Development and validation of a prediction model for opioid use disorder among youth.

BACKGROUND: Youth are vulnerable to opioid use initiation and its complications. With growing rates of opioid overdose, strategies to identify youth at risk of opioid use disorder (OUD) to efficiently focus prevention interventions are needed. This study developed and validated a prediction model of OUD in youth aged 14-18 years. METHODS: The model was developed in a Colorado healthcare system (derivation site) using Cox proportional hazards regression analysis. Model predictors and outcomes were identified using electronic health record data. The model was externally validated in a separate Denver safety net health system (validation site). Youth were followed for up to 3.5 years. We evaluated internal and external validity using discrimination and calibration. RESULTS: The derivation cohort included 76,603 youth, of whom 108 developed an OUD diagnosis. The model contained 3 predictors (smoking status, mental health diagnosis, and non-opioid substance use or disorder) and demonstrated good calibration (p = 0.90) and discrimination (bootstrap-corrected C-statistic = 0.76: 95 % CI = 0.70, 0.82). Sensitivity and specificity were 57 % and 84 % respectively with a positive predictive value (PPV) of 0.49 %. The validation cohort included 45,790 youth of whom, 74 developed an OUD diagnoses. The model demonstrated poorer calibration (p < 0.001) but good discrimination (C-statistic = 0.89; 95 % CI = 0.84, 0.95), sensitivity of 87.8 % specificity of 68.6 %, and PPV of 0.45 %. CONCLUSIONS: In two Colorado healthcare systems, the prediction model identified 57-88 % of subsequent OUD diagnoses in youth. However, PPV < 1% suggests universal prevention strategies for opioid use in youth may be the best health system approach.

Vaccine adverse events in a safety net healthcare system and a managed care organization.

BACKGROUND: The Institute of Medicine, in a 2013 report, recommended that the Vaccine Safety Datalink (VSD) expand collaborations to include more diversity in the study population. Kaiser Permanente Colorado (KPCO), an established VSD site, partnered with Denver Health (DH), an integrated safety net healthcare system, to demonstrate the feasibility of integrating DH data within the VSD. Prior to incorporating the data, we examined the identification of specific vaccine associated adverse events (VAEs) in these two distinct healthcare systems. METHODS: We conducted retrospective cohort analyses within KPCO and DH to compare select VAEs between the two populations. We examined the following associations between January 1, 2004 and December 31, 2013: Measles, Mumps, and Rubella (MMR) vaccine and febrile seizures in children 2years and younger, intussusception after rotavirus vaccine in infants 4-34weeks, syncope after adolescent vaccines (Tetanus, Diphtheria, acellular Pertussis; Meningococcal and Human Papillomavirus) in adolescents 13-17years and medically attended local reactions after pneumococcal polysaccharide (PPSV23) vaccine in adults 65years and older. Both sites used similar data procurement methods and chart review processes. RESULTS: For seizures after MMR vaccine (KPCO - 3.15vs. DH - 2.97/10,000 doses) and syncope after all adolescent vaccines (KPCO - 3.0vs. DH - 2.37/10,000 doses), the chart confirmed rates were comparable at the two sites. However, for medically attended local reactions after PPSV23, there were differences in chart confirmed rates between the sites (KPCO - 31.65vs. DH - 14.90/10,000 doses). For intussusception after rotavirus vaccine, the number of cases was too low to make a valid comparison (KPCO - 0vs. DH - 0.13/10,000 doses). CONCLUSION: We demonstrated that data on important targeted VAEs can be captured at DH and rates appear similar to those at KPCO. Work is ongoing on the optimal approach to assimilate DH data as a potential safety net healthcare system in the VSD.

Screening for and Diagnosis of Depression Among Adolescents in a Large Health Maintenance Organization.

OBJECTIVE: The aim of this analysis was to determine changes in patterns of depression screening and diagnosis over three years in primary and specialty mental health care in a large health maintenance organization (HMO) as part of a project to develop quality measures for adolescent depression treatment. METHODS: Two series of aggregate data (2010-2012) were gathered from the electronic health records of the HMO for 44,342 unique adolescents (ages 12 to 21) who had visits in primary and mental health care. Chi square tests assessed the significance of changes in frequency and departmental location of Patient Health Questionnaire-9 (PHQ-9) administration, incidence of depression symptoms, and depression diagnoses. RESULTS: There was a significant increase in PHQ-9 use, predominantly in primary care, consistent with internally generated organizational recommendations to increase screening with the PHQ-9. The increase in PHQ-9 use led to an increase in depression diagnoses in primary care and a shift in the location of some diagnoses from specialty mental health care to primary care. The increase in PHQ-9 use was also linked to a decrease in the proportion of positive PHQ-9 results that led to formal depression diagnoses. CONCLUSIONS: The rate of depression screening in primary care increased over the study period. This increase corresponded to an increase in the number of depression diagnoses made in primary care and a shift in the location in which depression diagnoses were made, from the mental health department to primary care. The frequency of positive PHQ-9 administrations not associated with depression diagnoses also increased.