Person:

Zafar, Sahar

BACKGROUND/OBJECTIVES: No data exist regarding the validity of International Classification of Disease (ICD)-10 dementia diagnoses within Medicare claims data. We examined the accuracy of claims-based diagnoses with respect to expert clinician adjudication using a novel database with individual-level linkages between electronic health record (EHR) and claims. DESIGN: In this retrospective observational study, two neurologists and two psychiatrists performed a standardized review of patients’ medical records from January-2016 to December-2018, and adjudicated dementia status. We measured the accuracy of three claims-based definitions of dementia against the reference standard. SETTING: Mass-General-Brigham Healthcare (MGB), Massachusetts, USA. PARTICIPANTS: From an eligible population of 40,690 fee-for-service (FFS) Medicare beneficiaries, aged 65-years and older, within the MGB Accountable Care Organization (ACO), we generated a random sample of 1,002 patients, stratified by the pretest likelihood of dementia using administrative surrogates. INTERVENTION: None. MEASUREMENTS: We evaluated the accuracy (area-under-receiver-operating-curve [AUROC]) and calibration (calibration-in-the-large [CITL] and calibration slope) of three ICD-10 claims-based definitions of dementia against clinician-adjudicated standards. We applied inverse probability weighting to reconstruct the eligible population and reported the mean and 95% confidence interval (95% CI) for all performance characteristics, using 10-fold cross-validation (CV). RESULTS: Beneficiaries had an average age of 75.3-years and were predominately female (59%) and non-Hispanic white (93%). The adjudicated prevalence of dementia in the eligible population was 7%. The best performing definition demonstrated excellent accuracy (CV-AUC 0.94; 95% CI 0.92-0.96) and was well-calibrated to the reference standard of clinician-adjudicated dementia (CV-CITL <0.001, CV-slope 0.97). CONCLUSION: This study is the first to validate ICD-10 diagnostic codes against a robust and replicable approach to dementia ascertainment using a real-world clinical reference standard. The best performing definition includes diagnostic codes with strong face validity and outperforms an updated version of a previously validated ICD-9 definition of dementia.

Importance: There is limited evidence concerning optimal seizure prophylaxis after spontaneous intracerebral hemorrhage (sICH). Objective: To evaluate which of four seizure prophylaxis strategies provides the greatest net benefit for sICH patients. Design, Setting, and Participants: Decision model simulating four common scenarios: 1) 60-year-old male with low early- (≤ 7 days post-stroke) (10%) and late-seizure risks (3.6% or 9.8%), and average short- and long-term adverse drug reaction (ADR) risks (9% and 30%, respectively); 2) 80-year-old female with low early- (10%) and late-seizure risks (3.6% or 9.8%), and high short- and long-term ADR risks (24% and 80%); 3) 55-year-old male with high early- (19%) and late-seizure risks (34.8% or 46.2%), and low short- and long-term ADR risks (9% and 30%); and 4) 45-year-old female with high early- (19%) and late-seizure risks (34.8% or 46.2%), and high short- and long-term ADR risks (18% and 60%). Interventions: Four antiseizure drug strategies: 1) Conservative: short-term (7-day) secondary early-seizure prophylaxis with long-term therapy after late-seizure; 2) Moderate: long-term secondary early- or late-seizure prophylaxis; 3) Aggressive: long-term primary prophylaxis; 4) Risk-guided: short-term secondary early-seizure prophylaxis among low-risk patients (2HELPS2B score), short-term primary prophylaxis among higher-risk patients, and long-term late-seizure secondary therapy. Main Outcomes and Measures: Quality-adjusted life years (QALYs). Results: For scenario 1, risk-guided strategy was preferred over conservative, moderate, and aggressive (QALYs = 8.13, 8.08, 8.07, and 7.88, respectively). For scenario 2, conservative and risk-guided strategies performed comparably and were favored over moderate and aggressive (QALYs = 2.18, 2.17, 2.09, 1.15). For scenario 3, aggressive strategy was preferred over moderate, risk-guided and conservative (QALY = 9.21, 8.93, 8.98, 8.77). For scenario 4, risk-guided strategy was preferred over conservative, moderate, and aggressive (QALY = 11.53, 11.23, 10.93, 8.08). Sensitivity analyses suggested that short-term strategies are preferred under most scenarios, and the risk-guided strategy performs comparably or better than alternative strategies in most settings. Conclusions and Relevance: Our model indicates that short-term (7-day) prophylaxis dominates longer-term therapy following sICH. Implementation of the 2HELPS2B score to guide clinical decisions for initiation of short-term primary versus secondary early-seizure prophylaxis should be considered for all patients after sICH.