In this insurance coverage dispute, Continental Casualty Company ("Continental") seeks to recoup from Hartford Fire Insurance Company ("Hartford") a portion of a settlement of a medical malpractice lawsuit against The Yater Medical Group and Dr. Howard Smith which were insured by the two companies during consecutive time periods. The district court granted summary judgment for Hartford upon concluding that no reasonable jury could find that the proximate cause of Dr. Smith's decision to deliver a baby prematurely was testing performed when Hartford's insurance policy was in effect. Continental contends that the district court ignored substantial evidence showing that negligence occurring during Hartford's policy period substantially contributed to the doctor's decision to deliver the baby prematurely, and failed to recognize that such negligence could, and in fact did, constitute a concurrent proximate cause of the injuries. We agree, and accordingly reverse and remand the case to the district court to determine the proper apportionment of liability for the settlement.

Both Continental and Hartford issued medical malpractice policies to The Yater Medical Group ("Yater"), for which Dr. Howard Smith, an obstetrician, was a named insured. The policies covered consecutive periods of time and included different exposure provisions as well as different "other insurance" provisions. Continental issued both a primary policy and a separate excess policy for the period of January 1, 1981, to January 1, 1982. The primary policy limited coverage to $1 million per claim and $1 million in the aggregate for Yater and its insured physicians. The primary policy provided that Continental would

pay on behalf of the Insured all sums which the Insured shall become legally obligated to pay as damages because of ... [i]njury arising out of the rendering or of failure to render, during the policy period, professional services....

[325 U.S.App.D.C. 313] (emphasis added). It did not contain an "other insurance" provision. The excess policy, however, which provided $10 million coverage jointly to Yater and Dr. Smith over the limits of the primary policy, contained an "other insurance" provision that denied coverage if the insured had other insurance to cover a loss.1

Hartford's policy covered the eight-month period from May 1, 1980, to January 1, 1981, which immediately preceded the Continental policy period. The limit of liability under Hartford's policy was $2 million for each medical incident and $2 million in the aggregate. Hartford's policy provided that Hartford would pay on behalf of the insured

[a]ll sums which the insured shall become legally obligated to pay as damages because of injury, to which this insurance applies, to any person caused by a medical incident which occurs during the policy period....

(emphasis added). The policy defined the term "medical incident" as "any act or omission in the furnishing of professional or dental services to any person...." Its "other insurance" provision included a pro rata limitation.2

The instant insurance coverage litigation arises from a medical malpractice action brought by Tracina Woods and her parents against Yater and Dr. Howard Smith, the obstetrician who delivered Tracina by cesarean section on February 19, 1981, after monitoring her development in utero from the early months of Mrs. Woods' pregnancy.3 The Woods alleged in their complaint that Dr. Smith had negligently delivered their baby prematurely and that, as a result, she had developed serious and permanent injuries, including cerebral palsy, spastic diplegia, and brain damage. The complaint also alleged that during his care and treatment of Mrs. Woods and the baby, Dr. Smith failed to take a thorough medical history, to conduct appropriate and careful physical examinations, to utilize appropriate laboratory and ancillary procedures, and to interpret accurately physical examinations and findings.

Shortly after the Woods filed their complaint, Yater gave notice of the action to Hartford. In response, Hartford informed Yater that it would join with Continental in providing a defense.4 Acknowledging that the Woods "may be able to tie in both liability and causation" during Hartford's policy period, Hartford agreed with Continental to pay 50% of the defense costs. Hartford also stated that if after the Woods' experts were deposed the Woods were "unable to tie in both liability and causation during [its] period of coverage," Hartford would "withdraw from paying any additional attorneys fees and related expenses and of course [would] not indemnify."

The undisputed evidence produced during discovery showed that Dr. Smith first saw Mrs. Woods as a patient on September 8, 1980. At that time, he confirmed that she was pregnant and recorded her last menstrual period as May 2, 1980, a date suggesting that the fetus was at least seventeen weeks old. Based on a pelvic examination, however, Dr. Smith found the size of the uterus to be [325 U.S.App.D.C. 314] consistent with a 12-week fetus. Because Dr. Smith was unsure of Mrs. Woods' date of conception, he ordered a sonogram. The sonogram, which was performed two days later, indicated that the fetus was approximately six weeks old. The discrepancy between Mrs. Woods' menstrual period and the sonogram led Dr. Smith to have "concerns" about the pregnancy, particularly that Mrs. Woods was "small for [the] dates" and the fetus thus might suffer from Intrauterine Growth Retardation ("IUGR"), a condition marked by lagging growth throughout the pregnancy.5 Nevertheless, Dr. Smith did not order any additional tests at that time, and provided routine prenatal care during three additional visits in 1980.

