Malpractice Concerns, Defensive Medicine, and the Histopathology Diagnosis of Melanocytic Skin Lesions

Linda J. Titus, PhD; Lisa M. Reisch, PhD; Anna N. A. Tosteson, ScD; Heidi D. Nelson, MD, MPH; Paul D. Frederick, MPH, MBA; Patricia A. Carney, PhD; Raymond L. Barnhill, MD, MSc; David E. Elder, MBChB, FRCPA; Martin A. Weinstock, MD, PhD; Michael W. Piepkorn, MD, PhD; Joann G. Elmore, MD, MPH

Disclosures

Am J Clin Pathol. 2018;150(4):338-345. 

In This Article

Results

As shown in Table 1, of the 207 participating pathologists, 57 (27.5%) reported they were influenced toward a more severe MSL diagnosis by concerns about medical malpractice, and 150 (72.5%) were not influenced toward a more severe diagnosis.

Results of Unadjusted Logistic Regression Models

Pathologists' demographic and clinical characteristics are shown in Table 1. Most pathologists (54.1%) were at least 50 years of age and male (59.4%). Most had residency training in anatomic or clinical pathology (89.9%). Fewer than half were fellowship trained or board certified in dermatopathology (39.1%) or had 20 or more years of experience interpreting MSL (30.9%). Most reported reviewing 35 or more MSLs per month (63.3%), and nearly all assessed some of their MSL caseload as borderline (89.8%). A minority (5.8%) had been named in an MSL-related malpractice lawsuit.

The results of unadjusted logistic regression models suggested that pathologists with dermatology residency training, compared with pathology residency, were more likely to render a more severe MSL diagnosis due to concern about medical malpractice (odds ratio [OR], 2.17; 95% confidence interval [CI], 0.86–5.47). Two variables, older age and more years of experience interpreting MSL, strongly reduced the likelihood of rendering a more severe diagnosis due to malpractice concerns; the unadjusted ORs were 0.42 (95% CI, 0.23–0.79) for age 50 years or older, compared with a younger age, and 0.24 (95% CI, 0.10–0.56) for at least 20 years of MSL experience, compared with fewer years. Age and years of MSL interpretative experience were highly correlated (r = 0.72; P < .001), precluding simultaneous multivariable assessment; we elected years of experience for inclusion in the multivariable model due to its greater clinical relevance. The remaining pathologist characteristics, including sex, two variables representing the volume of MSL caseload, assessing lesions as borderline, dermatopathology fellowship training and/or board certification, and having been named previously in an MSL-related lawsuit, were not associated (P > .10) with rendering a more severe diagnosis due to malpractice concerns.

Survey items assessing pathologists' perceptions about interpreting MSL are summarized in Table 2. Most pathologists (69.6%) found interpreting MSLs enjoyable, and nearly all (96.1%) found MSL lesions challenging. Most pathologists reported being more nervous interpreting MSLs than other pathology (70.5%), were concerned about patient safety and potential harm to patients resulting from their MSL assessments (74.9%), and were confident in their MSL assessments (86%). Fewer than half (40.6%) believed too many MSL were biopsied, and slightly more than half (52.7%) thought pathologists were overcalling some benign lesions as melanoma. Most pathologists agreed that obtaining a second opinion protected them from lawsuits (86.0%).

The results of unadjusted logistic regression models suggested that being more nervous about interpreting MSLs (OR, 2.03; 95% CI, 0.98–4.23), concern about patient safety and harm (OR, 2.07; 95% CI, 0.94–4.53), and agreeing that too many skin lesions are biopsied (OR, 1.78; 95% CI, 0.96–3.29) were positively associated with being influenced toward a more severe MSL diagnosis by malpractice concerns at P ≤ .10. Due to a strong correlation between being more nervous about interpreting MSLs and concern about patient safety/harm (r = 0.38; P < .001), precluding simultaneous assessment of these variables, we elected concern about patient safety/harm for inclusion in the multivariable model due to its greater clinical significance and relevance to medical malpractice. The remaining variables, including pathologists' perceptions that interpreting MSLs is enjoyable and challenging, overcalling benign lesions as melanoma, pathologists' confidence when assessing MSLs, and perceiving second opinions as protecting against malpractice lawsuits, were not associated (P > .10) with the outcome.

Assurance behaviors practiced to reduce concerns about malpractice when interpreting MSL are summarized in Table 3. Due to concerns about malpractice, 87.0% of pathologists reported they requested additional slides, 66.2% requested additional surgical sampling, 92.3% requested a second opinion, and 62.3% ordered additional molecular testing. Nearly all, 95.2%, of pathologists reported practicing at least one of the four measured assurance behaviors.

In unadjusted logistic regression models, three of the four assurance practices were associated (P ≤ .10) with rendering a more severe MSL diagnosis due to malpractice concerns; the ORs were 8.02 (95% CI, 1.46–44.11) for requesting additional slides cut from the block, 2.03 (95% CI, 1.01–4.07) for recommending additional surgical sampling, and 4.31 (95% CI, 0.74–24.92) for requesting second opinions. Ordering additional tests was not associated (P > .10) with the outcome. The composite assurance variable in which "strongly agree" was compared with all other response categories was significantly associated with being influenced toward a more severe MSL diagnosis by malpractice concerns (OR, 2.76; 95% CI, 1.47–5.15).

Figure 1 displays the proportion of pathologists whose MSL diagnoses were influenced (or not) by malpractice concerns according to their use (yes) or nonuse (no) of assurance behaviors. As shown, for each of the four individual assurance behaviors, the proportion of pathologists reporting that their MSL diagnoses were influenced toward a more severe diagnosis was consistently higher among those who practice the assurance behavior than among those who do not. The difference in proportions was greatest for requesting additional slides from the tissue block; influence toward a more severe MSL diagnosis due to medical malpractice concerns was reported by 3.7% of those who do not request additional slides due to concerns about medical malpractice and by 31.1% of those who do request additional slides.

Figure 1.

Pathologists' use of assurance behaviors (no/yes) due to malpractice concerns and the influence of malpractice concerns on the direction of their melanocytic skin lesion diagnosis, based on self-report (n = 207). aIncludes three pathologists who reported influence toward a less severe diagnosis. b"No" is defined by Likert responses "slightly disagree," "disagree," and "strongly disagree." "Yes" is defined by Likert responses "strongly agree," "agree," and "slightly agree."

Results of the Multivariable Logistic Regression Model

To control for possible confounding and identify pathologist characteristics that were independently related to rendering a more severe diagnosis, variables associated with the outcome at P ≤ .10 in simple, unadjusted logistic models were entered into a multivariable model Table 4. These variables included dermatology residency training, years of experience interpreting MSLs (≥20 vs <20), concern about patient safety/harm that might result from their assessment of MSLs, being of the opinion that too many MSLs are biopsied, and the composite variable representing the use of at least one assurance behavior.

The results of the multivariable model indicated that pathologists with at least 20 years of experience interpreting MSLs, compared to those with less than 20 years' experience, were substantially less likely to report rendering a more severe MSL diagnosis due to concerns about medical malpractice (OR, 0.29; 95% CI, 0.12–0.69; P = .005) (Table 4). In addition, pathologists who strongly agreed that they practiced at least one of four assurance behaviors, compared with those who did not, were nearly three times as likely to report being influenced toward a more severe MSL diagnosis (OR, 2.72; 95% CI, 1.41–5.26; P = .003). Other factors that were associated with rendering a more severe MSL diagnosis due to malpractice concerns in the simple, unadjusted models were not associated with the outcome when mutually adjusted.

Finally, due to our special interest in malpractice, we conducted a separate multivariable analysis to assess past experience with MSL malpractice litigation in relation to upgrading to a more severe diagnosis due to malpractice concerns. In this analysis, experience with MSL-related malpractice litigation was compared with a reference group containing those without malpractice experience or with malpractice experience unrelated to MSL. After adjustment for the same variables described above, excluding years of MSL experience, with which malpractice involvement was correlated (r = 0.19; P = .006), the OR was 0.82 (95% CI, 0.21–3.25) for the association between past malpractice experience and upgrading to a more severe MSL diagnosis.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....