Dataset Information

Modeling Heterogeneity in Healthcare Utilization Using Massive Medical Claims Data.

ABSTRACT: We introduce a modeling approach for characterizing heterogeneity in healthcare utilization using massive medical claims data. We first translate the medical claims observed for a large study population and across five years into individual-level discrete events of care called utilization sequences. We model the utilization sequences using an exponential proportional hazards mixture model to capture heterogeneous behaviors in patients' healthcare utilization. The objective is to cluster patients according to their longitudinal utilization behaviors and to determine the main drivers of variation in healthcare utilization while controlling for the demographic, geographic, and health characteristics of the patients. Due to the computational infeasibility of fitting a parametric proportional hazards model for high-dimensional, large sample size data we use an iterative one-step procedure to estimate the model parameters and impute the cluster membership. The approach is used to draw inferences on utilization behaviors of children in the Medicaid system with persistent asthma across six states. We conclude with policy implications for targeted interventions to improve adherence to recommended care practices for pediatric asthma.

SUBMITTER: Hilton RP

PROVIDER: S-EPMC6167939 | biostudies-literature | 2018

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Modeling Heterogeneity in Healthcare Utilization Using Massive Medical Claims Data.

Hilton Ross P RP Zheng Yuchen Y Serban Nicoleta N

Journal of the American Statistical Association 20170626 521

We introduce a modeling approach for characterizing heterogeneity in healthcare utilization using massive medical claims data. We first translate the medical claims observed for a large study population and across five years into individual-level discrete events of care called <i>utilization sequences</i>. We model the utilization sequences using an exponential proportional hazards mixture model to capture heterogeneous behaviors in patients' healthcare utilization. The objective is to cluster p ...[more]

PMID: 30294054

Similar Datasets

Project description:BackgroundMyotonic dystrophy (DM) is a rare, inherited disorder with multi-systemic effects that impact the skeletal muscles, eyes, heart, skin and gastrointestinal, endocrine, respiratory, and central nervous systems. DM is divided into two subtypes: DM1 can present from early childhood through adulthood and also has a congenital form (cDM) while DM2 typically manifests during mid-adulthood. Both forms are progressive with no approved treatments, and unmet need for disease-modifying therapies remains high. This study interrogated health insurance claims data to explore the clinical experience, healthcare resource utilization (HCRU), and all-cause costs for DM.ResultsA total of 8541 patients with DM and 242 patients with cDM and their matched controls were selected from a database of over 200 million claimants. HCRU and all-cause costs, including pharmacy, outpatient, and inpatient services, were analyzed across four years in 12-month follow-up periods. Mean all-cause costs per DM patient were high in each of the four periods (range $14,640-$16,704) and showed a steady increase from 13 to 23 months on, while the control group mean costs declined from $9671 in the first 12 months after the index event, to approach the US population average ($5193) over time. For cDM, the highest mean costs were in the first 12-months ($66,496 vs. $2818 for controls), and remained high (above $17,944) across all subsequent periods, while control mean costs approached $0. For DM and cDM, HCRU was higher compared to controls across all study periods and all-cause healthcare costs were mostly driven by inpatient and outpatient encounters. Analysis of all diagnosis codes over the study period (comorbidities) demonstrated an elevated comorbidity profile consistent with the clinical profile of DM.ConclusionsThis study is among the first to utilize claims data to increase understanding of the clinical experience and health economic outcomes associated with DM. The markedly elevated HCRU patterns and comorbidity profile presented here add to the broad body of scientific and clinical knowledge on DM. These insights can inform clinical care and support the development of disease modifying and/or symptom-targeting therapies that address the multi-systemic, progressive nature of DM.

Project description:BackgroundBecause pregnancy complications, including gestational diabetes mellitus (GDM) and hypertensive disorders in pregnancy, are risk factors for diabetes and cardiovascular disease, post-delivery follow-up is recommended.ObjectiveTo determine predictors of post-delivery primary and obstetric care utilization in women with and without medical complications.Research designFive-year retrospective cohort study using commercial and Medicaid insurance claims in Maryland.Subjects7,741 women with a complicated pregnancy (GDM, hypertensive disorders and pregestational diabetes mellitus [DM]) and 23,599 women with a comparison pregnancy.MeasuresWe compared primary and postpartum obstetric care utilization rates in the 12 months after delivery between the complicated and comparison pregnancy groups. We conducted multivariate logistic regression to assess the association between pregnancy complications, sociodemographic predictor variables and utilization of care, stratified by insurance type.ResultsWomen with a complicated pregnancy were older at delivery (p < 0.001), with higher rates of cesarean delivery (p < 0.0001) and preterm labor or delivery (p < 0.0001). Among women with Medicaid, 56.6% in the complicated group and 51.7% in the comparison group attended a primary care visit. Statistically significant predictors of receiving a primary care visit included non-Black race, older age, preeclampsia or DM, and depression. Among women with commercial health insurance, 60.0% in the complicated group and 49.5% in the comparison group attended a primary care visit. Pregnancy complication did not predict a primary care visit among women with commercial insurance.ConclusionsWomen with pregnancy complications were more likely to attend primary care visits post-delivery compared to the comparison group, but overall visit rates were low. Although Medicaid expansion has potential to increase coverage, innovative models for preventive health services after delivery are needed to target women at higher risk for chronic disease development.

Project description:ObjectiveThis study aimed to analyze the healthcare resource use (HCRU) and associated costs of multiple myeloma (MM) using German claims data.MethodsAnonymized claims data from one of the largest sickness funds in Germany were analyzed. Costs and HCRU were calculated from the perspective of the statutory health insurance. To analyze MM-associated incremental costs in a most recent calendar period for an overall MM population (31 March 2018-31 March 2019), a prevalent cohort of MM patients (continuously insured during 01 January 2010 until 31 March 2019 or death; two or more outpatient and/or one inpatient MM diagnoses [ICD-10: C90.0-] and alive on 31 March 2018) was compared with a control group (not diagnosed with MM) employing propensity-score matching. Additionally, to describe MM-associated HCRU and costs for treated patients per line of treatment (lot), a cohort of newly treated patients was considered (≥ 12 months' pre-index period without MM treatment). Therapy lines were determined based on observed days of medication supply, treatment switches, and treatment discontinuations.ResultsOverall, 2523 prevalent MM patients (52.0% female, mean age: 71.3 years) and 1673 newly treated MM patients (50.2% female, mean age: 73.0 years) met the selection criteria and were included in the analyses. After matching, a non-MM counterpart could be identified for 2474 prevalent out of 2523 MM patients. MM-associated incremental HCRU was characterized by an increased number of hospitalizations and a higher number of outpatient specialist visits (per patient-year [ppy] 0.48 additional hospitalizations and 3.80 additional specialist visits; p < 0.001), being also drivers of the associated total incremental add-on costs (15,381.09 € ppy, p < 0.001). Among newly treated patients, total direct costs ppy increased as patients received subsequent treatments (1st lot: 67,681,55 €; 4th lot+: 114,934.01 €), driven by outpatient MM prescriptions (1st lot: 28,692.32 €; 4th lot+: 62,980.72 €).ConclusionThe economic burden of MM is driven by outpatient prescriptions, inpatient hospitalizations, and outpatient specialist visits. Treatment and overall costs increase substantially when patients move to later lines of treatment.

Project description:IntroductionAlopecia areata (AA) is an autoimmune disorder causing sudden, non-scarring hair loss. There are currently no drugs approved for AA treatment. This study assessed prevalence of comorbidities, treatments, and healthcare costs and resource utilization among patients with AA in the USA.MethodsPatients diagnosed with AA between January 2011 and December 2018 were identified in IBM MarketScan® Research Databases. Eligible patients had no other hair loss-related disorders and were continuously enrolled with medical and pharmacy benefits at least 12 months before and after AA diagnosis. Descriptive statistics were used to summarize comorbid conditions, treatments related to AA or other autoimmune/inflammatory conditions, and all-cause and AA-specific healthcare costs and resource utilization identified from claims data.ResultsA total of 68,121 patients with AA were identified. Mean (SD) age was 40.3 (17.8) years and 61.0% were female. The most common comorbidities included hyperlipidemia (22.4%), hypertension (21.8%), thyroid disorders (13.1%), contact dermatitis or eczema (10.8%), depression (9.5%), and anxiety (8.4%). Comorbid autoimmune diseases included atopic dermatitis (2.8%), psoriasis (2.1%), chronic urticaria (1.5%), and rheumatoid arthritis (1.1%). During the 12-month follow-up period, 37,995 patients (55.8%) were prescribed treatment for their AA or other comorbid autoimmune/inflammatory disease; 44.9% of treated patients were prescribed therapy within 7 days of AA diagnosis. Of patients receiving treatment, 80.3% received topical steroids and 30.0% received oral steroids. Mean (SD) total healthcare costs were $11,241.21 ($43,839.69) for all-causes and $419.12 ($1534.99) for AA. AA-related expenses were driven by outpatient and prescription costs.ConclusionPatients with AA have a high comorbidity burden and lack of treatment. Current AA treatments, including systemic therapies other than oral steroids, were not frequently utilized in this study population. Healthcare costs incurred by patients with AA went beyond AA-related expenses. Longitudinal data are needed to better understand treatment trajectories and the disease burden in patients with AA.

Project description:The prevalence of nonalcoholic fatty liver disease (NAFLD) is increasing. The health care burden resulting from the multidisciplinary management of this complex disease is unknown. We assessed the total health care cost and resource utilization associated with a new NAFLD diagnosis, compared with controls with similar comorbidities. We used OptumLabs Data Warehouse, a large national administrative claims database with longitudinal health data of over 100 million individuals enrolled in private and Medicare Advantage health plans. We identified 152,064 adults with a first claim for NAFLD between 2010 and 2014, of which 108,420 were matched 1:1 by age, sex, metabolic comorbidities, length of follow-up, year of diagnosis, race, geographic region, and insurance type to non-NAFLD contemporary controls from the OptumLabs Data Warehouse database. Median follow-up time was 2.6 (range 1-6.5) years. The final study cohort consisted of 216,840 people with median age 55 (range 18-86) years, 53% female, 78% white. The total annual cost of care per NAFLD patient with private insurance was $7,804 (interquartile range [IQR] $3,068-$18,688) for a new diagnosis and $3,789 (IQR $1,176-$10,539) for long-term management. These costs are significantly higher than the total annual costs of $2,298 (IQR $681-$6,580) per matched control with similar metabolic comorbidities but without NAFLD. The largest increases in health care utilization that may account for the increased costs in NAFLD compared with controls are represented by liver biopsies (relative risk [RR] = 55.00, 95% confidence interval [CI] 24.48-123.59), imaging (RR = 3.95, 95% CI 3.77-4.15), and hospitalizations (RR = 1.87, 95% CI 1.73-2.02). Conclusion: The costs associated with the care for NAFLD independent of its metabolic comorbidities are very high, especially at first diagnosis. Research efforts shouldfocus on identification of underlying determinants of use, sources of excess cost, and development of cost-effective diagnostic tests.

Dataset Information

Modeling Heterogeneity in Healthcare Utilization Using Massive Medical Claims Data.

Publications

Modeling Heterogeneity in Healthcare Utilization Using Massive Medical Claims Data.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets