Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.787669
Title: New uses for old drugs : investigating whether antihypertensives can be repurposed for the prevention of dementia
Author: Walker, Venexia
ISNI:       0000 0004 7972 780X
Awarding Body: University of Bristol
Current Institution: University of Bristol
Date of Award: 2019
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
Drug repurposing applies existing drugs to novel indications to identify potential treatments in a more rapid and cost-effective manner than traditional drug development. Antihypertensive drugs are priority repurposing candidates for dementia prevention, however current evidence is inconclusive. This thesis addressed five aims related to the novel use of instrumental variable analysis methods to assess this hypothesis. Existing power calculators for instrumental variable analysis studies do not allow for the structure of pharmacoepidemiological research questions. Consequently, I derived and validated a power formula for studies using a single binary instrument to analyse the causal effect of a binary exposure on a continuous outcome in this context. There has been relatively little discussion focussed on using genetic instrumental variable analysis, namely Mendelian randomization, for drug repurposing to date. I therefore reviewed several examples drawn from the existing literature and considered the strengths and limitations of this method for this purpose. Breaks in the prescription of existing drugs for dementia could be exploited as natural experiments for pharmacoepidemiology studies. I examined the impact of regulatory guidance and patent expiry on dementia drug prescribing to determine if any such breaks occurred and found that prescriptions have increased, without breaks, between June 1997 and December 2015. Physicians' prescribing preference can be used as an instrument to investigate whether antihypertensive drugs have a causal effect on incident dementia using electronic health record data. Implementing this method in the Clinical Practice Research Datalink, I found the magnitude of the differences between antihypertensive drug classes for dementia prevention were smaller than those previously reported. Single nucleotide polymorphisms that mimic the action of antihypertensive targets can be used as an instrument in Mendelian randomization to investigate this same hypothesis. Using this method, I found little evidence that lowering systolic blood pressure, via antihypertensive drug classes, affected Alzheimer's disease risk.
Supervisor: Davies, Neil ; Kehoe, Patrick ; Martin, Richard Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.787669  DOI: Not available
Share: