Background

Myopia is a significant public health problem and its prevalence is increasing over time [1]. By the year 2020, it is estimated that 2.5 billion people – one-third of the world’s population – will be affected by myopia [2]. Furthermore, the prevalence of myopia has been shown to vary widely with geographic location3. In European and North American adult populations, the prevalence of myopia is reported to be between 20% and 40% [3, 4]. In Asia, the prevalence of myopia among teenagers and young adults exceeds 70% [5, 6].

The etiology of myopia is multifactorial and both genes and the environment play important roles [7, 8]; myopia results from complex genetics [2, 9,10,11,12]. It has been shown that in young adults, education appear to cause increases in axial length and shifts toward increased myopia [13]. The high prevalence of myopia and high numbers of myopic university students pose particularly important public health and social problems [14]. Ocular risks associated with myopia should not be underestimated, and there is a public health need to prevent myopia onset and progression.

Based on the above research, there have been numerous studies both on myopia prevalence and associated factors. Therefore, we used a large sample to confirm the prevalence of myopia and factors associated with myopia among all students at an Inner Mongolian medical university.

Methods

Data source

Cross-sectional censuses of the physical and mental health of university students were conducted in 2011(6044) and 2013(6109) among medical students residing on campus at the Inner Mongolia Medical College of China. The censuses included students enrolled from 2007 to 2012, covering 6 years. Some students (1015) resided on campus in 2011 and 2013; therefore, they participated in both censuses. The total number of students is 11,138. In our school system, some students reside at school for 3 years, and others reside at school for 4 years [15]; therefore, we conducted the census twice. The two censuses employed a self-administered questionnaire. To make data expression clearer, we defined the factors newly appear in this paper, except those which had been defined in our previous studies [11, 15,16,17]. We also conducted a pre-survey to determine whether each factor could be accurately understood by the students and the factors which are easy to confound the data were modified. We also explained these factors in detail to the students in our census. The test-retest reliability was 0.96, which was calculated through a randomly sampled 100 students in the census [15].

Census contents

Subjects evaluated as myopic were those who used myopic spectacles or contact lenses to look at objects and gave details about the age at which they started to wear spectacles or contact lenses [18].

We investigated factors including basic information: area (urban/rural), year of study (1, 2, 3, 4 and 5), and sex (male/female); genetic factors: family members’ myopic statuses (both parents, one parent, or no parent); and environmental factors: if they frequently see green (yes/no), perform eye exercises (insist on performing/ sometimes /rarely/ will not perform), use eye drops (yes/no), have an inadequate diet (yes/no), take breaks after reading 1 h (yes/no), use a lamp (yes/no), stay up late (yes/no), are affected by people around them staying up late (yes/no), stay up late for homework (yes/no), stay up late for study section review (yes/no), stay up late because of pressure to study (yes/no), search for information online (yes/no), when they started using a computer (primary school, high school, university), how often they used a computer (every day, 2–4 times a week, 1 time per week, almost none), how long they used a computer per day (less than 1 h, 1–3 h, more than 3 h), bedtime (before 22:00, 22:00–00:00, after 00:00, no regular time), read for long durations (yes/no), read while lying down (yes/no), read under dim light (yes/no), suffer from depression (yes/no), and if they rubbed their eyes (yes/no).

The dim light was lighting levels be below 30 footcandles (incandescent light bulbs below 40w) where the students usually reads and writes [19]. Participants excluded from the analyses included those reporting a history of cataract and/or glaucoma.

Statistical analysis

The chi-squared test was used to test for differences in myopic prevalence between 2011 and 2013 in relation to various parameters. Because there was no significant difference in myopia prevalence between 2011 and 2013, we explored factors related to myopia prevalence by merging the two censuses. Prevalence was calculated for each investigated factor and various myopia occurrence times among students according to different parental myopia statuses. Multiple-factor non-conditional logistic regression analysis was used to evaluate the significance of each factor of myopia after adjusted for possible confounding factors. Dependent variables fell into two categories: myopic and non-myopic. Independent variables on the dependent variable in the model included all investigated factors. The odds ratio (OR) and corresponding 95% CI were calculated. In the model, ORs >1.0 designated increased myopic risk and ORs <1.0 indicated protective factors.

Microsoft Excel and SPSS 13.0 statistical software were used for data management and analysis. A statistical significance level of p ≤ 0.05 was used throughout the study.

Results

A total of 11,138 students enrolled from 2007 to 2012 completed the questionnaire, of whom 7980 (27.3%) were men and 3149 (72.7%) were women. The mean age of the participants was 21.08 ± 1.57. The prevalence of myopia in 2011 and 2013 was 70.50 and 69.21, respectively, and no statistically significant difference existed between the two censuses (χ2 = 2.4, p = 0.12). One thousand fifteen students participated in both censuses, in which 694 myopic students were assessed in 2011, and only four myopic students were added in 2013.

Table 1 shows the baseline characteristics of the study participants and prevalence of myopia in relation to each census item. There was no difference among the various year of study in 2011 or 2013. Students’ myopic prevalence when both parents were myopic was over 90%; the prevalence when one parent was myopic was nearly 80%; and the prevalence when both parents were non-myopic was less than 70% (χ2 trend test = 18.23, p < 0.001). Myopic prevalence increased with an increased number of myopic parents according to the χ2 trend test. The prevalence of myopia was higher among women living in the city. The prevalence of myopia was also higher among students with staying up late, using a computer, lack of concern for eye health, lying down while reading, reading for a long duration, and going to bed after 10:00.

Table 1 Prevalence of myopia among Inner Mongolia Medical University students in relation to various parameters
Full size table

Table 2 shows the myopia statuses of the students. The results suggest that nearly 80% began wearing spectacles in middle school. Regarding the type of glasses, more than 80% wore frame glasses and more than half chose them in an eyeglasses store. Among myopic students, 53.44% envied normal vision and 56.91% felt eye fatigue. The attitude of 60% of myopic students was open to trying treatment.

Table 2 The myopia statuses of students at Inner Mongolia Medical University
Full size table

We included all factors in binary logistic regression models (Table 3). Students with one or two myopic parents were at high risk for myopia. Women who lived in the city with staying up late, using a computer more than 3 h per day, not performing eye exercises, using eye drops, rubbing their eyes were at high risk for myopia. Taking a break after reading for 1 h and not reading under a dim lamp were protective factors.

Table 3 Results of the logistic regression analysis on myopia among medical students
Full size table

Table 4 shows the time at which myopia occurred among students with different parental myopia statuses. The occurrence time of student myopia was earliest in kindergarten and primary school when both parents were myopic. The occurrence time of student myopia ranked second in middle school when one parent was myopic. The occurrence time was latest in university when neither parent was myopic.

Table 4 Student myopia occurrence time among different parental myopia statuses
Full size table

Discussion

Compared with the reported prevalence of myopia among the general population in Western countries, the prevalence of myopia in our study was considerably higher [20, 21]. Compared with medical students in other countries, the prevalence of myopia in our study was also higher [22, 23].

We performed two censuses of all students residing on the university campus in 2011 and 2013. There was no statistically significant difference between the two censuses in myopic prevalence. Further, the myopic prevalence of students who participated in both censuses was nearly unchanged. The prevalence of myopia was not significantly different from year of study 1 to year of study 5 in 2011 or 2013. The results suggest that myopic status was stable and did not significantly change during the university period. A study among university students was consistent with our result [14]. The results may be explained by genetic factors. Students’ myopic prevalence when both parents were myopic, when one parent was myopic, and when both parents were non-myopic showed a dose-dependent relationship. It showed that the majority of myopia cases within populations are caused by hereditary factors. In addition, the occurrence time of student myopia was the earliest in kindergarten and primary school when both parents were myopic. The occurrence time of student myopia was second earliest in middle school when one parent was myopic. Student myopia occurred latest in university when no parent was myopic. It clarified that student myopia occurs earlier with an increased number of myopic parents. Several studies have suggested relationships between heredity and myopia [2, 9]. Our results are consistent with their conclusions and confirmed that people were more likely to develop myopia earlier because of heredity from myopic parents [9].

While genetic factors play a major role, environmental factors also play a role in lens thickness changes, but do not change myopic status [12]. A previous study confirmed that environmental change causes myopia [12]. We further explored myopia-related environmental factors among 11,138 students. In our study, taking a break after reading 1 h and not studying under a dim lamp had protective effects on eye health. It clarified that a healthy lifestyle played a protective role in university students. On the contrary staying up late, using a computer more than 3 h per day, not performing eye exercises, using eye drops, and rubbing the eyes could increase the thickness of the glasses. Moreover, it was demonstrated that some protective measures were useful for medical university students and could prevent further increases in the thickness of their glasses.

Conclusions

Myopic status was stable during the university period. Genetic factors play a major role in myopia. Taking a break after reading 1 h and not studying under a dim lamp had protective effects on eye health. Staying up late, using a computer more than 3 h per day, not performing eye exercises, using eye drops, and rubbing the eyes could increase the thickness of the glasses and all above can effectively change through education, so university administrators should provide systematic education to enhance it in university students.

Limitation

In our study, we did not perform an eye examination for all students; thus, “myopic” was defined according to the individual student’s report that they “myopic used spectacles or contact lenses either occasionally or frequently” during our study. Therefore, the prevalence of myopia may be lower because some slightly myopic students may choose not to wear glasses.