The aim of this study was to compare the effectiveness of two interventions, namely providing free spectacles or only a prescription to students with insufficiently corrected or uncorrected significant refractive errors.

Introduction Worldwide uncorrected significant refractive errors (RE) are among the commonest causes of visual impairment. This combined with the effectiveness and simplicity of correcting RE by providing spectacles has made the correction of RE one of the priorities of Vision 2020. However, little information is available on effective strategies for identifying and delivering optical services to affected individuals in developing countries

Aim The aim of this study was to compare the effectiveness of two interventions, namely providing free spectacles or only a prescription to students with insufficiently corrected or uncorrected significant refractive errors

Objectives

1. To estimate the prevalence and type of significant refractive errors in secondary school children in Tanzania, and to determine the unmet need for spectacles
2. To compare the effectiveness of providing free spectacles vs. offering only a prescription. This was judged as the proportion of students needing spectacles who were wearing spectacles at 3 and at 6 months after prescription
3. To explore attitudes to spectacles and reasons for non-compliance

Methods The study took place in Dar es Salaam, Tanzania between January and September 2004. 52 schools were randomly selected into 2 intervention groups (Group A and B) using random selection stratified by school status (government vs. private). Students were screened for significant refractive errors. A socio-economic questionnaire was administered to all the students. Students who needed spectacles in Group A schools received free spectacles (intervention A), while students in group B schools received only a prescription for spectacles (intervention B). Students were followed up 3 and 6 months after spectacles provision/prescription to check their compliance and to explore reasons for non-compliance.

Main analyses of data were the prevalence of visual impairment and its causes, risk factors for myopia, and compliance with spectacles at 3 and at 6 months.

Results

1. Prevalence of impaired visual acuity and causes 6,904 students were screened by the eye team. 2.9% of students had uncorrected poor VA, 2.3% had poor presenting VA and 0.6% still had poor VA with best correction. The main cause of visual impairment was significant refractive error (84%).

   The prevalence of uncorrected significant RE in secondary school students in Dar es Salaam was low at 1.8%. Even assuming that all students with unconfirmed poor presenting screening eyesight (0.8%) had uncorrected significant RE which is unlikely the maximum prevalence would be estimated at 2.6%.

2. Compliance with spectacles

   Students who had actually purchased their spectacles were up to twice as likely to wear them at 3 and at 6 months as students who had received free spectacles, but only a small proportion had purchased them (3 months: 30%; 6 months: 45%). Only 46% of students who had received free spectacles wore them or had them at school at 3 months and 53% at 6 months. Significant independent predictors of students wearing or having their spectacles at school 3 to 6 months after prescription or provision of free spectacles were:

   • Having to purchase spectacles (less likely, OR=0.07 to 0.42)
   • Bilateral good presenting vision, such as mild refractive errors and undercorrected significant errors (less likely than unilateral visual impairment) and unilateral visual impairment (less likely than bilateral visual impairment; OR=0.47 for each step)
   • Myopia compared to astigmatism and hyperopia, both severe enough to impair distance vision (more likely, OR= 68 -infinity) The majority of students who were going to purchase spectacles seemed to purchase them soon after prescription.

3. Risk factors for myopia Female sex, non-African ethnicity and a professional mother were independent significant risk factors for myopia in this population. These findings are similar to risk factors for myopia found in other populations.

Conclusions/ Recommendation

1. Significant refractive errors were the most common cause of avoidable visual impairment in Tanzanian secondary school students in Dar es Salaam. However, the prevalence of uncorrected bilateral and unilateral significant refractive errors in Form 1 students was low (less than 3%) which makes the value of screening for RE in this population questionable. Screening seems even less justified if one considers that only about 46% of students who had been provided with free spectacles wore spectacles or had them at school three months later. If students were asked to purchase spectacles, this proportion was even lower at 26%.
2. Similar studies in different cultural settings are needed to explore the compliance with and reasons of non-compliance with spectacles in children, as poor compliance makes the value of any screening programme questionable.

Inclusion Criteria:

• clinical diagnosis of un- or undercorrected significant refractive error
• form 1 student in participating secondary schools

Exclusion Criteria:

• presenting VA of 6/12 or better in both eyes and no hyperopia of +2D or higher