Health indicators of pregnant women in tonkolili district, rural sierra leone

Despite having reported one of the highest maternal mortality ratios and neonatal mortality rates in the world, surprisingly little is known about the general health status of pregnant women in rural parts of Sierra Leone. Malaria, anaemia and malnutrition are known contributors to adverse pregnancy outcomes. Although their prevalence is known to be high, the burden of these conditions in the rural pregnant population remains unknown. Our study aimed to gain more insight into the health status of pregnant women. An observational retrospective descriptive study was conducted at the Lion Heart Medical Centre using antenatal care (ANC) registers. The study revealed high prevalence of malaria (35.2%), maternal undernutrition (10.4%) and anaemia (65.9%). The proportion of teenage pregnancies in the ANC population was 16.4%. Both malaria and anaemia were more prevalent in this group, with odds ratios of 2.1 and 1.7, respectively. The findings reveal alarming high rates of anaemia, acute undernutrition and malaria among pregnant women and high numbers of pregnancy among adolescents, with increased health risks. These results will be used to advocate for a malnutrition program, specifically for pregnant women. Our study further emphasises the importance of preventing malaria and anaemia in pregnant women. The study took place at the LHMC, a 70-bed referral hospital located in Tonkolili district, Gbonkolenken chiefdom. Surrounded by 15 primary healthcare units, the direct catchment area of the hospital consists of approximately 100,000 inhabitants. Pregnant women receive routine ANC from their nearest health centre and are invited to visit the LHMC for additional screening in the 2nd trimester. This service is covered by HBVP, funded by the Dutch Lion Heart Foundation, which aims to increase coverage and quality of ANC services by identifying and following up high risk cases, whereas uncomplicated cases are referred back to the nearest primary healthcare unit (PHU). The screening consists of anthropometric measurements, history taking, physical examination, laboratory investigations and ultrasound imaging. The HBVP includes provision for transport from and back to the PHU, and all services are performed free of charge. A retrospective analysis of the ANC records was conducted with the aim to map baseline characteristics and determine the prevalence of maternal malnutrition, adolescent pregnancies, malaria, HIV, syphilis and anaemia in pregnant woman. Approval for the study was obtained from the Sierra Leone Ethics and Scientific Review Committee. Additionally, the board of the LHMC gave institutional approval. Since the study involved a retrospective analysis of existing records, patient consent was not obtained. The desired sample size was calculated using Cochran formula [17] (Equation (1)). When a confidence interval (Z) of 95% is used to estimate the precision (e), the sample sized is calculated using expected prevalence (p), indicated by findings from the DHS [2]. The analysis showed that a sample size of 380 or more would give statistically significant results for malaria, anaemia and malnutrition, with precision rates (e) of 5%, 5% and 1%, respectively. Subsequently, a desired sample size of 500 first ANC visits was chosen. A time frame of 5 months was selected, and all data between August and December 2018 were collected. Data were extracted from existing ANC records by the corresponding author and transferred into a digital database (Microsoft Excel®, Microsoft, Redmond, WA, USA). The second author checked data entry. Each visit was encoded using a study number, personal data were not included to ensure privacy. The following information was recorded in the database: age, gravidity, parity, height, weight, MUAC, estimated gestational age (EGA) by ultrasound findings. Furthermore, laboratory results were matched with ANC visits and included in the database: haemoglobin (Hb) level, HIV, syphilis and rapid diagnostic test (RDT) result for infection with P. falciparum. The Hb levels in the study population were routinely checked using the HemoCue® 301 (HemoCue AB, Angelholm, Sweden). All women were screened for malaria, irrespective of complaints, with an RDT (CareStart™ Malaria HRP2, Access Bio, Somerset, NJ, USA). During pregnancy, the use of RDTs is preferred over light microscopy due to the possibly diminished peripheral parasite density caused by sequestration of the parasites in the placenta [16]. The SD Bioline HIV/Syphilis duo test (Standard Diagnostics, Gheung-gu, Republic of Korea) was used routinely to detect antibodies against HIV-1/2 and Treponema pallidum. MUAC was routinely measured for all women with an adult MUAC tape. The MUAC is rather insensitive with respect to the physical changes during pregnancy and therefore preferred [18,19]. According to international consensus, undernutrition (body mass index (BMI) < 18.5) in pregnancy correlates with MUAC < 23 cm in adults, obesity (BMI > 30) is reflected in a MUAC reading >30 cm [18]. The women in our study were grouped according to their nutritional status. Since these cut-off values are only validated for adults, adolescents and clients without a recorded age were excluded from further analysis. Only data from first antenatal visits were included in further analysis to prevent double counting and overrepresentation of complicated cases that would come for follow-up visits. For statistical analysis, the data were imported into an SPSS database (IBM® SPSS® statistics 25, Chicago, IL, USA). Prevalence values were determined using the frequencies analysis function for the entire study population as well as relevant subgroups (adolescents versus adults; primipara versus multipara; under-nourished versus well nourished). To allow further comparison between subgroups, the odds ratio (OR) was calculated by comparing the probability of malaria and anaemia (outcome) between subgroups with different characteristics (exposure). The 95% confidence interval (CI) was calculated to estimate the precision of the odds ratio, and when the CI did not overlap with the null value (OR = 1), statistical significance was assumed [20,21,22]. In the occasion of any missing values, the case was excluded for that particular analysis.