This report is part of a cross-sectional study designed to quantify near-vision anomalies (refractive and accommodative–vergence) and their association with symptoms in order to develop strategies on how to identify and treat them. The focus of this study is on near-vision measures in children aged 13–18 years.

The target population was black high school students in uMhlathuze municipality, and the study sample comprised students selected from 13 high schools out of a total sample frame of 60 high schools in this municipality of the province of KwaZulu-Natal, South Africa. Participants were selected using stratified, multistage cluster, random sampling (from the municipality to classroom levels). Students of African descent and of either gender were eligible to participate in the study. From the case history, participants were excluded if they had any systemic conditions such as diabetes, hypertension or multiple sclerosis and were on any systemic medication. Also excluded were children with amblyopia, suppression, strabismus, ocular diseases or nystagmus.

The study design, including the sampling and sample size calculation and details of testing procedures, are described in previous publications;^43,44 details are omitted here as the techniques are conventional optometric procedures.⁶

The school principals provided the venues where the visual examinations were conducted. The purpose and procedure for every technique was carefully explained to each participant prior to the eye screening. To ensure that the participants understood all the instructions, trial testing was performed for complex tests such as accommodative facility and fusional vergences. The test procedures and testing conditions were standardised and an average of three test measurements were taken for near tests. The focus of this study was on near-vision functions; therefore, far-distance variables are not described in detail.

Preliminary tests performed included visual acuity measurements and ocular health status assessment; suppression was evaluated at near using the Worth four-dot test and stereoacuity using the Randot Stereotest. Ocular motility was evaluated using the broad H-test. Refractive errors were determined objectively (without cycloplegia) with an autorefractor (Huvitz MRK-3100, Seoul, South Korea) and subjectively using the phoropter. Heterophoria was evaluated over the participants’ subjective refraction and was evaluated using the von Graefe prism dissociation method after strabismus had been ruled out using the unilateral cover test. The fusional ranges were evaluated using a trial frame and prism bars. Negative fusional vergence (NFV) was routinely measured first, before positive fusional vergences (PFV), as the convergence responses stimulated during the base-out measurements may produce the adaptation of tonic vergence (related to the physiological position of rest and because of the extraocular muscles’ tonus), which may bias the subsequent base-in values.^6,7,11

All data were analysed by a statistician using the Statistical Package for Social Sciences (SPSS) version 21. Descriptive statistics were presented as means, standard deviations and medians. The Kolmogorov–Smirnov (K-S) test was used to test for normality of data. Chi-square tests and logistic regression models were performed to test for differences in proportions for categorical variables among groups. The Kruskal–Wallis equality-of-population rank non-parametric test was applied to compare differences in means for related measures among groups, and independent t-tests were used to compare the means of gender, age groups and grade levels. The non-parametric Spearman’s rho correlation test was used to determine the strength of correlations between heterophoria and fusional vergences, while the chi-square test was applied to determine associations between phoria and refractive errors. The distributions of variables were presented using tables, graphs and proportions, and corresponding 95% confidence intervals were presented as an estimate of the prevalence.

The classification criteria (Table 1) for heterophoria⁴⁵ and fusional vergences followed the criteria used in other studies,^43,45 while vertical phoria followed the recommendation by Von Noorden.¹ Another criteria considered was the Sheard’s criterion proposes that for an individual to be comfortable (without experiencing symptoms), the fusional reserve opposing the heterophoria should be at least twice the size of the heterophoria: base out (PFV) for exophoria and base in (negative fusional vergence) for esophoria.^6,7 Thus, failing the Sheard’s criterion suggests that a patient may be symptomatic.^6,7,46 The fusion reserve ratio was calculated as the fusional convergence reserve divided by the prism value from the phoria test,²³ which corresponds to ≥ 2 based on Sheard’s criterion^7,23 when converted into a continuous variable, which implies that Sheard’s criterion is not met when values are lower than 2.

Written informed consents for access to the schools were obtained from the Department of Education and the school principals, and consents and assents were obtained from parents and children, respectively. The study protocol was approved by the Biomedical Research Ethics Committee (ethical clearance reference number: BE 177/12) of the University of KwaZulu-Natal, South Africa. The conduct of the study complied with the Declaration of Helsinki regarding research on human subjects.⁴⁷

A total of 1211 individuals returned their consent and assent forms, but 10 students were excluded (7 had amblyopia, 1 was diabetic, 1 had glaucoma and another had corneal scars because of trauma), leaving 1201. A further 145 children aged 19 and above were excluded from this study, as we intended to report only on participants of the official high school age range of 13–18 years. Thus, data were analysed for 1056 participants (a response rate of 87%), with a mean age of 15.89 ± 1.58 years and a median age of 16 years. There were 403 (38%) males and 653 (62%) females. The sample comprised 781 (73.96%) students from grades 8–10 (lower grade level) and 275 (26.07%) from grades 11–12 (higher grade level).

