Dataset Information

Comparative Effectiveness of New Approaches to Improve Mortality Risk Models From Medicare Claims Data.

ABSTRACT:

Importance

Risk adjustment models using claims-based data are central in evaluating health care performance. Although US Centers for Medicare & Medicaid Services (CMS) models apply well-vetted statistical approaches, recent changes in the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) coding system and advances in computational capabilities may provide an opportunity for enhancement.

Objective

To examine whether changes using already available data would enhance risk models and yield greater discrimination in hospital-level performance measures.

Design, setting, and participants

This comparative effectiveness study used ICD-9-CM codes from all Medicare fee-for-service beneficiary claims for hospitalizations for acute myocardial infarction (AMI), heart failure (HF), or pneumonia among patients 65 years and older from July 1, 2013, through September 30, 2015. Changes to current CMS mortality risk models were applied incrementally to patient-level models, and the best model was tested on hospital performance measures to model 30-day mortality. Analyses were conducted from April 19, 2018, to September 19, 2018.

Main outcomes and measures

The main outcome was all-cause death within 30 days of hospitalization for AMI, HF, or pneumonia, examined using 3 changes to current CMS mortality risk models: (1) incorporating present on admission coding to better exclude potential complications of care, (2) separating index admission diagnoses from those of the 12-month history, and (3) using ungrouped ICD-9-CM codes.

Results

There were 361 175 hospital admissions (mean [SD] age, 78.6 [8.4] years; 189 225 [52.4%] men) for AMI, 716 790 hospital admissions (mean [SD] age, 81.1 [8.4] years; 326 825 [45.6%] men) for HF, and 988 225 hospital admissions (mean [SD] age, 80.7 [8.6] years; 460 761 [46.6%] men) for pneumonia during the study; mean 30-day mortality rates were 13.8% for AMI, 12.1% for HF, and 16.1% for pneumonia. Each change to the models was associated with incremental gains in C statistics. The best model, incorporating all changes, was associated with significantly improved patient-level C statistics, from 0.720 to 0.826 for AMI, 0.685 to 0.776 for HF, and 0.715 to 0.804 for pneumonia. Compared with current CMS models, the best model produced wider predicted probabilities with better calibration and Brier scores. Hospital risk-standardized mortality rates had wider distributions, with more hospitals identified as good or bad performance outliers.

Conclusions and relevance

Incorporating present on admission coding and using ungrouped index and historical ICD-9-CM codes were associated with improved patient-level and hospital-level risk models for mortality compared with the current CMS models for all 3 conditions.

SUBMITTER: Krumholz HM

PROVIDER: S-EPMC6647547 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundElectronic health record (EHR) prediction models may be easier to use in busy clinical settings since EHR data can be auto-populated into models. This study assessed whether adding functional status and/or Medicare claims data (which are often not available in EHRs) improves the accuracy of a previously developed Veterans Affairs (VA) EHR-based mortality index.MethodsThis was a retrospective cohort study of veterans aged 75 years and older enrolled in VA primary care clinics followed from January 2014 to April 2020 (n = 62,014). We randomly split participants into development (n = 49,612) and validation (n = 12,402) cohorts. The primary outcome was all-cause mortality. We performed logistic regression with backward stepwise selection to develop a 100-predictor base model using 854 EHR candidate variables, including demographics, laboratory values, medications, healthcare utilization, diagnosis codes, and vitals. We incorporated functional measures in a base + function model by adding activities of daily living (range 0-5) and instrumental activities of daily living (range 0-7) scores. Medicare data, including healthcare utilization (e.g., emergency department visits, hospitalizations) and diagnosis codes, were incorporated in a base + Medicare model. A base + function + Medicare model included all data elements. We assessed model performance with the c-statistic, reclassification metrics, fraction of new information provided, and calibration plots.ResultsIn the overall cohort, mean age was 82.6 years and 98.6% were male. At the end of follow-up, 30,263 participants (48.8%) had died. The base model c-statistic was 0.809 (95% CI 0.805-0.812) in the development cohort and 0.804 (95% CI 0.796-0.812) in the validation cohort. Validation cohort c-statistics for the base + function, base + Medicare, and base + function + Medicare models were 0.809 (95% CI 0.801-0.816), 0.811 (95% CI 0.803-0.818), and 0.814 (95% CI 0.807-0.822), respectively. Adding functional status and Medicare data resulted in similarly small improvements among other model performance measures. All models showed excellent calibration.ConclusionsIncorporation of functional status and Medicare data into a VA EHR-based mortality index led to small but likely clinically insignificant improvements in model performance.

