Iron prophylaxis in infants: maternal perceptions, continuation, and barriers. A cross-sectional study in an urban Turkish tertiary-care setting

Iron deficiency anaemia in infancy is a common and preventable cause of impaired neurodevelopment. Türkiye provides free iron drops to all term infants through a national programme, yet many mothers stop supplementation before the recommended endpoint. We described maternal perceptions of, continuation of, and barriers to infant iron prophylaxis, and examined factors associated with stopping prophylaxis and with considering it necessary. In this single-centre cross-sectional study, 310 mothers of infants aged 4–12 months attending the paediatric outpatient clinics of a tertiary training and research hospital were interviewed with a structured questionnaire. Each candidate factor was examined using univariate logistic regression, and multivariable logistic regression models were then fitted for both outcomes using complete cases (two items were unanswered). To account for age-dependent exposure time, we performed a time-to-event analysis in which the event was early discontinuation (stopping before the recommended 12-month endpoint), with infants who continued to 12 months censored at completion; a model counting any stop as an event was run as a sensitivity analysis. The mean infant age was 9.44 ± 2.28 months. Most mothers reported at least one perceived benefit of prophylaxis (287/310, 92.6%; growth and development was the most frequently named, cited by 76.7% of these 287 respondents), yet 187 (60.3%) had stopped it. Mothers could report more than one reason for stopping; the commonest were forgetfulness (38.5%) and a doctor’s advice to stop (35.3%), while an adverse effect was reported by 31.6% and was the sole reported reason for 26.2%. In the multivariable model, only older infant age was independently associated with stopping (adjusted OR [aOR] 1.33 per month, 95% CI 1.18–1.49); low income was not independent (aOR 2.68, 95% CI 0.83–8.68). In the time-to-event analysis the median time to early discontinuation was approximately 9 months, though imprecise (few infants remained at risk beyond 8 months). In the primary Cox model the only independent correlate of a lower hazard was maternal employment (HR 0.56, 95% CI 0.34–0.94), but this was not robust to the event definition (HR 0.75, p = 0.13 when any stop was counted); in an exploratory model, breastfeeding for ≥ 6 months was also associated with a lower hazard (HR 0.48, 95% CI 0.30–0.76), although breastfeeding was ascertained at interview and is therefore hypothesis-generating. Age at initiation was not independently associated with the hazard, and the apparent association between older infant age and stopping in the cross-sectional model reflected the longer exposure time of older infants rather than an independent effect. Maternal employment was also independently associated with considering prophylaxis necessary (aOR 2.82, 95% CI 1.20–6.63). Most mothers recognised the benefits of iron prophylaxis but did not continue it; mothers attributed roughly one third of discontinuations to a clinician’s advice. After adjustment, neither low income nor maternal education was an independent correlate of stopping, and the association with infant age reflected longer exposure time rather than a behavioural effect. As the data are from a single urban centre and were collected in 2021, the findings are hypothesis-generating.