A Quality Improvement Project to Investigate Individual Provider Variation in Opioid Prescribing From the Emergency Department

Opioid misuse is a known public safety threat and public health problem, both nationwide and in the Commonwealth of Massachusetts. As a part of these efforts, Emergency Department (ED) outpatient prescriptions have been identified as a potential area for improvement. The Massachusetts Hospital Association (MHA), in conjunction with other stakeholders, developed a set of recommendations to address opioid management and prescribing within the ED setting.

This study is timed to coincide with the rollout by UMassMemorial Health Care of a system-wide opioid prescribing practice guideline/policy which mirrors the Massachusetts Hospital Association's guideline. This rollout provides a unique opportunity for a natural experiment related to provider prescribing practices.

This is a quality improvement project with two primary objectives:

• Improve understanding of physician practice patterns related to emergency department (ED) opioid prescribing
• Evaluate the effectiveness of providing individual clinician profile data in influencing behavior change among ED providers, as it relates to their opioid prescribing patterns

There are two corresponding specific aims:

• Quantitatively describe the current state of opioid prescribing by ED clinicians at four UMass Memorial EDs (using prescriptions generated within the ED electronic health information system between February 2014 and the study implementation date). Descriptive statistics being considered for feasibility of study include:
• Number of opioid prescriptions per hundred patients seen
• Percentage of total prescriptions written that are opioids
• Mean or median quantity of pills dispensed per opioid prescription
• Number of prescriptions for long-acting opioid formulations
• Evaluate the effect on prescribing practices of providing individual clinicians with their profile data to illustrate their personal opioid prescribing patterns, relative to the de-identified distribution for all other clinicians in the ED group

This project is timed to coincide with the separately planned implementation of a system-wide opioid prescribing guideline. The guideline implementation is an independent event, but we feel that it presents a unique opportunity for a simultaneous experimental intervention to assess whether providing clinicians with their individual data alters their prescribing practices beyond any effect achieved simply by being subject to the new guideline. Data collected from the electronic medical record and clinician self-reported data about their perceptions of their prescribing practices will be used to evaluate both the effect of the guideline implementation (i.e. comparison of prescribing patterns for all clinicians before and after guideline implementation) and the combined effect of guideline implementation plus individual profile intervention (i.e. comparisons across intervention versus control groups).