Decision support intervention for clinicians helped decrease low-value care

A decision support tool aimed at increasing clinicians' attention to possible harms, social norms, and reputational concerns helped reduce care overuse among older patients compared with traditional case-based education alone, new results of a randomized controlled trial show.

Investigators carried out an 18-month, single-blind trial at 60 primary care internal medicine, family medicine, and geriatrics practices within a single health system from September 2020 to February 2022. A total of 371 primary care physicians and their older adult patients were included in the study. Of these, 187 clinicians from 30 clinics integrated a behavioral science-informed, point-of-care, clinical decision support tool into their care and completed brief case-based education. The control group only completed brief case-based education.

The decision support tool aimed to increase salience of potential harms of overtreatment, convey social norms, and promote accountability among clinicians. To test the tool's efficacy, researchers assessed rates of prostate-specific antigen (PSA) testing in men ages 76 years and older without a history of prostate cancer, urine testing for nonspecific reasons in women ages 65 years and older, and overtreatment of diabetes with hypoglycemic agents in patients ages 75 years and older with HbA1c levels of less than 7%. Findings were published by Annals of Internal Medicine on Feb. 5.

At the time of randomization, average rates of annual clinic PSA testing, unspecified urine testing, and diabetes overtreatment were 24.9, 23.9, and 16.8 per 100 patients, respectively. Following the intervention, the intervention group had lower adjusted difference-in-differences on all three measures (−8.7 [95% CI, −10.2 to −7.1], −5.5 [95% CI, −7.0 to −3.6], and −1.4 [95% CI, −2.9 to −0.03], respectively) versus the education-only group. There was also no increase in emergency care related to urinary tract infections or hyperglycemia. However, among patients with diabetes who were previously overtreated, an HbA1c over 9% was more common in the intervention group (adjusted difference-in-differences, 0.47 per 100 patients [95% CI, 0.04 to 1.20]).

Because the study did not compare a clinical decision support intervention that applied behavioral principles with one that did not, the researchers caution it's not clear whether the differences seen are reflective of the tool itself or the language and principles used to develop the intervention. The trial was also conducted at a single health system and findings may not be generalizable. However, "these findings suggest that point-of-care behaviorally informed interventions can reduce over testing and overuse among older patients of primary care clinicians while preserving clinician discretion," the authors wrote.

An accompanying editorial commended the researchers and noted that a mix of strategies may prove to be the best solution for overuse, including "clinical decision-support tools that 'force' the right decision, such as preventing the ordering of worthless laboratory tests in patients nearing the end of life" and training clinicians in "skills and strategies for discussions that maintain a trusting relationship with their patients while they guide them away from low-value services."