The role of clinical diagnosis criteria on the frequency of accommodative insufficiency

Abstract

Aim: To estimate and compare the frequency of accommodative insufficiency (AI) within the same clinical population sample depending on the type of clinical criteria used for diagnosis. Comparing the frequency within the same population would help to minimize bias due to sampling or methodological variability.

Methods: Retrospective study of 205 medical records of symptomatic subjects free of any organic cause and symptoms persisting despite optical compensation evaluated. Based on the most commonly clinical diagnostics criteria found in the literature, four diagnostics criteria were established for AI (I, II, III and IV) based on subjective accommodative tests: monocular accommodative amplitude two or more diopters below Hofstetter's minimum value [15-(0.25×age)] (I, II, III, IV); failing monocular accommodative facility with minus lens, establishing the cut-off in 0 cycles per minute (cpm) (I) and in 6 cpm (II, III); failing binocular accommodative facility with minus lens, establishing the cut-off in 0 cpm (I) and in 3 cpm (II).

Results: The proportion of AI (95%CI) for criteria I, II, III and IV were 1.95% (0.04%-3.86%), 2.93% (0.31%-4.57%), 6.34% (1.90%-7.85%) and 41.95% (35.14%-48.76%) respectively, with a statistically significant difference shown between these values (χ² =226.7, P<0.001). A pairwise multiple comparison revealed that the proportion of AI detected for criterion IV was significantly greater than the proportion for the rest of the criteria (P-adjusted<0.05 in all cases).

Conclusion: The prevalence of cases of AI within the same clinical population varies with the clinical diagnostic criteria selected. The variation is statistically significant when considering the monocular accommodative amplitude as the only clinical diagnostic sign.

Keywords: accommodative facility; amplitude of accommodation; epidemiology.

Figure 1. Results of the algorithms applied using the AA, MAF and BAF variables for each clinical diagnostics criteria.

Figure 2. Comparison of the prevalence of AI for each clinical diagnostic criterion

Full lines represent pairs of criteria in which no statistically significant differences were observed between the prevalence of these criteria. Dotted lines represent pairs of criteria with significant differences. The numbers on the segment indicates the P-adjusted value for each pairs of criteria with significant differences (P-adjusted<0.05). The number in each node represents the rate of subjects diagnosed with AI from total sample. A: Pairwise comparison criteria I-II, I-IV, I-III, II-III, II-IV, III-IV. Statistically significant differences were found for the criterion IV. B: Pairwise comparison criteria I-II, I-III, II-III, to eliminate the comparative analysis with criterion IV. Statistically significant differences were found for criterion III.