Malpractice claims against emergency physicians in Massachusetts: 1975–1993

Malpractice claims against emergency physicians in Massachusetts: 1975–1993

Original Contributions Malpractice Claims Against Emergency Physicians in Massachusetts: 1975-1993 ANITA KARCZ, MD, MBA,* ROBERT KORN, MD,t MARY C. B...

2MB Sizes 16 Downloads 133 Views

Original Contributions

Malpractice Claims Against Emergency Physicians in Massachusetts: 1975-1993 ANITA KARCZ, MD, MBA,* ROBERT KORN, MD,t MARY C. BURKE, MD,::I: RICHARD CAGGIANO, MD,:I: MICHAEL J. DOYLE, MD,:I:MICHAEL J. ERDOS, MD,::I: ERROL D. GREEN, MD,:I:: KENNETH WILLIAMS, MD::I: This study reviewed 549 malpractice claims filed against emergency physicians in Massachusetts from 1975 through 1993, with a total of $39,168,891 of indemnity and expense spent on the 549 closed claims. High-risk diagnostic categories (chest pain, abdominal pain, wounds, fractures, pediatric fever/meningitis, epiglottitis, central nervous system bleeding, and abdominal aortic aneurysm) accounted for 63.75% of all closed claims and 64.23% of the total indemnity and expense spent on closed claims. Missed myocardial infarction (chest pain) claims accounted for 25.47% of the total cost of closed claims but only 10.38% of closed claims. The number of claims for missed myocardial infarction increased in the post-1988 closed claim group compared to the pre-1988 group; fractures and wounds were significantly less frequent in the post-1988 group. The frequency of high-risk claims decreased in the post-1988 group, largely because of the decline in fracture and wound claims. The category of missed myocardial infarction had a larger percentage of claims closed with indemnity payment than without indemnity payment. This parameter may serve as a marker for the overall seriousness of claims associated with a particular allegation, unlike the average cost per claim, which may be skewed by a few large awards. (Am J Emerg Med 1996;14:341-345. Copyright © 1996 by W.B. Saunders Company) Because the Massachusetts College of Emergency Physicians (MACEP) has had a longstanding cooperative relationship with the major malpractice insurer of Massachusetts physicians, ProMutual (formerly the Joint Underwriters Association and t h e M a s s a c h u s e t t s Medical Professional Insurance Association), we have been privileged with access to claims data regarding malpractice claims flied against emergency physicians and have published two prior studies of claims closed from 1980 through 19871 and claims closed from 1988 through 1990. 2 This article describes the charac-

From *Healthcare Opportunities, Inc, Watertown, MA, and the Department of Emergency Medicine, Metrowest Medical Center, Framingham, MA; tthe Department of Emergency Medicine, Lawrence and Memorial Hospital, New London, CT; and :~theClosed Claims Study Group of the Massachusetts College of Emergency Physicians, Plainville, MA. Manuscript received November 12, 1995, returned December 1, 1995; revision received January 3, 1996, accepted January 24, 1996. Address reprint requests to Dr Karcz, 89 Bailey Road, Watertown, MA 02172. Key Words: Malpractice, risk management. Copyright © 1996 by W.B. Saunders Company 0735-6757/96/1404-0001 $5.00/0

teristics of malpractice claims filed against emergency physicians in Massachusetts from 1975 through 1993. We undertook this study of a much larger claims universe to delineate claims characteristics that might be useful in identifying changing patterns of claims; understanding claims patterns is essential for directing efforts to prevent malpractice claims.

