scholarly journals Assessing the Value of Unsupervised Clustering in Predicting Persistent High Health Care Utilizers: Retrospective Analysis of Insurance Claims Data (Preprint)

2021 ◽  
Author(s):  
Raghav Ramachandran ◽  
Michael J McShea ◽  
Stephanie N Howson ◽  
Howard S Burkom ◽  
Hsien-Yen Chang ◽  
...  
Keyword(s):  
Mental Health ◽  
Health Care ◽  
Retrospective Analysis ◽  
Otitis Media ◽  
Regression Models ◽  
Claims Data ◽  
Complex Model ◽  
Insurance Claims Data ◽  
Study Population ◽  
Full Population

BACKGROUND A high proportion of health care services are persistently utilized by a small subpopulation of patients. To improve clinical outcomes while reducing costs and utilization, population health management programs often provide targeted interventions to patients who may become persistent high users/utilizers (PHUs). Enhanced prediction and management of PHUs can improve health care system efficiencies and improve the overall quality of patient care. OBJECTIVE The aim of this study was to detect key classes of diseases and medications among the study population and to assess the predictive value of these classes in identifying PHUs. METHODS This study was a retrospective analysis of insurance claims data of patients from the Johns Hopkins Health Care system. We defined a PHU as a patient incurring health care costs in the top 20% of all patients’ costs for 4 consecutive 6-month periods. We used 2013 claims data to predict PHU status in 2014-2015. We applied latent class analysis (LCA), an unsupervised clustering approach, to identify patient subgroups with similar diagnostic and medication patterns to differentiate variations in health care utilization across PHUs. Logistic regression models were then built to predict PHUs in the full population and in select subpopulations. Predictors included LCA membership probabilities, demographic covariates, and health utilization covariates. Predictive powers of the regression models were assessed and compared using standard metrics. RESULTS We identified 164,221 patients with continuous enrollment between 2013 and 2015. The mean study population age was 19.7 years, 55.9% were women, 3.3% had ≥1 hospitalization, and 19.1% had 10+ outpatient visits in 2013. A total of 8359 (5.09%) patients were identified as PHUs in both 2014 and 2015. The LCA performed optimally when assigning patients to four probability disease/medication classes. Given the feedback provided by clinical experts, we further divided the population into four diagnostic groups for sensitivity analysis: acute upper respiratory infection (URI) (n=53,232; 4.6% PHUs), mental health (n=34,456; 12.8% PHUs), otitis media (n=24,992; 4.5% PHUs), and musculoskeletal (n=24,799; 15.5% PHUs). For the regression models predicting PHUs in the full population, the F1-score classification metric was lower using a parsimonious model that included LCA categories (F1=38.62%) compared to that of a complex risk stratification model with a full set of predictors (F1=48.20%). However, the LCA-enabled simple models were comparable to the complex model when predicting PHUs in the mental health and musculoskeletal subpopulations (F1-scores of 48.69% and 48.15%, respectively). F1-scores were lower than that of the complex model when the LCA-enabled models were limited to the otitis media and acute URI subpopulations (45.77% and 43.05%, respectively). CONCLUSIONS Our study illustrates the value of LCA in identifying subgroups of patients with similar patterns of diagnoses and medications. Our results show that LCA-derived classes can simplify predictive models of PHUs without compromising predictive accuracy. Future studies should investigate the value of LCA-derived classes for predicting PHUs in other health care settings.

Profiling Primary Care Physicians for a New Managed Care Network

1997 ◽  
Vol 10 (3) ◽  
pp. 173-186 ◽  
Author(s):  
R. J. Ozminkowski ◽  
M. Noether ◽  
P. Nathanson ◽  
K. M. Smith ◽  
B. E. Raney ◽  
...  
Keyword(s):  
Health Care ◽  
Primary Care Physicians ◽  
Claims Data ◽  
Severity Of Illness ◽  
Policy Implications ◽  
Case Mix ◽  
Reform Strategies ◽  
Insurance Program ◽  
Insurance Claims Data ◽  
Care Costs

We developed methods for comparing physicians who would be selected to participate in a major employer's self-insurance program. These methods used insurance claims data to identify and profile physicians according to deviations from prevailing practice and outcome patterns, after considering differences in case-mix and severity of illness among the patients treated by those providers. The discussion notes the usefulness and limitations of claims data for this and other purposes. We also comment on policy implications and the relationships between our methods and health care reform strategies designed to influence overall health care costs.

scholarly journals Direct and Indirect Costs of Asthma Burden in Abu Dhabi: A Retrospective Analysis of Insurance Claims Data from 2015 to 2018

2021 ◽  
Vol Volume 13 ◽  
pp. 969-980
Author(s):  
Khulood Al Mazrouei ◽  
Asma Ibrahim Almannaei ◽  
Faiza Medeni Nur ◽  
Nagham Bachnak ◽  
Ashraf Alzaabi
Keyword(s):  
Retrospective Analysis ◽  
Claims Data ◽  
Indirect Costs ◽  
Abu Dhabi ◽  
Insurance Claims ◽  
Direct And Indirect Costs ◽  
Insurance Claims Data

scholarly journals Decomposition of outpatient health care spending by disease - a novel approach using insurance claims data

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Michael Stucki ◽  
Janina Nemitz ◽  
Maria Trottmann ◽  
Simon Wieser
Keyword(s):  
Health Care ◽  
Health Insurance ◽  
Claims Data ◽  
Health Care Spending ◽  
Insurance Claims ◽  
Diagnostic Coding ◽  
Insurance Claims Data ◽  
Health Insurance Claims Data ◽  
Health Insurance Claims ◽  
Disease Specific

Abstract Background Decomposing health care spending by disease, type of care, age, and sex can lead to a better understanding of the drivers of health care spending. But the lack of diagnostic coding in outpatient care often precludes a decomposition by disease. Yet, health insurance claims data hold a variety of diagnostic clues that may be used to identify diseases. Methods In this study, we decompose total outpatient care spending in Switzerland by age, sex, service type, and 42 exhaustive and mutually exclusive diseases according to the Global Burden of Disease classification. Using data of a large health insurance provider, we identify diseases based on diagnostic clues. These clues include type of medication, inpatient treatment, physician specialization, and disease specific outpatient treatments and examinations. We determine disease-specific spending by direct (clues-based) and indirect (regression-based) spending assignment. Results Our results suggest a high precision of disease identification for many diseases. Overall, 81% of outpatient spending can be assigned to diseases, mostly based on indirect assignment using regression. Outpatient spending is highest for musculoskeletal disorders (19.2%), followed by mental and substance use disorders (12.0%), sense organ diseases (8.7%) and cardiovascular diseases (8.6%). Neoplasms account for 7.3% of outpatient spending. Conclusions Our study shows the potential of health insurance claims data in identifying diseases when no diagnostic coding is available. These disease-specific spending estimates may inform Swiss health policies in cost containment and priority setting.

Use of Insurance Claims Data in Measuring Quality of Care

1990 ◽  
Vol 6 (2) ◽  
pp. 263-271 ◽  
Author(s):  
Kathleen N. Lohr
Keyword(s):  
Health Care ◽  
Quality Of Care ◽  
Administrative Data ◽  
Claims Data ◽  
Outcome Analysis ◽  
Process Of Care ◽  
Service Information ◽  
Insurance Claims Data ◽  
Data Banks

AbstractThis article discusses data that might be used for measuring quality of care, for health care administrative purposes, and for tracking the use of technologies. The advantages and limitations of administrative data banks for research purposes and some process-of-care and outcome analysis are noted. Three important obstacles to their use—reliability of diagnosis and service information, unique patient identifiers, and provider identifiers—are discussed briefly.

