Dataset Information

Multiple imputation for harmonizing longitudinal non-commensurate measures in individual participant data meta-analysis.

ABSTRACT: There are many advantages to individual participant data meta-analysis for combining data from multiple studies. These advantages include greater power to detect effects, increased sample heterogeneity, and the ability to perform more sophisticated analyses than meta-analyses that rely on published results. However, a fundamental challenge is that it is unlikely that variables of interest are measured the same way in all of the studies to be combined. We propose that this situation can be viewed as a missing data problem in which some outcomes are entirely missing within some trials and use multiple imputation to fill in missing measurements. We apply our method to five longitudinal adolescent depression trials where four studies used one depression measure and the fifth study used a different depression measure. None of the five studies contained both depression measures. We describe a multiple imputation approach for filling in missing depression measures that makes use of external calibration studies in which both depression measures were used. We discuss some practical issues in developing the imputation model including taking into account treatment group and study. We present diagnostics for checking the fit of the imputation model and investigate whether external information is appropriately incorporated into the imputed values.

SUBMITTER: Siddique J

PROVIDER: S-EPMC4596762 | biostudies-literature | 2015 Nov

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Multiple imputation for harmonizing longitudinal non-commensurate measures in individual participant data meta-analysis.

Siddique Juned J Reiter Jerome P JP Brincks Ahnalee A Gibbons Robert D RD Crespi Catherine M CM Brown C Hendricks CH

Statistics in medicine 20150621 26

There are many advantages to individual participant data meta-analysis for combining data from multiple studies. These advantages include greater power to detect effects, increased sample heterogeneity, and the ability to perform more sophisticated analyses than meta-analyses that rely on published results. However, a fundamental challenge is that it is unlikely that variables of interest are measured the same way in all of the studies to be combined. We propose that this situation can be viewed ...[more]

PMID: 26095855

Similar Datasets

Project description:BACKGROUND:Some evidence suggests that chronic kidney disease is a risk factor for lower-extremity peripheral artery disease. We aimed to quantify the independent and joint associations of two measures of chronic kidney disease (estimated glomerular filtration rate [eGFR] and albuminuria) with the incidence of peripheral artery disease. METHODS:In this collaborative meta-analysis of international cohorts included in the Chronic Kidney Disease Prognosis Consortium (baseline measurements obtained between 1972 and 2014) with baseline measurements of eGFR and albuminuria, at least 1000 participants (this criterion not applied to cohorts exclusively enrolling patients with chronic kidney disease), and at least 50 peripheral artery disease events, we analysed adult participants without peripheral artery disease at baseline at the individual patient level with Cox proportional hazards models to quantify associations of creatinine-based eGFR, urine albumin-to-creatinine ratio (ACR), and dipstick proteinuria with the incidence of peripheral artery disease (including hospitalisation with a diagnosis of peripheral artery disease, intermittent claudication, leg revascularisation, and leg amputation). We assessed discrimination improvement through c-statistics. FINDINGS:We analysed 817?084 individuals without a history of peripheral artery disease at baseline from 21 cohorts. 18?261 cases of peripheral artery disease were recorded during follow-up across cohorts (median follow-up was 7·4 years [IQR 5·7-8·9], range 2·0-15·8 years across cohorts). Both chronic kidney disease measures were independently associated with the incidence of peripheral artery disease. Compared with an eGFR of 95 mL/min per 1·73 m2, adjusted hazard ratios (HRs) for incident study-specific peripheral artery disease was 1·22 (95% CI 1·14-1·30) at an eGFR of 45 mL/min per 1·73 m2 and 2·06 (1·70-2·48) at an eGFR of 15 mL/min per 1·73 m2. Compared with an ACR of 5 mg/g, the adjusted HR for incident study-specific peripheral artery disease was 1·50 (1·41-1·59) at an ACR of 30 mg/g and 2·28 (2·12-2·44) at an ACR of 300 mg/g. The adjusted HR at an ACR of 300 mg/g versus 5 mg/g was 3·68 (95% CI 3·00-4·52) for leg amputation. eGFR and albuminuria contributed multiplicatively (eg, adjusted HR 5·76 [4·90-6·77] for incident peripheral artery disease and 10·61 [5·70-19·77] for amputation in eGFR <30 mL/min per 1·73 m2 plus ACR ?300 mg/g or dipstick proteinuria 2+ or higher vs eGFR ?90 mL/min per 1·73 m2 plus ACR <10 mg/g or dipstick proteinuria negative). Both eGFR and ACR significantly improved peripheral artery disease risk discrimination beyond traditional predictors, with a substantial improvement prediction of amputation with ACR (difference in c-statistic 0·058, 95% CI 0·045-0·070). Patterns were consistent across clinical subgroups. INTERPRETATION:Even mild-to-moderate chronic kidney disease conferred increased risk of incident peripheral artery disease, with a strong association between albuminuria and amputation. Clinical attention should be paid to the development of peripheral artery disease symptoms and signs in people with any stage of chronic kidney disease. FUNDING:American Heart Association, US National Kidney Foundation, and US National Institute of Diabetes and Digestive and Kidney Diseases.

Dataset Information

Multiple imputation for harmonizing longitudinal non-commensurate measures in individual participant data meta-analysis.

Publications

Multiple imputation for harmonizing longitudinal non-commensurate measures in individual participant data meta-analysis.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets