Clinician personality and the evaluation of higher-risk patient symptoms

Abstract

Background: : Acute myocardial infarction, breast cancer, and colorectal cancer are among the most commonly misdiagnosed conditions in primary care, and there is little information regarding physician decision making in this area.

Methods: : We measured clinician risk tolerance and stress from uncertainty among 193 primary care clinicians and determined their association with management of patients presenting with chest pain (n = 700), breast lumps (n = 630), or rectal bleeding (n = 470). We used the 6-item Jackson Personality Inventory (JPI) and the 13-item Stress from Uncertainty Scale (SUS). Extended evaluation was defined as performance of electrocardiogram, cardiac stress testing, or emergency department triage for patients with chest pain; mammography, ultrasound, or surgical referral for patients with breast lumps; and colonoscopy or gastroenterology referral for patients with rectal bleeding. We fit multivariable logistic regression models to determine the association between survey scores and performance of extended evaluation.

Results: : The response rate was 79% for the JPI survey and 73% for the SUS survey. The mean JPI score was 8.9 (SD, 4.5) and the mean SUS score was 32.6 (SD, 10.4). Extended evaluation occurred for 60% of patients with chest pain, 93% of patients with breast lumps, and 63% of patients with rectal bleeding. Scores in JPI and SUS were not associated with rates of extended evaluation.

Conclusions: : Primary care clinician risk tolerance and stress from uncertainty were not associated with evaluation decisions. Future research is needed to improve the consistency of evaluations to improve ambulatory patient safety.