METHODS A computer data file was obtained from ProMutual of all the claims filed against emergency physicians since the formation of the Massachusetts Joint Underwriting Association in 1975. This file was merged with clinical data from MACEP's two prior closed claims studies; additional data was obtained by review of claims closed in the years 1991, 1992, and 1993. Closed claim review was performed by physicians board-certified in emergency medicine, using the same data work sheets and dictation setup used in MACEP's previous closed claim study.2 The review was performed during the 1994 calendar year. Detailed analysis of the clinical content of the claim files is still in progress and was not the focus of this article. The ProMutual computer data file contained a brief allegation for each claim. When the actual claim file could not be obtained, this allegation was used to classify the claim if the nature of the allegation was clear from the data base description. Claims without available files were not included as high-risk claims unless the data base description definitively placed them in a high-risk group. Essential data were crosschecked in the computer data file against file review information to help assure data accuracy. All data were analyzed using commercially available database, spreadsheet, and statistical software (Dataease, QuattroPro, and Statistics on Software). All statistical tests were one-tailed, with a significance level of .05.

RESULTS There were a total of 747 claim files in the ProMutual database that were filed against emergency physicians from 1975 through the end o f 1993; 132 of these claim files were open at the end of 1993. The 615 closed claim files were designated in the ProMutual database as claims (279), suits (270), depositions only (48), and unclassified (18). The deposition-only and unclassified files accounted for $93,331 of expense and no indemnity payment. The data analysis was performed on the remaining 549 claim files. 341

342

AMERICAN JOURNAL OF EMERGENCY MEDICINE • Volume 14, Number 4 • July 1996

There were 549 closed claims, accounting for $39,168,891 of indemnity and expense. Indemnity is the amount paid to the plaintiff. Expense includes costs incurred by the insurer related to claim defense: legal fees, expert witness fees, court costs, etc. The graphical distribution of claims closed per year and the total cost of the claims closed per year are shown in Figure 1. The average amount spent (indemnity and expense) per closed claim for all 549 closed claims was $71,346 (standard deviation of 203,672). A comparison of claims closed before 1988 (pre-1988) and those closed during or after 1988 (post-1988) is shown in Table 1. The year 1988 was chosen to divide the two groups in an attempt to create two approximately equal-size claim groups and to maintain congruence with our previously reviewed groups of claims closed between 1980 and 1987 and from 1988 to 1990. h2 Claims closed before 1988 (total 275 claims) cost less per claim than claims closed during or after 1988 (total 274 claims), with an average indemnity and expense per claim of $42,920 (standard deviation of 116,413) and $99,875 (standard deviation of 260,556), respectively. The closed claims were also separated into high-risk and non-high-risk diagnostic groups. The high-risk group included claims related to chest pain, abdominal pain, wounds, fractures, pediatric fever/meningitis, central nervous system bleeding, abdominal aortic aneurysm, and epiglottitis; these designations have been previously described as particular liability problems in emergency medicine. 1-3 The high-risk

TABLE1. ClosedClaims 1975-1993 Claims Closed ClaimsClosed Before 1988 Duringand After 1988 Total Avg I&E per claim As percentage of all claims Non-high-risk High-risk MI Wounds Fractures Abdominal pain CNS bleeding CWOP Non-high-risk High-risk CWIP Non-high-risk High-risk

275 42,920 87(31.64%) 188(68.36%) 21 ( 7 . 6 4 % ) 69(25.09%) 57(20.73%) 26(9.45%) 10(3.63%) 155(56.36%) 51 (32.90%) 104(67.10%) 120(43.64%) 36(30.00%) 84(70.00%)

274 99,875 112(40.88%) 162(59.12%) 36(13.14%) 40(14.60%) 40(14.60%) 29(10.58%) 10(3.65%) 171 (62.41%) 77(45.03%) 94(54.97%) 103(37,59%) 35(33.98%) 68(66.02%)

claims group accounted for 63.75% of all closed claims and for 64.23% of the total indemnity and expense spent on closed claims. There was no difference between the highrisk and non-high-risk groups in average cost per claim (t test, P = .47). The high-risk claims significantly decreased from 63.36% of all claims closed pre-1988 to 59.12% of all claims closed

$7,000,000

60

$6,000,000 50

$5,OOO,OOO 40

~:;~::::::t AII C$a i m s #

$4,000,000

~ H i

Risk Clairns #

30

o

~ I & E $3,0OO,OOO ~ 20 $2,000,000

10 $1,000,000

$-

FIGURE 1. Yearlyclaims volume and costs.

Log. (Hi Risk Claims) Log. (AlI Claims)

KARCZ ET AL • MASSACHUSETTS MALPRACTICE CLAIMS 1975-1993

post-1988. Missed myocardial infarction (MI) accounted for a significantly higher percentage of claims in the post-1988 group. Fractures and wounds accounted for a significantly lower percentage of claims in the post-1988 group (Table 1). Missed myocardial infarction (chest pain) accounted for the highest percentage indemnity and expense, 25.47%, of any single category of high-risk malpractice claim (Table 2). Wounds and fractures accounted for the largest percentages, 19.85% and 17.67%, respectively, of claims. Out of 549 closed claims, there were 6 cases with indemnity payments of $1 million or more. Three of these cases involved missed cervical and thoracic spine fractures resulting in paralysis. One claim allegation was failure to admit a patient who later committed murder. The other two allegations involved failure to diagnose cerebral hemorrhage and delay in diagnosis and treatment of cerebral hemorrhage. The closed claim data was also segmented into claims closed with indemnity payment (CWIP) and claims closed without indemnity payment (CWOP) (Table 3). The mean expense per claim differed between the CWIP and CWOP categories (t test, P = .02), with the average expense per claim greater in the CWIP group. The non-high-risk claims group had a higher percentage of claims closed without indemnity payment compared to the high-risk claims group (two sample proportions test, P = .036). However, there was no difference between the average indemnity and expense per claim for the high-risk claims closed with indemnity payment and the non-high-risk claims closed with indemnity payment (t test, P = .29). A distribution of the high-risk categories by indemnity payment status is shown in Table 4. Of the high-risk diagnostic groups, missed MI was the only diagnostic group with more than a very few claims (eg, pediatric fever/ meningitis and epiglottitis) that had a larger percentage of claims closed with indemnity payment than closed without indemnity payment.

DISCUSSION Malpractice claims are not trivial; they consume a large amount of financial resources within the medical system, generating many billions of dollars in defensive medical practice in addition to the directly measurable Costs. 4'5 The clinical aspects of medical malpractice are greatly complicated by the tort system and its often seemingly capricious variability. Malpractice claims contribute greatly to physi-

343

TABLE3. Closed Claims 1975-1993 All Closed Claims Number Percentage CWOP Number Percentage Avg I SD Avg E SD Avg I&E SD Total I Total E Total I&E Percent I&E of all claims CWlP Number Percentage Avg I SD Avg E SD Avg I&E SD Total l Total E Total I&E Percent I&E of all claims

High-Risk Claims

Non-HighRisk Claims

549

350 63.75

326 59.38 0 0 8,184 14,593 8,184 14,593 0 2,667,937 2,667,937 6.81

198 56.57 0 0 8,257 15,413 8,257 15,413 0 1,634,965 1,634,965 6.50

199 36.25 128 64.32 0 0 8,070 13,224 8,070 13,224 0 1,032,971 1,032,971 7.37

223 152 71 40.62 43.43 35.68 147,312 142,281 158,081 256,228 214,347 328,139 16,370 12,475 24,707 58,861 15,376 101,360 163,681 154,756 182,788 295,728 223,158 409,308 32,850,536 21,626,780 11,223,755 3,650,419 1,896,198 1,754,220 36,500,955 23,522,978 12,977,975 93.19 93.50 92.63

cian stress, both by the practitioner's perception of the likelihood of being sued and by the personal upheaval that occurs when a physician is sued. 6-9 Even with legal constraint of the statute of limitations to 3 years, the average time in our study from an incident to a claim closing is nearly 4 years, a long period of stress in a physician's practice lifetime. Clinical and demographic data derived from malpractice claim reviews may provide useful information for planning strategies to decrease malpractice claims incidence. Malpractice claim files are not designed to be educational or research documents, but are functional tools used by the insurer for managing claims. Although the detailed information that clinicians may wish to glean from these files is often not

TABLE2. Closed Claims 1975-1993: Distribution of Claims and Cost of Claims No. of Claims Percentage of Total Claims Total Indemnity Total Expense Chest pain Wounds Fractures Abdominal pain* CNS bleeding Pediatric fever/meningitis Epiglottitis Total high-risk Non-high-risk All closed claims

57 109 97 55 20 8 4 350 199 549

10.38 19.85 17.67 10.02 3.64 1.46 0.73 63.75 36.25 100

*Abdominal pain claims include claims for abdominal aortic aneurysm.

9,020,355 665,483 6,002,277 1,605,695 2,120,000 1,368,675 844,295 21,626,780 11,223,755 32,850,535

954,492 570,114 831,036 578,432 323,312 237,512 36,266 3,531,164 2,787,192 6,318,356

Total I&E

Percentage of Total I&E

9,974,847 1,235,597 6,833,313 2,184,128 2,443,312 1,606,187 880,561 25,157,944 14,010,947 39,168,891

25.47 3.15 17.45 5.58 6.24 6.24 4.10 64.23 35.77 100

344

AMERICAN JOURNAL OF EMERGENCY MEDICINE • Volume 14, Number 4 • July 1996

TABLE 4. Closed Claims 1975-1993: High-Risk Categories by Indemnity Payment

All claims CWOP CWIP Total Wounds CWOP CWIP Total CNS bleeding CWOP CWIP Total Pediatric fever/meningitis CWOP CWIP Total Chest pain CWOP CWIP Total Fractures CWOP CWIP Total

No. of Claims

Percentage Claims

326 223 549

59,38 40.62 100.00

69 40 109

63.30 36.70

13 7 20

65.00 35.00

4 4 8

50.00 50.00

18 39 57

31.58 68.42

63 34 97

64.95 35.05

0 4 4

0.00 100.00

31 24 55

56.36 43.64

Epiglottitis CWOP CWlP Total Abdominal pain CWOP CWIP Total

present, review of closed claims may produce valuable clinical data. 10-13 Segmenting claims by whether or not an indemnity payment was made may be a useful indicator of severity of injury and/or negligence. Indemnity payment may be determined by expediency and practicality of settlement versus the cost of a trial and the risks of a jury verdict. However, the segmentation of claims by whether or not indemnity payment was made is traditionally used by the insurance industry as an indicator of claim severity. One study has reported a small but significant association of severity of injury with payment and found that defensibility of the case, ie, adherence to standards of care, predominantly influenced whether an indemnity payment was made. 14 In other words, even claims settled to avoid exposure to a potentially more expensive jury trial may indicate some deviation from practice standards. Claims closed without indemnity payment accounted for only 6.81% of the total indemnity and expense for all closed claims in our sample, even though the overall percentage of claims closed without indemnity payment was 59.38%. Our results are not dissimilar from other reports. The Physicians Insurance Association of America (PIAA) reports that 865 of 1,251 (69.14%) claims from 1985 through 1993 against emergency physicians were closed without indemnity payment and that these claims accounted for 13.6% of total

