Sign In

Diabetes Study of Northern California

methods

​Methods

​Setting

All subjects are members of Kaiser Permanente Northern California, a large, integrated healthcare delivery system. Kaiser Permanente provides comprehensive medical services to over 3.6 million members in Northern California, approximately a third of the region’s population. The membership’s ethnic and socioeconomic distribution closely approximates that of the general population of the region, except for the extreme tails of income distribution. Approximately 4,400 physicians provide integrated care at 17 hospitals and 152 medical offices.

Data Sources

Kaiser Permanente Electronic Medical Record

Since 2004, Kaiser Permanente has operated HealthConnect, an integrated electronic medical record (EMR) system designed by the Epic Corporation to automate its patient files and make documentation of care more efficient and complete. HealthConnect replaced many of the core utilization and clinical documentation legacy applications, including ambulatory visit check-in, hospital-based utilization, and clinical documentation of diagnoses, orders for tests and procedures, and prescribed inpatient medications. In addition to replacing many existing legacy functions, HealthConnect has significantly enhanced the scope and detail of available data.

Kaiser Permanente Northern California Diabetes Registry

As of January 2014, the registry identified about 264,000 patients with clinically recognized diabetes. The Kaiser Permanente Northern California Diabetes Registry was first established in 1993 using standardized criteria to identify and prospectively follow members with diabetes, to measure prevalence and incidence of diabetes and its comorbidities, understand factors associated with disease progression and complications, and evaluate processes of care and outcomes.

We conducted the first survey of the registry from 1994 to 1997 among all registry members over 19 years of age. The primary goal was to capture individual-level information on the clinical characteristics of diabetes, age at diagnosis, ethnicity, education, health-related behaviors, and diabetes family history. With 77,726 respondents (83% response rate among eligible members), that survey cohort has been the basis for numerous publications regarding the epidemiological and health services aspects of diabetes.

DISTANCE Survey

In 2005-2006, DISTANCE researchers conducted a second survey, among an ethnically stratified, random sample of diabetic members of known (white, African American, Latino, Asian) and unknown ethnicity receiving care from Kaiser Permanente. The DISTANCE Survey assessed a wide range of behavioral, social, and economic factors (ethnicity, educational attainment, health literacy, health behaviors, diabetes knowledge, perceptions, attitudes and psychosocial attributes) that may explain observed variations in the incidence of diabetes complications and death. Understanding factors that are predictive of poor health outcomes may facilitate the design of targeted interventions to reduce health disparities. We hypothesized that these factors may differ substantively in prevalence or effect size across ethnic groups or educational levels.

The DISTANCE Survey was completed by 20,188 patients (62% response rate among eligible members): 3,420 African American (16.9%), 2,312 Asian (11.4%), 4,602 white (22.8%), 2,404 Filipinos (11.9%), 3,717 Latinos (18.4%), 2,222 multi-racial (11.0%) and 1,511 other (7.5%). Respondents varied widely in terms of clinical and behavioral profiles, as well as education, income, wealth, occupation, place of birth, and neighborhood characteristics. In most cases, findings from these survey responders should generalize to non-respondents as relatively few baseline variables differed between respondents and nonrespondents. Moreover, analyses of response bias focusing on associations between race or education and poor glycemic control (A1C > 7%) detected no significant differences (p = 0.55 and 0.28 for survey response interaction with race and education, respectively). The complete survey is available online (DIST​​​ANCE Survey) and a published description of methods is available at International Journal of Epidemiology or PubMed Central.