Estimation of age of transition from children's to adult healthcare for young people with long term conditions using linked routinely collected healthcare data

IntroductionHealthcare transitions, including from paediatric to adult services, can be disruptive and cause a lack of continuity in care. Existing research on the paediatric-adult healthcare transition often uses a simple age cut-off to assign transition status. This risks misclassification bias, reducing observed changes at transition (adults are included in the paediatric group and vice versa) possibly to differing extents between groups that transition at different ages. ObjectiveTo develop and assess methods for estimating the transition point from paediatric to adult healthcare from routine healthcare records. MethodsA retrospective cohort of young people (12 to 23 years) with long term conditions was constructed from linked primary and secondary care data in England. Inpatient and outpatient records were classified as paediatric or adult based on treatment and clinician specialities. Transition point was estimated using three methods based on record classification (First Adult: the date of first adult record; Last Paediatric: date of last paediatric record; Fitted: a date determined by statistical fitting). Estimated transition age was compared between methods. A simulation explored impacts of estimation approaches compared to a simple age cut-off when assessing associations between transition status and healthcare events. ResultsSimulations showed using an age-based cut-off at 16 or 18 years as transition point, common in research on transition, may underestimate transition-associated changes. Many health records for those aged 14 years were classified as adult, limiting utility of the First Adult approach. The Last Paediatric approach is least sensitive to this possible misclassification and may best reflect experience of the transition. ConclusionsEstimating transition point from routine healthcare data is possible and offers advantages over a simple age cut-off. These methods, adapted as necessary for data from other countries, should be used to reduce risk of misclassification bias in studies of transition in nationally representative data.

A new generic method to improve machine learning applications in official statistics

The use of machine learning algorithms at national statistical institutes has increased significantly over the past few years. Applications range from new imputation schemes to new statistical output based entirely on machine learning. The results are promising, but recent studies have shown that the use of machine learning in official statistics always introduces a bias, known as misclassification bias. Misclassification bias does not occur in traditional applications of machine learning and therefore it has received little attention in the academic literature. In earlier work, we have collected existing methods that are able to correct misclassification bias. We have compared their statistical properties, including bias, variance and mean squared error. In this paper, we present a new generic method to correct misclassification bias for time series and we derive its statistical properties. Moreover, we show numerically that it has a lower mean squared error than the existing alternatives in a wide variety of settings. We believe that our new method may improve machine learning applications in official statistics and we aspire that our work will stimulate further methodological research in this area.

Exposure misclassification bias in the estimation of vaccine effectiveness

In epidemiology, a typical measure of interest is the risk of disease conditional upon exposure. A common source of bias in the estimation of risks and risk ratios is misclassification. Exposure misclassification affects the measurement of exposure, i.e. the variable one conditions on. This article explains how to assess biases under non-differential exposure misclassification when estimating vaccine effectiveness, i.e. the vaccine-induced relative reduction in the risk of disease. The problem can be described in terms of three binary variables: the unobserved true exposure status, the observed but potentially misclassified exposure status, and the observed true disease status. The bias due to exposure misclassification is quantified by the difference between the naïve estimand defined as one minus the risk ratio comparing individuals observed as vaccinated with individuals observed as unvaccinated, and the vaccine effectiveness defined as one minus the risk ratio comparing truly vaccinated with truly unvaccinated. The magnitude of the bias depends on five factors: the risks of disease in the truly vaccinated and the truly unvaccinated, the sensitivity and specificity of exposure measurement, and vaccination coverage. Non-differential exposure misclassification bias is always negative. In practice, if the sensitivity and specificity are known or estimable from external sources, the true risks and the vaccination coverage can be estimated from the observed data and, thus, the estimation of vaccine effectiveness based on the observed risks can be corrected for exposure misclassification. When analysing risks under misclassification, careful consideration of conditional probabilities is crucial.

Combining self-reported and objectively measured survey data to improve hypertension prevalence estimates: Portuguese experience

Background Accurate data on hypertension is essential to inform decision-making. Hypertension prevalence may be underestimated by population-based surveys due to misclassification of health status by participants. Therefore, adjustment for misclassification bias is required when relying on self-reports. This study aims to quantify misclassification bias in self-reported hypertension prevalence and prevalence ratios in the Portuguese component of the European Health Interview Survey (INS2014), and illustrate application of multiple imputation (MIME) for bias correction using measured high blood pressure data from the first Portuguese health examination survey (INSEF). Methods We assumed that objectively measured hypertension status was missing for INS2014 participants (n = 13,937) and imputed it using INSEF (n = 4910) as auxiliary data. Self-reported, objectively measured and MIME-corrected hypertension prevalence and prevalence ratios (PR) by sex, age group and education were estimated. Bias in self-reported and MIME-corrected estimates were computed using objectively measured INSEF data as a gold-standard. Results Self-reported INS2014 data underestimated hypertension prevalence in all population subgroups, with misclassification bias ranging from 5.2 to 18.6 percentage points (pp). After MIME-correction, prevalence estimates increased and became closer to objectively measured ones, with bias reduction to 0 pp - 5.7 pp. Compared to objectively measured INSEF, self-reported INS2014 data considerably underestimated prevalence ratio by sex (PR = 0.8, 95CI = [0.7, 0.9] vs. PR = 1.2, 95CI = [1.1, 1.4]). MIME successfully corrected direction of association with sex in bivariate (PR = 1.1, 95CI = [1.0, 1.3]) and multivariate analyses (PR = 1.2, 95CI = [1.0, 1.3]). Misclassification bias in hypertension prevalence ratios by education and age group were less pronounced and did not require correction in multivariate analyses. Conclusions Our results highlight the importance of misclassification bias analysis in self-reported hypertension. Multiple imputation is a feasible approach to adjust for misclassification bias in prevalence estimates and exposure-outcomes associations in survey data.