Factors associated with anaemia in pregnancy: A retrospective cross-sectional study in the Bolgatanga Municipality, northern Ghana

**Background:** Anaemia in pregnancy (AIP) is a severe public health problem associated with adverse maternal and perinatal outcomes including intrauterine growth restriction, preterm birth, stillbirth, postpartum haemorrhage, and maternal mortality. In Ghana, AIP prevalence was reported as 45% in national surveys (GDHS 2014, GMS 2017), with a prior study in the Bolgatanga Municipality reporting 50.4%. While ANC provides iron folate supplementation, infection prevention, and haemoglobin monitoring at three mandatory time points (registration, 28 weeks, 36 weeks), few studies have assessed haemoglobin levels and anaemia status at all three recommended points in the same cohort.

**Methods:** A retrospective cross-sectional study was conducted in 28 health facilities providing ANC in the Bolgatanga Municipality, northern Ghana. Using ANC registers as the sampling frame, 372 pregnant women between 36 and 40 weeks of gestation were randomly sampled (from an initial target of 411; 91.2% response rate). Data were collected via clinical records review and a structured questionnaire between October and November 2020. Haemoglobin levels were extracted from maternal and child health record books and laboratory slips. AIP was defined using Ghana's national cut-off of 11.0 g/dL, and the WHO-recommended adjustment of 10.5 g/dL was applied for women in the second trimester to account for haemodilution. Binary logistic regression was used to identify factors associated with AIP at each of the three stages, with results presented as adjusted odds ratios (AOR) with 95% confidence intervals.

**Key Results:** The mean age of participants was 27 years; 47.0% were aged 26–35 years, 84.1% were Christian, 94.1% were married, 44.9% had basic education only, and 44.3% were unemployed. At ANC registration, AIP prevalence was 35.8% (95% CI: 30.9–40.9) using the 11.0 g/dL cut-off and 25.3% (95% CI: 20.9–30.0) after haemodilution adjustment. At 28 weeks, prevalence was 53.1% (95% CI: 45.8–60.3) using 11.0 g/dL and 37.5% (95% CI: 30.6–44.8) after adjustment. At 36 weeks, prevalence was 44.8% (95% CI: 39.2–50.4). At booking, late registration (after first trimester) was associated with increased odds of AIP (AOR = 1.87, 95% CI: 1.17–2.98, p = 0.009), as was registering at a regional hospital (AOR = 2.25, 95% CI: 1.02–4.97, p = 0.044), while registering at a private hospital was protective (AOR = 0.32, 95% CI: 0.11–0.92, p = 0.035). At 28 weeks, protective factors included age 26–35 years (AOR = 0.46, 95% CI: 0.21–0.98, p = 0.044), Christianity (AOR = 0.34, 95% CI: 0.31–0.89, p = 0.028), highest wealth quintile (AOR = 0.27, 95% CI: 0.09–0.83, p = 0.022), and tertiary education (AOR = 0.09, 95% CI: 0.02–0.54, p = 0.009). At 36 weeks, late booking remained a risk factor (AOR = 1.72, 95% CI: 1.05–2.84, p = 0.033), while high wealth (AOR = 0.44, 95% CI: 0.20–0.99, p = 0.049), age 26–35 years (AOR = 0.38, 95% CI: 0.21–0.68, p = 0.001), age 36–49 years (AOR = 0.35, 95% CI: 0.13–0.90, p = 0.024), and secondary education of spouse (AOR = 0.35, 95% CI: 0.14–0.88, p = 0.026) were protective.

**Clinical Implications:** The study demonstrates that AIP prevalence worsens across pregnancy when haemodilution is accounted for, contrary to the expectation that ANC interventions would reduce anaemia over time. This suggests current ANC-based anaemia control strategies may be insufficient. The authors recommend that Ghana adopt the WHO-recommended haemoglobin cut-off of 10.5 g/dL for the second trimester to avoid overdiagnosis due to physiological haemodilution. Interventions should prioritise early ANC registration (first trimester), improved household wealth, and sustained female education. The differing correlates at each gestational stage suggest that anaemia prevention and treatment should be tailored to individual risk profiles. Limitations include the use of different haemoglobin measurement devices across facilities, incomplete data (only 192 of 372 women had haemoglobin recorded at 28 weeks), and the retrospective design which precluded assessment of variables such as gestational weight and adherence to iron supplementation.