Doctors With Multiple Malpractice Claims, Disciplinary Actions, and Complaints: What Do We Know?
Lightning never strikes the same place twice. This familiar proverb suggests that when an uncommon event occurs once, it is unlikely to do so again—even less likely than its initial occurrence. For many physical and social phenomena, however, the opposite is true. From earthquakes to bankruptcy to school attendance and criminal activity (and, yes, even lightning strike points), a more apt proverb would be past as prelude. Does physicians' involvement in medicolegal events—unsolicited patient complaints, disciplinary action by medical boards, and malpractice claims—belong on this list?
Physicians often complain that medicolegal events strike randomly and arbitrarily. But there are well-documented risk factors, both on the physician side (e.g., communication skill  and male gender ) and the patient side (e.g., serious adverse outcomes of care ). Whether prior medicolegal events constitute a risk factor has been the subject of some debate. The debate flared in the early 1990s with the launch of the National Practitioner Data Bank, the federal repository of information on paid malpractice claims and disciplinary actions against health practitioners.(4-7)
A founding premise of the National Practitioner Data Bank was that arming health care institutions with information about the medicolegal histories of their affiliated physicians would help improve the quality and safety of care. But if those histories were an unreliable indicator of current and future medicolegal risk, the premise was questionable. Commentators from within the liability industry highlighted the fact that meritless malpractice claims were often settled when the costs of defending them were likely to exceed their payment (or "payoff") value.(4,7,8) Proponents of the National Practitioner Data Bank, on the other hand, called for its expansion beyond paid claims to include information on matters proceeding through the litigation process.(8)
A close reading of the literature from this period is illuminating. Skepticism centered on the predictive value of small malpractice payments, not medicolegal events in general. More importantly, a good deal of the disagreement and uncertainty over the predictive value of physicians' medicolegal event history appears to have stemmed from the conflation of two related but distinct questions: (i) Do prior medicolegal events signal elevated risk of future ones? And (ii) Is it feasible to accurately predict medicolegal events at the physician level in large groups of physicians? These are distinct questions because a characteristic can be strongly associated with an outcome, yet it still may be statistically infeasible to use the characteristic as a basis for prospectively identifying individuals in a population who will experience the outcome. A good example is the relationship between depression and suicide. People who die by suicide are much more likely to be depressed than people who do not. But it has not proven feasible to accurately predict suicide using information about patient histories of depression, largely because depression is prevalent in the population and suicide is very rare.
With respect to the first question posed above, there is compelling evidence that the risk of future medicolegal events is significantly higher among physicians with markedly worse medicolegal event track records than their peers.(5,9-11) Past isn't always prelude, but it's a useful marker of future risk.
The answer to the second question is less clear. A series of studies in the 1980s and early 1990s explored whether physicians' medicolegal event history could be combined with other readily observable traits, such as gender and specialty, to flag with reasonable accuracy physicians who would go on to experience malpractice claims.(12,13) Although the broad conclusion of this research was that reliable prediction of claims was not possible, several recent studies of patient complaints (14,15) and malpractice claims (9) use different statistical approaches and provide greater cause for optimism. These findings, coupled with the growing availability of data on large groups of physicians and the rapidly advancing field of machine learning, suggest that regulators, industry, and researchers should revisit the feasibility of medicolegal event prediction. Presently, however, liability insurers and medical regulators generally do not use individual-level predictive modeling to guide their work.
A corollary of the fact that prior medicolegal events are strongly and positively associated with risk of future medicolegal events is that these events tend to cluster in the physician workforce, with some physicians accounting for a disproportionately large share of all medicolegal events.(9-11,14,16) In fact, medicolegal events appear to be extremely maldistributed. For example, one recent study of paid malpractice claims in the United States between 2005 and 2014 found that 1% of all physicians accounted for one-third of all claims (14); this is remarkably similar to the level of concentration detected in a national study of patient complaints in Australia.(9)
What do we know about physicians who accumulate multiple medicolegal events? Many studies have profiled physicians who experience medicolegal events, typically by comparing them to peers with clean records. While it is tempting to extrapolate from these studies, it is important to recognize that physicians who accumulate unusually large numbers of medicolegal events are a small subset of all physicians who experience medicolegal events, and they may be distinctive. Surprisingly, few studies have focused explicitly on characteristics of medicolegal event–prone physicians. Available evidence suggests that, compared to physicians who experience no or few events, medicolegal event–prone physicians are more likely to be male, to be international medical graduates, and to work in certain high-risk specialties (e.g., orthopedic surgery, obstetrics and gynecology).(9,14,15,17) Findings are mixed regarding whether physician age is a risk factor.
