Premium
Myopia and cognitive dysfunction among elderly Chinese adults: a propensity score matching analysis
Author(s) -
Sun HongPeng,
Liu Hu,
Xu Yong,
Pan ChenWei
Publication year - 2016
Publication title -
ophthalmic and physiological optics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.147
H-Index - 66
eISSN - 1475-1313
pISSN - 0275-5408
DOI - 10.1111/opo.12248
Subject(s) - propensity score matching , medicine , confidence interval , confounding , odds ratio , covariate , observational study , logistic regression , statistics , mathematics
Abstract Purpose The association between myopia and cognitive dysfunction among elderly adults was assessed by applying a Propensity Score Matching ( PSM ) approach. This is a statistical method which allows investigators to estimate causal treatment effects using observational or nonrandomised data. Methods The study was designed as a community‐based cross‐sectional study based on a Chinese cohort aged 60 years or older in China. Objective refraction was measured using an autorefractor and subjective refraction was used to refine vision, using the results of the objective refraction as the starting point. Myopia was defined as a spherical equivalent value of less than −0.50 dioptre (D) in the right eye. The Abbreviated Mental Test ( AMT ) was used for cognitive assessment. The propensity scores for myopia were formulated using 13 potential confounders. We matched the propensity scores for subjects with and without myopia within a caliper of 0.01 of logit function of propensity scores. Results About 4123 elderly adults who successfully completed the AMT were included in this analysis. The odds ratio ( OR ) of cognitive dysfunction for myopia before matching was 1.98 (95% confidence interval [ CI ] 1.61, 2.44; p < 0.001). There were significant covariate imbalances between comparison groups and after propensity score matching, covariate imbalance was significantly reduced. After propensity score matching, the OR of cognitive dysfunction was marginally significant and the magnitude of association was reduced ( OR : 1.31 95% CI 1.00, 1.71; p = 0.05). Traditional multivariate logistic regression modelling found an OR of 1.52 (95% CI 1.23, 2.06; p < 0.001) after adjusting for the 13 potential confounders. Conclusions Myopia was associated with a higher prevalence of cognitive dysfunction among elderly Chinese aged 60 years or older in China. The PSM approach may be a useful method to address selection bias in observational studies when randomised trials cannot ethically be conducted.