Dataset Information

Adjustment for survey non-participation using record linkage and multiple imputation: A validity assessment exercise using the Health 2000 survey.

ABSTRACT:

Aims

It is becoming increasingly possible to obtain additional information about health survey participants, though not usually non-participants, via record linkage. We aimed to assess the validity of an assumption underpinning a method developed to mitigate non-participation bias. We use a survey in Finland where it is possible to link both participants and non-participants to administrative registers. Survey-derived alcohol consumption is used as the exemplar outcome.

Methods

Data on participants (85.5%) and true non-participants of the Finnish Health 2000 survey (invited survey sample N=7167 aged 30-79 years) and a contemporaneous register-based population sample (N=496,079) were individually linked to alcohol-related hospitalisation and death records. Applying the methodology to create synthetic observations on non-participants, we created 'inferred samples' (participants and inferred non-participants). Relative differences (RDs) between the inferred sample and the invited survey sample were estimated overall and by education. Five per cent limits were used to define acceptable RDs.

Results

Average weekly consumption estimates for men were 129 g and 131 g of alcohol in inferred and invited survey samples, respectively (RD -1.6%; 95% confidence interval (CI) -2.2 to -0.04%) and 35 g for women in both samples (RD -1.1%; 95% CI -2.4 to -0.8%). Estimates for men with secondary levels of education had the greatest RD (-2.4%; 95% CI -3.7 to -1.1%).

Conclusions

The sufficiently small RDs between inferred and invited survey samples support the assumption validity and use of our methodology for adjusting for non-participation. However, the presence of some significant differences means caution is required.

SUBMITTER: Mcminn MA

PROVIDER: S-EPMC7614246 | biostudies-literature | 2023 Mar

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Adjustment for survey non-participation using record linkage and multiple imputation: A validity assessment exercise using the Health 2000 survey.

Mcminn Megan A MA Martikainen Pekka P Härkänen Tommi T Tolonen Hanna H Pitkänen Joonas J Leyland Alastair H AH Gray Linsay L

Scandinavian journal of public health 20210814 2

<h4>Aims</h4>It is becoming increasingly possible to obtain additional information about health survey participants, though not usually non-participants, via record linkage. We aimed to assess the validity of an assumption underpinning a method developed to mitigate non-participation bias. We use a survey in Finland where it is possible to link both participants and non-participants to administrative registers. Survey-derived alcohol consumption is used as the exemplar outcome.<h4>Methods</h4>Da ...[more]

PMID: 34396808

Similar Datasets

Project description:BackgroundAdvancements in linking publicly available census records with vital and administrative records have enabled novel investigations in epidemiology and social history. However, in the absence of unique identifiers, the linkage of the records may be uncertain or only be successful for a subset of the census cohort, resulting in missing data. For survival analysis, differential ascertainment of event times can impact inference on risk associations and median survival.MethodsWe modify some existing approaches that are commonly used to handle missing survival times to accommodate this imperfect linkage situation including complete case analysis, censoring, weighting, and several multiple imputation methods. We then conduct simulation studies to compare the performance of the proposed approaches in estimating the associations of a risk factor or exposure in terms of hazard ratio (HR) and median survival times in the presence of missing survival times. The effects of different missing data mechanisms and exposure-survival associations on their performance are also explored. The approaches are applied to a historic cohort of residents in Ambler, PA, established using the 1930 US census, from which only 2,440 out of 4,514 individuals (54%) had death records retrievable from publicly available data sources and death certificates. Using this cohort, we examine the effects of occupational and paraoccupational asbestos exposure on survival and disparities in mortality by race and gender.ResultsWe show that imputation based on conditional survival results in less bias and greater efficiency relative to a complete case analysis when estimating log-hazard ratios and median survival times. When the approaches are applied to the Ambler cohort, we find a significant association between occupational exposure and mortality, particularly among black individuals and males, but not between paraoccupational exposure and mortality.DiscussionThis investigation illustrates the strengths and weaknesses of different imputation methods for missing survival times due to imperfect linkage of the administrative or registry data. The performance of the methods may depend on the missingness process as well as the parameter being estimated and models of interest, and such factors should be considered when choosing the methods to address the missing event times.

Project description:Linkage of medical databases, including insurer claims and electronic health records (EHRs), is increasingly common. However, few studies have investigated the behavior and output of linkage software. To determine how linkage quality is affected by different algorithms, blocking variables, methods for string matching and weight determination, and decision rules, we compared the performance of 4 nonproprietary linkage software packages linking patient identifiers from noninteroperable inpatient and outpatient EHRs. We linked datasets using first and last name, gender, and date of birth (DOB). We evaluated DOB and year of birth (YOB) as blocking variables and used exact and inexact matching methods. We compared the weights assigned to record pairs and evaluated how matching weights corresponded to a gold standard, medical record number. Deduplicated datasets contained 69,523 inpatient and 176,154 outpatient records, respectively. Linkage runs blocking on DOB produced weights ranging in number from 8 for exact matching to 64,273 for inexact matching. Linkage runs blocking on YOB produced 8 to 916,806 weights. Exact matching matched record pairs with identical test characteristics (sensitivity 90.48%, specificity 99.78%) for the highest ranked group, but algorithms differentially prioritized certain variables. Inexact matching behaved more variably, leading to dramatic differences in sensitivity (range 0.04-93.36%) and positive predictive value (PPV) (range 86.67-97.35%), even for the most highly ranked record pairs. Blocking on DOB led to higher PPV of highly ranked record pairs. An ensemble approach based on averaging scaled matching weights led to modestly improved accuracy. In summary, we found few differences in the rankings of record pairs with the highest matching weights across 4 linkage packages. Performance was more consistent for exact string matching than for inexact string matching. Most methods and software packages performed similarly when comparing matching accuracy with the gold standard. In some settings, an ensemble matching approach may outperform individual linkage algorithms.

Dataset Information

Adjustment for survey non-participation using record linkage and multiple imputation: A validity assessment exercise using the Health 2000 survey.

Aims

Methods

Results

Conclusions

Publications

Adjustment for survey non-participation using record linkage and multiple imputation: A validity assessment exercise using the Health 2000 survey.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets