Using big data to strengthen the evidence about effects of prescribing and deprescribing of medications in older people with limited life-expectancy
Description: Generally, nursing home (NH) residents use many medications. Medications are beneficial and needed to treat symptoms and diseases, but some medications have questionable benefits at the end of life. These medications with questionable benefits are suitable for deprescribing. Deprescribing means stopping or tapering a medication.
Up to now, we do not know the effects of changes in medication use (e.g. deprescribing medications with questionable benefits and initiating beneficial medications) at the end of life. In this project, we aim to evaluate these effects on the quality of life of NH residents with limited life-expectancy, using innovative data techniques. We will use data on quality of life and physical and psychosocial health of NH residents with limited life-expectancy collected in an ongoing data implementation project (BelRAI 2.0). These data are linked to administrative databases (“Big Data”) including reimbursed treatment and medication data of the whole Belgian population. Using these data we can approximate an RCT and measure effects of changes in medication use on quality of life, susceptibility to disease and mortality by comparing people for whom use of a specific medication has changed (exposure group) to people for whom use of this medication has not changed (control group), without putting them at actual risks of e.g. dying sooner by actually stopping a medication in real-life.
We address the aim of this project by 4 studies: (1) a qualitative study to explore the factors that facilitate or hinder deprescribing of medications from physicians’, nurse’ and pharmacists’ point of view, (2) an interrupted time series study to evaluate the effects of clinical practice deprescribing guidelines on actual changes in medication use, (3) a prospective cohort study to describe health an medication changes of a cohort of NH residents with limited life-expectancy, and (4) quasi-experimental matched cohort analyses to measure the effects of changes in medication use on quality of life, morbidity and mortality.
Joachim Cohen (promotor)
Tinne Dilles (promotor)
Kristel Paque (lead)
PhD student (researcher)
Research Foundation Flanders (FWO - G023620N)