Improving Pain Management in Primary Care

A Clinical Reminder Coupled with Feedback of Patient Information to Improve Pain Management in Primary Care 

  Principal Investigator:  Matthew J.  Bair, MD

Project Number HFP 05-187 funded by the VA Health Services Research & Development

October 1, 2005-December 31, 2007

Because of the potential benefit of computerized clinical reminders (CR), the National VA Pain Management Coordinating Committee has encouraged the development of pain reminders and reminder dialogs to improve pain treatment. Our research team has already developed a CR with the following goals: (a) to identify and target primary care patients with severe pain, who are at greatest risk for the adverse effects of ineffective pain treatment (e.g., dissatisfaction with care, depression, and functional loss); (b) to encourage patients to be active participants in their pain management; and (c) to be acceptable and useful to busy primary care providers.

The objective of the project is to conduct a pilot study comparing a simple intervention consisting of an CR coupled with feedback of both patient and pain-specific information vs. usual care. For this study we propose two Specific Aims: 

(1)        Evaluate the acceptability of the intervention to patients and providers

(2)        Assess the effect of the intervention on processes of care and patient outcomes

This pilot study was designed to evaluate the acceptability and effectiveness of a three-part intervention: (1) an already developed, computerized clinical reminder (CR) that identifies patients with severe pain, coupled with (2) assessment of pain-relevant information (e.g., pain-related concerns and expectations, pain severity and interference, and functional status); and (3) feedback of this patient information to primary care providers. Details of intervention: The CR will be “triggered” when primary care patients report severe pain (≥ 7 on the 0 to 10 scale). Such patients will also complete an assessment about their pain-related concerns and expectations and the domains mentioned above. The pain severity score along with this pain-specific information will be attached as a brief summary to the patient’s chart as “just in time” feedback to the provider. Providers in the usual care clinics will receive the single-item pain score (0 to 10) routinely administered, but no clinical reminder or patient-specific feedback. Since our five medicine clinics are organized by firms we will randomly assign clinics to either the intervention or control arms. Clinics, rather than providers, will be the unit of randomization in this pilot study in order to minimize contamination.

Outcome assessments will be done at baseline (i.e., prior to clinic visit) as well as 1-week and 3-months post-visit by a research assistant blinded to the intervention status. Baseline assessment will be done in-person and follow-up assessments by telephone. The primary process outcomes will be initiation or change in pain treatment. Patient outcomes will include satisfaction with pain treatment, unmet expectations, and pain severity and interference. Secondarily, depression (assessed by PHQ-9), HRQL, and functional limitations will be measured. Providers will be surveyed at baseline to obtain demographics, education and training, and pain management knowledge, attitudes, and beliefs. At the end of the study patients and providers will be asked to rate the acceptability of the intervention.

For phase 2, semi-structured interviews of providers will be conducted to assess reasons for failing to initiate or change pain treatment in patients with severe pain. These interviews will be done while referring to the medical record to discuss the care of specific patients.