Lessons in Liability: Examining Medical Malpractice Suits Against General Surgeons in Maryland

Abstract

Background:

Although medical malpractice lawsuits pose a significant burden, there is a paucity of research on physician-specific characteristics influencing lawsuits against surgeons. Our objective was to identify factors associated with general surgeons being named in malpractice cases.

Methods:

This was a cross sectional study of Maryland general surgeons, using malpractice data from a publicly accessible judiciary database. Case number per decade and lifetime lawsuit status were modeled with linear and logistic regression.

Results:

Male surgeons had a higher average lawsuit volume (p=0.002) and were more likely to be named in a malpractice case (p<0.001). In regression analysis, a second graduate degree was a predictor of average cases per 10 years (p=0.008) and male gender predicted lifetime lawsuit status (OR=1.73, p=0.046).

Conclusions:

Male gender was associated with increased odds of being named in a malpractice lawsuit. Identifying this difference is a preliminary step in developing interventions to reduce lawsuits amongst surgeons.

Introduction

Medical malpractice is defined in the state of Maryland as “an action for damages for an injury arising out of the rendering of or failure to render professional services by a health care provider.”¹ Although malpractice suits were designed to mitigate lapses in patient safety, the costs associated with medical malpractice lawsuits account for a considerable portion of yearly healthcare spending in the U.S. The Harvard University School of Public Health² estimates an annual cost of medical malpractice suits approaching $55.6 billion, accounting for 2.4% of all American healthcare spending. The cost attributed to practicing defensive medicine, which is the change in physician behavior in response to the threat of a lawsuit, is estimated at an even more staggering $60–108 billion.³ On top of the financial burden of malpractice suits, the mental and physical impacts these suits have on physicians are also considerable. For instance, studies have shown that physicians who have recently been defendants in malpractice suits are more likely to experience symptoms of burnout, depression, and suicidal ideation.^4,5 This constellation of symptoms has become common enough that it is termed Medical Malpractice Stress Syndrome, which has been linked to cardiovascular complications and other health problems.⁶

Previous studies have demonstrated that the incidence of malpractice suits differs based on physician specialty, with surgical specialties facing more claims than non-surgical specialties.^7,8 A 2016 survey conducted by the American Medical Association found that 34% of all US physicians have been sued for malpractice at least once, with the greatest incidence among general surgery and obstetrics/gynecology (63% each).⁸ Prior work investigating the impact of physician demographics on the likelihood of lawsuits has had conflicting results; some studies have found that male physicians are more likely to be sued^9–11, while others showed no statistically significant difference by sex.^12–14 Additionally, studies have evaluated the impact of graduating from a foreign medical school¹⁵ and age^7,11,16 on medicolegal action, with mixed findings regarding whether each is a protective or risk factor. Further work is needed to clarify the impact of demographic characteristics on malpractice trends amongst surgeons.

The purpose of our study was to characterize demographic and training factors associated with being named as a defendant in medical malpractice suits. We hypothesized that reported gender of the surgeon would be associated with lawsuit occurrence. Better understanding the factors involved in malpractice cases may help in identifying the reasons for any discrepancies in lawsuit trends and in developing targeted interventions to prevent lawsuits and their downstream sequelae.

Methods

Data Source and Study Population

This is a cross-sectional study performed using a database generated to capture all general surgeons practicing in Maryland. The study population was composed of members of the American College of Surgeons (ACS) online “Find a Surgeon” database who were listed in the “general surgeon” or “colorectal” specialty categories, whose practicing address was within Maryland, and who specified being of either female or male gender. This group approximates the population of general surgeons in Maryland, given that most practicing general surgeons subscribe to ACS membership.¹⁷

We collected additional information about each surgeon using a combination of the ACS registry and other online sources, including Doximity¹⁸ and the surgeon’s current hospital affiliation website. Medical school rankings were based on the U.S. News and World Report Medical School Research ranking¹⁹ and residency rankings were based on the Doximity Residency Navigator search by reputation. All surgeons identified on the ACS website were included in the initial cohort for data collection. Surgeons were excluded from the final cohort if it was not possible to determine how long they practiced in Maryland, how long ago they completed their highest level of training, or where they attended medical school or residency.

We used the publicly accessible Maryland Judiciary Case Search to obtain data on malpractice suits. This tool provides online access to case records filed in the state of Maryland, including traffic, criminal, and civil case records at the District and Circuit Court levels. The database was established in 2006 and is comprised of cases dating back to as early as 1978 or as late as 1999, depending on the county. It does not contain any protected records or records that have been expunged. All cases in which a surgeon was listed as the defendant and that were either labeled as “medical malpractice” or included details describing the case as a malpractice lawsuit were included.

Three investigators collected the surgeon data and two investigators collected the malpractice lawsuit data. Inter-observer reliability was ensured by selecting a small subset of the study population and repeating data collection amongst all investigators to confirm consistent results. Collection of malpractice data occurred over a four-week time period from November-December 2021 to preserve the cross-sectional study design.

Predictor and Outcome Variables

Surgeon demographics (gender, date of birth), training information (degree type, second degree, international medical graduate, medical school name and ranking, residency name and ranking, year of completion of highest level of training, fellowship type) and practice information (current hospital information, years of practice in Maryland, academic rank, and academic institution rank, if applicable) were collected. Number of malpractice lawsuits each surgeon was named in, as well as filing date, case status, and case disposition for each suit, were also obtained. To account for years of practice, we derived our primary outcome variable—cases per 10 years—by dividing the number of cases for each surgeon by the number of years in which they practiced in Maryland and multiplying by 10. We chose to examine cases per 10 years rather than an annual measure because of the overall low rate of lawsuits and to avoid numbers that were too small for reasonable interpretation. An additional outcome variable was a binary indicator of lifetime lawsuit status, indicating whether the surgeon had ever been sued in their career.

Statistical Analysis

We used descriptive statistics to identify demographic, training, and practice characteristics of the entire study population. Characteristics were compared across gender cohorts using a Wilcoxon Rank Sum test for continuous variables and a Chi Squared test for categorical variables. We performed an unadjusted analysis comparing gender and lawsuits using a Wilcoxon Rank Sum test for cases per 10 years and a Chi Squared test for lifetime lawsuit status. A multiple linear regression (MLR) model was created for the cases per 10 years outcome metric, which included all variables that were statistically significantly between genders on bivariate analysis. A second MLR model was then produced only including the variables that were significant on initial regression, with the model being reduced in size by backwards selection. We also created a multiple logistic regression model, which used our binary outcome variable of presence or absence of any career lawsuits as the response variable. An additional reduced model was produced using backwards selection for our logistic regression. All statistical analyses were conducted using R Studio Statistical Software (version 1.4.1717).²⁰

Results

586 surgeons were identified from the ACS website, and 92 surgeons were excluded from the study due to inability to obtain information on their medical school and residency programs, highest level of training, or length of practice. The final study cohort consisted of 351 male and 143 female surgeons practicing in general or colorectal surgery in Maryland. The comparisons of demographic characteristics based on gender can be found in Table 1. Female surgeons were more likely to have a second graduate degree (24.5% vs. 11.4%; P < 0.001), less likely to have received their medical education outside of the United States (4.9% vs. 29.6%; P < 0.001), more likely to have completed a fellowship (72.0% vs. 49.0%; P < 0.001), and had been practicing for fewer years (13.0 vs 24.8 years, P < 0.001). For the entire cohort, the average number of medical malpractice cases for a surgeon was 0.672, with male and female surgeons having been the defendant in cases an average of 0.803 and 0.350 times, respectively.

Table 1.

Demographic characteristics of general and colorectal surgeons in Maryland registered as fellows of the American College of Surgeons

Demographic Characteristics	Male (n=351)		Female (n=143)		P value*
	Mean (SD)	# (%)	Mean (SD)	# (%)
Cases Per 10 Years	0.401 (0.93)		0.298 (0.79)		0.002
Lifetime Cases † , Absolute	0.803 (1.59)		0.350 (1.10)		<0.001
Surgeons with ≥1 lifetime case, median (minimum, maximum)	1 (1, 12)		1 (1, 10)
Lifetime Cases † , Binary					<0.001
Yes		129 (36.8)		28 (19.6)
No		222 (63.2)		115 (80.4)
MD Degree					0.415
Yes		325 (92.6)		136 (95.1)
No		26 (7.4)		7 (4.9)
Second Degree					<0.001
Yes		40 (11.4)		35 (24.5)
No		311 (88.6)		108 (75.5)
International					<0.001
Yes		104 (29.6)		7 (4.9)
No		247 (70.4)		136 (95.1)
Fellowship					<0.001
No Fellowship Training		179 (51.0)		40 (28.0)
Completed Fellowship Training		172 (49.0)		103 (72.0)
Years Since Completion of Highest Level of Training	24.78 (13.57)		13.02 (10.04)		<0.001
Years Practiced in Maryland	21.12 (13.95)		10.52 (9.48)		<0.001
Medical School Rank					<0.001
Unranked		139 (39.6)		28 (19.6)
Ranked
1–10		31 (8.8)		10 (7.0)
11–25		21 (6.0)		11 (7.7)
26–50		72 (20.5)		44 (30.8)
51+		88 (25.1)		50 (35.0)
Residency Reputation Rank					<0.001
Unranked		80 (22.8)		8 (5.6)
Ranked
1–10		35 (10.0)		15 (10.5)
11–50		93 (26.5)		39 (27.3)
51–100		36 (10.3)		31 (21.7)
101–150		37 (10.5)		7 (4.9)
151–200		47 (13.4)		34 (23.8)
201+		23 (6.6)		9 (6.3)

^*P values of continuous and categorical variables correspond to Wilcoxon rank sum tests and chi-squared tests, respectively

^†Lifetime cases defined as whether a surgeon had been involved in one or more malpractice cases during their career practicing in Maryland

The average number of malpractice cases per 10 years was compared between gender cohorts, with female surgeons having been named in fewer cases than men (0.298 vs. 0.401 cases/10 years, P = 0.002). When comparing lawsuits as a binary variable, there was a statistically significant difference between women and men, with 36.8% of male surgeons having been sued at least once, compared to 19.6% of female surgeons (P < 0.001).

The two cohorts were then compared leveraging multiple linear regression, the results of which are shown in Table 2. For the initial model, the significant predictors were having a second graduate degree (P = 0.008) and residency ranking of 151–200 (P = 0.017), with gender nearly reaching statistical significance (P = 0.062). In the reduced model (also shown in Table 2) derived from backwards selection, gender became a significant predictor (P = 0.038). Based on this analysis, male surgeons, all else being equal, had 0.189 more malpractice cases per decade than female surgeons. Additionally, those with second graduate degrees had a higher number of cases per decade than those without by 0.31.

Table 2.

Initial and reduced multiple linear regression models for malpractice cases per 10 years associated with demographic and training factors in Maryland general surgeons

	Initial Model			Reduced Model
Characteristic	β	95% CI	P-Value	β	95% CI	P-Value
Intercept	0.252	(−0.17, 0.67)	0.240	0.127	(−0.04, 0.29)	0.128
Gender
Female	(Reference)	---	---	(Reference)	---	---
Male	0.189	(−0.01, 0.39)	0.062	0.185	(0.01, 0.36)	0.038
Second Degree
No	(Reference)	---	---	(Reference)	---	---
Yes	0.311	(0.08, 0.54)	0.008	0.300	(0.08, 0.52)	0.007
International
No	(Reference)	---	---
Yes	0.160	(−0.13, 0.45)	0.276
Fellowship
No	(Reference)	---	---
Yes	−0.089	(−0.27, 0.09)	0.337
Years Since Completion of Training	−0.002	(−0.01, 0.00)	0.517
Medical School Rank
1–10	(Reference)	---	---
11–25	−0.171	(−0.59, 0.24)	0.416
26–50	−0.036	(−0.37, 0.30)	0.834
51+	0.002	(−0.33, 0.34)	0.990
Unranked	−0.187	(−0.56, 0.19)	0.331
Residency Reputation Rank *
1–10	(Reference)	---	---
11–50	0.006	(−0.30, 0.31)	0.971
51–100	−0.002	(−0.34, 0.34)	0.989
101–150	0.144	(−0.24, 0.53)	0.458
151–200	0.415	(0.07, 0.76)	0.017	0.411	(0.20, 0.62)	<0.001
201+	0.215	(−0.21, 0.64)	0.316
Unranked	−0.072	(−0.41, 0.26)	0.674

Predictors removed by backwards selection.

^*Residency reputation rank included in the final model is a binary predictor for a residency rank between 151 and 200.

Finally, we compared the two cohorts using a multiple logistic regression model, the results of which are shown in Table 3. The significant predictors in the initial model were gender (P = 0.046) and the number of years the physician had practiced in Maryland (P < 0.001). In the reduced model, gender again trended towards significance (P = 0.095), but only the number of years the surgeon had practiced in Maryland was a significant predictor of an increased odds of ever being named in a medical malpractice case (P < 0.001).

Table 3.

Initial and reduced multiple logistic regression models of lifetime malpractice cases associated with demographic and training factors in Maryland general surgeons

Characteristic	Initial Model			Reduced Model
	OR Estimate	95% CI	P-Value	OR Estimate	95% CI	P-Value
Intercept	0.143	(0.05, 0.42)	<0.001	0.149	(0.09, 0.23)	<0.001
Gender
Female	(Reference)	---	---	(Reference)	---	---
Male	1.730	(1.02, 2.99)	0.046	1.534	(0.94, 2.56)	0.095
Second Degree
No	(Reference)	---	---
Yes	1.439	(0.77, 2.64)	0.244
International
No	(Reference)	---	---
Yes	1.281	(0.60, 2.86)	0.533
Fellowship
No	(Reference)	---	---
Yes	0.956	(0.60, 1.53)	0.848
Years Practiced in Maryland	1.047	(1.03, 1.07)	<0.001	1.043	(1.03, 1.06)	<0.001
Medical School Rank
1–10	(Reference)	---	---
11–25	0.675	(0.21, 2.08)	0.499
26–50	1.060	(0.45, 2.62)	0.896
51+	1.153	(0.49, 2.85)	0.752
Unranked	0.598	(0.21, 1.68)	0.326
Residency Reputation Rank
1–10	(Reference)	---	---
11–50	0.742	(0.33, 1.67)	0.464
51–100	0.830	(0.33, 2.07)	0.689
101–150	1.489	(0.57, 3.92)	0.416
151–200	1.138	(0.48, 2.78)	0.774
201+	1.554	(0.53, 4.56)	0.420
Unranked	0.723	(0.31, 1.73)	0.461

Predictors removed by backwards selection

Discussion

In this study, we sought to understand the physician-specific factors that were predictive of a general surgeon having a lawsuit filed against them. We did so by developing a database of surgeons in Maryland with an ACS membership, allowing comparison of malpractice suits based on a variety of demographic factors. Significant findings included the differential rate of lawsuit filing by gender, with male surgeons being more likely than female surgeons to have been named in a lawsuit and to be named in more yearly lawsuits. This finding was robust when controlling for the number of years in practice. Additionally, we found that surgeons with a second graduate degree had a greater number of lawsuits filed against them.

The gender difference in malpractice suits was our main topic of inquiry, and when controlling for years in practice, male gender remained a significant predictor of lawsuits against surgeons. This finding is consistent with prior research on medical malpractice suits, which has generally shown that male physicians are more likely to be sued. This discrepancy may be, in part, explained by the physician-specific qualities that influence a patient to file a malpractice lawsuit. Many studies have suggested that female physicians are more empathetic²¹ and have higher rates of patient-centered communication.^22,23 Patients are less likely to report medical errors when physicians express empathy and apologize for their mistakes^24–26; thus, an empathetic physician may be less likely to be sued. In recent years, medical schools have had an increased focus on training students in the humanistic aspects of patient care through efforts like patient simulations or didactic courses dedicated to improving patient care.²⁷ As female surgeons in our cohort completed their training more recently than male surgeons, this could contribute to the differences observed between groups. Continuing to promote such curricula and targeting them towards groups at greater risk for lawsuits may serve to decrease malpractice occurrence.

Alternatively, lawsuit filing may be based primarily on the medical facts of a given case. Again, female physicians display behaviors that may lend themselves to superior medical care and therefore fewer malpractice cases, including the tendency of female physicians to follow guideline recommendations more often than their male counterparts,^28,29 to spend more time with their patients and collect more information in their histories,^30,31 and to provide preventative care more frequently.^32,33 Some studies also suggest that female surgeons tend to have better outcomes than male surgeons.^29,34 Given these facts, it is possible that the observed discrepancy in lawsuit number is more related to better patient outcomes amongst certain surgeons.

On multivariable analysis, physician-level factors predictive of higher lawsuit volume also included possessing a second graduate degree. Within our study, the most common second graduate degrees were an MBA, MPH, MS, or PhD, and female surgeons were significantly more likely to have a second degree compared to their male counterparts. Given the additional years of training and expense associated with these degrees, holding multiple degrees may indicate that the surgeon has extra-clinical pursuits, including administration, policy, and research activities. Given the well-established volume-outcome relationships in surgery,^35,36 a potential explanation for the increased number of lawsuits among surgeons with multiple degrees is that extra-clinical interests prevent surgeons from reaching desired volume thresholds and thus impair outcomes. Interestingly, though, that theory would contradict a previous finding from Seaburg et al. (2016),³⁷ which found that an increased number of publications positively, though marginally, impacted clinical performance. Additionally, holding multiple degrees may be associated with clinical practice in different settings, such as academic versus private hospitals, which could mediate the relationship between second degrees and lawsuit volume. Unfortunately, the sample size in our study prohibited further exploration into these relationships, including the differential impact of individual graduate degree types on likelihood of lawsuits.

Compared to prior work, our study is strengthened by our unique approach to data collection. In creating a comprehensive database of Maryland surgeons, we captured a thorough cross-sectional snapshot of malpractice suits in Maryland. Previous studies have analyzed lawsuit data using resources such as Westlaw, which contain only a selection of relevant case law on a particular topic, or with survey data, which is limited by self-report and response bias. Through our use of the Maryland Judiciary Case Search, we better identified all lawsuits occurring in the fixed time period since the database was created.

This study has some limitations. First, the dataset was limited to surgeons from the state of Maryland with an active ACS membership who identified as general or colorectal surgeons, which reduces the generalizability of our findings. We selected only general and colorectal surgeons for this study as they were the specialty categories available within the ACS database that would include only surgeons trained in a general surgery residency program; this allowed us to compare surgeons who had undergone similar training and might have similar patterns of lawsuits. Second, we only evaluated lawsuit number and not lawsuit outcome. As a result of ongoing litigation, incomplete or unclear records, and settlements outside of trial, the details of the outcomes of lawsuits, and therefore the relative merits of the suits being brought against different surgeons, could not be assessed using our database. Thus, in our study, lawsuits that were quickly dismissed or in which the physician was found to be not guilty were treated the same as cases in which the physician demonstrated gross negligence. As a result, our analysis could not differentiate between legitimate or frivolous lawsuits. This is especially important given the fact that recent work has shown that almost 60% of all medical malpractice suits are abandoned by the plaintiff³⁸ and as many as 70% of cases end with the patient recovering no damages.³⁹ Finally, because our study excluded surgeons whose pertinent demographic and training information was not identifiable online, we likely excluded a non-random group of physicians with a diminished internet presence, particularly older surgeons. As risk of being sued increases with age⁴⁰, we may be underestimating lawsuit rates and thus biasing our results towards the null.

Future work could expand on our findings. It would be useful to examine a broader range of surgical specialties, including fields such as obstetrics and gynecology that have a higher proportion of female providers, to understand how these findings differ amongst types of surgeons. It will also be important to conduct additional studies that can assess the legitimacy and outcomes of lawsuits and whether those findings alter the results seen in this study. Considering the other factors involved in malpractice cases, including patient and structural factors, would further bolster future research on this topic.

Conclusions

Male surgeons were more likely to have been named in a malpractice lawsuit than female surgeons and were named in more lawsuits, even when standardizing by the number of years in practice. Surgeons with a second graduate degree were also more likely to have been named in a lawsuit. Identifying these physician-specific factors may aid in the development of interventions to mitigate lawsuit occurrence among surgeons.

• This study used a statewide judiciary database to quantify malpractice cases

• Male surgeons were more likely to be named in malpractice lawsuits

• A second graduate degree predicted more yearly malpractice cases