indemnity and expense. 15 Other sources have cited similar percentages of claims closed without indemnity payment. A study of New Jersey claims occurring from 1977 to 1991, with 162 emergency medicine cases, found that 61.1% of claims were closed without indemnity payment. 16 A large national sample of claims closed in 1984 reported that 75.5% of claims against emergency physicians were closed without indemnity payment.Iv The fact that there is no difference between high-risk and non-high-risk claims in average indemnity and expense is not surprising. The high-risk claims encompass a range of clinical diagnoses included as high-risk claims because of high frequency (despite relatively low cost per claim) or high severity (despite relatively low frequency). The highrisk group did have a higher percentage of cases closed with indemnity payment to the plaintiff; this means that a high-risk case is more likely to result in a successful claim by the plaintiff. The non-high-risk cases have a lower likelihood of successful claim. The largest payout for a single diagnostic category was missed MI claims, which accounted for 25.47% of total indemnity and expense while accounting for only 10.38% of all closed claims. Missed MI was documented as the leading dollar-loss category of claims closed between 1977 and 1987 in a national study, but accounted for only 8.3% of all ED claims. 3 A study of Ohio closed claims occurring between 1985 and 1988 found that missed MI accounted for 11% of all claims.IS The average indemnity and expense for missed MI claims in our sample was $174,997 (standard deviation of 198,116). A national insurer reported 176 claims for missed MI closed in 1991 and 1992, with an average cost per claim of $134,799.19 One study has reported an average insurance loss per claim for missed MI cases closed between 1981 and 1985 of $113,806 (standard deviation of 178,330)12; this study included 65 claims closed with indemnity payment or with legal fees of greater than $1,000. In our study, missed MI claims were the only high-risk diagnostic category (with more than just a few cases) where the percentage of claims closed with indemnity payment (68.42%) exceeded the percentage of claims closed without indemnity payment (31.58%). To our knowledge, only one previous study reported a CWIP percentage of 55% to 61.4% for 64 missed MI claims closed between 1977 and 19871°; this study reviewed missed MI claims alone, without the context of other diagnostic categories. On the basis of our data, we propose that the percentage of cases closed with indemnity payment is a marker for the seriousness of claims associated with a specific allegation. The average cost per claim or the average cost per claim closed with indemnity payment may be inflated because of a few high awards (see discussion of fractures, below). Additionally, as can be seen by the discussion above, sources variously report average cost per claim based on all claims, claims closed with indemnity payment, or, as cited above, claims closed without indemnity payment plus those with legal fees exceeding $1,000. Comparing different studies is difficult, with additional confounding factors of claims closed in different time periods, where inflation may affect the comparisons of average cost per claim, and various state legal

KARCZ ET AL • MASSACHUSETTS MALPRACTICE CLAIMS 1975-1993

constraints on awards, which also may affect average cost per claim. Claims closed without indemnity payment were significantly more frequent after 1988 than before 1988. High-risk claims accounted for a lower percentage of total closed claims after 1988 compared to claims closed before 1988. This is largely accounted for by a decrease in the percentage of fracture and wound claims. Missed MI claims increased in frequency from the pre-1998 to the post-1988 period. There was no change in the percentage of abdominal pain closed claims between the pre-1988 and post-1988 closed claims in our study. This finding is of particular interest given changes in clinical management of abdominal pain over the years. Admission rates for abdominal pain have decreased from 27.4% of all emergency department (ED) cases in 1972 to 18.3% in 1993; ED volumes have increased dramatically over that time period, with patients with abdominal complaints remaining at about 5% of the ED population, z° Even though these conclusions regarding abdominal pain demographics cannot be generalized to all EDs, this data suggests that within the last 20 years many more cases of abdominal pain are being evaluated in EDs, with far fewer admissions and with no change in the incidence of closed claims. Historically, missed fracture claims have been frequent occurrences with a low cost per claim. Our first study of Massachusetts claims closed between 1980 and 1987 exhibited a mean cost per claim of $25,510 (standard deviation of 75,661). 1 However, since that time there have been three closed claims resulting in indemnity payments of $1 million for missed cervical or thoracic spine fractures resulting in paralysis. Because of these three high-dollar-loss claims, the average cost per fracture claim in the current study population of 97 closed claims was $70,447 (standard deviation of 202,543); the average cost per claim with indemnity payment was $189,640 (standard deviation of 307,904). Eighteen of the 97 closed claims (18.6%) were spine fractures, with an average cost per claim of $237,564 (standard deviation of 383,136); 15 of 97 closed claims (15.5%) were related to hip fractures, with an average cost per claim of $66,693 (standard deviation of 121,015). Thirty-five of the 109 wound claims (32.1%) involved retained foreign bodies, and 37 (33.9%) alleged missed nerve and/or tendon injury. CONCLUSIONS

Claims closed without indemnity payment were significantly more frequent after 1988 than before 1988. High-risk claims accounted for a lower percentage of total closed claims during 1988 through 1993 compared to claims closed before 1988. This is largely accounted for by a decrease in the percentage of fracture and wound claims. Missed MI claims increased in frequency from the pre-1988 to the post-1988 group. Missed M1 is the major dollar-loss category of claims filed against emergency physicians, accounting for one quarter of all money spent on claims. Missed MI was the only diagnostic category with a larger percentage of claims

345

closed with indemnity payment than without indemnity payment. We propose that the percentage of claims closed with indemnity payment is a marker for the seriousness of claims associated with a specific allegation. Average cost per claim is a useful parameter but is more subject to influence by a few cases with particularly bad outcomes and subject to other methodological drawbacks. The authors thank Maureen Mender, the Director of Risk Management at ProMutual, for facilitating our study, and Barbara Ricci and Elizabeth Kelly of the Risk Management Department Staff for their invaluable assistance with claim file review and data entry. REFERENCES 1. Karcz A, Holbrook J, Auerbach BS, et ah Preventability of malpractice claims in emergency medicine: A closed claims study. Ann Emerg Med 1990;19:865-873 2. Karcz A, Holbrook J, Burke MC, et al: Massachusetts emergency medicine closed malpractice claims: 1988-1990. Ann Emerg Med 1993;22:553-559 3. American College of Emergency Physicians: Comprehensive Guide to Effective Practice Management. Dallas, TX, ACEP, 1986 4. DeLew N, Greenberg G, Kinchen K: A layman's guide to the U.S. health care system. Health Care Financing Review 1992;14:151169 5. Reynolds RA, Rizzo JA, Gonzalez ML: The cost of medical professional liability. JAMA 1987;257:2776-2781 6. Charles SC, Wilbert JR, Kennedy EC: Physicians' self-reports of reactions to malpractice litigation. Am J Psychiatry 1984;141:563565 7. Charles SC, Wilbert J, Franke KJ: Sued and non-sued physicians' self-reported reactions to malpractice litigation. Am J Psychiatry 1985;142:437-440 8. Shapiro RS, Simpson DE, Lawrence SL, et al: A survey of sued and non-sued physicians and suing patients. Arch Intern Med 1989;149:2190-2196 9~ Martin CA, Wilson JF, Fiebelman ND III, et al: Physicians' psychologic reactions to malpractice litigation. South Med J 1991;84: 1300-1304 10. Pelberg AL: Missed myocardial infarction in the emergency room. J Am Coil Utilization Rev Phys 1989;4:39-41 11. Trautlein JJ, Lambert RL, Miller J: Malpractice in the emergency department--Review of 200 cases. Ann Emerg Med 1984;13: 709-711 12. Rusnak RA, Stair TO, Hansen K, et al: Litigation against the emergency physician: Common features in cases of missed myocardial infarction. Ann Emerg Med 1989; 18:1029-1934 13. Rusnak RA, Borer JM, Fastow JS: Misdiagnosis of acute appendicitis: Common features discovered in cases after litigation. Am J Emerg Med 1994;12:397-402 14. Taragin MI, Willett LR, Wilczek AP, et al: The influence of standard of care and severity of injury on the resolution of medical malpractice claims. Ann Intern Med 1992;117:789-784 15. Physicians Insurance Association of America: Data Sharing System, June 1995 16. Taragin MI, Sonnenberg FA, Karns ME, et al: Does physician performance explain interspecialty differences in malpractice claim rates. Med Care 1994;32:661-667 17. US GeneralAccounting Office: Medical malpractice: Characteristics of claims closed in 1984. Washington DC, US General Accounting Office, 1987, HRD-87-55 18. Sites RL: Emergency room closed malpractice claims (from 1985-1988 closed claims involving Ohio hospitals). Columbus, OH, Ohio Hospital Association December 29, 1989, Bulletin 89-99-A 19. 1993 Annual Report to Policyholders: Physicians and Surgeons. St. Paul, MN, St. Paul Fire and Marine Insurance Company, 1993 20. Powers RD, Guertler AT: Abdominal pain in the ED: Stability and change over 20 years. Am J Emerg Med 1995; 13:301-303