Search

Slide presentation to accompany manuscript. Background: The inability to identify dates of death in insurance claims data is a major limitation to retrospective claims-based research. Deaths likely result in disenrollment; however, disenrollment may also reflect a change in insurance provider. We aim to develop a user-friendly public algorithm to predict death within the year of disenrollment using an administrative claims database. Methods: We identified adults (18+ years) with at least 1 year of continuous enrollment prior to disenrollment in 2007-2018. Using Social Security Death Index, inpatient discharge status, and death indicators in the administrative data as the gold standard, we used claims in the prior year to predict death. Models including candidate predictors for age, sex, Census region, month of disenrollment, chronic condition indicators (components of the Elixhauser score), and prior healthcare utilization were estimated using used elastic net regression tuned by 5-fold cross-validation and final models evaluated in an independent testing set. Weighted analysis adjusts for rare outcome (i.e., class imbalance). Sensitivity and specificity associated with various thresholds of predicted probability to classify death at disenrollment were calculated. Results: We identified 13,360,460 beneficiaries who disenrolled during the study period, with 5% of patients who died within the 61 days of disenrollment. The strongest predictors of death were age at disenrollment, diagnosis of metastatic cancer in the year prior to death, and type of care received (e.g., inpatient stay, hospice care). Using a prediction threshold of 30%, the algorithm classified death at disenrollment with a sensitivity of 0.684 and specificity of 0.985 (ROC=0.97). Conclusions: Our algorithm uses publicly defined chronic conditions and utilization patterns that are easy to implement in claims data and predicts death at disenrollment. Users can easily implement the algorithm and can choose the prediction threshold (balancing sensitivity and specificity) to

Background: The inability to identify dates of death in insurance claims data is a major limitation to retrospective claims based research. If not an outcome, death is a competing risk and poses a threat to validity when treated as non-informative right censoring. Objectives: We aim to develop a user-friendly public algorithm to predict death within the year of disenrollment using an administrative claims database. Methods: We identified adults (18+ years) with at least 2 years of continuous enrollment prior to disenrollment between 01/2007 and 01/2018. Leveraging unique linkages in addition to data that are typically unavailable in the publicly licensed data, we ascertained date of death from the Social Security Death Index, inpatient discharge status, and death indicators in the administrative data. Models including candidate predictors for age, sex, Census region, month of disenrollment, year of disenrollment, chronic condition indicators (components of the Elixhauser score), and prior healthcare utilization were estimated using used elastic net regression tuned by 5-fold cross-validation and final models evaluated in an independent testing set. Weighted analysis adjusts for rare outcome (i.e., class imbalance). Sensitivity, specificity, and ROC associated with various thresholds of predicted probability to classify death at disenrollment were calculated. Results: Overall, we identified 13,360,460 beneficiaries who disenrolled during the study period, with 5% of patients who died within the year of disenrollment. The strongest predictors of death were age at disenrollment, diagnosis of metastatic cancer in the year prior to death, and type of care received (e.g., inpatient stay, hospice care). Using a prediction threshold of 30%, the algorithm classified death at disenrollment with a sensitivity of 0.684 and specificity of 0.985 (ROC=0.97. At the same prediction threshold, the weighted algorithm classified death with a sensitivity of .947 and a specificity of 0.898 (ROC=.973). Conclusions: Our algorithm uses publicly defined chronic conditions and utilization patterns that are easy to implement in claims data and predicts death at disenrollment with high specificity and varying sensitivity depending on the chosen prediction threshold. Users can easily implement the algorithm and can choose the prediction threshold (balancing sensitivity and specificity) to meet the needs of the specific study at hand.

The Mental Health Parity and Addiction Equity Act requires insurance parity for mental health/substance use disorder and general medical services. Previous research found that parity did not increase mental health/substance use disorder spending and lowered out-of-pocket spending. Whether parity's effects differ by diagnosis is unknown. The authors examined this question in the context of parity implementation in the Federal Employees Health Benefits (FEHB) Program.The authors compared mental health/substance use disorder treatment use and spending before and after parity (2000 and 2002, respectively) for two groups: FEHB enrollees diagnosed in 1999 with bipolar disorder, major depression, or adjustment disorder (N=19,094) and privately insured enrollees unaffected by the policy in a comparison national sample (N=10,521). Separate models were fitted for each diagnostic group. A difference-in-difference design was used to control for secular time trends and to better reflect the specific impact of parity on spending and utilization.Total spending was unchanged among enrollees with bipolar disorder and major depression but decreased for those with adjustment disorder (-$62, 99.2% CI=-$133, -$11). Out-of-pocket spending decreased for all three groups (bipolar disorder: -$148, 99.2% CI=-$217, -$85; major depression: -$100, 99.2% CI=-$123, -$77; adjustment disorder: -$68, 99.2% CI=-$84, -$54). Total annual utilization (e.g., medication management visits, psychotropic prescriptions, and mental health/substance use disorder hospitalization bed days) remained unchanged across all diagnoses. Annual psychotherapy visits decreased significantly only for individuals with adjustment disorders (-12%, 99.2% CI=-19%, -4%).Parity implemented under managed care improved financial protection and differentially affected spending and psychotherapy utilization across groups. There was some evidence that resources were preferentially preserved for diagnoses that are typically more severe or chronic and reduced for diagnoses expected to be less so.