Risk factors for xerophthalmia in the Republic of Kiribati.

OBJECTIVES: To identify risk factors for xerophthalmia in the Republic of Kiribati.

DESIGN: Case-control study.

SETTING: The Republic of Kiribati.

SUBJECTS: 666 xerophthalmic preschool children (cases) and 816 children without xerophthalmia (controls) from a population-based sample of 4619 children who participated in a xerophthalmia prevalence survey.

MAIN OUTCOME MEASURES: Clinical signs of xerophthalmia (night blindness, Bitot's spots, corneal xerophthalmia and keratomalacia).

RESULTS: Multivariable logistic regression models showed older age [1.35(1.24, 1.47)] [odds ratio (95% confidence interval)]; male sex [1.32(1.05, 1.67)]; recent diarrhea [1.45(1.10, 1.89)]; severe [3.82(2.73, 5.35)], moderate [3.55(2.04, 6.18)], and mild [3.07(2.33, 4.04)] protein-energy malnutrition; current breast-feeding [0.30(0.19, 0.46)]; higher frequency of consumption of carotenoid-containing fruits and vegetables [0.93(0.80, 0.96)]; and the presence of a Foundation for the Peoples of the South Pacific garden project [0.70(0.52, 0.93)] were each independently associated with xerophthalmia. A recent history of measles was associated with corneal xerophthalmia [7.73(1.78, 33.65)].

CONCLUSIONS: These data provide further evidence of the relationship between xerophthalmia and factors that may be amenable to intervention, and suggest that greater availability and consumption of provitamin A carotenoids is associated with decreased risk of xerophthalmia among preschool children.