<p>Health indicator recording in UK primary care electronic health records: key implications for handling missing data</p>

Abstract

Background Clinical databases are increasingly used for health research; many of them capture information on common health indicators including height, weight, blood pressure, cholesterol level, smoking status, and alcohol consumption. However, these are often not recorded on a regular basis; missing data are ubiquitous. We described the recording of health indicators in UK primary care and evaluated key implications for handling missing data. Methods We examined the recording of health indicators in The Health Improvement Network (THIN) UK primary care database over time, by demographic variables (age and sex) and chronic diseases (diabetes, myocardial infarction, and stroke). Using weight as an example, we fitted linear and logistic regression models to examine the associations of weight measurements and the probability of having weight recorded with individuals’ demographic characteristics and chronic diseases. Results In total, 6,345,851 individuals aged 18–99 years contributed data to THIN between 2000 and 2015. Women aged 18–65 years were more likely than men of the same age to have health indicators recorded; this gap narrowed after age 65. About 60–80% of individuals had their height, weight, blood pressure, smoking status, and alcohol consumption recorded during the first year of registration. In the years following registration, these proportions fell to 10%–40%. Individuals with chronic diseases were more likely to have health indicators recorded, particularly after the introduction of a General Practitioner incentive scheme. Individuals’ demographic characteristics and chronic diseases were associated with both observed weight measurements and missingness in weight. Conclusion Missing data in common health indicators will affect statistical analysis in health research studies. A single analysis of primary care data using the available information alone may be misleading. Multiple imputation of missing values accounting for demographic characteristics and disease status is recommended but should be considered and implemented carefully. Sensitivity analysis exploring alternative assumptions for missing data should also be evaluated.

<p>Handling Missing Values in Interrupted Time Series Analysis of Longitudinal Individual-Level Data</p>

“…For example, in clinical practice, younger women are more likely than younger men to have weight recorded when they consult their family physician (general practitioner). 12 In other words, the distribution of missing data in weight depends on the individual's sex, so weight is missing at random (MAR) given sex. The same will apply to other partially observed outcomes that are MAR.…”

Physical health monitoring in dementia and associations with ethnicity: a descriptive study using electronic health records

“…The study found less than half of people with dementia had a weight and/or BMI recorded, consistent with another THIN study reporting 47% had this recorded in 2010–2013, 25 and other reports of poor weight recording in primary care. 31 Eating and drinking difficulties are common in dementia. The risk of malnutrition increases as dementia progresses and people with dementia have 10 times more malnutrition or dehydration-related hospital admissions compared with age-matched controls.…”

Analyzing Missing Data in Perinatal Pharmacoepidemiology Research: Methodological Considerations to Limit the Risk of Bias

“…This contrasts with studies set in birth cohorts, teratology information services, or general practice databases, which often have to contend with much higher levels of missingness and with patterns that are likely to be informative. The substantial problem of missing data in these study types has promoted important methodological research on the topic, 23,24 as well as a greater uptake of multiple imputation methods by researchers using this type of data (Table I). Simpler approaches to handle missing data such as indicator variables were often not reported in the articles we evaluated, which is encouraging given the well-established shortcomings of the method.…”