Study Setting and Subjects
Healthy Hearts Northwest (H2N) took place within 209 smaller (10 or fewer clinicians) primary care practices across Washington State, Oregon and Idaho and the results of this randomized trial have been previously reported.26 All practices were supported by a practice facilitator (PF) for 15 months. In addition, practices were randomized to receive educational outreach visits (based on principle of academic detailing), and/or shared learning opportunities. The published trial results adhered to CONSORT guidelines for reporting clinical trials.
Development of a QI Capacity Assessment Instrument
To identify and define core concepts or “domains” that contribute to QI capacity two members of the study team (KC, LM) along with two outside experts with considerable prior experience in patient centered medical home transformation efforts was convened. They first reviewed the literature related to the definitions and characteristics of QI capacity,5 principles of transformation to a patient-centered medical home,42,43 and elements of the chronic care model.44,45 Because efforts to develop QI capacity are closely related to the goals of medical home transformation, the review was rooted in medical home change concepts previously described in the Safety Net Medical Home Initiative.46 From this review, this group identified a parsimonious set of domains that would inform activities where changes could be made within the 15 month study intervention window.41 These domains were:
- Embed clinical evidence into daily work to guide how care is delivered to patients
- Utilize data to understand and improve clinical performance measures
- Establish a regular QI process involving cross-functional teams
- Identify at-risk patients through pro-active population management for outreach
- Define roles and responsibilities across the team to improve care
- Deepen patient self-management support to improve clinical outcomes
- Link patients to resources outside of the clinic to support patients
To assist both practices and their PF in assessing their current state of QI capacity, a survey was developed by selecting items relevant to each of the seven change concepts. The Safety Net Medical Home Initiative had previously validated the 36-item Patient-Centered Medical Home Assessment tool for medical home transformation.47 Study team members (MLP, KC) reviewed these items and found that nineteen aligned well with the seven change concepts described above. A twentieth item was created by study team members (MLP, KC) to assess the availability of a standard method or tool to stratify patients by level of clinical risk. This new item was included with three other items to help assess the fourth domain listed above, identification of at-risk patients. Consistent with the PCMH-A instrument,47 and with the more widely used and validated Assessment of Chronic Illness Care (ACIC) instrument,27,48 each item on the QICA is rated on a 1 to 12 scale with higher scores indicating higher QI capacity (see Appendix). The 20 items were grouped into seven subscales corresponding to the change concepts identified above. Each subscale corresponded to one of the seven change concepts and included one to four items.
Measures and Data Collection
During a face-to-face visit by the PF at the start of the study intervention, the QICA survey was completed individually by clinicians and staff in each H2N practice. With the PF present, team members discussed their response to each item and came to a consensus on a practice-level response. These meetings took place between December 2015 and July of 2016. Prior to this meeting, practice managers completed a survey describing practice characteristics, which included the Change Process Capacity Questionnaire (CPCQ).39 In addition, each practice was asked to submit clinical quality measures (CQMs) data derived from the electronic health record on three CVD risk factors in the year prior to completing the QICA: appropriate aspirin use, blood pressure control, and tobacco screening/cessation.49,50 After the in-person visit to facilitate completion of the QICA, PFs kept detailed field notes describing the encounter.
Analysis
We first described practice characteristics, and summarized baseline QICA scores and clinic performance on each CQM. We then used Kruskal-Wallis tests to assess the association between total QICA score and practice characteristics including practice size, ownership and location (rural v. urban). To examine the reliability of the QICA, we estimated Cronbach’s alpha for the total scale and each subscale. We assessed scale validity by reporting the Spearman correlation between the total score and subscale scores, with the CPCQ total score, and practice performance on each of the three CQMs in the year prior to administering the QICA survey. Finally, we examined qualitative PF notes from clinics in the top and bottom 10% of the distribution of QICA scores.