In January 1981, Dr. Smith ordered a second sonogram from which he concluded that there was a strong possibility that the fetus suffered from IUGR. Dr. Smith began to see Mrs. Woods with greater frequency, examining her on six occasions between January 9, 1981, and February 11, 1981, and ordering a third sonogram during this period. His concern about IUGR also led him to monitor Mrs. Woods' serum estriol level.6 Three samples collected on February 4, 6, and 11 indicated a constant estriol level of 3.3 NG/ML. On February 13 there was a thirty percent drop to 2.3 NG/ML. Concerned that the fetus might be in jeopardy, Dr. Smith admitted Mrs. Woods to the Washington Hospital Center on February 18 for additional monitoring and bed rest, so that he could deliver the baby "if there were other further reasons to be concerned regarding [IUGR]."

Upon her admission to the hospital, Dr. Smith directed Dr. Henry Sobel to perform a fourth sonogram and an amniocentesis. Dr. Sobel reported that the sonogram indicated the presence of a low level of amniotic fluid, which prevented the successful performance of the amniocentesis. "On the basis of the falling estriols and the decrease in the amniotic fluid," Dr. Smith concluded that "we, indeed, had [IUGR] and that this baby was very compromised and should be delivered the following morning." Believing that the fetus was approximately thirty-six to thirty-seven weeks old (i.e., three to four weeks short of full term), Dr. Smith delivered the baby by cesarean section on February 19, 1981. After birth, the baby experienced respiratory problems, and several years later, she was diagnosed as having cerebral palsy, spastic diplegia, and other physical disorders.

Expert evidence produced during discovery highlighted the parties' differing interpretations of the evidence. The Woods presented experts who disputed Dr. Smith's conclusion that the fetus was not developing properly and suffered from IUGR. In these experts' opinion, the baby's premature delivery caused her to develop cerebral palsy and other serious disorders. The experts identified a number of errors by Dr. Smith that led him to deliver the baby prematurely. First, because Dr. Smith failed to ascertain that Mrs. Woods' periods were irregular, he placed undue emphasis on the date of her last menstrual period in assessing fetal age. Second, given the conflicting signals as to fetal age provided by the last menstrual period, the initial sonogram, and a pelvic examination, as well as the fact that the best time to evaluate fetal age is early in the pregnancy, Dr. Smith should have taken additional steps early on to determine the age of the fetus with greater accuracy. This included ordering follow-up sonograms in the first month after Mrs. Woods' initial visit and carefully monitoring early "landmarks" in fetal development, such as the first detection of fetal heartbeat and fetal movement. Third, Dr. Smith erroneously concluded that the fourth sonogram showed a low level of amniotic fluid corroborating an IUGR diagnosis. Fourth, in focusing on the falling estriol levels, Dr. Smith relied on an "insufficient [325 U.S.App.D.C. 315] basis" to perform an elective cesarean section.

Yater and Dr. Smith responded with their own expert evidence that the baby's injuries had nothing to do with Dr. Smith or the cesarean section, but instead resulted from a genetic defect, IUGR, or care provided after birth by the Washington Hospital Center. One expert opined that the baby's disabilities resulted from developmental problems in utero and that she suffered from IUGR, although he did not know with reasonable medical certainty the specific cause of the injuries. A second expert opined that the state of the placenta and umbilical cord, as well as the estriol levels and the baby's birth weight, indicated placental insufficiency and IUGR.

At the conclusion of discovery, defense counsel estimated that the insureds had no chance of winning at trial, and that the likely jury verdict would be between $6 and $10 million. Based on counsel's recommendation and Dr. Smith's request that the case be settled, Continental entered into a settlement agreement with the Woods for $4,127,554 on the day set for jury selection. Hartford had previously advised the district court that because in its view there was no coverage under its policy, it was withdrawing from the defense. Hartford therefore declined to participate in the settlement and Continental funded it in its entirety.

