Design, methodology, and preliminary results of the follow-up of a population-based cohort study in rural area of northern China: Handan Eye Study
10.1097/CM9.0000000000000418
- Author:
Kai CAO
1
;
Jie HAO
2
;
Ye ZHANG
3
;
Ai-Lian HU
1
;
Xiao-Hui YANG
1
;
Si-Zhen LI
4
;
Bing-Song WANG
1
;
Qing ZHANG
1
;
Jian-Ping HU
1
;
Cai-Xia LIN
3
;
Mayinuer YUSUFU
1
;
Ning-Li WANG
1
Author Information
1. Beijing Institute of Ophthalmology, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, Beijing 100730, China
2. Clinical Research Center, Beijing Institute of Ophthalmology, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, Beijing 100730, China
3. Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, Beijing 100730, China
4. Nanjing Aier Eye Hospital, Nanjing, Jiangsu 210000, China
- Collective Name:the Handan Eye Study Group
- Publication Type:Journal Article
- Keywords:
Cohort study;
Rural population;
Methodology;
Follow-up;
Bias
- From:
Chinese Medical Journal
2019;132(18):2157-2167
- CountryChina
- Language:English
-
Abstract:
Background:Handan Eye Study (HES), a large population-based cohort study in rural area of northern China, was one of the few studies focusing on the major eye diseases of rural Chinese population. The aim of this study was to introduce the design, methodology and to assess the data quality of the follow-up phase of HES.
Methods:All participants were recruited in Yongnian county of Handan city between 2012 and 2013. Main outcomes were measured by visual quality scales and ocular examinations. We performed the Chi-square test to make comparison of categorical data among groups, One-way analysis of variance and Kruskal-Wallis test was applied to make comparison of continuous data among groups, a post-hoc test was done to make further pairwise comparison. Inter-class correlation coefficients (ICCs) and Kappa coefficients were used to evaluate the consistency between different operators. Logistic regression was used to explore the influence factors of death, odds ratio (OR) and 95% confidence interval (CI) were used to estimate the effect size of each influence factor.
Results:The follow-up rate was 85.3%. Subjects were classified into three groups: the follow-up group (n = 5394), the loss to follow-up group (n = 929), and the dead group (n = 507), comparison of their baseline information was done. Compared with the other two groups, age of the dead group (66.52 ± 10.31 years) was the oldest (Z = 651.293, P < 0.001), male proportion was the highest (59.0%) (χ2 = 42.351, P < 0.001), only 65.9% of the dead finished middle school education (Z = 205.354, P < 0.001). The marriage percentage, body mass index (BMI), best-corrected visual acuity (BCVA), and intra-ocular pressure of the dead group was the lowest either. Spherical equivalent error (SER) of the dead group was the highest. Besides, history of smoking, hypertension, diabetes, and heart disease were more common in the dead group. Multivariate analysis showed that age (OR = 1.901, 95% CI: 1.074–1.108), gender (OR = 0.317, 95% CI: 0.224–0.448), and BCVA (OR = 0.282, 95% CI: 0.158–0.503) were associated with death. While between the follow-up group and the loss to follow-up group, there was only difference on age, gender, BMI, systolic blood pressure and SER. The Cronbach coefficients of all scales used in the follow-up were ≥0.63 and the cumulative variances were ≥0.61, indicating good reliability and validity. The ICCs and Kappa coefficients between different operators were ≥0.69.
Conclusions:HES has a high follow-up rate and a low risk of loss to follow-up bias. Age, gender, and BCVA are influence factors of death. Specifically, male subjects are at a higher risk of death than female, age is a risk factor of death while BCVA is a protective factor for death.
