• Cardiometabolic Safety of Atypical Antipsychotic Medications among Publicly Insured U.S. Youth

      Burcu, Mehmet; Zito, Julie Magno; 0000-0003-4572-0987 (2017)
      Background: The use of atypical antipsychotics (AAPs) among publicly insured U.S. youth has substantially increased in the past two decades. Furthermore, more than half of AAP-treated youth have concomitant antidepressant or stimulant use, although the cardiometabolic effects of such combinations are largely unknown. Methods: The main focus of this dissertation was to evaluate the risk of incident type 2 diabetes mellitus (T2DM) and adverse cardiovascular events in AAP-treated youth according to the concomitant use of stimulants or serotonin reuptake inhibitors (SSRI/SNRIs)-the leading antidepressant subclass. The risk of T2DM and adverse cardiovascular events were assessed using discrete time failure models. To adjust for confounding, disease risk score methodology was employed using >125 baseline and time-dependent covariates. Medication use was assessed using four time-varying exposure measures: current/former/non-use, duration of use, cumulative dose, and average daily dose. Results: In a large regionally diverse cohort of Medicaid-insured youth, AAP use was associated with an increased risk of T2DM that increased with duration of AAP use and cumulative AAP dose. Further, in AAP-treated youth, concomitant SSRI/SNRI use was associated with an additional increased risk of T2DM, which intensified with duration of SSRI/SRNI use and SSRI/SNRI dose. In a separate set of analyses that focused on youth who initiated antidepressant treatment (regardless of AAP use), an increased risk of T2DM was also observed for SSRI/SNRIs. Finally, following treatment initiation with AAPs, current AAP use was also associated with an increased risk of incident cardiovascular events that led to hospitalizations or emergency department visits. This increased risk also intensified with increasing AAP dose and when SSRI/SNRIs were used concomitantly with AAPs. By contrast, in AAP-treated youth, concomitant use of stimulants was not associated with an increased risk of T2DM or cardiovascular events. Conclusions: In view of the growing complexity of atypical antipsychotic regimens in Medicaid-insured youth and low rates of baseline metabolic monitoring in youth initiating AAP treatment, these findings suggest that complex AAP regimens should be used judiciously with appropriate cardiometabolic monitoring. Continued efforts are warranted to support Medicaid oversight policies that assure safe and effective use of complex AAP regimens in youth populations.
    • Community-level and Individual-level Predictors of Variation in Rates of Homelessness among Youth Transitioning Out of Foster Care

      Sneddon, Dori; Bright, Charlotte Lyn (2019)
      Youth who age out of foster care are a known high-risk subgroup for homelessness. Studies estimate between 19% and 36% of youth experience homelessness shortly after emancipation. This study examined homelessness among youth transitioning out of foster care by incorporating individual-level and county-level influences to better understand the risk of homelessness among this population. Multilevel models and generalized estimating equation models were constructed to include both individual- and county-level variables. Data were obtained from multiple national datasets: the 2011-2015 National Youth in Transition Database (NYTD), 2011 Adoption and Foster Care Analysis and Reporting System (AFCARS), University of Wisconsin’s County Health Rankings & Roadmaps Data and the 2013 U.S. Department of Agriculture (USDA) Rural-Urban Continuum Code data file. The analytic sample included 3,968 youth who responded to the NYTD Wave 1 (age 17), Wave 2 (age 19) and Wave 3 (age 21) survey. Of the sample, 35.3% experienced homelessness between 17 – 21 years old. Findings indicated statistically significant variation between counties in the proportion of youth who become homeless. Multiple individual-level factors were found to predict homelessness between ages 17 – 21. Prior homelessness, substance use history, and incarceration had a positive relationship with risk of homelessness. Connection with a caring adult, enrollment in school, and employment were inversely related to risk of homelessness. Specific to foster care experience, number of placements and age of entry had a positive relationship with risk of homelessness. Being in foster care at age 19 and at age 21 were related to a reduced risk of homelessness. Not as hypothesized, receipt of independent living services had a positive relationship with homelessness. None of the county-level indicators had a statistically significant relationship to the homelessness outcome. Policy and practice implications for child welfare include extending foster care, capturing housing histories and prioritizing housing plans for youth, and targeting intensive services to youth at the highest risk of homelessness. Future research to further examine socioeconomic community- and state-level predictors of homelessness among this population inform homelessness prevention and housing strategies for youth aging out of foster care. Suggested areas for improvement in NYTD data are also discussed.
    • Longitudinal Patterns of Early Mental Health Service Utilization in a Medicaid-insured Birth Cohort and the Impact of Continuity of Care on the Quality of Pediatric Mental Health Treatment

      Pennap, Dinci; Zito, Julie Magno (2018)
      Background: The prevalence of pediatric mental health (MH) diagnosis and treatment have expanded in the U.S. We assessed the longitudinal patterns of incident diagnosis and new psychotropic medication use in a Medicaid-insured birth cohort. Additionally, continuity and quality of MH service utilization were assessed in a publicly-insured pediatric population. Quality care was defined by the 2009 Children's Health Insurance Program Reauthorization Act (CHIPRA) mandated children's health care quality measures. Methods: We applied longitudinal designs to Medicaid claims data from a Mid-Atlantic state (2007-2014). Using Kaplan-Meier estimators we assessed the cumulative incidence of MH service use in a cohort of newborns (aim 1). We assessed the association between relational patient-provider continuity of care and: 1) emergency department (ED) visits or hospitalizations in the 12 months following first MH diagnosis among 3-16 year olds (aim 2); and 2) the quality of follow-up care among 6-12 year old new users of ADHD medications (aim 3), using logistic regression models. Quality was defined as having ≥1 follow-up outpatient visit in the 30 days following medication initiation and ≥2 follow-up visits in the 270 days after the first follow-up visit, with a total medication supply of ≥210 days. Results: By age 8, 19.7% and 10.2% of the birth cohort (n=35,244) had received a MH diagnosis or psychotropic medication, respectively. Among medication users, 80.5% received monotherapy, 16.4% received 2 medication classes, and 4.3% received ≥3 medication classes concomitantly for ≥60 days. Compared to children with high CoC, the odds of ED visits was significantly higher among youths with low CoC [Odds Ratio(OR)=1.27; 95% CI=1.13-1.41] and low CoC was associated with greater odds of hospitalization [OR=1.17; 95% CI=1.06-1.29]. Compared to those with low CoC, children with higher continuity of care had greater odds of meeting CHIPRA initiation- [OR=1.41; 95% CI=1.25-1.60] and continuation-phase [OR=1.45; 95% CI=1.29-1.64] visit-based measures. Conclusions: Early exposure to psychotropic medications and prolonged duration of use have implications for long-term safety, highlighting the need for safety and outcomes research in pediatric populations. Our findings suggest a need for more research in the areas of quality assessment and continuity of care among youths with mental health conditions.