Menu
Menu
Menu
Menu
Background Despite growing evidence of pregnancy preparation benefits, there is little knowledge on how women in developing countries prepare for pregnancy and factors influencing their preparedness for pregnancy. Here, we determine how women in Malawi prepare for pregnancy and factors associated with pregnancy preparation. Methods We used data from a previous cohort study comprising 4,244 pregnant mothers, recruited between March and December 2013 in Mchinji district, Malawi. Associations of pregnancy preparation with socio-demographic and obstetric factors were tested for using mixed effects ordinal regression, with the likelihood ratio and Wald’s tests used for variable selection and independently testing the associations. Results Most mothers (63.9%) did not take any action to prepare for their pregnancies. For those who did (36.1%), eating more healthily (71.9%) and saving money (42.8%) were the most common forms of preparation. Mothers who were married (adjusted odds-ratio (AOR 7.77 (95% CI [5.31, 11.25]) or with no or fewer living children were more likely to prepare for pregnancy (AOR 4.71, 95% CI [2.89,7.61]. Mothers with a period of two to three years (AOR 2.51, 95% CI [1.47, 4.22]) or at least three years (AOR 3.67, 95%CI [2.18, 6.23]) between pregnancies were more likely to prepare for pregnancy than women with first pregnancy or shorter intervals. On the other hand, teenage and older (≥ 35 years old) mothers were less likely to prepare for pregnancy (AOR 0.61, 95%CI [0.47, 0.80]) and AOR 0.49 95%CI [0.33, 0.73], respectively). Conclusion While preconception care may not be formally available in Malawi, our study has revealed that over a third of mothers took some action to prepare for pregnancy before conception. Although this leaves around two thirds of women who did not make any form of pregnancy preparation, our findings form a basis for future research and development of a preconception care package that suits the Malawian context.
We used data from a previous prospective cohort study of pregnant mothers in Mchinji, a rural district in central Malawi. The initial study aimed to explore relationships between pregnancy intention and key maternal and neonatal outcomes [16]. Full details of the initial cohort study design and setting are published elsewhere [16]. Briefly, pregnant women were recruited over a period of nine months between March and December 2013 from 25 randomly selected area blocks (out of 49 area blocks of approximately equal population size) of Mchinji district. The twenty-five blocks, which covered about half of the district, were randomly selected and grouped into three zones based on location. Pregnant mothers were identified through key informants, who had village registers and enumerated every household and its members for an ongoing district-wide pneumococcal vaccine surveillance programme. All pregnant mothers from the selected 25 areas were eligible to participate in the study if they were aged 15 years or older and provided informed consent. The degree of pregnancy intention was measured using the validated Chichewa (Malawi’s local language) version of the London Measure of Unplanned Pregnancy (LMUP) [20,21]. The LMUP is a psychometrically validated measure of the degree of intention of a current or recent pregnancy, consisting of six questions covering contraception use, timing of pregnancy, intention, desire for a baby, discussion with a partner and pre-conception preparation. Responses to each question are scored as zero, one or two. The overall degree of pregnancy intention is measured on a scale of zero to 12 in order of increasing degree of pregnancy intention [22]. Participating mothers were asked all six questions on the LMUP and a further set of demographic and obstetric history questions during pregnancy. The focus of the present study was to carry out a detailed analysis of the participants’ responses to question six of the LMUP, which asks about the mothers’ preconception actions in preparation for their pregnancy, in relation to their demographic and obstetric characteristics (http://www.lmup.com) [20]. The dependent variable was mothers’ preparation for pregnancy. It was measured by the participants’ responses to question six of LMUP. Responses to LMUP question six were summarised into three categories: “No preparation” (if participants did not do any of the actions), “some preparation” (if the participant did any one action) and “prepared” (if they took any two or more actions). The independent variables considered were socio-demographic and obstetric characteristics and previous history of depression (Table 1). In this study, possible episodes of depression before pregnancy were screened by asking pregnant women whether (a) they felt down, depressed or hopeless (low mood) or (b) if they had felt no interest or having little pleasure in doing things (anhedonia) in the year before pregnancy [16]. Affirmative responses to at least one of the questions were put in three categories including: (1) yes to either one or both questions, but episodes only lasted for less than two weeks; (2) yes to either question with episodes lasting for more than two weeks; and (3) yes to both questions and episodes lasting for over two weeks (Table 1). Mothers’ socio-economic status was determined by an asset-based approach whereby data was collected on variables that reflected the mothers’ living standards, including characteristics of their houses, access to utilities and durable assets, such as bicycle or radio owned by their households. These variables were then converted into a single variable of socio-economic status by principal component analysis, which was then divided to group women into the socio-economic quintiles”. We performed exploratory and descriptive analyses to identify frequencies of respondents, preparedness categories, variable correlations and other background characteristics in relation to the outcome of pregnancy preparation. As the dependent variable was ordinal, we fitted an ordinal regression model for univariate analysis of the association between the dependent variable and each of the independent variables. Likelihood ratio and Wald’s tests were then used to identify independent variables to adjust for and test for their association with pregnancy preparation. We selected all variables that were significant at a 20% significance level for inclusion in the multivariable ordinal model. The final multivariable model included the following socio-demographic and obstetric factors: mother’s age at last birthday, marital status, mother’s education, wealth status, distance to closest health facility, time interval between pregnancies number of live children and history of depression prior to the pregnancy. Both the univariate and multivariable ordinal regressions were run as mixed effects models with geographical cluster included as a random effect. We reported crude and adjusted odds ratios, with their 95% confidence intervals (CI) as measure of uncertainty. Ethical approval, including the approach to include pregnant women aged 15 and over, was provided by the University College London Research Ethics Committee and the College of Medicine Research Ethics Committee at the University of Malawi (approval numbers 3974/001 and P.03/12/1273 respectively).
N/A