Pattern of utilization of pegfilgrastim in patients with chemotherapy-induced neutropenia: A retrospective analysis of administrative claims data

2009 ◽  
Vol 27 (15_suppl) ◽  
pp. 9624-9624
Author(s):  
A. T. Skarin ◽  
F. Vekeman ◽  
F. Laliberté ◽  
O. Afonja ◽  
M. Lafeuille ◽  
...  
Keyword(s):  
Retrospective Analysis ◽  
Claims Data ◽  
Real Life ◽  
Cytotoxic Chemotherapy ◽  
Administrative Claims ◽  
Cancer Type ◽  
Insurance Claims Data ◽  
Health Insurance Claims Data ◽  
Long Acting ◽  
Practice Methods

9624 Background: Pegfilgrastim is a long-acting granulocyte colony-stimulating factor (G-CSF) used to prevent or treat febrile neutropenia associated with myelosuppressive anticancer therapies. According to the prescribing information, pegfilgrastim should not be administered within 14 days before or 24 hours after cytotoxic chemotherapy because of the potential for myeloid toxicity. This study examined use patterns of pegfilgrastim in real-life practice. Methods: Analysis of health insurance claims data in 2000- 2007 from > 35 large health plans across the US was conducted. Patients who had a cancer diagnosis and chemotherapy within 120 days of their first pegfilgrastim injection were identified. The proportion of pegfilgrastim injections that were followed by administration of chemotherapy within 11 and 9 days was calculated. Analysis was also stratified by cancer type [Non-Hodgkin's lymphoma (NHL), lung, breast]. Results: A total of 13,526 cancer patients received 57,118 pegfilgrastim injections. NHL, lung, and breast cohorts comprised 2,722, 2,772, and 4,955 patients, respectively. Mean age (SD) was 55.0 (11.6) and women represented 65.9% of study population. Among all cancer types, 19.2% of pegfilgrastim injections had a chemotherapy claim within the following 11 days. This pattern of use was the highest in NHL (18.9%), followed by lung (17.1%), and breast (16.2%). Similar results were observed in the 9-day sensitivity analysis (see Table ). Conclusions: Based on the retrospective analysis of this administrative claims database, the use of pegfilgrastim within 11 days of an administration of chemotherapy was observed in 15–20% of cases which is inconsistent with the recommended guidelines. Pegfilgrastim use in these situations may have the potential to increase sensitivity of rapidly dividing myeloid cells to cytotoxic chemotherapy. Further research is being conducted to assess the related clinical and economic impact of this pattern of usage. [Table: see text] [Table: see text]

Associations between perceptions of care experiences and receipt of mental health care among older adults with cancer.

2020 ◽  
Vol 38 (29_suppl) ◽  
pp. 175-175
Author(s):  
Lisa M Lines ◽  
Daniel H Barch ◽  
Diana Zabala ◽  
Michael T. Halpern ◽  
Paul Jacobsen ◽  
...  
Keyword(s):  
Mental Health ◽  
Older Adults ◽  
Health Care ◽  
Linear Regression ◽  
Mental Health Care ◽  
Mood Disorders ◽  
Regression Models ◽  
Case Mix ◽  
Linear Regression Models ◽  
Fair Poor

175 Background: Older adults with cancer and worse self-rated mental health report worse care experiences. We hypothesized that, controlling for health and demographic characteristics, older adults with cancer who received care for anxiety or mood disorders would report better care experiences. Methods: We used SEER-CAHPS data to identify Medicare beneficiaries, aged 66 and over, diagnosed from August 2006 through December 2013 with one of the 10 most prevalent solid tumor malignancies. To identify utilization for anxiety or mood disorders (screening, diagnosis, or treatment), we analyzed inpatient, outpatient, home health, physician, and prescription drug claims from 12 months before through up to 5 years after cancer diagnosis. Outcomes of interest were global care experience ratings (Overall Care, Personal Doctor, and Specialist; rated on a 0-10 scale) and composite measures (Getting Needed Care, Getting Care Quickly, and Doctor Communication; scored from 0-100). We estimated linear regression models and also used a Bayesian Model Averaging approach, adjusting for standard case-mix adjustors (including sociodemographics and self-reported general health and mental health status [MHS]) and other characteristics, including cancer site and stage at diagnosis. We also included interaction terms between mental health care utilization and MHS. Results: Approximately 22% of the overall sample (n = 4,998) had both cancer and a claim for an anxiety or mood disorder, and of those individuals, 18% reported fair/poor MHS. Only 7% of those in the cancer-only cohort reported fair/poor MHS. Before adjusting for mental health utilization, worse MHS was significantly associated with worse experience of care. After accounting for anxiety/mood disorder-related utilization, linear regression models showed no significant associations between fair/poor MHS and worse care experiences, while Bayesian models found that reliable associations remained between worse MHS and lower global ratings of Overall Care and Specialist. Conclusions: Utilization for anxiety/mood disorders mediates the association between fair/poor MHS and worse care experiences. Although MHS is a case-mix adjustor for CAHPS public reporting, it is important to recognize that care for anxiety or mood disorders may improve care experiences among seniors with cancer.

Using Medicaid Claims Data to Identify Service Gaps for High-Need Clients: The NYC Mental Health Care Monitoring Initiative

2011 ◽  
Vol 62 (1) ◽  
pp. 9-11 ◽  
Author(s):  
Thomas E. Smith ◽  
Anita Appel ◽  
Sheila A. Donahue ◽  
Susan M. Essock ◽  
Carlos T. Jackson ◽  
...  
Keyword(s):  
Mental Health ◽  
Health Care ◽  
Mental Health Care ◽  
Claims Data ◽  
Health Care Monitoring

scholarly journals Design, Recruitment, and Baseline Characteristics of a Virtual 1-Year Mental Health Study on Behavioral Data and Health Outcomes: Observational Study

10.2196/17075 ◽  
2020 ◽  
Vol 7 (7) ◽  
pp. e17075
Author(s):  
Shefali Kumar ◽  
Jennifer L A Tran ◽  
Ernesto Ramirez ◽  
Wei-Nchih Lee ◽  
Luca Foschini ◽  
...  
Keyword(s):  
Mental Health ◽  
Health Care ◽  
Health Care Utilization ◽  
Health Outcomes ◽  
Behavioral Data ◽  
Care Utilization ◽  
Depression And Anxiety ◽  
Activity Tracker ◽  
Study Population ◽  
Study Participants

Background Depression and anxiety greatly impact daily behaviors, such as sleep and activity levels. With the increasing use of activity tracking wearables among the general population, there has been a growing interest in how data collected from these devices can be used to further understand the severity and progression of mental health conditions. Objective This virtual 1-year observational study was designed with the objective of creating a longitudinal data set combining self-reported health outcomes, health care utilization, and quality of life data with activity tracker and app-based behavioral data for individuals with depression and anxiety. We provide an overview of the study design, report on baseline health and behavioral characteristics of the study population, and provide initial insights into how behavioral characteristics differ between groups of individuals with varying levels of disease severity. Methods Individuals who were existing members of an online health community (Achievement, Evidation Health Inc) and were 18 years or older who had self-reported a diagnosis of depression or anxiety were eligible to enroll in this virtual 1-year study. Participants agreed to connect wearable activity trackers that captured data related to physical activity and sleep behavior. Mental health outcomes such as the Patient Health Questionnaire (PHQ-9), the Generalized Anxiety Disorder Questionnaire (GAD-7), mental health hospitalizations, and medication use were captured with surveys completed at baseline and months 3, 6, 9, and 12. In this analysis, we report on baseline characteristics of the sample, including mental health disease severity and health care utilization. Additionally, we explore the relationship between passively collected behavioral data and baseline mental health status and health care utilization. Results Of the 1304 participants enrolled in the study, 1277 individuals completed the baseline survey and 1068 individuals had sufficient activity tracker data. Mean age was 33 (SD 9) years, and the majority of the study population was female (77.2%, 994/1288) and identified as Caucasian (88.3%, 1137/1288). At baseline, 94.8% (1211/1277) of study participants reported experiencing depression or anxiety symptoms in the last year. This baseline analysis found that some passively tracked behavioral traits are associated with more severe forms of anxiety or depression. Individuals with depressive symptoms were less active than those with minimal depressive symptoms. Severe forms of depression were also significantly associated with inconsistent sleep patterns and more disordered sleep. Conclusions These initial findings suggest that longitudinal behavioral and health outcomes data may be useful for developing digital measures of health for mental health symptom severity and progression.

