Search

In clinical studies, multi‐state model (MSM) analysis is often used to describe the sequence of events that patients experience, enabling better understanding of disease progression. A complicating factor in many MSM studies is that the exact event times may not be known. Motivated by a real dataset of patients who received stem cell transplants, we considered the setting in which some event times were exactly observed and some were missing. In our setting, there was little information about the time intervals in which the missing event times occurred and missingness depended on the event type, given the analysis model covariates. These additional challenges limited the usefulness of some missing data methods (maximum likelihood, complete case analysis, and inverse probability weighting). We show that multiple imputation (MI) of event times can perform well in this setting. MI is a flexible method that can be used with any complete data analysis model. Through an extensive simulation study, we show that MI by predictive mean matching (PMM), in which sampling is from a set of observed times without reliance on a specific parametric distribution, has little bias when event times are missing at random, conditional on the observed data. Applying PMM separately for each sub‐group of patients with a different pathway through the MSM tends to further reduce bias and improve precision. We recommend MI using PMM methods when performing MSM analysis with Markov models and partially observed event times. [ABSTRACT FROM AUTHOR]

Hypertensive disorders of pregnancy (HDP) are associated with an increased risk of cardiovascular disorders, with recent evidence linking pre‐eclampsia with vascular dementia. We examined associations of HDP with cognitive performance measured in midlife, in a prospective cohort study, the Avon Longitudinal Study of Parents and Children. Six cognitive function domains were measured 20 years after pregnancy at a mean age of 51 years. The cognition tests were repeated at clinics in the following two years. Cognitive function domains measured were immediate and delayed verbal episodic memory, working memory, processing speed, verbal intelligence, and verbal fluency. Exposures were pre‐eclampsia, gestational hypertension (GH), and a combined category of any HDP, all compared to normotensive pregnancy. Of 3393 pregnancies included in the analysis, GH was experienced by 417 (12.3%) and pre‐eclampsia by 57 (1.7%). GH was associated with lower verbal episodic memory, in the delayed logic memory test (‐0.16 SDs; 95% CI ‐0.30, ‐0.03; p =.015) and there was weak evidence of an association with the immediate logic memory test (‐0.13 SDs; ‐0.27, 0.001; p =.058). However, we did not see steeper declines by age for women with GH and there was no evidence of associations with other cognitive domains or for pre‐eclampsia with any domains. Results were not substantially changed after controlling for midlife blood pressure. Our findings suggest that a history of GH is associated with slightly reduced episodic memory 20 years after pregnancy, but we found no evidence of a quicker age‐related decline compared to women with normotensive pregnancies. [ABSTRACT FROM AUTHOR]

Summary We explored the effects of daylight saving time clock changes on sleep duration in a large accelerometer dataset. Our sample included UK Biobank participants (n = 11,780; aged 43–78 years) with accelerometer data for one or more days during the 2 weeks surrounding the Spring and Autumn daylight saving time transitions from October 2013 and November 2015. Between‐individual t‐tests compared sleep duration on the Sunday (midnight to midnight) of the clock changes with the Sunday before and the Sunday after. We also compared sleep duration on all other days (Monday–Saturday) before and after the clock changes. In Spring, mean sleep duration was 65 min lower on the Sunday of the clock changes than the Sunday before (95% confidence interval −72 to −58 min), and 61 min lower than the Sunday after (95% confidence interval −69 to −53). In Autumn, the mean sleep duration on the Sunday of the clock changes was 33 min higher than the Sunday before (95% confidence interval 27–39 min), and 38 min higher than the Sunday after (95% confidence interval 32–43 min). There was some evidence of catch‐up sleep after both transitions, with sleep duration a little higher on the Monday–Friday than before, although this was less pronounced in Autumn. Future research should use large datasets with longer periods of accelerometer wear to capture sleep duration before and after the transition in the same individuals, and examine other aspects of sleep such as circadian misalignment, sleep fragmentation or daytime napping. [ABSTRACT FROM AUTHOR]

In patient follow‐up studies, events of interest may take place between periodic clinical assessments and so the exact time of onset is not observed. Such events are known as "bounded" or "interval‐censored." Methods for handling such events can be categorized as either (i) applying multiple imputation (MI) strategies or (ii) taking a full likelihood‐based (LB) approach. We focused on MI strategies, rather than LB methods, because of their flexibility. We evaluated MI strategies for bounded event times in a competing risks analysis, examining the extent to which interval boundaries, features of the data distribution and substantive analysis model are accounted for in the imputation model. Candidate imputation models were predictive mean matching (PMM); log‐normal regression with postimputation back‐transformation; normal regression with and without restrictions on the imputed values and Delord and Genin's method based on sampling from the cumulative incidence function. We used a simulation study to compare MI methods and one LB method when data were missing at random and missing not at random, also varying the proportion of missing data, and then applied the methods to a hematopoietic stem cell transplantation dataset. We found that cumulative incidence and median event time estimation were sensitive to model misspecification. In a competing risks analysis, we found that it is more important to account for features of the data distribution than to restrict imputed values based on interval boundaries or to ensure compatibility with the substantive analysis by sampling from the cumulative incidence function. We recommend MI by type 1 PMM. [ABSTRACT FROM AUTHOR]