A Quasi-Experimental Trial of the NutriAide App for Anaemia in Pregnancy

Background

Anaemia in pregnancy is a major problem in India — NFHS-5 puts the prevalence at 52.2% — and adherence to routine iron–folic acid (IFA) supplementation is often poor because of side effects. Mobile-health (mHealth) tools that support dietary awareness and supplement reminders have been proposed as a way to improve behaviour and outcomes. This study tested whether the NutriAide app (developed by the National Institute of Nutrition), used for daily dietary self-logging, adds measurable benefit over standard antenatal care in anaemic pregnant women.

Approach

Conducted as an ICMR Short-Term Studentship project (ref. STS2024-3Y-03247), this was a quasi-experimental trial in the antenatal clinic of a tertiary-care Obstetrics & Gynaecology department over two months. 59 pregnant women with mild-to-moderate anaemia (haemoglobin 8–10 g/dL) in the second or early third trimester, who owned smartphones, were allocated to an intervention group (n = 30) — NutriAide app for daily food logging plus WhatsApp reminders, on top of routine IFA and a diet chart — or a control group (n = 29) receiving IFA and the diet chart alone. Haemoglobin and serum iron, ferritin, folate and B12 were measured at baseline and after the intervention. I led the data analysis in R 4.3.2 (paired/Welch comparisons, group summaries).

What we found

• Haemoglobin improved in both groups over two months — but the app group did not do better than standard care.
• Mean haemoglobin rise was 1.26 ± 0.95 g/dL (intervention) vs 1.92 ± 1.42 g/dL (control) — no significant difference (p = 0.10); end-of-study haemoglobin was actually higher in the control group (11.49 vs 10.54 g/dL, p = 0.022).
• App adherence was the key limitation: only 60% (18/30) installed NutriAide, and of those just 66.7% (12) ever logged a meal.
• Conclusion: app-based dietary self-monitoring did not outperform routine supplementation and counselling within this short, small study.

Outputs & impact

This completed studentship analysis is a candid null result: the NutriAide app showed no significant advantage over standard antenatal anaemia care, with poor engagement, a short two-month follow-up, small sample, and unmeasured confounders (e.g., parenteral iron) limiting any effect. It offers a useful design lesson for future mHealth nutrition trials — larger, multi-centre samples, longer follow-up, and real-time adherence tracking — alongside a reproducible R analysis pipeline. The manuscript and dissemination are in preparation.