, incalculable but invaluable ‘benefits’ of parenthood far outweigh any of the mere monetary burdens involved.” Public Health Trust v. Brown, 388 So.2d 1084, 1085-86 (Fla.Dist.Ct.App.1980) (footnote deleted), review denied, 399 So.2d 1140 (Fla.1981); accord, e.g., Cockrum v. Baumgartner, 51 U.S.L.W. 2534, 2534 (Ill. Feb. 18, 1983) (“In a proper hierarchy of values, the benefit of life should not be outweighed by the expense of supporting it.”). Other courts have found that there are some cases in which the addition of a child constitutes an injury to the family. One court provided the following explanation:

To say that for reasons of public policy contraceptive failure can result in no damage as a matter of law ignores the fact that tens of millions of persons use contraceptives daily to avoid the very result which the defendant would have us say is always a benefit, never a detriment. Those tens of millions of persons, by their conduct, express the sense of the community.

Troppi v. Scarf, 31 Mich.App. 240, 253, 187 N.W.2d 511, 517, leave to appeal denied, 385 Mich. 753 (1971); see also Terrell v. Garcia, 496 S.W.2d 124, 131 (Tex.Civ.App.1973) (Cadena, J., dissenting) (“The birth of [an “ ‘unwanted’ ”] child may be a catastrophe not only for the parents and the child itself, but also for previously born siblings.”), cert. denied, 415 U.S. 927, 94 S.Ct. 1434, 39 L.Ed.2d 484 (1974).

Though we need not finally decide the question given our ultimate result, we suspect that allowing the plaintiff to prove t that raising a child constitutes damage is! the course of greater justice, and the one ' the District of Columbia courts may well < adopt. Usually, of course, it is true that ; the birth of a healthy child confers so substantial a benefit on its parents as to out- j, weigh the physical, emotional, and financial ¡ j burdens of bearing and raising it; “else, : | presumably, people would not choose to J multiply so freely,” Troppi, 31 Mich.App. at 254, 287 N.W.2d at 517. But when a couple has chosen not to have children, or not to have any more children, the suggestion ! arises that for them, at least, the birth of a, child would not be a net benefit. That is; their choice and the courts are required to respect it. Cf. Roe v. Wade, 410 U.S. 113, 153, 93 S.Ct. 705, 726-727, 35 L.Ed.2d 147 (1973) (woman’s right to abortion: “Maternity, or additional offspring, may force upon the woman a distressful life and future.”); Griswold v. Connecticut, 381 U.S. *1553 479, 485-86, 85 S.Ct. 1678, 1682, 1683, 14 L.Ed.2d 510 (1965) (couple’s right to use contraceptives).

Nevertheless, courts have recognized that there is an unusual difficulty in wrongful conception cases in setting the amount of compensation, because the extent, if any, to which the birth of a child is an injury to particular parents is not obvious but will vary depending on their circumstances and aspirations. See Troppi, 31 Mich.App. at 256-57, 187 N.W.2d at 518-19 (consequences of birth from failure of contraceptives will vary widely with purposes and circumstances of parents, comparing unmarried college student with honeymooning newlyweds). The parents may in fact have ended up with a child that they adore and that they privately consider to be, on balance, an overwhelming benefit to their lives. 9 This is because the parents may have sought to avoid conception for any of a number of reasons. They may have done so for socioeconomic reasons, seeking to avoid disruption of their careers or lifestyle, or to conserve family resources, see, e.g., Troppi, 31 Mich.App. at 244, 187 N.W.2d at 512 (after seven children, parents decided to limit size of family); Sherlock v. Stillwater Clinic, 260 N.W.2d 169, 171 (Minn.1977) (same); Betancourt v. Gaylor, 136 N.J.Super. 69, 74, 344 A.2d 336, 339 (1975) (parents sought to avoid expense of additional child); for eugenic reasons, seeking to avoid the birth of *1554 a handicapped child, see, e.g., Ochs v. Borrelli, 187 Conn. 253, 254-55, 445 A.2d 883, 883-84 (1982) (semble) (prior two children born with orthopedic defects); or for therapeutic reasons, seeking to avoid the dangers to the mother’s health of pregnancy and childbirth, see e.g., Wilczynski v. Goodman, 73 Ill.App.3d 51, 53, 29 Ill.Dec. 216, 218, 391 N.E.2d 479, 481 (1979) (therapeutic abortion); Christensen v. Thornby, 192 Minn. 123, 123, 255 N.W. 620, 621 (1934) (vasectomy sought because wife had been told another birth would be dangerous to her health). 10

When a couple chooses sterilization solely ;for therapeutic or eugenic reasons, it seems especially likely that the birth of a healthy child, although unplanned, may be, as it is for most parents, a great benefit to them. In such cases, a court will tend to feel that it is unjust to impose on the defendant doctor the often huge costs of raising the child, and will fear that a jury that did so was motivated by passion or anti-doctor prejudice. Thus, in considering the question of whether childrearing expenses may be recoverable, many courts and commentators have placed great emphasis on the couple’s reasons for undergoing sterilization. For example, the court in the earliest wrongful conception case, Christensen v. Thornby, made the point most clearly:

The purpose of the operation was to save the wife from the hazards to her life which were incident to childbirth. It was not the alleged purpose to save the expense incident to pregnancy and delivery. The wife has survived. Instead of losing his wife, the plaintiff has been blessed with the fatherhood of another child. The expenses alleged are incident to the bearing of a child, and their avoidance is remote from the avowed purpose of the operation.

Id. at 126, 255 N.W. at 622 (alternative holding). Other cases are to the same effect. See, e.g., Betancourt, 136 N.J.Super. at 72-75, 344 A.2d at 338-39 (distinguishing denial of recovery in Gleitman v. Cosgrove, 49 N.J. 22, 227 A.2d 689 (1967), on basis of eugenic rather than economic purpose there); Speck v. Finegold, 268 Pa.Super. 342, 358-59, 362, 408 A.2d 496, 505, 507 (1979) (discussing Christensen, Betancourt, and Gleitman), aff'd, 497 Pa. 77, 439 A.2d 110 (1981); id. at 374 n. 6, 408 A.2d at 513 n. 6 (Spaeth, J., concurring & dissenting) (where birth was unwanted because of risk that in the end did not materialize — such as danger to mother or fear of deformity — “arguably the damages should not include the expenses of raising the child”); Terrell, 496 S.W.2d at 130 (Cadena, J., dissenting) (distinguishing Hays v. Hall, 477 S.W.2d 402 (Tex.Civ.App.), rev’d, 488 S.W.2d 412 (Tex.1972), because of eugenic purpose there); Bishop v. Byrne, 265 F.Supp. 460, 463 (S.D.W.Va.1967) (under West Virginia law, victim was injured by failure of therapeutic sterilization “if the condition which [the operation] sought to avoid subsequently occurred”); see also Comment, Liability for Failure of Birth Control Methods, 76 Colum.L.Rev. 1187, 1197 (1976) [hereinafter cited as Columbia Note] (“If contraceptive measures fail here [where they have been used for therapeutic or eugenic purposes], but a normal child is born, damages might properly be denied on the theory that no injury was suffered.”); Recent Case, 28 DePaul L.Rev. 249, 257 (1978) (proposing use of special negligence instruction that would consider the purpose of the sterilization and family circumstances in determining the actual damage caused); Note, Wrongful Conception: Who Pays for Bringing up Baby?, 47 Fordham L.Rev. 418, 432 (1978) [hereinafter cited as Fordham Note] (“In wrongful conception cases, the ‘value’ of parenthood will vary according to the individual’s reasons for wanting the sterilization operation.”); Note, Wrongful Birth: A Child of Tort Comes of Age, 50 U.Cin.L.Rev. 65, 78 (1981) (“[I]f plaintiffs in a failed sterilization case hope to gain maximum recovery, they will have to prove that the purpose of steriliza *1555 tion was to prevent pregnancy and not possible injury to the woman because of pregnancy.”).

We tend to agree that a factfinder ¡I should place great weight on a couple’s I reason for undergoing sterilization in deciding whether the subsequent birth of a child, ,on balance, constitutes damage to the parents. Their reason for departing from the usual view that childrearing is a positive experience is in effect a calculation of the way in which they anticipate the costs of childbirth to outweigh the benefits. That calculation, untainted by bitterness and greed, or by a sense of duty to a child the parents have brought into the world, is usually the best available evidence of the extent to which the birth of the child has in ’’fact been an injury to them. Thus, for example, where a couple sought sterilization solely for therapeutic or eugenic rea j\ sons, there is a'presumption- raised that the uneventful birth of a' healthy child constitutes damage to the parents only to the extent that they experienced abnormal fear of harm to the mother or of the birth of a handicapped child. Courts and juries may assume that the parents treasure the child and that the usual expenses of raising it will be outweighed by the benefits derived.

The presumption raised by the evidence of the parents’ reason for seeking sterilization is, however, rebuttable. If it can be shown that the parents,"situation has somehow significantly changed since the sterilization — by reliance* on presumed infertility in making - -an income-reducing career j change, for example, or by a sudden in--| crease in wealth — it may be that the origi-1 nal calculation of anticipated injury has ■ changed for better or worse. See, e.g., Co- ’j lumbia Note, supra, at 1197 (“the motive for having a vasectomy or tubal ligation is relevant as evidence of lack of injury but should not be dispositive”); Fordham Note, supra, at 435 (where motives for sterilization are economic, “sudden relief from financial hardship or a reduction in family size” after operation may make benefits override burdens); Note, Wrongful Birth Damages: Mandate and Mishandling by Judicial Fiat, 13 Val.U.L.Rev. 127, 135 & n. 66 (1978) (parent may have taken early retirement). 11 Generally, however, the plaintiff’s ■ recovery will most accurately reflect the | amount of injury incurred if it is limited to paying for those risks that the plaintiff specifically sought to avoid and that came to pass. 12

In jury trials, of course, it is usually the task of a properly instructed jury to find such facts as the plaintiff’s motive in seeking sterilization and other facts reflect *1556 ing the degree to which childrearing ultimately constitutes injury to the plaintiff. Nevertheless, there are always cases in which a rational jury could find that the evidence suggested only certain facts, and in those cases such findings must be directed by the court, notwithstanding the jury’s verdict. On the issue of whether childrearing constituted injury to Hartke, the District Court, as we read its opinion, found this to be such a case. We agree.

The evidence here is overwhelming that Sandra Hartke sought to be sterilized for therapeutic reasons: she desperately feared that serious complications or even death would result from pregnancy and childbirth. The record is filled with testimony to this effect. For example, Hartke testified that she told Dr. McKelway “that I didn’t want to be pregnant, and I was terrified of being pregnant, and I was extremely concerned about doctors and hospitals. That was my overriding concern.” Transcript at July 23, 1981, at 85-86. She also said she told him

[t]hat I did not want to go through a pregnancy, that I was terrified of going through a pregnancy, and I felt I was going to die, and I had been advised by doctors that I would not make it through a pregnancy. I told him that I had peritonitis, which was gangrene, of the abdominal cavity in 1964 when I had my first daughter.

* * * * * *

... He said that he could perform an abortion in the hospital, and I said I didn’t want to be put in this position anymore. I couldn’t keep coping with the idea that I was going to die. I wanted him to help me find a way that I wouldn’t have to keep going back into the hospital for all these kinds of problems. He told me that he can sterilize me.

Transcript of July 21, 1981, at 8, 14. Other like testimony is printed in the margin. 13 *1557 The testimony of Dr. McKelway and of Hartke’s husband was to the same effect. 14

That the danger of childbirth remained her sole concern up to the time of her pregnancy is clear from her reasons for deciding not to have an abortion. Once the pregnancy was determined, by means of a sonogram, not to be ectopic, Hartke was advised that the risks of carrying the pregnancy to term were about the same as the risks of abortion. She testified that she decided not to have an abortion because, “[h]aving decided that the risks were equal either way, that to resterilize me Dr. Barter was going to open me back up, then I might as well try to carry the pregnancy to term, that Dr. Barter would help me through the pregnancy.” Transcript of July 21,1981, at 32. This is the only testimony regarding her reasons for carrying the