Approved Research

Pharmacoepidemiological research in the UK Biobank

Lay summary

Two pharmacoepidemiological projects, which might help to bridge the gaps between clinical trials and real-world data, are planned to be done within the scope of this research project within an estimated duration of 36 months.

It is not easy to investigate polypharmacy (most commonly defined as concomitant use of 5 medications or more) and potentially inappropriate medication (PIM, defined as prescriptions in adults which have a negative benefit-risk ratio and could be replaced by safer alternatives) for older adults in clinical trials because frail and highly co-morbid populations are mostly excluded. Although growing evidence from observational studies indicates polypharmacy and PIM use are significantly associated with increased mortality and cardiovascular events, most previous studies are vulnerable to indication bias, which occurs when the risk of an adverse event is related to the indication for medication use but not the use of the medication itself, and healthy-user/sick-stopper bias, which is described as the propensity for patients who receive one preventive therapy to also seek other preventive services or partake in other healthy behaviors. In this way, we aim to conduct a more well-designed study investigating whether polypharmacy and PIM use are significantly associated with adverse health outcomes independent from co-morbidity to provide evidence for clinical practice regarding de-prescribing.

Novel anti-diabetic drugs have entered the market in the last two decades (the dipeptidyl peptidase-4 inhibitors, sodium-glucose co-transporter-2 inhibitors, and glucagon-like peptide-1 receptor agonists). However, head-to-head comparisons regarding clinical effectiveness among these novel anti-diabetic drugs were not done in pivotal trials (typically a Phase III clinical trial intended to demonstrate and confirm the efficacy and safety of a treatment). Besides, short time span of clinical trials makes it difficult to observe long-term outcomes and potential adverse drug events. Thus, we aim to provide real-world data regarding head-to-head comparisons among these novel anti-diabetic drugs to provide more evidence for prescribing novel anti-diabetic drugs in clinical practice.

Current scope:

We intend to address the following two pharmacoepidemiological research aims:

1) To assess the associations of polypharmacy and potentially inappropriate medication (PIM) for older adults with cardiovascular events, all-cause, and cause-specific mortality.

2) To assess the associations of novel anti-diabetics with cardiovascular events.

New scope:

We intend to address the following two pharmacoepidemiological research aims:

1) To assess the associations of polypharmacy and potentially inappropriate medication (PIM) for older adults with cardiovascular events, all-cause, and cause-specific mortality.

2) To assess the associations of novel anti-diabetics with cardiovascular events.

3) To assess the associations of polypharmacy and frailty for middle-aged and older adults with cardiovascular events, all-cause, and cause-specific mortality.

4) To assess the associations of vitamin D supplementation with dementia outcomes.