But these are crude characterizations. Many fundamental questions about physicians with troubling medicolegal event records remain unanswered. One set of questions relates to their clinical practice trajectory. What proportion of medicolegal event–prone physicians continue to work, and where? Who credentials them? Who provides them with liability insurance? Are they more likely to work in solo practice, in areas of need, remain unaffiliated with a major hospital? Does their patient load dwindle as their medicolegal event track record grows?
Another set of questions relates to factors that drive recurrence. To what extent does a physician's accumulation of medicolegal events spring from a general weakness that cuts across multiple dimensions of care (e.g., poor communication skills, lack of insight), as opposed to more circumscribed shortcomings (e.g., lack of competence in a particular procedure)? And what does the periodicity of a typical medicolegal event series look like? A succession of medicolegal events accumulated in a short period of time may signal a different kind of quality problem (e.g., life crisis event) than a series accumulated steadily over an extended period. A physician-level study of event typologies and timing would be revealing.
Although these avenues of research have potential to advance understanding of the who, how, and why of physicians with multiple medicolegal events, making the link to quality improvement on the ground demands more. Medicolegal events are blunt measures; they could mark fairly isolated events. Alternatively, the may signal a much larger quality problem that afflicts many of the patients that medicolegal event–prone physicians treat. We need to know more about the extent to which medicolegal event–proneness correlates with other, more sensitive physician-level measures of quality and safety.(18)
Finally, a capacity to reliably identify physicians at high risk of medicolegal events would be a marvelous breakthrough: it has the potential to take medicolegal institutions off the sidelines of the quality improvement movement and cast them in a more central role in prevention. But prediction will have limited impact on quality and safety unless it can be linked to effective interventions. With a few notable exceptions (19), there is scant evidence about what works to restore medicolegal event–prone physicians to safe practice, or, when necessary, to shepherd them away from patient care.
David Studdert, LLB, ScD, MPH
Professor of Medicine and Law
Stanford University School of Medicine
Stanford Law School
1. Levinson W, Roter DL, Mullooly JP, Dull VT, Frankel RM. Physician–patient communication. The relationship with malpractice claims among primary care physicians and surgeons. JAMA. 1997;277:553-559. [go to PubMed]
8. National Practitioner Data Bank: Malpractice Reporting Requirements. Washington, DC: US Department of Health and Human Services, Office of Inspector General; 1992. Report No. OEI-OI-90-00521. [Available at]
9. Bismark MM, Spittal MJ, Gurrin LC, Ward M, Studdert DM. Identification of doctors at risk of recurrent complaints: a national study of healthcare complaints in Australia. BMJ Qual Saf. 2013;22:532-540. [go to PubMed]
15. Spittal MJ, Bismark MM, Studdert DM. The PRONE score: an algorithm for predicting doctors' risks of formal patient complaints using routinely collected administrative data. BMJ Qual Saf. 2015;24:360-368. [go to PubMed]
16. Schaffer AC, Jena AB, Seabury SA, Singh H, Chalasani V, Kachalia A. Rates and characteristics of paid malpractice claims among US physicians by specialty, 1992–2014. JAMA Intern Med. 2017;177:710-718. [go to PubMed]
19. Hickson GB, Pichert JW, Webb LE, Gabbe SG. A complementary approach to promoting professionalism: identifying, measuring, and addressing unprofessional behaviors. Acad Med. 2007;82:1040-1048. [go to PubMed]