Housing Insecurity Predictive of Poor Glycemic Control

People with type 2 diabetes who faced housing insecurity had worse glycemic outcomes, a new study found.

Among adults with type 2 diabetes in Northern California, those who had at least one address change in a year showed significantly higher rates of HbA1c over 9% compared with those with no address change (27.2% vs 21.4%, P=0001), reported Tainayah Thomas, PhD, of Kaiser Permanente Northern California Division of Research in Oakland.

In an adjusted model, this equated to a 14% higher relative risk of having an HbA1c above 9% (RR 1.14, 95% CI 1.05-1.25), Thomas explained at the virtual European Association for the Study of Diabetes 2020 meeting.

“Implications of our study are that an address change may be an early warning sign of housing insecurity or stress that could be used to trigger screening or other intervention,” she said during a press conference.

Likewise, those with one or more address changes during the span of 1 year also had a lower prevalence of “good” glycemic control — defined in the study as an HbA1c less than 8% (61.1% vs 67.2%, P=0.0001). That translated to a relative risk of 0.94 for maintaining good control (95% CI 0.90-0.97) compared with people who didn’t move.

Rates of emergency department usage were also higher for those with recent address changes (40.9% vs 27%, P=0.0001), for a relative risk of 1.44 (95% CI 1.35-1.53).

Additionally, only 49.2% of those who moved at least once received flu shots compared with 57.2% of those who didn’t move (RR 0.92, 95% CI 0.88-0.97).

For the cross-sectional study, Thomas and colleagues pulled data from members of the Kaiser Permanente East Bay health system in 2018. This included data on 25,614 members over the age of 18 with type 2 diabetes residing in the cities of Oakland, Richmond, Alameda, or Pinole in Northern California.

Men and women were equally represented in the cohort, and about 80% of the individuals were over age 50 (average age was 62). The cohort was also ethnically diverse — about 30% Black adults, 23% Asian, 22% white, and 21% Hispanic.

The researchers quantified “housing insecurity” based on at least one address change in the patients’ electronic medical record. There was a possible issue, however, with how “housing insecurity” was defined, Thomas said: “An address change may not signal housing insecurity. It could signal stress related to moving or changes in lifestyle behaviors associated with adverse outcomes.” Two or more address changes within the span of 12 months occurred very infrequently in this cohort, she added.

Regarding the next steps in the research, Thomas said: “One of the things we’re interested in and future directions of our project is thinking about predictive modeling — what are other predictors of housing insecurity that can easily be identified in an electronic health record … So once we build this, we can already identify a subset of patients who have a number of risk factors for housing insecurity that then can be screened, maybe not by a physician necessarily, but by someone in the health system, for intervention.”

She also noted plans for the team to delve into how the COVID-19 pandemic has affected housing insecurity and subsequent glycemic outcomes in people with type 2 diabetes.