The results of our previous exploratory study for age-related macular degeneration (AMD) suggested that two drugs used for the treatment of dementia, Donepezil and Memantine, might have a potential protective effect on patients being diagnosed with AMD. We plan to investigate the association between the use of these two drugs and the subsequent risk of developing AMD using more rigorous study designs which can better account for any other factors which might influence our findings. We will also explore whether the use of Donepezil/Memantine is linked with other closely related conditions, including cataracts, glaucoma, and deafness/hearing loss.
Aims:
To investigate whether the use of Donepezil (compared with Rivastigmine/Galantamine) or Memantine (compared with ) other dementia medications is associated with an increased or decreased risk of subsequent development of age-related macular degeneration (AMD), cataract, glaucoma, and deafness/hearing loss.
Methods/design:
We will perform two pharmacoepidemiological studies to investigate the effects of Donepezil and Memantine separately. A new-user design will be used for the comparison between Donepezil and Rivastigmine/Galantamine (Cohort One), while a prevalent-new user design will be employed to compare Memantine users and other dementia medications (Cohort Two). Standardised morality ratio (SMR) weighting will be applied to create a pseudo-unexposed group with similar characteristics to the exposed group. Competing risk Cox proportional hazard regression models will be used to calculate crude hazard ratios (crude HRs) and adjusted hazard ratios (adjusted HRs), together with their corresponding 95% CIs. Death during the follow-up period will be treated as a competing event.
Outcomes:
Primary outcome: Age-related macular degeneration (AMD); secondary outcomes: cataracts, glaucoma, and deafness/hearing loss.
Exposures:
Prescription of Donepezil in Cohort One and prescription of Memantine in Cohort Two.
Participants:
Individuals with a diagnosis of dementia (vascular dementia, non-vascular dementia, or Alzheimer’s, defined by relevant clinical codes) aged 40 years and over will be eligible for this study. All participants must have been registered with the general practice for at least one year before the index date (start of follow-up).
Covariates:
We will include sociodemographic characteristics, dementia-related characteristics, comorbidities, biomarkers, and drug prescriptions in both the SMR weighting and Cox regression model (double robustness principle) to account for residual confounding.
Primary outcome: Age-related macular degeneration (AMD).
Secondary outcomes: Cataracts, glaucoma, and deafness/hearing loss.
Nicola Adderley - Chief Investigator - University of Birmingham
Jingya Wang - Corresponding Applicant - University of Birmingham
Ji Eun Diana Han - Collaborator - University of Birmingham
Krishnarajah Nirantharakumar - Collaborator - University of Birmingham