Treatment recommendations made by a consultant psychiatrist to improve the quality of care in a collaborative mental health intervention in rural Nepal

BACKGROUND: The Collaborative Care Model (CoCM) for mental healthcare, where a consulting psychiatrist supports primary care and behavioral health workers, has the potential to address the large unmet burden of mental illness worldwide. A core component of this model is that the psychiatrist reviews treatment plans for a panel of patients and provides specific clinical recommendations to improve the quality of care. Very few studies have reported data on such recommendations. This study reviews and classifies the recommendations made by consulting psychiatrists in a rural primary care clinic in Nepal. METHODS: A chart review was conducted for all patients whose cases were reviewed by the treatment team from January to June 2017, after CoCM had been operational for 6 months. Free text of the recommendations were extracted and two coders analyzed the data using an inductive approach to group and categorize recommendations until the coders achieved consensus. Cumulative frequency of the recommendations are tabulated and discussed in the context of an adapted CoCM in rural Nepal. RESULTS: The clinical team discussed 1174 patient encounters (1162 unique patients) during panel reviews throughout the study period. The consultant psychiatrist made 214 recommendations for 192 (16%) patients. The most common recommendations were to revisit the primary mental health diagnosis (16%, n = 34), add or increase focus on counselling and psychosocial support (9%, n = 20), increase the antidepressant dose (9%, n = 20), and discontinue inappropriate medications (6%, n = 12). CONCLUSIONS: In this CoCM study, the majority of treatment plans did not require significant change. The recommendations highlight the challenge that non-specialists face in making an accurate mental health diagnosis, the relative neglect of non-pharmacological interventions, and the risk of inappropriate medications. These results can inform interventions to better support non-specialists in rural areas