Adoption of Opioid-Prescribing Guidelines in Primary Care: A Realist Synthesis of Contextual Factors

Context: Primary care clinics

Objective: As part of an effort to design an implementation strategy tailoring tool, our research group sought to understand what is known about how contextual factors and prescriber characteristics affect the adoption of guideline-concordant opioid-prescribing practices in primary care settings.

Study Design and Analysis: We conducted a realist synthesis of 71 articles.

Setting or Dataset: Global literature review

Population Studied: Patients on long-term opioid therapy for chronic pain

Intervention/Instrument: N/A

Outcome Measures: Adoption of clinical guidelines for opioid prescribing in primary care settings.

Results: We found that adoption is related to contextual factors at the individual, clinic, health system and environmental levels, which operate via intrapersonal, interpersonal, organisational and structural mechanisms.

Prescriber level factors tending towards adherence included: training in palliative care, pain management, mental health, or addiction medicine; frequent, high-volume opioid prescribing; medication for opioid use disorder prescribing; and knowing someone who has died of an opioid overdose. Prescriber level factors tending towards non-adherence included low-volume, low-dose prescribing; pharmaceutical company-sponsored training in opioids; and incentives from pharmaceutical companies. Clinic level factors tending towards adherence included a team-based culture; interdisciplinarity; and a culture of QI. Clinic level factors tending towards non-adherence included turf battles and clinician conflict about opioid prescribing. Health system factors tending towards adherence included clear policies toward opioid prescribing, including mandates; and resources devoted to supporting guideline adherence. Health system factors tending towards non-adherence included high demand/low support; workflows that do not accommodate guideline concordant care; administrative intrusion into clinical decision-making; and inadequate prescriber compensation. Environmental factors tending towards non-adherence included poverty and health disparities.

Conclusion: A static model cannot capture the complexity of the relationships between context, mechanisms and outcomes regarding adherence to opioid prescribing guidelines. Instead, deeper understanding requires dynamic models that conceptualize clusters of contextual factors and mechanisms that tend toward guideline concordance and clusters that tend toward non-concordance.