Dataset Information

Development and Validation of HealthImpact: An Incident Diabetes Prediction Model Based on Administrative Data.

ABSTRACT:

Objective

To develop and validate a model of incident type 2 diabetes based solely on administrative data.

Data sources/study setting

Optum Labs Data Warehouse (OLDW), a national commercial administrative dataset.

Study design

HealthImpact model was developed and internally validated using nested case-control study design; n = 473,049 in training cohort and n = 303,025 in internal validation cohort. HealthImpact was externally validated in 2,000,000 adults followed prospectively for 3 years. Only adults ?18 years were included.

Data collection/extraction methods

Patients with incident diabetes were identified using HEDIS rules. Control subjects were sampled from patients without diabetes. Medical and pharmacy claims data collected over 3 years prior to index date were used to build the model variables.

Principal findings

HealthImpact, scored 0-100, has 48 variables with c-statistic 0.80815. We identified HealthImpact threshold of 90 as identifying patients at high risk of incident diabetes. HealthImpact had excellent discrimination in external validation cohort (c-statistic 0.8171). The sensitivity, specificity, positive predictive value, and negative predictive value of HealthImpact >90 for new diagnosis of diabetes within 3 years were 32.35, 94.92, 22.25, and 96.90 percent, respectively.

Conclusions

HealthImpact is an efficient and effective method of risk stratification for incident diabetes that is not predicated on patient-provided information or laboratory tests.

SUBMITTER: McCoy RG

PROVIDER: S-EPMC5034198 | biostudies-literature | 2016 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Development and Validation of HealthImpact: An Incident Diabetes Prediction Model Based on Administrative Data.

McCoy Rozalina G RG Nori Vijay S VS Smith Steven A SA Hane Christopher A CA

Health services research 20160221 5

<h4>Objective</h4>To develop and validate a model of incident type 2 diabetes based solely on administrative data.<h4>Data sources/study setting</h4>Optum Labs Data Warehouse (OLDW), a national commercial administrative dataset.<h4>Study design</h4>HealthImpact model was developed and internally validated using nested case-control study design; n = 473,049 in training cohort and n = 303,025 in internal validation cohort. HealthImpact was externally validated in 2,000,000 adults followed prospect ...[more]

PMID: 26898782

Similar Datasets

Project description:ImportanceSystems-level barriers to diabetes care could be improved with population health planning tools that accurately discriminate between high- and low-risk groups to guide investments and targeted interventions.ObjectiveTo develop and validate a population-level machine learning model for predicting type 2 diabetes 5 years before diabetes onset using administrative health data.Design, setting, and participantsThis decision analytical model study used linked administrative health data from the diverse, single-payer health system in Ontario, Canada, between January 1, 2006, and December 31, 2016. A gradient boosting decision tree model was trained on data from 1 657 395 patients, validated on 243 442 patients, and tested on 236 506 patients. Costs associated with each patient were estimated using a validated costing algorithm. Data were analyzed from January 1, 2006, to December 31, 2016.ExposuresA random sample of 2 137 343 residents of Ontario without type 2 diabetes was obtained at study start time. More than 300 features from data sets capturing demographic information, laboratory measurements, drug benefits, health care system interactions, social determinants of health, and ambulatory care and hospitalization records were compiled over 2-year patient medical histories to generate quarterly predictions.Main outcomes and measuresDiscrimination was assessed using the area under the receiver operating characteristic curve statistic, and calibration was assessed visually using calibration plots. Feature contribution was assessed with Shapley values. Costs were estimated in 2020 US dollars.ResultsThis study trained a gradient boosting decision tree model on data from 1 657 395 patients (12 900 257 instances; 6 666 662 women [51.7%]). The developed model achieved a test area under the curve of 80.26 (range, 80.21-80.29), demonstrated good calibration, and was robust to sex, immigration status, area-level marginalization with regard to material deprivation and race/ethnicity, and low contact with the health care system. The top 5% of patients predicted as high risk by the model represented 26% of the total annual diabetes cost in Ontario.Conclusions and relevanceIn this decision analytical model study, a machine learning model approach accurately predicted the incidence of diabetes in the population using routinely collected health administrative data. These results suggest that the model could be used to inform decision-making for population health planning and diabetes prevention.

Project description:Aims/hypothesisWe aimed to report current rates of CVD in type 1 diabetes and to develop a CVD risk prediction tool for type 1 diabetes.MethodsA cohort of 27,527 people with type 1 diabetes without prior CVD was derived from the national register in Scotland. Incident CVD events during 199,552 person-years of follow-up were ascertained using hospital admissions and death registers. A Poisson regression model of CVD was developed and then validated in the Swedish National Diabetes Register (n = 33,183). We compared the percentage with a high 10 year CVD risk (i.e., ≥10%) using the model with the percentage eligible for statins using current guidelines by age.ResultsThe age-standardised rate of CVD per 100,000 person-years was 4070 and 3429 in men and women, respectively, with type 1 diabetes in Scotland, and 4014 and 3956 in men and women in Sweden. The final model was well calibrated (Hosmer-Lemeshow test p > 0.05) and included a further 22 terms over a base model of age, sex and diabetes duration (C statistic 0.82; 95% CI 0.81, 0.83). The model increased the base model C statistic from 0.66 to 0.80, from 0.60 to 0.75 and from 0.62 to 0.68 in those aged <40, 40-59 and ≥ 60 years, respectively (all p values <0.005). The model required minimal calibration in Sweden and had a C statistic of 0.85. Under current guidelines, >90% of those aged 20-39 years and 100% of those ≥40 years with type 1 diabetes were eligible for statins, but it was not until age 65 upwards that 100% had a modelled risk of CVD ≥10% in 10 years.Conclusions/interpretationA prediction tool such as that developed here can provide individualised risk predictions. This 10 year CVD risk prediction tool could facilitate patient discussions regarding appropriate statin prescribing. Apart from 10 year risk, such discussions may also consider longer-term CVD risk, the potential for greater benefits from early vs later statin intervention, the potential impact on quality of life of an early CVD event and evidence on safety, all of which could influence treatment decisions, particularly in younger people with type 1 diabetes.

Project description:Prediction models for population-based screening need, for global usage, to be resource-driven, involving predictors that are affordably resourced. Here, we report the development and validation of three resource-driven risk models to identify people with type 2 diabetes (T2DM) at risk of stage 3 CKD defined by a decline in estimated glomerular filtration rate (eGFR) to below 60 mL/min/1.73m2. The observational study cohort used for model development consisted of data from a primary care dataset of 20,510 multi-ethnic individuals with T2DM from London, UK (2007-2018). Discrimination and calibration of the resulting prediction models developed using cox regression were assessed using the c-statistic and calibration slope, respectively. Models were internally validated using tenfold cross-validation and externally validated on 13,346 primary care individuals from Wales, UK. The simplest model was simplified into a risk score to enable implementation in community-based medicine. The derived full model included demographic, laboratory parameters, medication-use, cardiovascular disease history (CVD) and sight threatening retinopathy status (STDR). Two less resource-intense models were developed by excluding CVD and STDR in the second model and HbA1c and HDL in the third model. All three 5-year risk models had good internal discrimination and calibration (optimism adjusted C-statistics were each 0.85 and calibration slopes 0.999-1.002). In Wales, models achieved excellent discrimination(c-statistics ranged 0.82-0.83). Calibration slopes at 5-years suggested models over-predicted risks, however were successfully updated to accommodate reduced incidence of stage 3 CKD in Wales, which improved their alignment with the observed rates in Wales (E/O ratios near to 1). The risk score demonstrated similar model performance compared to direct evaluation of the cox model. These resource-driven risk prediction models may enable universal screening for Stage 3 CKD to enable targeted early optimisation of risk factors for CKD.

Project description:BackgroundFrailty is the third most common complication of diabetes after macrovascular and microvascular complications. The aim of this study was to develop a validated risk prediction model for frailty in patients with diabetes.MethodsThe research used data from the China Health and Retirement Longitudinal Study (CHARLS), a dataset representative of the Chinese population. Twenty-five indicators, including socio-demographic variables, behavioral factors, health status, and mental health parameters, were analyzed in this study. The study cohort was randomly divided into a training set and a validation set at a ratio of 70 to 30%. LASSO regression analysis was used to screen the variables for the best predictors of the model based on a 10-fold cross-validation. The logistic regression model was applied to explore the associated factors of frailty in patients with diabetes. A nomogram was constructed to develop the prediction model. Calibration curves were applied to evaluate the accuracy of the nomogram model. The area under the receiver operating characteristic curve and decision curve analysis were conducted to assess predictive performance.ResultsOne thousand four hundred thirty-six patients with diabetes from the CHARLS database collected in 2013 (n = 793) and 2015 (n = 643) were included in the final analysis. A total of 145 (10.9%) had frailty symptoms. Multivariate logistic regression analysis showed that marital status, activities of daily living, waist circumference, cognitive function, grip strength, social activity, and depression as predictors of frailty in people with diabetes. These factors were used to construct the nomogram model, which showed good concordance and accuracy. The AUC values of the predictive model and the internal validation set were 0.912 (95%CI 0.887-0.937) and 0.881 (95% CI 0.829-0.934). Hosmer-Lemeshow test values were P = 0.824 and P = 0.608 (both > 0.05). Calibration curves showed significant agreement between the nomogram model and actual observations. ROC and DCA indicated that the nomogram had a good predictive performance.ConclusionsComprehensive nomogram constructed in this study was a promising and convenient tool to evaluate the risk of frailty in patients with diabetes, and contributed clinicians to screening the high-risk population.

Project description:BACKGROUND:Obesity is closely related to the development of insulin resistance and type 2 diabetes (T2D). The prevention of T2D has become imperative to stem the rising rates of this disease. Weight loss is highly effective in preventing T2D; however, the at-risk pool is large, and a clinically meaningful metric for risk stratification to guide interventions remains a challenge. The objective of this study is to predict T2D risk using full-information continuous analysis of nationally sampled data from white and black American adults age ?45 years. METHODS AND FINDINGS:A sample of 12,043 black (33%) and white individuals from a population-based cohort, REasons for Geographic And Racial Differences in Stroke (REGARDS) (enrolled 2003-2007), was observed through 2013-2016. The mean participant age was 63.12 ± 8.62 years, and 43.7% were male. Mean BMI was 28.55 ± 5.61 kg/m2. Risk factors for T2D regularly recorded in the primary care setting were used to evaluate future T2D risk using Bayesian logistic regression. External validation was performed using 9,710 participants (19% black) from Atherosclerotic Risk in Communities (ARIC) (enrolled 1987-1989), observed through 1996-1998. The mean participant age in this cohort was 53.86 ± 5.65 years, and 44.6% were male. Mean BMI was 27.15 ± 4.92 kg/m2. Predictive performance was assessed using the receiver operating characteristic (ROC) curves and area under the curve (AUC) statistics. The primary outcome was incident T2D. By 2016 in REGARDS, there were 1,602 incident cases of T2D. Risk factors used to predict T2D progression included age, sex, race, BMI, triglycerides, high-density lipoprotein, blood pressure, and blood glucose. The Bayesian logistic model (AUC = 0.79) outperformed the Framingham risk score (AUC = 0.76), the American Diabetes Association risk score (AUC = 0.64), and a cardiometabolic disease system (using Adult Treatment Panel III criteria) (AUC = 0.75). Validation in ARIC was robust (AUC = 0.85). Main limitations include the limited generalizability of the REGARDS sample to black and white, older Americans, and no time to diagnosis for T2D. CONCLUSIONS:Our results show that a Bayesian logistic model using full-information continuous predictors has high predictive discrimination, and can be used to quantify race- and sex-specific T2D risk, providing a new, powerful predictive tool. This tool can be used for T2D prevention efforts including weight loss therapy by allowing clinicians to target high-risk individuals in a manner that could be used to optimize outcomes.

Project description:Administrative data are used for research, quality improvement, and health policy in severe sepsis. However, there is not a sepsis-specific tool applicable to administrative data with which to adjust for illness severity. Our objective was to develop, internally validate, and externally validate a severe sepsis mortality prediction model and associated mortality prediction score.Retrospective cohort study using 2012 administrative data from five U.S. states. Three cohorts of patients with severe sepsis were created: 1) International Classification of Diseases, 9th Revision, Clinical Modification codes for severe sepsis/septic shock, 2) Martin approach, and 3) Angus approach. The model was developed and internally validated in International Classification of Diseases, 9th Revision, Clinical Modification, cohort and externally validated in other cohorts. Integer point values for each predictor variable were generated to create a sepsis severity score.Acute care, nonfederal hospitals in New York, Maryland, Florida, Michigan, and Washington.Patients in one of three severe sepsis cohorts: 1) explicitly coded (n = 108,448), 2) Martin cohort (n = 139,094), and 3) Angus cohort (n = 523,637) INTERVENTIONS: None.Maximum likelihood estimation logistic regression to develop a predictive model for in-hospital mortality. Model calibration and discrimination assessed via Hosmer-Lemeshow goodness-of-fit and C-statistics, respectively. Primary cohort subset into risk deciles and observed versus predicted mortality plotted. Goodness-of-fit demonstrated p value of more than 0.05 for each cohort demonstrating sound calibration. C-statistic ranged from low of 0.709 (sepsis severity score) to high of 0.838 (Angus cohort), suggesting good to excellent model discrimination. Comparison of observed versus expected mortality was robust although accuracy decreased in highest risk decile.Our sepsis severity model and score is a tool that provides reliable risk adjustment for administrative data.

Dataset Information

Development and Validation of HealthImpact: An Incident Diabetes Prediction Model Based on Administrative Data.

Objective

Data sources/study setting

Study design

Data collection/extraction methods

Principal findings

Conclusions

Publications

Development and Validation of HealthImpact: An Incident Diabetes Prediction Model Based on Administrative Data.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets