New Pattern in Esophoria Incidence Identified by Linear Splines

OBJECTIVES: Esophoria describes an imbalance in the ocular muscles most common in children that creates a tendency for one or both eyes to turn medially, leading to diplopia. More recently, esophoria has been linked to excessive screen-time in older children and adults. Recent studies have reported a consistent, population-wide increase in screen-time, particularly in younger age groups. We aimed to investigate whether there was a corresponding increase in incidence of esophoria over this same period.

METHODS: Annual incidence of esophoria in the UK was calculated from the Clinical Practice Research Datalink (CPRD) via the Livingstone online healthcare analytics platform. Patients diagnosed with esophoria reported between 2004-2019 were included and compared to the total research-quality GP-registered patients. To examine patterns in incidence rates, three regression models were compared: linear, exponential, and linear spline. The latter assesses two linear regressions connected by a ‘knot’, reflecting a change in increase in incidence. Model fits were compared via adjusted R2.

RESULTS: 52,634 cases were identified. Crude, annual incidence rates increased from 1.9 cases per 10,000 population (p10k) in 2004 to 2.4 in 2019; a 29.8% increase. Stratified by age, the biggest rises in incidence were seen in 0-4 (from 20.4 to 24.0 p10k) and 5-14 (from 6.0 to 7.4 p10k) age groups. Incidence rate was consistently best modelled by linear splines (average adjusted R2=0.65). Assessment of the knots indicated a weak, non-significant relationship between age and the time of inflection (R2=0.13; p=.30).

CONCLUSIONS: We observe an annual incidence of esophoria in younger people during a period in which screen-time is known to have increased. This, however, does not necessarily imply causality. The observed increase may be due to other factors or to a change in reporting and recording of diagnoses. These findings suggest a potential value for spline and knot analyses to elucidate novel epidemiological patterns.