Dataset Information

Using electronic health records to predict costs and outcomes in stable coronary artery disease.

ABSTRACT:

Objectives

To use electronic health records (EHR) to predict lifetime costs and health outcomes of patients with stable coronary artery disease (stable-CAD) stratified by their risk of future cardiovascular events, and to evaluate the cost-effectiveness of treatments targeted at these populations.

Methods

The analysis was based on 94 966 patients with stable-CAD in England between 2001 and 2010, identified in four prospectively collected, linked EHR sources. Markov modelling was used to estimate lifetime costs and quality-adjusted life years (QALYs) stratified by baseline cardiovascular risk.

Results

For the lowest risk tenth of patients with stable-CAD, predicted discounted remaining lifetime healthcare costs and QALYs were £62 210 (95% CI £33 724 to £90 043) and 12.0 (95% CI 11.5 to 12.5) years, respectively. For the highest risk tenth of the population, the equivalent costs and QALYs were £35 549 (95% CI £31 679 to £39 615) and 2.9 (95% CI 2.6 to 3.1) years, respectively. A new treatment with a hazard reduction of 20% for myocardial infarction, stroke and cardiovascular disease death and no side-effects would be cost-effective if priced below £72 per year for the lowest risk patients and £646 per year for the highest risk patients.

Conclusions

Existing EHRs may be used to estimate lifetime healthcare costs and outcomes of patients with stable-CAD. The stable-CAD model developed in this study lends itself to informing decisions about commissioning, pricing and reimbursement. At current prices, to be cost-effective some established as well as future stable-CAD treatments may require stratification by patient risk.

SUBMITTER: Asaria M

PROVIDER: S-EPMC4849559 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:PurposePatients with cancer often have compromised immune system which can lead to worse COVID-19 outcomes. The purpose of this study is to assess the association between COVID-19 outcomes and existing cancer-specific characteristics.Patients and methodsPatients aged 18 or older with laboratory-confirmed COVID-19 between June 1, 2020, and December 31, 2020, were identified (n = 314 004) from the Optum® de-identified COVID-19 Electronic Health Record (EHR) derived from more than 700 hospitals and 7000 clinics in the United States. To allow sufficient observational time, patients with less than one year of medical history in the EHR dataset before their COVID-19 tests were excluded (n = 42 365). Assessed COVID-19 outcomes including all-cause 30-day mortality, hospitalization, ICU admission, and ventilator use, which were compared using relative risks (RRs) according to cancer status and treatments.ResultsAmong 271 639 patients with COVID-19, 18 460 had at least one cancer diagnosis: 8034 with a history of cancer and 10 426 with newly diagnosed cancer within one year of COVID-19 infection. Patients with a cancer diagnosis were older and more likely to be male, white, Medicare beneficiaries, and have higher prevalences of chronic conditions. Cancer patients had higher risks for 30-day mortality (RR 1.07, 95% CI 1.01-1.14, P = 0.028) and hospitalization (RR 1.04, 95% CI 1.01-1.07, P = 0.006) but without significant differences in ICU admission and ventilator use compared to non-cancer patients. Recent cancer diagnoses were associated with higher risks for worse COVID-19 outcomes (RR for mortality 1.17, 95% CI 1.08-1.25, P<0.001 and RR for hospitalization 1.10, 95% CI 1.06-1.14, P<0.001), particularly among recent metastatic (stage IV), hematological, liver and lung cancers compared with the non-cancer group. Among COVID-19 patients with recent cancer diagnosis, mortality was associated with chemotherapy or radiation treatments within 3 months before COVID-19. Age, black patients, Medicare recipients, South geographic region, cardiovascular, diabetes, liver, and renal diseases were also associated with increased mortality.Conclusions and relevanceIndividuals with cancer had higher risks for 30-day mortality and hospitalization after SARS-CoV-2 infection compared to patients without cancer. More specifically, patients with a cancer diagnosis within 1 year and those receiving active treatment were more vulnerable to worse COVID-19 outcomes.

Project description:IntroductionType 2 diabetes mellitus (T2DM) is a major cause of blindness, kidney failure, myocardial infarction, stroke and lower limb amputation. We are still unable, however, to accurately predict or identify which patients are at a higher risk of deterioration. Most risk stratification tools do not account for novel factors such as sociodemographic determinants, self-management ability or access to healthcare. Additionally, most tools are based in clinical trials, with limited external generalisability.ObjectiveThe aim of this work is to design and validate a machine learning-based tool to identify patients with T2DM at high risk of clinical deterioration, based on a comprehensive set of patient-level characteristics retrieved from a population health linked dataset.Sample and designRetrospective cohort study of patients with diagnosis of T2DM on 1 January 2015, with a 5-year follow-up. Anonymised electronic healthcare records from the Whole System Integrated Care (WSIC) database will be used.Preliminary outcomesOutcome variables of clinical deterioration will include retinopathy, chronic renal disease, myocardial infarction, stroke, peripheral arterial disease or death. Predictor variables will include sociodemographic and geographic data, patients' ability to self-manage disease, clinical and metabolic parameters and healthcare service usage. Prognostic models will be defined using multidependence Bayesian networks. The derivation cohort, comprising 80% of the patients, will be used to define the prognostic models. Model parameters will be internally validated by comparing the area under the receiver operating characteristic curve in the derivation cohort with those calculated from a leave-one-out and a 10 times twofold cross-validation.Ethics and disseminationThe study has received approvals from the Information Governance Committee at the WSIC. Results will be made available to people with T2DM, their caregivers, the funders, diabetes care societies and other researchers.