Continental sued Hartford for contribution to recoup a portion of the settlement funds, in the full amount of Hartford's $2 million policy limit, and for prejudgment interest. Upon consideration of the parties' cross-motions for summary judgment, the district court granted judgment for Hartford, ruling that no reasonable jury could find that Hartford was obligated to contribute to the settlement. Assuming that Dr. Smith performed substandard care in 1980 in determining the correct fetal age during Hartford's policy period, the district court found that "the 1981 serum estriol level and amniotic fluid readings," independent of the fetal age assessment, led to Dr. Smith's decision to deliver the baby. The court concluded that because no medical conduct by Dr. Smith in 1980 could have been a proximate cause of the baby's injuries, Hartford could not have been found liable to the Woods and hence it had no obligation to contribute to the settlement. Continental appeals.

Summary judgment should be granted when the pleadings, depositions, answers to interrogatories, admissions, and affidavits, viewed in the light most favorable to the non-movant, show that there are no genuine issues of material fact and the moving party is entitled to judgment as a matter of law. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 250, 255, 106 S. Ct. 2505, 2511, 2513-14, 91 L. Ed. 2d 202 (1986); Ambrosini v. Labarraque, 101 F.3d 129, 132 (D.C.Cir.1996), cert. dismissed, --- U.S. ----,----, 117 S. Ct. 1572, 137 L. Ed. 2d 716 (1997). In other words, "[s]ummary judgment is appropriate ... where 'the evidence is such that a reasonable jury could not return a verdict for the nonmoving party.' " Washington Post Co. v. United States Dept. of Health & Human Serv., 865 F.2d 320, 325 (D.C.Cir.1989) (quoting Anderson, 477 U.S. at 248, 106 S.Ct. at 2510). Our review of the district court's grant of summary judgment is de novo. See Ambrosini, 101 F.3d at 132.

Continental acknowledges that the district court properly defined the determinative issue regarding Hartford's liability to be whether any action by Dr. Smith during Hartford's policy period was a proximate cause of the baby's injuries, and that the issue of proximate cause must be resolved by determining the basis for Dr. Smith's decision to perform the cesarean section. Continental contends, however, that in concluding that Dr. Smith's actions in 1980 during Hartford policy period could not, as a matter of law, constitute a proximate cause of the baby's injuries, the district court ignored the substantial evidence in the underlying malpractice litigation that Dr. Smith's negligence in 1980 during the early months of the pregnancy substantially contributed to his decision to deliver the baby prematurely in 1981. Continental further maintains that the district court erroneously failed to recognize [325 U.S.App.D.C. 316] that there could be, and in fact was, more than one proximate cause of the baby's injuries. Because there was more than sufficient evidence of Dr. Smith's negligence in 1980 during Hartford's policy period to go to a jury, Continental contends that Hartford is obligated to pay some portion of the settlement, and thus the district court erred in granting judgment for Hartford and should have granted judgment for Continental.

Under the law of the District of Columbia, proximate cause in insurance contract cases has been defined as follows:7

The proximate cause is the efficient cause, the one that necessarily sets the other causes in operation. The causes that are merely incidental or instruments of a superior or controlling agency are not the proximate causes and the responsible ones, though they may be nearer in time to the result.

Unkelsbee v. Homestead Fire Ins. Co., 41 A.2d 168, 171 (D.C.Mun.App.1945) (quoting Aetna Ins. Co. v. Boon, 95 U.S. 117, 130, 24 L. Ed. 395 (1877)); see also Quadrangle Dev. Corp. v. Hartford Ins. Co., 645 A.2d 1074, 1077 (D.C.1994). The District of Columbia Court of Appeals has explained that proximate cause " 'is the dominant cause, not the one which is incidental to that cause....' " Quadrangle, 645 A.2d at 1077 (quoting Unkelsbee, 41 A.2d at 171). However, rather than viewing proximate cause as a single act or omission that is most immediately responsible for an injury, the Court of Appeals has stated that " 'where there is a concurrence of two causes, the efficient cause--the one that sets the others in motion--is the cause to which the loss is to be attributed....' " Id. (quoting Frontis v. Milwaukee Ins. Co., 156 Conn. 492, 242 A.2d 749 (1968)). "[I]n considering what is the proximate and what [is] the remote cause of an injury, '[t]he inquiry must always be whether there was any intermediate cause disconnected from the primary fault, and self-operating, which produced the injury.' " Unkelsbee, 41 A.2d at 171 (quoting Aetna Ins. Co., 95 U.S. at 130 (quoting Milwaukee & St. Paul R. Co. v. Kellogg, 94 U.S. 469, 24 L. Ed. 256 (1876))).

We hold that the district court erred in ruling that a jury in the underlying medical malpractice litigation could not reasonably find that Dr. Smith's negligence in 1980 was a proximate cause of his decision to deliver Tracina Woods prematurely. Although, as the district court found, a jury could reasonably have found that Dr. Smith's interpretation of the serum estriol and amniotic fluid tests alone led him to perform the cesarean section, a jury also could reasonably have found that Dr. Smith's failure in 1980 to assess accurately the fetal age at the beginning of the pregnancy set in motion the course of events that led to the premature delivery in 1981. From his initial tests of Mrs. Woods in September 1980, because of his concern that Mrs. Woods was "small for [the] dates," Dr. Smith believed that the fetus might suffer from IUGR. Throughout the pregnancy, Dr. Smith remained concerned about the possibility of IUGR. Based upon the sonogram in January 1981, he concluded that there was a "strong possibility" that the fetus suffered from IUGR. In February 1981, the falling estriol levels and the decrease in amniotic fluid finally confirmed Dr. Smith's belief that the fetus suffered from IUGR and therefore had to be delivered immediately. In other words, the evidence would have supported a finding that Dr. Smith became concerned about IUGR in 1980 at the beginning of the pregnancy, and that his interpretations of all follow-up examinations were influenced by this concern. So viewed, because an IUGR diagnosis is, by definition, dependent on an assessment of fetal age, see supra n. 5, Dr. Smith's interpretation in 1981 of the falling estriol and amniotic fluid levels as conclusively establishing IUGR was an intermediate cause, not "self-operating" and "disconnected" from, but intertwined with and dependent on his initial evaluation of the fetus in 1980. See Unkelsbee, 41 A.2d at 171.

The facts of the instant case are similar to those in Glacier Gen. Assurance Co. v. Continental [325 U.S.App.D.C. 317] Cas. Co., 605 F. Supp. 126 (D.D.C.1985), where two companies had insured a doctor for consecutive policy periods, the doctor's alleged negligence spanned both policy periods, and the insurers settled the malpractice suit prior to a determination of liability or causation. Although the ultimate injury to the doctor's patient, the amputation of his legs, occurred during the second policy period, the parties "agree[d] that an earlier diagnosis [during the first policy period] ... might have prevented the amputation from becoming necessary." Id. at 128. Without pinpointing the period of time, the district court concluded "with some confidence that the conduct which proximately caused the injury ... overlapped the coverage period of both carriers." Id. at 129.

Here, by contrast with Glacier, the parties dispute whether a non-negligent assessment of fetal age during Hartford's policy period in 1980 might have prevented the premature delivery of the baby during Continental's policy period in 1981. Hartford's experts say no, but according to the Woods' experts, and Hartford does not contest this fact, Dr. Smith was negligent in conducting his initial evaluation of Mrs. Woods in 1980. Due to the irregularity of Mrs. Woods' menstrual periods, Dr. Smith should have discounted the significance of Mrs. Woods' last menstrual period as an accurate indicator of fetal age. Because he was confronted with conflicting signals as to the age of the fetus, Dr. Smith also should have performed serial sonograms early in the pregnancy and paid closer attention to early fetal landmarks in order to confirm the actual age of the fetus. Had he taken these steps, the Woods' experts opined, Dr. Smith would have realized that the fetus was developing properly in utero, that Mrs. Woods was not "small for [the] dates," and that the fetus did not suffer from IUGR. Because Dr. Smith failed in 1980 to take these steps, Dr. Smith was never able to get an accurate assessment of fetal age, and hence failed to rule out IUGR as a problem. In the opinion of the Woods' experts, the conduct that proximately caused the injuries thus overlapped both the Hartford and Continental policy periods.

Given the parties' differing views of the doctor's conduct in 1980, the difficulty for this court arises from the fact that there has been no final determination of the cause of the injury to the baby because Continental settled with the Woods prior to trial. Had Hartford denied that Dr. Smith was negligent at all, then it would be entitled to force Continental to prove that its insured was negligent, or at least that there be a trial on proximate cause. The parties, however, have litigated the coverage controversy as if the only issue to be decided is whether any of Dr. Smith's actions before January 1, 1981, was a proximate cause of the baby's injuries, and as if the complete record from which Hartford's liability is to be determined is the pleadings and discovery prior to settlement. Even though it is conceivable that a jury might reasonably have decided either way on proximate cause, given the manner in which the parties have litigated the coverage issue, it makes little sense for a new jury in the coverage case to speculate on how another jury would have decided the medical malpractice case. A jury would be asked to find facts without live testimony and little opportunity to reconcile any conflicts in testimony. Under these circumstances, we conclude that the question before the court is limited to whether the evidence available at the time of settlement could lead a reasonable jury to find that Dr. Smith's pre-1981 negligence was a proximate cause of the Woods' injuries. That is a question of law, and the answer is yes.

In relying on a portion of Dr. Smith's testimony suggesting that his actions in 1980 did not ultimately influence his decision to deliver the baby prematurely in 1981, the district court gave insufficient attention to the IUGR diagnosis. In response to a hypothetical question whether, if presented with the same circumstances except for gestational age, he would have delivered the baby even if he knew to a medical certainty that fetal age was twenty-six weeks, Dr. Smith testified, "[I]f this were the situation that presented itself at 26 weeks ..., I would have made the same decision." The district court interpreted this testimony to mean that Dr. Smith's estimate of fetal age did not [325 U.S.App.D.C. 318] influence his decision to deliver. But this interpretation fails to take into account that a critical element of the hypothetical was the assumption that the fetus suffered from IUGR. Because Dr. Smith's diagnosis of IUGR was tied to his assessment of fetal age, the district court clearly erred in finding that Dr. Smith's response to this hypothetical demonstrated that fetal age was irrelevant to his decision to deliver prematurely.

Without contesting Dr. Smith's alleged negligence in 1980, Hartford points to Dr. Smith's testimony that, absent the falling estriol and amniotic fluid levels, he would not have performed the cesarean section on February 19, 1981. Notwithstanding that it was in Dr. Smith's interest to downplay the significance of his conduct during the early months of Mrs. Woods' pregnancy and any confusion he might have had as to the age of the fetus, accepting Dr. Smith's testimony as true, it does not negate the causative significance of his undisputed negligence in 1980. For purposes of Hartford's liability, the determinative question is not whether the evidence conclusively establishes that in the absence of the two test results in 1981, Dr. Smith would have performed the cesarean section, but whether Dr. Smith's suspicion in 1980 of IUGR was " 'the efficient cause, the one that necessarily set[ ] the other causes in operation,' " Unkelsbee, 41 A.2d at 171 (citation omitted), which led him to perform the cesarean section on February 19, 1981. In view of the significance of the IUGR diagnosis to Dr. Smith's ultimate decision to deliver prematurely, a reasonable jury could find Dr. Smith's actions during Hartford's policy period were a proximate cause of his decision to deliver the baby prematurely.8

Having concluded, given the manner in which the parties have litigated the coverage issue, that there was dual coverage by Continental and Hartford of the injuries to the Woods' baby, the question remains of how to apportion the loss between the two companies. The determination would appear to turn on the "other insurance" provisions of the policies. See Keene Corp. v. Insurance Co. of North America, 667 F.2d 1034, 1050 (D.C.Cir.1981). While Continental's $1 million primary policy did not have an "other insurance" provision, its excess policy, with a $10 million ceiling, included an "other insurance" provision whereby the policy would only cover insurance liability "greater than the applicable limit of liability provided by the other insurance." Hartford's policy, with a $2 million ceiling, included an "other insurance" provision that Hartford "would not be liable for a greater proportion" of the loss than the limit of its policy to the aggregate limit of "all valid and collectible insurance."

Under District of Columbia law, where "other insurance" provisions in an excess insurance policy and a pro-rata policy can be reconciled to give effect to the intent of the contracting parties, the court will do so. Jones v. Medox, Inc., 430 A.2d 488, 493-94 (D.C.1981) (in banc). In other words, when more than one insurance policy covers an injury, a court should not impose its view of the appropriate apportionment of liability between the insurers but should "focus[ ] on the contractual provisions and the intent of the parties." Auger v. Tasea Inv. Co., 676 A.2d 18, 20 (D.C.1996) (quoting Jones, 430 A.2d at 494). But see Glacier, 605 F. Supp. at 130-32 (requiring equal contributions from two insurers, but not citing Jones). Because [325 U.S.App.D.C. 319] the parties have not briefed the effect of the "other insurance" provisions on liability under the policies, nor addressed the issue in their cross-motions for summary judgment, we remand the case to the district court to determine the proper allocation of the settlement and rule on Continental's request for prejudgment interest.

Accordingly, we reverse the grant of judgment to Hartford and remand the case to the district court.