Findings from a study of primary care physicians (PCPs) to measure the effectiveness of unsolicited electronic consultation (eConsult) to patients with type 2 diabetes (T2D) found PCPs prescribed significantly more glucose-lowering medications for patients with poorly controlled T2D than PCPs in a control group (26.2% vs 13.1%, P<.001). The results were recently published in the Journal of General Internal Medicine.
A group of 161 PCPs were randomly assigned to 2 groups: 81 to an intervention group and 80 to a control group. Approximately 130 patients with poorly controlled T2D using charts reviewed by endocrinologists were picked for eConsult and were drawn from data contained in electronic health records (EHRs). The eConsults took place between September 2018 and June 2019 using a primary care network affiliated with Massachusetts General Hospital. Patients were aged 26 to 80 years.
Following chart review, endocrinologists recommended 211 specific management interventions for patients treated by PCPs in the intervention group. The PCPs either accepted the endocrinologists’ recommendations provided by the unsolicited eConsult or did not. After 6 months, PCPs implemented only 80 of the 211 (38%) endocrinologist recommendations.
The primary outcome was the average change in glycated hemoglobin (HbA1c ) levels in patients from baseline after 6 months, while secondary outcomes included referral completion rate, rates of PCP prescription of glucose-lowering medications, number of in-person endocrinology visits, change in all glucose-lowering medications, and variances between all other intervention recommendations. Investigators also performed 12- and 18-month follow-ups measuring the change in HbA1c levels, which also showed no significant difference.
The most frequent recommendations included addition of a new medication (n=131) and dose adjustment of existing medication (n=45), an implementation rate of 28% and 53%, respectively.
When comparing the differences in prescription rates between the intervention and control groups, prescribed medications included the newer glucagon-like peptide-1 receptors (GLP-1 RA) or sodium-glucose cotransporter 2 inhibitors (SGLT2i) medications (19.3% vs 6.9%, P=.003), metformin (3.1% vs -3.1%, P=.03), and sulfonylurea (1.5% vs -6.9%, P=.03). After 6-months, patients treated by PCPs in the intervention group had 13 in-person endocrinology visits compared to 29 (P=.012) in the control group.
The PCPs barriers to recommendation of therapy included patients who did not return for a follow-up after the eConsult, disagreement with the recommendation of the endocrinologists, lack of familiarity with the recommended medications, patient refusal, and desire to try lifestyle changes or recheck HbA1c levels prior to initiation of medication.
Limitations of this study included the small sample size, eligibility for inclusion, and the length of time endocrinologists took to complete their part of the eConsult, which resulted in a 94-day time lag from list generation to completion. Additionally, patients were drawn from only from one single, urban, academic medical center, so results cannot be generalized.
“Unsolicited electronic consultation represents a novel care delivery tool,” the study authors concluded. “While implementation of such a program is feasible, it did not result in a clinically significant decrease in HbA1c compared to matched controls. Improvements in secondary process outcomes suggest the potential of this tool to improve care for patients with type 2 diabetes with improvements in program design and implementation.”
Disclosure: One study author declared affiliation with the pharmaceutical industry. Please see the original reference for a full list of authors’ disclosures.
Oseran AS, Rao K, Chang Y, et al. HbA1c-triggered endocrinology electronic consultation for type 2 diabetes management. J Gen Intern Med. Published online October 4, 2021. doi:10.1007/s11606-021-07157-x