Dataset Information

Predicting High Health Care Resource Utilization in a Single-payer Public Health Care System: Development and Validation of the High Resource User Population Risk Tool.

ABSTRACT: BACKGROUND:A large proportion of health care spending is incurred by a small proportion of the population. Population-based health planning tools that consider both the clinical and upstream determinants of high resource users (HRU) of the health system are lacking. OBJECTIVE:To develop and validate the High Resource User Population Risk Tool (HRUPoRT), a predictive model of adults that will become the top 5% of health care users over a 5-year period, based on self-reported clinical, sociodemographic, and health behavioral predictors in population survey data. RESEARCH DESIGN:The HRUPoRT model was developed in a prospective cohort design using the combined 2005 and 2007/2008 Canadian Community Health Surveys (CCHS) (N=58,617), and validated using the external 2009/2010 CCHS cohort (N=28,721). Health care utilization for each of the 5 years following CCHS interview date were determined by applying a person-centered costing algorithm to the linked health administrative databases. Discrimination and calibration of the model were assessed using c-statistic and Hosmer-Lemeshow (HL) ? statistic. RESULTS:The best prediction model for 5-year transition to HRU status included 12 predictors and had good discrimination (c-statistic=0.8213) and calibration (HL ?=18.71) in the development cohort. The model performed similarly in the validation cohort (c-statistic=0.8171; HL ?=19.95). The strongest predictors in the HRUPoRT model were age, perceived general health, and body mass index. CONCLUSIONS:HRUPoRT can accurately project the proportion of individuals in the population that will become a HRU over 5 years. HRUPoRT can be applied to inform health resource planning and prevention strategies at the community level.

SUBMITTER: Rosella LC

PROVIDER: S-EPMC6143224 | biostudies-literature | 2018 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Predicting High Health Care Resource Utilization in a Single-payer Public Health Care System: Development and Validation of the High Resource User Population Risk Tool.

Rosella Laura C LC Kornas Kathy K Yao Zhan Z Manuel Douglas G DG Bornbaum Catherine C Fransoo Randall R Stukel Therese T

Medical care 20181001 10

<h4>Background</h4>A large proportion of health care spending is incurred by a small proportion of the population. Population-based health planning tools that consider both the clinical and upstream determinants of high resource users (HRU) of the health system are lacking.<h4>Objective</h4>To develop and validate the High Resource User Population Risk Tool (HRUPoRT), a predictive model of adults that will become the top 5% of health care users over a 5-year period, based on self-reported clinic ...[more]

PMID: 29189576

Similar Datasets

Project description:BackgroundQuality of care public reporting provides structural, process and outcome information to facilitate hospital choice and strengthen quality competition. Yet, evidence indicates that patients rarely use this information in their decision-making, due to limited awareness of the data and complex and conflicting information. While there is enthusiasm among policy makers for public reporting, clinicians and researchers doubt its overall impact. Almost no study has analyzed how users behave on public reporting portals, which information they seek out and when they abort their search.MethodsThis study employs web-usage mining techniques on server log data of 17 million user actions from Germany's premier provider transparency portal Weisse-Liste.de (WL.de) between 2012 and 2015. Postal code and ICD search requests facilitate identification of geographical and treatment area usage patterns. User clustering helps to identify user types based on parameters like session length, referrer and page topic visited. First-level markov chains illustrate common click paths and premature exits.ResultsIn 2015, the WL.de Hospital Search portal had 2,750 daily users, with 25% mobile traffic, a bounce rate of 38% and 48% of users examining hospital quality information. From 2013 to 2015, user traffic grew at 38% annually. On average users spent 7 min on the portal, with 7.4 clicks and 54 s between clicks. Users request information for many oncologic and orthopedic conditions, for which no process or outcome quality indicators are available. Ten distinct user types, with particular usage patterns and interests, are identified. In particular, the different types of professional and non-professional users need to be addressed differently to avoid high premature exit rates at several key steps in the information search and view process. Of all users, 37% enter hospital information correctly upon entry, while 47% require support in their hospital search.ConclusionsSeveral onsite and offsite improvement options are identified. Public reporting needs to be directed at the interests of its users, with more outcome quality information for oncology and orthopedics. Customized reporting can cater to the different needs and skill levels of professional and non-professional users. Search engine optimization and hospital quality advocacy can increase website traffic.

Project description:ObjectiveTo compare different definitions of multimorbidity to identify patients with higher health care resource utilization.Patients and methodsWe used a multinational retrospective cohort including 147,806 medical inpatients discharged from 11 hospitals in 3 countries (United States, Switzerland, and Israel) between January 1, 2010, and December 31, 2011. We compared the area under the receiver operating characteristic curve (AUC) of 8 definitions of multimorbidity, based on International Classification of Diseases codes defining health conditions, the Deyo-Charlson Comorbidity Index, the Elixhauser-van Walraven Comorbidity Index, body systems, or Clinical Classification Software categories to predict 30-day hospital readmission and/or prolonged length of stay (longer than or equal to the country-specific upper quartile). We used a lower (yielding sensitivity ≥90%) and an upper (yielding specificity ≥60%) cutoff to create risk categories.ResultsDefinitions had poor to fair discriminatory power in the derivation (AUC, 0.61-0.65) and validation cohorts (AUC, 0.64-0.71). The definitions with the highest AUC were number of (1) health conditions with involvement of 2 or more body systems, (2) body systems, (3) Clinical Classification Software categories, and (4) health conditions. At the upper cutoff, sensitivity and specificity were 65% to 79% and 50% to 53%, respectively, in the validation cohort; of the 147,806 patients, 5% to 12% (7474 to 18,008) were classified at low risk, 38% to 55% (54,484 to 81,540) at intermediate risk, and 32% to 50% (47,331 to 72,435) at high risk.ConclusionOf the 8 definitions of multimorbidity, 4 had comparable discriminatory power to identify patients with higher health care resource utilization. Of these 4, the number of health conditions may represent the easiest definition to apply in clinical routine. The cutoff chosen, favoring sensitivity or specificity, should be determined depending on the aim of the definition.

Project description:BackgroundThe COVID-19 pandemic has highlighted the inability of health systems to leverage existing system infrastructure in order to rapidly develop and apply broad analytical tools that could inform state- and national-level policymaking, as well as patient care delivery in hospital settings. The COVID-19 pandemic has also led to highlighted systemic disparities in health outcomes and access to care based on race or ethnicity, gender, income-level, and urban-rural divide. Although the United States seems to be recovering from the COVID-19 pandemic owing to widespread vaccination efforts and increased public awareness, there is an urgent need to address the aforementioned challenges.ObjectiveThis study aims to inform the feasibility of leveraging broad, statewide datasets for population health-driven decision-making by developing robust analytical models that predict COVID-19-related health care resource utilization across patients served by Indiana's statewide Health Information Exchange.MethodsWe leveraged comprehensive datasets obtained from the Indiana Network for Patient Care to train decision forest-based models that can predict patient-level need of health care resource utilization. To assess these models for potential biases, we tested model performance against subpopulations stratified by age, race or ethnicity, gender, and residence (urban vs rural).ResultsFor model development, we identified a cohort of 96,026 patients from across 957 zip codes in Indiana, United States. We trained the decision models that predicted health care resource utilization by using approximately 100 of the most impactful features from a total of 1172 features created. Each model and stratified subpopulation under test reported precision scores >70%, accuracy and area under the receiver operating curve scores >80%, and sensitivity scores approximately >90%. We noted statistically significant variations in model performance across stratified subpopulations identified by age, race or ethnicity, gender, and residence (urban vs rural).ConclusionsThis study presents the possibility of developing decision models capable of predicting patient-level health care resource utilization across a broad, statewide region with considerable predictive performance. However, our models present statistically significant variations in performance across stratified subpopulations of interest. Further efforts are necessary to identify root causes of these biases and to rectify them.

Project description:BackgroundPreterm birth remains a significant burden to families, health systems and societies. The aim was to quantify the incremental prematurity-related public health expenditure in Hungary and to estimate the potential impact of a decrease in the prevalence of prematurity on the public payer's spending.MethodsOver a 6-year time horizon, public financing data of inpatient, outpatient and pharmaceutical care for children born at ≥ 25 weeks of gestation in 2009/2010 were retrieved from the Hungarian National Health Insurance Fund database. In descriptive analysis, the public payer's spending was given as cost/capita. The impact of a decrease in prematurity prevalence was specified as the total budget impact. An exchange rate of 294 Hungarian forint/Euro was applied.ResultsA total of 93,124 children (including 8.6% who were premature babies) were included in the analysis. A strong negative relationship was shown between gestational age and per capita cost. The 6-year cost of care for the cohort born at 26 weeks of gestation (28,470 Euro per capita) was 24 times higher than that for the cohort born at 40 weeks. First-year inpatient spending accounted for the largest proportion of total health care spending across all gestational ages. All investigated prematurity complications (retinopathy of prematurity, necrotizing enterocolitis, bronchopulmonary dysplasia, intraventricular cerebral bleeding and leukomalacia) resulted in additional significant incremental spending. If 70% of pregnancies ending with preterm birth could be prolonged by 1 week, the savings would be almost 7.0 million Euros in the first 6 years of life.ConclusionThis comprehensive analysis of prematurity-related health care spending confirmed that premature infants have much higher costs for care than those born at term in Hungary. These quantitative outcomes can provide essential inputs for the cost-effectiveness analysis of medical technologies and public health interventions that can decrease the prevalence of premature birth.Trial registrationNot applicable.

Project description:PurposeIn patients treated with chemoradiation for esophageal cancer (EC), randomized trial data demonstrate that proton beam therapy (PBT) reduces toxicities and postoperative complications (POCs) compared with intensity-modulated radiation therapy (IMRT). However, whether radiation therapy modality affects postoperative health care resource utilization remains unknown.Materials and methodsWe examined 287 patients with EC who received chemoradiation (prescribed 50.4 Gy/GyE) followed by esophagectomy, including a real-world observational cohort of 237 consecutive patients treated from 2007 to 2013 with PBT (n = 81) versus IMRT (n = 156); and an independent, contemporary comparison cohort of 50 patients from a randomized trial treated from 2012 to 2019 with PBT (n = 21) versus IMRT (n = 29). Postoperative complications were abstracted from medical records. Health care charges were obtained from institutional claims and adjusted for inflation (2021 dollars). Charge differences (Δ = $PBT - $IMRT) were compared by treatment using adjusted generalized linear models with the gamma distribution.ResultsBaseline PBT versus IMRT characteristics were not significantly different. In the observational cohort, during the neoadjuvant chemoradiation phase, health care charges were higher for PBT versus IMRT (Δ = +$71,959; 95% confidence interval [CI], $62,274-$82,138; P < .001). There was no difference in surgical charges (Δ = -$2234; 95% CI, -$6003 to $1695; P = .26). However, during postoperative hospitalization following esophagectomy, health care charges were lower for PBT versus IMRT (Δ = -$25,115; 95% CI, -$37,625 to -$9776; P = .003). In the comparison cohort, findings were analogous: Charges were higher for PBT versus IMRT during chemoradiation (Δ = +$61,818; 95% CI, $49,435-$75,069; P < .001), not different for surgery (Δ = -$4784; 95% CI, -$6439 to $3487; P = .25), and lower for PBT postoperatively (Δ = -$27,048; 95% CI, -$41,974 to -$5300; P = .02). Lower postoperative charges for PBT were especially seen among patients with any POCs in the contemporary comparison (Δ = -$176,448; 95% CI, -$209,782 to -$78,813; P = .02).ConclusionHigher up-front chemoradiation resource utilization for PBT in patients with EC was partially offset postoperatively, moderated by reduction in POC risks. Results extend existing clinical evidence of toxicity reduction with PBT.

Project description:ImportanceDespite research supporting the benefits of early palliative care, timely initiation by gynecologic oncology patients is reportedly low, which may limit the effectiveness of palliative care.ObjectiveTo investigate the association of the timing of palliative care initiation with the aggressiveness of end-of-life care using established quality indicators among patients with ovarian cancer.Design, setting, and participantsThis population-based retrospective cohort study of ovarian cancer decedents used linked administrative health care data to identify palliative care provision across all health care sectors and health care professionals (specialist and nonspecialist) and end-of-life quality indicators in Ontario, Canada, from 2006 to 2018. Data analyses were performed July 12, 2024.Main outcomes and measuresThe primary outcome was the associations between the timing of palliative care and end-of-life quality indicators, including emergency department use, hospital or intensive care unit admission in the last 30 days of life, chemotherapy in last 14 days of life, death in the hospital, and a composite measure of aggressive care. Late palliative care was defined as 3 months or less prior to death.ResultsThere were 8297 ovarian cancer decedents. Their mean (SD) age at death was 69.6 (13.1) years, and their mean (SD) oncologic survival was 2.8 (3.9) years. Among 3958 patients with known cancer stage, 3495 (88.3%) presented with stage III or IV disease. One-third of patients (2667 [32.1%]) received late palliative care in the final 3 months of life. Results of multivariable regression analysis indicated that any palliative care initiated earlier than 3 months before death was associated with lower rates of aggressive end-of-life care (odds ratio [OR], 0.47 [95% CI, 0.37-0.60]), death in hospital (OR, 0.54 [95% CI, 0.45-0.65]), and intensive care unit admission (OR, 0.46 [95% CI, 0.27-0.76]). Specialist palliative consultation from 3 months up to 6 monts before death was associated with decreased likelihood of late chemotherapy (OR, 0.46 [95% CI, 0.24-0.88]).ConclusionsFindings from this cohort study suggested that early palliative care may be associated with less-aggressive end-of-life care than late palliative care. Implementation strategies for early palliative care initiation are needed to optimize care quality and health resource utilization at the end of life.

Project description:Background. The complexity of decision science models may prevent their use to assist in decision making. User-centered design (UCD) principles provide an opportunity to engage end users in model development and refinement, potentially reducing complexity and increasing model utilization in a practical setting. We report our experiences with UCD to develop a modeling tool for cancer control planners evaluating cancer survivorship interventions. Design. Using UCD principles (described in the article), we developed a dynamic cohort model of cancer survivorship for individuals with female breast, colorectal, lung, and prostate cancer over 10 y. Parameters were obtained from the National Program of Cancer Registries and peer-reviewed literature, with model outcomes captured in quality-adjusted life-years and net monetary benefit. Prototyping and iteration were conducted with structured focus groups involving state cancer control planners and staff from the Centers for Disease Control and Prevention and the American Public Health Association. Results. Initial feedback highlighted model complexity and unclear purpose as barriers to end user uptake. Revisions addressed complexity by simplifying model input requirements, providing clear examples of input types, and reducing complex language. Wording was added to the results page to explain the interpretation of results. After these updates, feedback demonstrated that end users more clearly understood how to use and apply the model for cancer survivorship resource allocation tasks. Conclusions. A UCD approach identified challenges faced by end users in integrating a decision aid into their workflow. This approach created collaboration between modelers and end users, tailoring revisions to meet the needs of the users. Future models developed for individuals without a decision science background could leverage UCD to ensure the model meets the needs of the intended audience. Highlights Model complexity and unclear purpose are 2 barriers that prevent lay users from integrating decision science tools into their workflow. Modelers could integrate the user-centered design framework when developing a model for lay users to reduce complexity and ensure the model meets the needs of the users.

Dataset Information

Predicting High Health Care Resource Utilization in a Single-payer Public Health Care System: Development and Validation of the High Resource User Population Risk Tool.

Publications

Predicting High Health Care Resource Utilization in a Single-payer Public Health Care System: Development and Validation of the High Resource User Population Risk Tool.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets