Current Practices in Missing Data Handling for Interrupted Time Series Studies Performed on Individual-Level Data: A Scoping Review in Health Research

Abstract

Objective Missing data can produce biased estimates in interrupted time series (ITS) analyses. We reviewed recent ITS investigations on health topics for determining 1) the data management strategies and statistical analysis performed, 2) how often missing data were considered and, if so, how they were evaluated, reported and handled. Study Design and Setting This was a scoping review following standard recommendations from the PRISMA Extension for Scoping Reviews. We included a random sample of all ITS studies that assessed any intervention relevant to health care (eg, policies or programmes) with individual-level data, published in 2019, with abstracts indexed on MEDLINE. Results From 732 studies identified, we finally reviewed 60. Reporting of missing data was rare. Data aggregation, statistical tools for modelling population-level data and complete case analyses were preferred, but these can lead to bias when data are missing at random. Seasonality and other time-dependent confounders were rarely accounted for and, when they were, missing data implications were typically ignored. Very few studies reflected on the consequences of missing data. Conclusion Handling and reporting of missing data in recent ITS studies performed for health research have many shortcomings compared with best practice.

Design and statistical analysis reporting among interrupted time series studies in drug utilization research: a cross-sectional survey

“…First, most of the studies did not mention the missing data. A study mentioned that most of the study used data aggregated at the population level, but it will lead to bias when data are missing at random at the individual level [ 11 ]. In a simulated scenario in this study, if the outcome is missing at random for male but is fully observed for female, the aggregated data will show a wrong seasonal pattern.…”

“…Several studies have systematically reviewed methodological issues regarding the design and statistical analysis of the ITS study [ 11 , 21 – 25 ]. All of the previous reviews pointed out that the considerations of autocorrelation, non-stationary and seasonality were limited, which was aligned with our study.…”

“…Several important methodological issues need to be considered when conducting ITS studies, such as time period selection, sample size, missing data, autocorrelation, and non-stationary and seasonality, which have been described in previous tutorials [5,[11][12][13][14][15]. Three issues newly emerging in recent years also require additional methodological considerations.…”

“…Previous studies might not have comprehensively addressed these methodological issues [11,[21][22][23][24][25]. The last survey on the ITS studies in drug utilization research was published in 2015 and did not cover the new methodological issues mentioned above [21].…”