There was a correlation between the mean spherical equivalents of the left and right eyes (r = 0.645, p = 0.01). The descriptive statistics for refractive errors, heterophoria and fusional vergences are shown in Table 2. Overall, data for refractive errors and phoria were not normally distributed, whereas data for fusional vergences were.

The frequencies of distance phoria are presented (Figure 1) with 95% confidence intervals (CI) in text. For distance lateral phoria, orthophoria was the most frequent at 80.1% (CI: 77.7% – 82.6%); exophoria was 13.9% (CI: 11.9% – 16.0%), with the majority being 1 pd; while esophoria was 6.0% (CI: 4.5% – 7.4%), with the majority being 1 pd.

For lateral phoria at near (Figure 1), exophoria was the most frequent (51.30%; CI: 48.3% – 54.2%). The majority were between 2 and 3 pd, while esophoria was 12.1% (CI: 10.2% – 14.0%) and mainly 2 pd.

For vertical phoria, hyperphoria at distance was 2.7% and 3.6% at near, while orthophoria was 97.3% and 96.4%, respectively.

For the distribution of phorias according to demographic variables, the male participants were significantly more exophoric (mean 2.88 ± 3.66 pd, median 2 pd, frequency 57%) than females (mean 2.25 ± 3.22 pd, median 2 pd, frequency 48.8%) (OR, 2.73; p = 0.01, logistic regression). Neither age (p = 0.48) nor grade level (p = 0.64) influenced the distribution of exophoria. The prevalence of near esophoria was significantly higher among children aged 13 years compared to others aged 14–18 years (χ² = 18.884, p = 0.02; mean value = F = 2.382, df = 1047, p = 0.03) and school grade (χ² = 6.578, p = 0.01). There was no gender difference in the mean distance lateral phoria (p = 0.09), while mean distance vertical phoria was higher in older age group (p = 0.02).

For the distribution of near heterophoria and fusional vergences, the mild categories of exophoria and PFV and normal NFV were most prevalent (Table 3).

The associations between fusional vergences and heterophoria were specific to low magnitudes of fusional vergences (Table 4).

There was an inverse correlation between near exophoria and PFV break (r = −0.095, p = 0.01) and recovery (r = −0.138, p = 0.01), whereas there was a positive correlation between exophoria and NFV (r = 0.148, p = 0.01). The PFV break and recovery values were reduced (poorer) with an increase in the severity of the heterophoria. However, the association was significant only for mild PFV (F = 2.639, df = 928, p = 0.04). Participants with severe exophoria had significantly worse PFV values than did those in the mild and moderate categories (Table 4).

Esophoria was inversely correlated with NFV break values (r = −0.136, p = 0.01) and recovery (r = 0.136, p = 0.01). For both severity of esophoria (> 2 and ≤ 2 pd), an increase in the magnitude was associated with poorer (lower) NFV values. The difference was significant only for anomalous NFV break (p = 0.01). A similar trend was observed for NFV recovery. With PFV, esophoria was positively correlated with PFV break (r = 0.132, p = 0.03) and PFV recovery (r = 0.07, p = 0.02). Positive fusional vergence values were higher with an increase in the magnitude of the esodeviation, and the association was significant only for mild PFV. A similar trend was observed for PFV recovery.

There was no significant correlation between orthophoria and PFV break (r = −0.028, p = 0.36), but there was negative correlation with recovery (r = −0.07, p = 0.02) and highly positive correlation with NFV break (r = 0.134, p = 0.01) and recovery (r = 0.143, p = 0.01).

There was no significant correlation between hyperphoria and PFV values (break = r = −0.031, p = 0.85; recovery: r = −0.03, p = 0.82), whereas hyperphoria > 0.5 pd was inversely correlated with NFV (r = −0.31, p = 0.05). The mean NFV break values were significantly poorer (lower) with an increase in the severity of the hyperdeviation. However, the difference was significant only for anomalous NFV break values. A similar trend was observed for NFV recovery (t = 5.37, df = 1044, p = 0.01)

A total of 6.45% of the children with exophoria failed the Sheard’s criterion. There was no significant gender (χ² = 1.96, df = 1, p = 0.16), age group (χ² = 2.176, df = 1, p = 0.14) or grade level (χ² = 0.79, df = 1, p = 0.37) difference in the failure of Sheard’s criteria. The mean fusional convergence ratio was 4.67, with a median of 1.16. The Mann–Whitney test showed no significant difference in fusional convergence ratio in gender (p = 0.15) and age (p = 0.98).

Exophoria was significantly associated with astigmatism (χ² = 11.1, df = 3, p = 0.01); while esophoria was significantly associated with myopia (χ² = 25.7, df = 1, p = 0.01), astigmatism (χ² = 34.2, df = 1, p = 0.01) and anisometropia (χ² = 38.7, df = 1, p = 0.01).

There was a weak but significant correlation between phoria and fusional vergences, which may be related to their associations being specific to categories, for example, mild PFV. Similar to the present study, most studies reported weak associations between heterophoria and fusional vergences,^23,24,32 while others found no correlation.^30,31,35 Anderson et al.²⁸ found significant associations, although the extent of the associations were not quantified. Associations between phoria and fusional vergence are expected, given that heterophoria is compensated by fusional vergences, which vary with the direction of the phoria.^34,35 Accordingly, exophoria exists because of a partial failure of base-out prism adaptation, while esophoria is because of a partial failure to adapt to base-in prism.^17,18

The association between phorias and fusional vergences becomes clinically important in cases of decompensation. Thus, a knowledge of the magnitudes or severities of the vergence reserves and demand give useful clues as to the possibility of having comfortable binocular vision. Invariably, the report of symptoms is important in optometric practice, as most patients who consult do so because of the symptoms they are experiencing and an analysis of patients’ symptoms guides a clinician’s treatment plans. Aligned to heterophoria and fusional vergences is Sheard’s criterion, which is a good indicator of symptoms.⁴⁶ The 6.5% failure rate of Sheard’s criterion in this study is lower than the 12.5% found by Myklebust et al.⁴² in Norwegian schoolchildren, while the 4.6 mean convergence ratio found is comparable to 4.45 pd by Lanca and Rowe in Portugal.²³

The findings on the association of phorias and refractive errors from previous studies are diverse,^{37,38,39,40,41,42} which may be related to different study designs and criteria applied to define variables. Studies of the association between phoria and refractive errors of African schoolchildren could not be found. However, other studies found associations between near phoria and refractive errors.^13,16,36,38 In the present study, esophoria was associated with myopia, which agrees with the findings by Chung and Chong³⁶ but contrasts with those of Ditmars,³⁷ where myopic patients tended to be exophoric, whereas hyperopes tended to be esophoric. Li-Ju et al.³⁸ also found a significant association between exophoria and myopia in Taiwanese schoolchildren. Similar to the present study, Walline et al.⁴ reported no association between hyperopia and esophoria. The absence of significant associations between near esophoria and hyperopia may be related to the mainly low magnitudes of hyperopia found in the present study.

In general, the association between heterophorias and refractive errors may be considered from different perspectives. In esophoria resulting from high latent hyperopia, the esophoria may be such that the motor and sensory fusion cannot cope, resulting in decompensation and tropia.³⁹ The association may be because of the need for greater accommodative efforts to overcome hyperopia, which in conjunction with convergence causes an esodeviation.¹⁶ Furthermore, near esophoria is associated with the development and increased rates of myopic progression in children^33,36,40 and is related to vergence adaptation to a prolonged exposure to near tasks, which causes a convergence; this may even be a possible cause of myopia progression.^33,34

In summary, this study extends our understanding and knowledge of the distribution of heterophoria and its associations with fusional vergences and refractive errors in a sample of South African schoolchildren. Given that heterophoria is symptomatic only when it is decompensated, the significant correlations with fusional vergences, as well as the percentage of children who failed Sheard’s criterion, deserve attention, as the failure of Sheard’s criterion is regarded as a good indicator of symptoms.⁴⁶ The participants in the study may not have reported symptoms at the time of vision screening. However, with changes in phoria associated with increase in age and near demands, the subsequent decrease in PFV values may result in phoria being decompensated. This is considerable, especially as such a compromise can occur with severe phoria and lower levels of fusional vergences. This study provides epidemiological data on heterophoria and fusional vergence for high school children. The findings reported can guide clinicians and researchers on the expected magnitudes of heterophorias and fusional vergences where phoria is likely to become decompensated under increased near tasks.

The strengths of this study include the random sampling employed, the use of many schools with a relatively large sample size and the collection of data by only one examiner, suggesting that the findings from this study may be considered representative and could be extrapolated to the entire population of black African schoolchildren in the municipality. However, there are some limitations that may influence the interpretation of the findings. For instance, although the von Graefe technique is a commonly used phoria measurement technique, a limitation is that it may give relatively higher exophoria values.⁴⁹ The use of this technique to assess heterophoria and fusional vergence would have allowed for consistency in technique. The intention in the present study was to apply techniques as close as possible to those used in the convergence insufficiency projects⁵⁰ that formed the basis of the present study. However, other authors^51,52 used the von Graefe technique (for phoria) and the trial frame and prism bar (for fusional reserves), as in this study, and this allowed for comparison. The prism bar on a trial frame is a preferred tool in the screening setting as it is quicker, allows the view of eye movement and makes it easier for school-aged children to follow instructions.⁴³ Furthermore, even when the Risley prism is used to measure fusional vergences, the prism power must still be set to zero while the examiner notes the appropriate compensating prism direction in mind, before the vergence measurements are commenced. That approach may be compared to starting the fusional measurement with a zero diopter prism. Another concern is the difference in the field of view between phoropter and prism bar, but it is unknown whether that could greatly influence findings. It might also be argued that other tests, such as vergence facility testing, negative relative accommodation, binocular accommodative facility with plus lenses and near point of convergence tests that also measure vergence functions, apply different testing techniques yet give gross information of the vergence functions.