Lhasa childhood eye study: the rationale, methodology, and baseline data of a 5 year follow-up of school-based cohort study in the Tibetan plateau region of Southwest China

The LCES is a school-based cohort study mainly designed to longitudinally observe the occurrence and development of different ocular diseases especially myopia in school-age children. The study adheres to the STROBE guidelines and the principles of the Declaration of Helsinki. The study has finished the clinical registration on http://www.chictr.org.cn (ChiCTR1900026693). Ethics committee approval was obtained from the Institutional Review Board of Beijing Tongren Hospital, Capital Medical University (TRECKY2019–146). Informed consent forms signed by the parents of all the participants were obtained prior to the start of the program. The LCES would be conducted from 2019 to 2024. We randomly selected a group of grade 1 students from the elementary schools in Lhasa and will follow them until they enter different middle schools.

Voluntary grade one students, living in Lhasa city for at least half a year and planning to continue to live there for at least 5 years, are legible for LCES. Individuals, suffering from mental illness or other medical conditions, were unable to cooperate with the baseline survey and follow-up would be excluded. According to the report from ACES which was conducted in central China recently, the 5-year cumulative incidence of myopia in school-age children was calculated to be 58.37% [7]. Considering that the incidence of myopia in Lhasa might be lower than in Anyang, we used the cumulative incidence of 40% to calculate the sample size. Assuming a design effect of 2.0, a tolerated error of 0.1 times the myopia incidence, and a loss of follow-up of 20%, 1382 grade one students would be needed. 27 out of the 28 elementary schools in Lhasa available joining the LCES were stratified into three levels based on the evaluation of local government. Finally, 1943 grade 1 students of 7 elementary schools were randomly sampled by stratified cluster sampling.

A coordination meeting was held by the local city government before the recruitment. Officials from city governments, the health, and the education departments of Lhasa summoned the principals of the selected schools and inform them of the importance and significance of the LCES. Local media such as television and newspapers were also used to publicize the study. The teachers in charge of each class were trained by the ophthalmologists from LCES to seek mutual understanding of the program and to ascertain the eligibility of each child. With the support of the principals and teachers, parents’ meetings were held to explain the importance and significance of the LCES. And then we invited them to participate in the program. Written informed consent forms were interpreted in detail and issued for signatures. The contact information of the Project Manager and Assistant Project Manager were given to facilitate communication between the study staff and the participants. Participants could withdraw from the study anytime. For those children and parents who didn’t agree to participate, at least three telephone calls were performed to seek their understanding. The children would be invited in-class-unit arranged by teachers in charge of a free check-up in the health examination station of LCES at an appointed time if eligibility criteria were fulfilled.

Only parent questionnaires were surveyed in LCES. Because most of the children were unable to fulfill the student questionnaires after we did a pilot study in one of the class chose. The questionnaires used in the LCES were mainly derived from the parent questionnaires used in ACES to facilitate comparisons between the two studies. We discussed with the representatives of local teachers, parents, and ophthalmologists to make sure the questionnaires culturally appropriate and linguistically accurate and no change needed. As reported in ACES the questionnaires’ reliability and validity were verified. In general, the questionnaires were designed to collect information about work, outdoor activity, date of birth and ophthalmic treatment, reading, living, as well as eating habits, and parents’ information such as refractive and socioeconomic status, education, pregnancy history, and medical records.

Since more than half of the parents were Tibetans and did not know Chinese, the questionnaires were translated and explained in detail by the trained teachers and the LCES staffs familiar with Tibetan and Chinese at the parents’ meeting. Difficulties about the questionnaire were resolved immediately during the meeting to make sure the results accurate. The fulfilled questionnaires were inspected by experienced staff of LCES according to the standardized protocol. Any equivocal points found from the finished questionnaires were clarified by interviewing the parents again. The questionnaire survey would be administered annually during the follow-up. The student questionnaires would be added from the one-year-follow-up on as the children might be able to understand and fulfill the questionnaires after one-year study.

We performed all the examinations for the students in the health examination station of Lhasa Maternal and Child Health Care Center. The procedures were outlined in Fig. 1. Basic systematic examinations were composed of height, weight, body circumference, heart rate, blood pressure, oxygen saturation, etc. Standardized ocular examinations included distant, near and pin-hole visual acuity, identification of amblyopia and strabismus, ocular biometry, optical coherence tomography, retinal photography, cycloplegic autorefraction, intraocular pressure, stereo acuity, and ocular dominance. Examiners would perform the examinations according to the standard operation procedures (online supplementary).

The procedures of quality control were implemented throughout the study. The standard operating procedures were produced for staff training before data collection. All examinations were required to be performed following the procedures. For the fieldwork, two ophthalmologists (FJ, CWW) worked at the field site to ensure that the protocols were followed strictly. The examination handbook and questionnaires of every student were verified at least twice to evaluate integrity and precision. The students with missing data or erroneous data, would be asked to come back for ocular examinations or their parents would be contacted again for questionnaire. After verification of double data entry, 5% of the database was randomly selected to identify any inconsistency between the original documents and electronic records.

All the data were filled in forms and were independently entered into the database using Epidata software 3.1 (The Epidata Association, Odense, Denmark) by two individuals. The OCT results and Retinal photographs were stored in digital format for further post-processing and grading. Statistical analysis was performed using SAS software (version 9.4, SAS Inc., Cary, NC, USA). Age- and sex-specific prevalence estimates of myopia, amblyopia, strabismus and other ocular diseases, and the incidence of myopia would be calculated and analyzed. A general linear model (analysis of covariance) and a linear regression model would be used to evaluate the associations between risk factors and myopia at baseline. The Cox model would be used to analyze the risk factor of time to develop myopia during follow-up. Generalized estimating equation models would be used for eye-specific analyses.

Recruitment details were shown in Fig. 2. Among the 1942 sampled students, 40 were ineligible for LCES according to the inclusion and exclusion criteria. Among the remaining 1902 eligible individuals, 46 did not attend the baseline exam, leaving 1856 grade one students who completed all the examinations during the period from 21st October to 18th November 2019, with an overall response rate of 97.58%; 97.14% for boys and 98.09% for girls.

For the 1856 participants in LCES, the mean age was 6.83 ± 0.46 years and 984 (53.02%) were males. Overall participant characteristics were shown in Table 1. 1762(94.93%) of the 1856 participants were Tibetan, while 85 (4.58%) were Han, 9 (0.49%) were other nationalities.

Among the 1856 grade one students examined, 1853 (99.84%) had cycloplegic autorefraction performed on both of their eyes. Cycloplegic spherical equivalent had a median of + 1.13 D (range − 9.00D to + 8.25 D) (Table 2). The prevalence rates of myopia and high myopia were 3.94 and 0.16%, respectively. The prevalence rates of emmetropia and hyperopia were 89.15 and 6.91%, respectively. The prevalence was based on data from the right eyes. A significant difference was found in the overall comparison of the refractive status between LCES and ACES using Fisher’s accurate test. Comparisons of refractive status in detail were shown in Table 2.