Evaluating Predictors of Participation in Telephone-Based Social-Connectedness Interventions for Older Adults: A Dual Machine-Learning and Regression Approach.

Social isolation is a well-documented contributor to poor mental and physical health, and interventions promoting social connectedness have been associated with various health benefits. This study examined predictors of participation in a telephone-based social connectedness intervention for socially isolated older adults. Data were obtained from a social-connectedness intervention that paired college students with Houston-area, community-dwelling adults aged 65 years and older and enrolled in Medicare Advantage plans. We combined machine learning and regression techniques to identify significant predictors of program participation. The following machine-learning methods were implemented: (1) k-nearest neighbors, (2) decision tree and ensembles of decision trees, (3) gradient-boosted decision tree, and (4) random forest. The primary outcome was a binary flag indicating participation in the telephone-based social-connectedness intervention. The most predictive variables in the ML models, with scores corresponding to the 90th percentile or greater, were included in the regression analysis. The predictive ability of each model showed high discriminative power, with test accuracies greater than 95%. Our findings suggest that telephone-based social-connectedness interventions appeal to individuals with disabilities, depression, arthritis, and higher risk scores. scores. Recognizing features that predict participation in social-connectedness programs is the first step to increasing reach and fostering patient engagement.

Charlson Comorbidity Index: ICD-9 Update and ICD-10 Translation.

BACKGROUND: The original Charlson Comorbidity Index (CCI) encompassed 19 categories of medical conditions that were identifiable in medical records. Subsequent publications provided scoring algorithms based on International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes. The recent adoption of International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) codes in the United States created a need for a new scoring scheme. In addition, a review of existing claims-based scoring systems suggested 3 areas for improvement: the lack of explicit identification of secondary diabetes, the lack of differentiation between HIV infection and AIDS, and insufficient guidance on scoring hierarchy. In addition, addressing the third need raised the issue of disease severity in renal disease. OBJECTIVES: This initiative aimed to create an expanded and refined ICD-9 scoring system for CCI, addressing the classification of issues noted above, create a corresponding ICD-10 system, assess the comparability of ICD-9- and ICD-10-based scores, and validate the new scoring scheme. METHODS: We created ICD-9 and ICD-10 code tables for 19 CCI medical conditions. The new scoring scheme was labeled CDMF CCI and was tested using claims-based data for individuals aged ≥65 years who participated in a Humana Medicare Advantage plan during at least 1 of 3 consecutive 12-month periods. Two 12-month periods were during the ICD-9 era and the third 12-month period was during the ICD-10 era. Because many individuals were counted in more than one 12-month period, we described the study population as comprising 3 panels. We used regression models to analyze the association between the CCI score and same-year inpatient admissions and near-term (90-day) mortality. Additional testing was done by comparing the mean CCI score or disease prevalence in the 3 subpopulations of people with HIV/AIDS, renal disease, or diabetes. Finally, we calculated area under the receiver operating characteristics (AUC-ROC) curve values by applying the Deyo system and our ICD-9 and ICD-10 scoring systems. RESULTS: The CDMF ICD-9 and ICD-10 scoring scheme yielded comparable scores across the 3 panels, and inpatient admissions and mortality rates consistently increased in each panel as the CCI score increased. Comparisons of the performance of the Deyo system and our proposed CDMF ICD-9 system in the 3 key subpopulations showed that the CDMF ICD-9 system produced a lower CCI score in the presence of HIV infection without AIDS, achieved similar detection ability of diabetes, and allowed good differentiation between mild-to-moderate and severe renal disease. AUC-ROC values were similar between the CDMF ICD-9 coding system and the Deyo system. CONCLUSION: Our results support the implementation of the CDMF CCI scoring instrument to triage individual patients for disease- and care-management programs. In addition, the CDMF scheme allows for a more precise understanding of chronic disease at a population level, thus allowing health systems and plans to design services and benefits to meet multifactorial clinical needs. Preliminary validation sets the stage for further testing using long-term follow-up data and for the adaptation of this coding scheme to a chart review instrument.

Diabetes Complications Severity Index (DCSI)-Update and ICD-10 translation.

AIMS: The Diabetes Complications Severity Index (DCSI) converts diagnostic codes and laboratory results into a 14-level metric quantifying the long-term effects of diabetes on seven body systems. Adoption of the International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) necessitates translation from ICD-9-CM and creates refinement opportunities. METHODS: ICD-9 codes for secondary and primary diabetes plus all five ICD-10 diabetes categories were incorporated into an updated tool. Additional modifications were made to improve the accuracy of severity assignments. SUBJECTS: The tools were tested in a Medicare Advantage population. RESULTS: In the type 2 subpopulation, prevalence steadily declined with increasing score according to the updated DCSI tool, whereas the original tool resulted in an aberrant local prevalence peak at DCSI = 2. In the type 1 subpopulation, score prevalence was greater in type 1 versus type 2 subpopulations (3 versus 0) according to both instruments. Both instruments predicted current-year inpatient admissions risk and near-future mortality, using either purely ICD-9 data or a mix of ICD-9 and ICD-10 data. DISCUSSION: While the performance of the tool with purely ICD-10 data has yet to be evaluated, this updated tool makes assessment of diabetes patient severity and complications possible in the interim.