Browsing by Author "Upadhyay, Navneet"
Now showing 1 - 2 of 2
- Results Per Page
- Sort Options
Item Geographic Access to Providers and Pediatric Mental Health Care(2018-12) Upadhyay, Navneet; Chen, Hua; Aparasu, Rajender R.; Rowan, Paul J.; Fleming, Marc L.; Balkrishnan, RajeshIntroduction: Given the severe shortage of mental health care providers in the US and in Texas, variations in the distribution of the providers and its interaction with race/ethnicity and poverty level exist. Therefore, geographic access to provider could significantly affect the pediatric mental health care. This study aimed to comprehensively investigate the effect of geographic access to mental health care providers and its interaction with race and socioeconomic characteristics on the different level (identification of mental disorders, treatment engagement, and quality of treatment) of pediatric mental health care. Objectives: The main objective of this study was to examine the association of geographic access to providers (primary care provider (PCP): primary care physicians, physician assistants, registered nurse, and mental health care provider: psychiatrists, psychologists, and psychiatric nurse) with the following: (a) likelihood of receiving screening, in the pediatric population aged 4-18 years (Aim 1); (b) likelihood of engaging in the treatment(≥2 treatment-related visits) and receiving minimum guideline-concordant care in the pediatric depression patients (Aim 2) and (c) likelihood of engaging in the treatment and medication adherence in the pediatric ADHD patients of Medicaid managed care plan of Texas (Aim 3). Material and Methods: A retrospective cohort study was conducted using multiple data sources. An administrative claims data for the years 2013-2016 from a large pediatric Medicaid Managed Care Plan was the primary data source, that has individual level information such as age, gender, race, zip code of residence, health-related and provider- related information. This was linked to 2010 US Census data with a zip code of enrollees to get the neighborhood level information such as population, poverty level, racial/ethnic composition. It was further linked with the National Provider Identifier (NPI) Registry using NPI codes to ascertain provider location and provider specialty. Three geographic access measures were used (a) Travel distance to the nearest provider for aim 1, travel distance to the provider who diagnosed mental disorders/initiated the treatment for aim 1 and aim 2. (b) PCP density per 10,000 individuals within 5-mile of travel distance, and (c) specialists density per 10,000 individuals within 5-mile of travel distance. For aim 1, the study period was of two years in which study population was all the enrollees (aged 4-18 years) that were enrolled for at least 20 months within two years in the health plan and did not have any mental disorders in the first year. Behavioral disorder screening claims were identified using CPT codes in the second year of the study period. Aim 2 included children and adolescents (4-18 years) who were newly diagnosed with depression (identified using ICD codes) and did not have any mental disorder related claims in previous 6 months and had received at least one treatment for the depression. Outcomes in this aim were treatment engagement (defined as ≥2 antidepressants prescription or psychotherapy sessions for depression) and minimally adequate guideline-concordant treatment ( defined as ≥8 sessions of psychotherapy within 84 days post-index treatment date or ≥ 84 days of continuous treatment with antidepressants out of 114 days post-index treatment date). Aim 3 included children and adolescents (4-18 years) who were newly diagnosed with ADHD (identified using ICD codes) and did not have any mental disorder related claims in the previous 6 months. Outcomes in this aim were treatment engagement (≥2 ADHD medication prescription or psychotherapy sessions for ADHD) and medication adherence (PDC ≥ 0.8) during 300 days post-index treatment. Medication adherence was only assessed in the patients who engaged in the treatment. Covariates included in the study were age, gender, race, Medicaid eligibility categories (family income), specialty of the provider who gave the index diagnosis (for the analysis of treatment engagement) or specialty of the provider who prescribed the index prescription (for Aim 2 and Aim 3), neighborhood poverty level, and urban influence on a zip code. Single level multivariable logistic regression was used in the study. ArcGIS® was used for all geographic information system (GIS) functions, except travel distance that was calculated using Google Map®. All statistical analyses were carried out using SAS®. This study was approved by the Institutional Review Board at the University of Houston. Results: For Aim 1, Behavioral disorder screening rate was 12.6% among 457,870 children and adolescents who met the inclusion criteria. Multivariable analysis stratified by patient race/ethnicity revealed that that travel distance was negatively associated with screening engagement only among Hispanics (10-20 miles vs. 0-10 miles: OR=0.78, 95% CI [0.71-0.86]; 20-30 miles vs. 0-10 miles: OR=0.35, 95% CI [0.23-0.54]). In the subgroup that had access to at least one PCP within 10 miles of travel distance, minorities were more sensitive to increase in PCP density compare to Whites for the likelihood of screening. For Aim 2, A total of 3,472 children and adolescents with newly diagnosed depression with at least one treatment-related visit were identified. Seventy percent of the MDD treatment was initiated by mental health specialists, and 25% by PCPs. The treatment engagement rate was 65%. Among those who engaged in the treatment, 13.63% met the minimum adequacy of treatment criteria. Travel distance to the provider who initiated the treatment was negatively associated with the likelihood of treatment engagement among Hispanics (Travel distance 5.0 - 14.9 miles vs 0 - 4.9 miles: OR=0.74, 95% CI [0.54-0.88], p value=0.03; Travel distance ≥15 miles vs 0 - 4.9 miles: OR=0.82, 95% CI [0.56-0.97], p value=0.048). Mental health specialist density was positively associated with treatment engagement among Blacks (specialist’ density per 10,000 in 5-mile travel distance 1.00-1.99 vs < 1.00: OR=1.63, 95% CI[1.03-4.51], p value=0.03; 2.00-4.99 specialists vs < 1.0 specialists: OR=2.28, 95% CI[1.21-7.11], p value<0.01; and ≥5.00 specialists vs <1 specialists: OR=1.74, 95% CI[1.11-5.53], p value=0.02). Regarding treatment completion, travel distance to the provider who initiated treatment had a statistically significant effect on all racial/ethnic groups. Those who lived 15 miles or more away from the provider who initiated the treatment were 22% less likely to complete the treatment as compared to those who had to travel less than 5 miles (OR=0.78, 95% CI [0.55-0.93], p value=0.03). For Aim 3, A total of 10,206 children and adolescents with an incident ADHD diagnosis were identified, of which 70% of were diagnosed by non-mental health specialists and among 68% of ADHD cases, medication was initiated by PCPs. The treatment engagement rate was 55%. Among those engaged with treatment, mean adherence (PDC) was 0.54 (0.24) and 27.5% with PDC ≥ 0.70. None of the geographic access measures were associated with the odds of treatment engagement and treatment adherence. Factors positively associated with treatment engagement were being White (Blacks vs Whites OR=0.63, 95% CI [0.54-0.73]; Hispanics vs Whites OR=0.43, 95% CI [0.37-0.49]), and diagnosed by mental health specialists (specialists vs PCP OR=1.20, 95% CI [1.05-1.37]), and factors positively associated with medication adherence were being White (Blacks vs Whites OR=0.40, 95% CI [0.31-0.52]; Hispanics vs Whites OR=0.43, 95% CI [0.33-0.55]), and treatment initiated by mental health specialists (specialists vs PCP OR=1.68, 95% CI [1.42-1.98]). Conclusion: Hispanics and Blacks were the most sensitive to geographic access to providers in receiving behavioral disorder screening and depression treatment. However, the study did not find any association of treatment engagement and medication adherence with the geographic access to providers in the ADHD subgroup. The findings of this study suggest that the involvement of PCP in mental health care may provide a solution to geographic access disparity.Item Risk of Substance Use in Attention Deficit/Hyperactivity Disorder with Predominately Inattentive Symptoms(2014-08) Upadhyay, Navneet; Chen, Hua; Aparasu, Rajender R.; Mgbere, OsaroObjective: Primary objective of the study was to compare the risk of substance use associated with various attention deficit/hyperactivity disorder (ADHD) subtypes in a large, nationally representative adolescent sample. Secondary objective was to explore the extent of this association being modified by pharmaceutical interventions especially in ADHD with predominately inattentive symptoms (ADHD/I subtype). Methods: National Comorbidity survey-Adolescent supplement (NCS-A), a nationally representative sample of adolescents (age 13-18) was used to assess the wide range of mental disorders and substance use information. Childhood ADHD subtypes and age at onset were calculated retrospectively with the variables of the ADHD DSM-IV criteria symptoms. Substance use information was assessed using the age at first use, frequency of use information about the illicit drugs, alcohol and tobacco. Pharmaceutical intervention and onset of therapy were collected from the parent’s questionnaire about the prescription medication used for relieving the symptoms of ADHD. Association between ADHD subtype and substance use was examined using the multivariate logistic regression analysis. Results: In a nationally representative adolescent sample of US, our study found that individuals with ADHD with predominately hyperactivity symptoms (ADHD/H) and ADHD with predominately hyperactivity and inattention symptoms (ADHD/C) subtype has similar future risk of using substance as ADHD/I subtypes [ADHD/H (OR=0.91, 95% C.I.=0.38-2.17) and ADHD/C (OR=1.49, 95 %CI=0.90-0.34)], after adjusting for socio-demographic factors, comorbid conditions, environmental factors and ADHD pharmacotherapy. ADHD pharmacotherapy is effective in reducing the risk of substance use in ADHD/C and ADHD/H subtype, whereas it was not effective for the purpose in the ADHD/I subtype (in ADHD/I with mood disorder, OR=0.483, 95% C.I.=0.1-2.34, p value=0.36 and in ADHD/I without mood disorder, OR=0.544, 95% C.I.=0.28, p value=0.29). Conclusion: ADHD/I have a significant risk of substance use which is comparable to other subtypes. Still pharmacotherapy is seem to be not effective intervention strategy in ADHD/I subtype. More research is required for examining the effectiveness of behavioral therapy in reducing risk of substance use in the individuals with ADHD/I subtype.