Dataset Information

Characterising and predicting persistent high-cost utilisers in healthcare: a retrospective cohort study in Singapore.

ABSTRACT: OBJECTIVE:We aim to characterise persistent high utilisers (PHUs) of healthcare services, and correspondingly, transient high utilisers (THUs) and non-high utilisers (non-HUs) for comparison, to facilitate stratifying HUs for targeted intervention. Subsequently we apply machine learning algorithms to predict which HUs will persist as PHUs, to inform future trials testing the effectiveness of interventions in reducing healthcare utilisation in PHUs. DESIGN AND SETTING:This is a retrospective cohort study using administrative data from an Academic Medical Centre (AMC) in Singapore. PARTICIPANTS:Patients who had at least one inpatient admission to the AMC between 2005 and 2013 were included in this study. HUs incurred Singapore Dollar 8150 or more within a year. PHUs were defined as HUs for three consecutive years, while THUs were HUs for 1 or 2?years. Non-HUs did not incur high healthcare costs at any point during the study period. OUTCOME MEASURES:PHU status at the end of the third year was the outcome of interest. Socio-demographic profiles, clinical complexity and utilisation metrics of each group were reported. Area under curve (AUC) was used to identify the best model to predict persistence. RESULTS:PHUs were older and had higher comorbidity and mortality. Over the three observed years, PHUs' expenditure generally increased, while THUs and non-HUs' spending and inpatient utilisation decreased. The predictive model exhibited good performance during both internal (AUC: 83.2%, 95%?CI: 82.2% to 84.2%) and external validation (AUC: 79.8%, 95%?CI: 78.8% to 80.8%). CONCLUSIONS:The HU population could be stratified into PHUs and THUs, with distinctly different utilisation trajectories. We developed a model that could predict at the end of 1?year, whether a patient in our population will continue to be a HU in the next 2?years. This knowledge would allow healthcare providers to target PHUs in our health system with interventions in a cost-effective manner.

SUBMITTER: Ng SHX

PROVIDER: S-EPMC6955475 | biostudies-literature | 2020 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Characterising and predicting persistent high-cost utilisers in healthcare: a retrospective cohort study in Singapore.

Ng Sheryl Hui Xian SHX Rahman Nabilah N Ang Ian Yi Han IYH Sridharan Srinath S Ramachandran Sravan S Wang Debby Dan DD Khoo Astrid A Tan Chuen Seng CS Feng Mengling M Toh Sue-Anne Ee Shiow SES Tan Xin Quan XQ

BMJ open 20200106 1

<h4>Objective</h4>We aim to characterise persistent high utilisers (PHUs) of healthcare services, and correspondingly, transient high utilisers (THUs) and non-high utilisers (non-HUs) for comparison, to facilitate stratifying HUs for targeted intervention. Subsequently we apply machine learning algorithms to predict which HUs will persist as PHUs, to inform future trials testing the effectiveness of interventions in reducing healthcare utilisation in PHUs.<h4>Design and setting</h4>This is a ret ...[more]

PMID: 31911514

Similar Datasets

Project description:BackgroundSenior high cost health care users (HCU) are a priority for many governments. Little research has addressed regional variation of HCU incidence and outcomes, especially among incident HCU. This study describes the regional variation in healthcare costs and mortality across Ontario's health planning districts [Local Health Integration Networks (LHIN)] among senior incident HCU and non-HCU and explores the relationship between healthcare spending and mortality.MethodsWe conducted a retrospective population-based matched cohort study of incident senior HCU defined as Ontarians aged ?66 years in the top 5% most costly healthcare users in fiscal year (FY) 2013. We matched HCU to non-HCU (1:3) based on age, sex and LHIN. Primary outcomes were LHIN-based variation in costs (total and 12 cost components) and mortality during FY2013 as measured by variance estimates derived from multi-level models. Outcomes were risk-adjusted for age, sex, ADGs, and low-income status. In a cost-mortality analysis by LHIN, risk-adjusted random effects for total costs and mortality were graphically presented together in a cost-mortality plane to identify low and high performers.ResultsWe studied 175,847 incident HCU and 527,541 matched non-HCU. On average, 94 out of 1000 seniors per LHIN were HCU (CV?=?4.6%). The mean total costs for HCU in FY2013 were 12 times higher that of non-HCU ($29,779 vs. $2472 respectively), whereas all-cause mortality was 13.6 times greater (103.9 vs. 7.5 per 1000 seniors). Regional variation in costs and mortality was lower in senior HCU compared with non-HCU. We identified greater variability in accessing the healthcare system, but, once the patient entered the system, variation in costs was low. The traditional drivers of costs and mortality that we adjusted for played little role in driving the observed variation in HCUs' outcomes. We identified LHINs that had high mortality rates despite elevated healthcare expenditures and those that achieved lower mortality at lower costs. Some LHINs achieved low mortality at excessively high costs.ConclusionsRisk-adjusted allocation of healthcare resources to seniors in Ontario is overall similar across health districts, more so for HCU than non-HCU. Identified important variation in the cost-mortality relationship across LHINs needs to be further explored.

Project description:OBJECTIVES:To describe healthcare use and spending before and on becoming a new (incident) senior high-cost user (HCU) compared with senior non-HCUs; to estimate the incremental costs, overall and by service category, attributable to HCU status; and to quantify its monetary impact on the provincial healthcare budget in Ontario, Canada. DESIGN:We conducted a retrospective, population-based comparative cohort study using administrative healthcare records. Incremental healthcare utilisation and costs were determined using the method of recycled predictions allowing adjustment for preincident and incident year values, and covariates. Estimated budget impact was computed as the product of the mean annual total incremental cost and the number of senior HCUs. PARTICIPANTS:Incident senior HCUs were defined as Ontarians aged ?66 years who were in the top 5% of healthcare cost users during fiscal year 2013 (FY2013) but not during FY2012. The incident HCU cohort was matched with senior non-HCUs in a ratio of 1 HCU:3 non-HCU. RESULTS:Senior HCUs (n=175 847) reached the annual HCU threshold of CAD$10 192 through different combinations of incurred costs. Although HCUs had higher healthcare utilisation and costs at baseline, HCU status was associated with a substantial spike in both, with prolonged hospitalisations playing a major role. Twelve per cent of HCUs reached the HCU expenditure threshold without hospitalisation. Compared with non-HCUs (n=5 27 541), HCUs incurred an additional CAD$25 527 per patient in total healthcare costs; collectively CAD$4.5 billion or 9% of the 2013 Ontario healthcare budget. Inpatient care had the highest incremental costs: CAD$13 427, 53% of the total incremental spending. CONCLUSIONS:Costs attributable to incident senior HCU status accounted for almost 1/10 of the provincial healthcare budget. Prolonged hospitalisations made a major contribution to the total incremental costs. A subgroup of patients that became HCU without hospitalisation requires further investigation.

Project description:INTRODUCTION:Senior high-cost users (HCUs) are estimated to represent 60% of all HCUs in Ontario, Canada's most populous province. To improve our understanding of individual and health system characteristics related to senior HCUs, we will examine incident senior HCUs to determine their incremental healthcare utilisation and costs, characteristics of index hospitalisation episodes, mortality and their regional variation across Ontario. METHODS AND ANALYSIS:A retrospective, population-based cohort study using administrative healthcare records will be used. Incident senior HCUs will be defined as Ontarians aged ≥66 years who were in the top 5% of healthcare cost users during fiscal year 2013 but not during fiscal year 2012. Each HCU will be matched to three non-HCUs by age, sex and health planning region. Incremental healthcare use and costs will be determined using the method of recycled predictions. We will apply multivariable logistic regression to determine patient and health service factors associated with index hospitalisation and inhospital mortality during the incident year. The most common causes of admission will be identified and contrasted with the most expensive hospitalised conditions. We will also calculate the ratio of inpatient costs incurred through admissions of ambulatory care sensitive conditions to the total inpatient expenditures. The magnitude of variation in costs and health service utilisation will be established by calculating the extremal quotient, the coefficient of variation and the Gini mean difference for estimates obtained through multilevel regression analyses. ETHICS AND DISSEMINATION:This study has been approved by Hamilton Integrated Research Ethics Board (ID#1715-C). The results of the study will be distributed through peer-reviewed journals. They also will be disseminated at research events in academic settings, national and international conferences as well as with presentations to provincial health authorities.

Project description:BackgroundSince the introduction of higher valency pneumococcal conjugate vaccines in 2009, recent estimates on the economic burden of pediatric pneumococcal disease (PD) in Germany have been lacking. This study estimates healthcare resource utilization (HCRU) and medical cost associated with PDs in children < 16 years old in Germany from 2014-2019.MethodsA nationally representative sample from the Institute for Applied Health Research (InGef) German claims database was used, covering approximately 5% of the total German population. Episodes of pneumococcal pneumonia (PP), all-cause pneumonia (ACP), invasive pneumococcal disease (IPD), and acute otitis media (AOM) in children aged < 16 years were identified using ICD-10-GM codes. HCRU was estimated from annual rates of outpatient visits, outpatient antibiotic prescriptions and inpatient admissions, divided by person-years (PY) at-risk. Average direct medical costs per episode were estimated as the total cost of all HCRU, divided by the total number of episodes. The Mann-Kendall test was used to assess monotonic time trends from 2014-2019.Results During 2014-2019, 916,805 children aged < 16 years were followed up for a total of 3,608,716 PY. The average costs per episode for out-versus inpatient care associated with PP and ACP were €67 (95% CI 58-76) versus €2,606 (95% CI 1,338-3,873), and €63 (95% CI 62-63) versus €620 (95% CI 598-641), respectively. For IPD, the average medical cost per episode for out-versus inpatients were €30 (95% CI 19-42) versus €6,051 (95% CI 3,323-8,779), respectively. There were no significant trends in HCRU or costs for IPD or pneumonia over the study period, except for a significant reduction in ACP outpatient visits. A significant decrease in rate of outpatient visits and antibiotic prescribing for recurrent AOM was observed, in addition to an increase in rates of hospital admissions for simple AOM. This was paralleled by a significant increase in inpatient costs per episode for treating AOM overall, and simple AOM, over the study period.Conclusions The HCRU and cost per episode of pneumonia and IPD did not vary significantly from 2014-2019, but increased for AOM. The economic burden of pneumonia, IPD, and AOM remains substantial in Germany.

Project description:IntroductionAtopic dermatitis (AD) data are scarce in Dubai [United Arab Emirates (UAE)]. Therefore, this study aimed at understanding real-world healthcare resource utilization (HCRU) and related costs, specialties, treatment landscape, consultation-based prevalence and incidence, and patient characteristics.MethodsThis retrospective, longitudinal, insurance e-claims (Dubai Private Insurance-insured expatriates) database studied AD in Dubai between 1 January 2014 and 31 March 2020. Two cohorts of patients based on treatment status as the eligibility criteria were selected from 442,956 patients with at least two AD diagnosis claims: treated AD [mild to moderate (10,134 patients) and moderate to severe (3515 patients)] and untreated or on drugs not included in the treated AD cohort (10,806 patients).ResultsAcross treated AD (mild to moderate and moderate to severe) and untreated AD cohorts, mean age was ~ 29 years; the majority were from dermatology (65-44%) and pediatrics (29-32%) specialty. Key HCRU cost contributors were hospitalizations and outpatient visits in both the treated AD groups. Mean annual disease-specific HCRU cost per patient was highest for the moderate-to-severe treated (531.5 USD) cohort, followed by the mild-to-moderate treated (378.4 USD) cohort, and lowest for the untreated (144.0 USD) cohort; patients with AD with any infection, asthma, or allergic rhinitis showed a similar trend. However, AD-diagnosed patients with Staphylococcus infection had the highest mean HCRU cost among the mild-to-moderate treated AD cohort, followed by the moderate-to-severe treated AD cohort.ConclusionThis study indicated AD to be a common skin disease with a prevalence rate of 4-5% in Dubai (UAE), with the majority of patients (about 90%) being treated by specialists. However, there is a significant underuse of newer innovative therapies (including biologics). Also, disease severity (moderate-to-severe AD) was associated with high direct medical cost, which could be controlled by early intervention. Furthermore, AD treatment choice could focus on major direct HCRU cost contributors such as hospitalizations, comorbid conditions, and infections.

Dataset Information

Characterising and predicting persistent high-cost utilisers in healthcare: a retrospective cohort study in Singapore.

Publications

Characterising and predicting persistent high-cost utilisers in healthcare: a retrospective cohort study in Singapore.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets