The Effect Of Mental Health Specialist Use On Guideline-Concordant Diabetes Care Among Adults With Co-Occurring Diabetes And Depression

Diabetes and depression are two of most common chronic conditions that tend to co-occur. Yet like most patients with multiple chronic conditions (MCC), these individuals are less likely to receive guideline-concordant care compared with those with either diabetes or depression only. The two major types of provider for depression care are primary care physicians (PCP) and mental health specialists (MH), including psychiatrists and other non-prescribers, such as psychologists, social workers, family and marriage therapists, and counselors. Research has demonstrated that MH care leads to 10%-20% higher levels of guideline-concordant depression care than PCP in individuals with depression. While depression leads to lower guideline-concordant diabetes care, effect of MH treating depression on guideline-concordant diabetes care is still inconclusive. Further, MH and PCP were not directly compared in individuals with co-occurring diabetes and depression in the past research, and evidence of MH care on guideline-concordant diabetes care in this specific population is less clear.

This study will examine the effect of MH use on guideline-concordant outpatient diabetes care. The conceptual framework, built on the Grossman health production model, illustrates why the effect of MH care on healthcare utilization is ambiguous while diabetes and depression are competing for limited time and money budget. If MH and PCP are somehow substitutional in depression care, MH visit might reduce primary care visit and further lower probability of receiving guideline-concordant diabetes care. On the other hand, if MH and PCP are complementary, MH visit might increase PCP visit and further improves quality of diabetes care. Therefore, this study aims to empirically examine the effect of mental health specialist care.

This study uses North Carolina Medicaid Claims data and Health Profession Data System (HPDS) from 2006 to 2011. Medicaid beneficiaries with diabetes and depression are included. Two-Stage Residual Inclusion will be estimated to correct potential endogeneity of MH use due to reverse causality between MH use and outcome and omitted variables affecting MH use and outcome simultaneously. Two sets of instrumental variable (IV) will be used: the first set is area-mental health specialist supply, including total supply variables provided by HPDS, and Medicaid provider supply derived from Medicaid claims; the second set is county-level rate of mental health specialist use among Medicaid beneficiaries. The guideline-concordant diabetes care indicators are eye exam at least once a year, HbA1c check twice a year, and LDL test once a year.

In light of the shortage of mental health specialist and the shifting of depression treatment modality towards pharmacotherapy, defining responsibilities of PCP and MH in caring for persons with depression is challenging. Many studies have suggested that depression can be effectively treated as a chronic condition in primary care settings. As individuals with diabetes and depression are mainly managed at primary care, examining the role of mental health specialist among this population became critical. Results from this study provide implication for the future role of mental health specialist for caring individuals with MCC.