Differential Impact of Homelessness on Glycemic Control in Veterans with Type 2 Diabetes Mellitus

Abstract

<div class="section"> <a class="named-anchor" id="d1529218e255"> <!-- named anchor --> </a> <h5 class="section-title" id="d1529218e256">BACKGROUND</h5> <p id="Par1">Veterans with evidence of homelessness have high rates of mental health and substance abuse disorders, but chronic medical conditions such as diabetes are also prevalent. </p> </div><div class="section"> <a class="named-anchor" id="d1529218e260"> <!-- named anchor --> </a> <h5 class="section-title" id="d1529218e261">OBJECTIVE</h5> <p id="Par2">We aimed to determine the impact of homelessness on glycemic control in patients with type 2 diabetes mellitus. </p> </div><div class="section"> <a class="named-anchor" id="d1529218e265"> <!-- named anchor --> </a> <h5 class="section-title" id="d1529218e266">DESIGN</h5> <p id="Par3">Longitudinal analysis of a retrospective cohort.</p> </div><div class="section"> <a class="named-anchor" id="d1529218e270"> <!-- named anchor --> </a> <h5 class="section-title" id="d1529218e271">SUBJECTS</h5> <p id="Par4">A national cohort of 1,263,906 Veterans with type 2 diabetes. Subjects with evidence of homelessness were identified using a combination of diagnostic and administrative codes. </p> </div><div class="section"> <a class="named-anchor" id="d1529218e275"> <!-- named anchor --> </a> <h5 class="section-title" id="d1529218e276">MAIN MEASURES</h5> <p id="Par5">Odds for poor glycemic control using hemoglobin A1C (HbA1C) cutoff values of 8 % and 9 %. Homeless defined as a score based on the number of indicator variables for homelessness within a veterans chart. </p> </div><div class="section"> <a class="named-anchor" id="d1529218e280"> <!-- named anchor --> </a> <h5 class="section-title" id="d1529218e281">KEY RESULTS</h5> <p id="Par6">Veterans with evidence of homelessness had a significantly greater annual mean HbA1C ≥ 8 (32.6 % vs. 20.43 %) and HbA1C ≥ 9 (21.4 % vs. 9.9 %), tended to be younger (58 vs. 67 years), were more likely to be non-Hispanic black (39.1 %), divorced (43 %) or never married (34 %), to be urban dwelling (88.8 %), and to have comorbid substance abuse (46.7 %), depression (42.3 %), psychoses (39.7 %), liver disease (18.8 %), and fluid/electrolyte disorders (20.4 %), relative to non-homeless veterans (all <i>p</i> &lt; 0.0001). Homelessness was modeled as an ordinal variable that scored the number of times a homelessness indicator was found in the Veterans medical record. We observed a significant interaction between homelessness and race/ethnicity on the odds of poor glycemic control. Homelessness, across all racial-ethnic groups, was associated with increased odds of uncontrolled diabetes at a cut-point of 8 % and 9 % for hemoglobin A1C ; however, the magnitude of the association was greater in non-Hispanic whites [8 %, OR 1.55 (1.47;1.63)] and Hispanics [8 %, OR 2.11 (1.78;2.51)] than in non-Hispanic blacks [8 %, OR 1.22 (1.15;1.28)]. </p> </div><div class="section"> <a class="named-anchor" id="d1529218e288"> <!-- named anchor --> </a> <h5 class="section-title" id="d1529218e289">CONCLUSIONS</h5> <p id="Par7">Homelessness is a significant risk factor for uncontrolled diabetes in Veterans, especially among non-Hispanic white and Hispanic patients. While efforts to engage homeless patients in primary care services have had some success in recent years, these data suggest that broader efforts targeting management of diabetes and other chronic medical conditions remain warranted. </p> </div><div class="section"> <a class="named-anchor" id="d1529218e293"> <!-- named anchor --> </a> <h5 class="section-title" id="d1529218e294">Electronic supplementary material</h5> <p id="d1529218e296">The online version of this article (doi:10.1007/s11606-016-3786-z) contains supplementary material, which is available to authorized users. </p> </div>