Project description:ObjectivesAdministrative claims data sets are often used for emergency care research and policy investigations of healthcare resource utilization, acute care practices, and evaluation of quality improvement interventions. Despite the high profile of emergency department (ED) visits in analyses using administrative claims, little work has evaluated the degree to which existing definitions based on claims data accurately captures conventionally defined hospital-based ED services. We sought to construct an operational definition for ED visitation using a comprehensive Medicare data set and to compare this definition to existing operational definitions used by researchers and policymakers.MethodsWe examined four operational definitions of an ED visit commonly used by researchers and policymakers using a 20% sample of the 2012 Medicare Chronic Condition Warehouse (CCW) data set. The CCW data set included all Part A (hospital) and Part B (hospital outpatient, physician) claims for a nationally representative sample of continuously enrolled Medicare fee-for-services beneficiaries. Three definitions were based on published research or existing quality metrics including: 1) provider claims-based definition, 2) facility claims-based definition, and 3) CMS Research Data Assistance Center (ResDAC) definition. In addition, we developed a fourth operational definition (Yale definition) that sought to incorporate additional coding rules for identifying ED visits. We report levels of agreement and disagreement among the four definitions.ResultsOf 10,717,786 beneficiaries included in the sample data set, 22% had evidence of ED use during the study year under any of the ED visit definitions. The definition using provider claims identified a total of 4,199,148 ED visits, the facility definition 4,795,057 visits, the ResDAC definition 5,278,980 ED visits, and the Yale definition 5,192,235 ED visits. The Yale definition identified a statistically different (p < 0.05) collection of ED visits than all other definitions including 17% more ED visits than the provider definition and 2% fewer visits than the ResDAC definition. Differences in ED visitation counts between each definition occurred for several reasons including the inclusion of critical care or observation services in the ED, discrepancies between facility and provider billing regulations, and operational decisions of each definition.ConclusionCurrent operational definitions of ED visitation using administrative claims produce different estimates of ED visitation based on the underlying assumptions applied to billing data and data set availability. Future analyses using administrative claims data should seek to validate specific definitions and inform the development of a consistent, consensus ED visitation definitions to standardize research reporting and the interpretation of policy interventions.

Project description:ImportancePredicting payments for particular conditions or populations is essential for research, benchmarking, public reporting, and calculations for population-based programs. Centers for Medicare & Medicaid Services (CMS) models often group codes into disease categories, but using single, rather than grouped, diagnostic codes and leveraging present on admission (POA) codes may enhance these models.ObjectiveTo determine whether changes to the candidate variables in CMS models would improve risk models predicting patient total payment within 30 days of hospitalization for acute myocardial infarction (AMI), heart failure (HF), and pneumonia.Design, setting, and participantsThis comparative effectiveness research study used data from Medicare fee-for-service hospitalizations for AMI, HF, and pneumonia at acute care hospitals from July 1, 2013, through September 30, 2015. Payments across multiple care settings, services, and supplies were included and adjusted for geographic and policy variations, corrected for inflation, and winsorized. The same data source was used but varied for the candidate variables and their selection, and the method used by CMS for public reporting that used grouped codes was compared with variations that used POA codes and single diagnostic codes. Combinations of use of POA codes, separation of index admission diagnoses from those in the previous 12 months, and use of individual International Classification of Diseases, Ninth Revision, Clinical Modification codes instead of grouped diagnostic categories were tested. Data analysis was performed from December 4, 2017, to June 10, 2019.Main outcomes and measuresThe models' goodness of fit was compared using root mean square error (RMSE) and the McFadden pseudo R2.ResultsAmong the 1 943 049 total hospitalizations of the study participants, 343 116 admissions were for AMI (52.5% male; 37.4% aged ≤74 years), 677 044 for HF (45.5% male; 25.9% aged ≤74 years), and 922 889 for pneumonia (46.4% male; 28.2% aged ≤74 years). The mean (SD) 30-day payment was $23 103 ($18 221) for AMI, $16 365 ($12 527) for HF, and $17 097 ($12 087) for pneumonia. Each incremental model change improved the pseudo R2 and RMSE. Incorporating all 3 changes improved the pseudo R2 of the patient-level models from 0.077 to 0.129 for AMI, from 0.042 to 0.129 for HF, and from 0.114 to 0.237 for pneumonia. Parallel improvements in RMSE were found for all 3 conditions.Conclusions and relevanceLeveraging POA codes, separating index from previous diagnoses, and using single diagnostic codes improved payment models. Better models can potentially improve research, benchmarking, public reporting, and calculations for population-based programs.

Project description:BackgroundWith increasing use of real world data in observational health care research, data quality assessment of these data is equally gaining in importance. Electronic health record (EHR) or claims datasets can differ significantly in the spectrum of care covered by the data.ObjectiveIn our study, we link provider specialty with diagnoses (encoded in International Classification of Diseases) with a motivation to characterize data completeness.MethodsWe develop a set of measures that determine diagnostic span of a specialty (how many distinct diagnosis codes are generated by a specialty) and specialty span of a diagnosis (how many specialties diagnose a given condition). We also analyze ranked lists for both measures. As use case, we apply these measures to outpatient Medicare claims data from 2016 (3.5 billion diagnosis-specialty pairs). We analyze 82 distinct specialties present in Medicare claims (using Medicare list of specialties derived from level III Healthcare Provider Taxonomy Codes).ResultsA typical specialty diagnoses on average 4,046 distinct diagnosis codes. It can range from 33 codes for medical toxicology to 25,475 codes for internal medicine. Specialties with large visit volume tend to have large diagnostic span. Median specialty span of a diagnosis code is 8 specialties with a range from 1 to 82 specialties. In total, 13.5% of all observed diagnoses are generated exclusively by a single specialty. Quantitative cumulative rankings reveal that some diagnosis codes can be dominated by few specialties. Using such diagnoses in cohort or outcome definitions may thus be vulnerable to incomplete specialty coverage of a given dataset.ConclusionWe propose specialty fingerprinting as a method to assess data completeness component of data quality. Datasets covering a full spectrum of care can be used to generate reference benchmark data that can quantify relative importance of a specialty in constructing diagnostic history elements of computable phenotype definitions.