Table 2.

Study Measures, Variables and Data Sources

Variable/DefinitionData SourceData Collection/Analysis
Assess the Feasibility, Appropriateness and Acceptability of INTEGRATE-D
Acceptability – extent to which intervention is agreeable, palatable, satisfactoryAssessed via survey15 and semi-structured interview. Survey had four questions per variable. See Online Appendix for items.Collected from clinic members exposed to the intervention ICs at the end of the intervention.
Descriptive analysis for ICs only. Survey scores, which ranged from 1 (strongly disagree) to 5 (strongly agree) for each clinic member response were averaged at the clinic-level.
Appropriateness – extent to which intervention fits and is compatible for addressing issue or problem
Feasibility – extent to which an intervention can be successfully used or carried out
Compare changes in use of quality-improving strategies
Clinics’ ability to implement quality improving strategies related integrated type 2 diabetes careAssessed via survey using 14 items from the Change Process Capability Questionnaire (CPCQ).30 See Appendix 2 for items.One person at IC and CCs completed the survey at the same time, pre- and post-intervention.
Survey scores, which ranged from −2 (strongly disagree) to +2 (strongly agree), were summed for each clinic with possible sums ranging from −28 to +28. Average aggregate scores were compared between IC versus CCs
Compare changes in process of care screening rates
A1C screening – binary variable indicating whether the patient was screened at least once during the periodElectronic Health Record (EHR) data abstracted through manual and automated methodsOperationalized at the patient level for pre- and post-intervention periods, means aggregated at the clinic-level for IC and CC. Pre-intervention defined as the time during the 12 months before the intervention; post-intervention defined as any time during the 12 months following the start of the intervention.
Cholesterol screening - binary variable indicating whether the patient was screened at least once during the period
Nephropathy screening - binary variable indicating whether the patient was screened at least once during the period
Psychosocial screenings – binary variables indicating whether the patient was screened at least once during the period for depression (PHQ-2 and/or PHQ-9) and/or for diabetes distress (DD)1
Compare changes in clinical outcomes (PHQ-9 scores and A1C levels)
Behavioral health – change in symptoms (PHQ-9) for patients with depression symptoms (PHQ-9 > 9); data abstracted through manual and automated methodsOperationalized at the patient level for pre- and post-intervention periods, means aggregated for intervention and control clinics; pre-intervention defined as score closest to the intervention start date; post-intervention defined as score closest to the intervention end date.
Diabetes Management – change in A1C for patients diagnosed with type 2 diabetes
Confounding Variables - Patient socio-demographics, comorbidity, insurance, and utilization
Age, gender, language preference, race/ethnicity, income/federal poverty level, insurance type, physical, mental/behavioral health comorbidity, healthcare utilizationEHR data abstracted through manual and automated methodsOperationalized at the patient level
  • 1 Diabetes distress (DD) screening was implemented by IC; clinics did not screen for this pre-intervention.

  • Abbreviations: IC, intervention clinics; CC, control clinics.