scholarly journals The importance of health insurance claims data in creating learning health systems: evaluating care for high-need high-cost patients using the National Patient-Centered Clinical Research Network (PCORNet)

2019 ◽  
Vol 26 (11) ◽  
pp. 1305-1313 ◽  
Author(s):  
Maureen A Smith ◽  
Mary S Vaughan-Sarrazin ◽  
Menggang Yu ◽  
Xinyi Wang ◽  
Peter A Nordby ◽  
...  
Keyword(s):  
Health Care ◽  
Health Insurance ◽  
Case Management ◽  
Health Care Utilization ◽  
Health Systems ◽  
Claims Data ◽  
Care Utilization ◽  
Insurance Claims Data ◽  
Health Insurance Claims Data ◽  
Health Insurance Claims

Abstract Objective Case management programs for high-need high-cost patients are spreading rapidly among health systems. PCORNet has substantial potential to support learning health systems in rapidly evaluating these programs, but access to complete patient data on health care utilization is limited as PCORNet is based on electronic health records not health insurance claims data. Because matching cases to comparison patients on baseline utilization is often a critical component of high-quality observational comparative effectiveness research for high-need high-cost patients, limited access to claims may negatively affect the quality of the matching process. We sought to determine whether the evaluation of programs for high-need high-cost patients required claims data to match cases to comparison patients. Materials and Methods A retrospective cohort study design with multiple measures of before-and-after health care utilization for 1935 case management patients and 3833 matched comparison patients aged 18 years and older from 2011 to 2015. EHR and claims data were extracted from 3 health systems participating in PCORNet. Results Without matching on claims-based health care utilization, the case management programs at 2 of 3 health systems were associated with fewer hospital admissions and emergency visits over the subsequent 12 months. With matching on claims-based health care utilization, case management was no longer associated with admissions and emergency visits at those 2 programs. Discussion The results of a PCORNet-facilitated evaluation of 3 programs for high-need high-cost patients differed substantially depending on whether claims data were available for matching cases to comparison patients. Conclusions Partnering with learning health systems to rapidly evaluate programs for high-need high-cost patients will require that PCORNet facilitates comprehensive and timely access to both electronic health records and health insurance claims data.

Identifying Potential Care Management Candidates: The Use of Diagnosis-Based Models to Predict 5-Year Health Care Costs

2016 ◽  
Vol 17 (2) ◽  
pp. 105-111
Author(s):  
John Robst
Keyword(s):  
Health Care ◽  
Case Management ◽  
Regression Models ◽  
Care Management ◽  
Predictive Power ◽  
Service Use ◽  
Claims Data ◽  
Individual Characteristics ◽  
Estimation And Prediction ◽  
Care Costs

Objective: This article examined individual characteristics associated with having higher costs in a 5-year period to identify patients that may potentially benefit from case management.Methods: Florida Medicaid claims data from 2005 to 2010 were used to examine the characteristics, diagnoses, and services (in 2005) associated with individual costs in 5 future years (2006–2010). The data were divided into estimation and prediction samples with regression models estimated using diagnoses and service use in 2005 to predict future costs. Predictive power was assessed by applying the model results to the prediction sample and comparing predicted costs to actual costs.Results: Demographics, service use, and diagnosis in 2005 were associated with costs in the following 5 years. Models were predictive of future costs with a significant relationship between the predicted costs and actual costs.Conclusion: Diagnosis-based models in conjunction with prior costs can predict future costs. Individuals predicted to have higher costs may be candidates for case management to potentially avoid reduce costs.

Sign in / Sign up

Export Citation Format

Share Document