Physical activity level during pregnancy in south africa: A facility-based cross-sectional study

Physical activity participation during pregnancy confers many maternal and foetal health benefits to the woman and her infant and is recommended by various health bodies and institutions. However, in South Africa, scant information exists about the physical activity status and its determinants among pregnant women. The aim of this study was to assess the physical activity level and associated factors among pregnant women. This cross-sectional study enrolled 1082 pregnant women attending public health facilities in Buffalo City Municipality, Eastern Cape, South Africa. Information on socio-demographic and maternal characteristics were obtained, and the Pregnancy Physical Activity Questionnaire was used to assess physical activity during pregnancy. Multiple logistic regression analyses were used to assess associations between physical activity and the predictor variables during pregnancy. Adjusted odds ratios with 95% confidence interval were applied to estimate factors associated with physical activity levels. Statistical significance was set at p < 0.05. Only 278 of the women (25.7%) met recommendations for prenatal activity (≥150 min moderate intensity exercise per week). The average time spent in moderate–vigorous physical activity was 151.6 min (95% CI: 147.2–156.0). Most of the women participated in light exercises with a mean of 65.9% (95% CI: 64.8–67.0), and 47.6% (95% CI: 46.3–48.9) participated in household activities. The majority of the women did not receive physical activity advice during prenatal care sessions (64.7%). Factors negatively associated with prenatal physical activity were lower age (<19 years) (adjusted odd ratio (AOR) = 0.3; CI: 0.16–0.76), semi-urban residence (AOR = 0.8; CI: 0.55–1.03), lower educational level (AOR = 0.5; CI: 0.20–0.71), unemployment (AOR = 0.5; CI: 0.29–0.77) and nulliparity (AOR = 0.6; CI: 0.28–1.31). However, prenatal physical activity was positively associated with starting physical activity in the first trimester (AOR = 1.9; CI: 1.06–3.31) compared to other trimesters. The findings of this study demonstrated low levels of physical activity during pregnancy in South Africa. The majority of women did not meet the recommendation of 150 min of moderate intensity activity per week. Light intensity and household activities were the most preferred form of activity. The factors affecting physical activity of women in this present study include lower age, semi-urban setting, low educational level, unemployment and nulliparity. In order to increase activity levels, future work should seek to improve knowledge, access and support for physical activity in pregnant women in South Africa. This should include education and advocacy regarding physical activity for professionals involved in maternal health provision. This was a cross-sectional descriptive study conducted among pregnant women in 12 primary health centres in Buffalo City Municipality, in the Eastern Cape Province, South Africa. Buffalo City Municipality is situated on the East Coast of the Eastern Cape Province. The details of the study setting have been described in a recent publication [41]. Briefly, the municipality is economically, one of the poorest provinces among the nine provinces in South Africa. The Buffalo City Metropolitan Municipality accounts for a total population of 884,000, or 12.2% of the total population in the Eastern Cape Province. In total, 460,000 (51.99%) of the total population are females and 424,000 (48.01%) males [42]. Buffalo City Municipality has two provincial Hospitals (Frere and Cecilia Makhiwane hospitals), and two district hospitals (Bhisho and Grey hospitals). There are five community health centres, 72 primary health clinics, and 12 mobile health services [42]. In addition, all the community health centres and primary health clinics offer antenatal healthcare services freely to all pregnant women regardless of their geographical residence, ethnic, and socio-economic background. The community and primary health facilities deliver antenatal care services every working day. Personal communication with a health facility manager revealed that, on average, the clinics register 5–6 new pregnant women who visit the primary health centres per day. Therefore, annually, an estimated 17,000 pregnant women visit the 12 selected primary health clinics for antenatal services. We applied the Sarmah et al. [43] formulae for an infinite population to calculate the sample size at a confidence level of 95%, with the precision level of ±3%, and at a prevalence of physical activity or exercise during pregnancy of 50% (p = 0.5) as follows: p = 0.5 and hence q = 1 − 0.5 = 0.5; e = 0.03; z = 1.96 So, n0 = (1.96)2 (0.5)(0.5)(0.3)2=1067=1067. However, adding 10% non-response, the final sample size was 1215 women, to account for possible attrition and to protect against possible data loss. We applied a two-stage sampling technique to select pregnant women, regardless of the gestation period. Firstly, using a simple random procedure, 12 antenatal primary health centres were selected to participate in the study and, secondly, participants who meet the inclusion criteria were conveniently selected because of cost and easy accessibility, since the study was conducted at the health facilities. Pregnant women were included in the study if 18 years or older, receiving antenatal care, having a single pregnancy (not multiple ones), and could read or understand the IsiXhosa, Afrikaans or English languages. Women with disabilities or reasons to cease exercise at the time of recruitment, such as “persistent excessive shortness of breath that does not resolve on rest, severe chest pain, regular and painful uterine contractions, vaginal bleeding, persistent loss of fluid from the vagina indicating rupture of the membranes, and persistent dizziness or faintness that does not resolve on rest” [44], were excluded. Detailed information about the recruitment of the participants is shown in Figure 1. Flow diagram of sample selection and participation. The University of Fort Hare Health Research Ethics Committee approved the study protocol (Ref#2019 = 06 = 009 = OkaforUB). In addition, permission was obtained from the Eastern Cape Department of Health and all the selected health facilities. Informed consent was obtained from the pregnant women prior to data collection. Data collection was conducted between July to October 2019. To ensure the required sample size, data collection was carried out at each antenatal health clinic on pre-specified days, in a designated room allocated to the primary researcher by the health facility manager. All eligible pregnant women attending their antenatal care visits at selected health facilities during the study period were randomly approached to participate in the study, after signing an informed consent form. The Pregnant Physical Activity Questionnaire (PPAQ) [45] was used to assess the level, type and intensity of prenatal activity. The primary outcome measure was active and inactive participation of pregnant women in physical activity during pregnancy. The PPAQ is a validated and reliable tool, widely used across countries to measure prenatal physical activity [45,46,47]. The PPAQ is comprised of 32 physical activities, which include household and care-giving (13 activities), occupational (five activities), sports and exercise (eight activities), transportation (three activities), and inactivity (three activities). We solicited participants’ participation on these different activities, and the type, intensity, duration and frequency of physical activity recorded as hours and minutes per day. To maximise the accuracy and ensure completeness of information, the PPAQ was interviewer-administered to the participants at 12 selected primary health centres during their antenatal visit and took approximately 20 to 25 min. The Metabolic Equivalent Task (MET) of each activity was categorised as sedentary (<1.5 METs), low or light (1.5– ≤3.0 METs), moderate (3.0–6.0 METs), and vigorous intensity (>6.0 METs) [45]. We developed a structured questionnaire to solicit information on socio-demographic, obstetrics and lifestyle behaviours of the participants. As with the PPAQ, this aspect of the questionnaire was interviewer-administered to obtain information on age, residence, ethnicity, marital status, level of education, employment status, religion, family support, and behavioural and lifestyle characteristics, which include, current exposure to alcohol and smoking. We categorised women as ‘smokers’, if they reported smoking any number of cigarettes during pregnancy, ‘non-smokers’ (reported not smoking), ‘drinkers’, as those who reported any use of alcohol during pregnancy, and ‘non-drinkers’ (reported no-drinking). Other information included whether participants had had antepartum haemorrhage in their first trimester, perceived health condition in pregnancy (women were asked how they perceive their general health: very good, good, or bad), whether participants received prenatal physical activity advice from health professionals, and had engaged in physical activity before and during pregnancy. We obtained information on parity, mode of pregnancy delivery, and pregravid body mass index from the antenatal records of the participants. We adopted the Institute of Medicine (IOM) recommended BMI cut-off values to classify underweight (<18.5 kg/m2), normal weight (18.5–24.9 kg/m2), overweight (25.0–29.9 kg/m2), and obese (>30.0 kg/m2) [48]. Descriptive statistics, including mean and standard deviation (SD), median and inter-quartile range (IQR) or as proportions was applied. The Centers for Disease Control and Prevention (CDC) recommendations was used to classify women as ‘inactive’ (reporting 0–149 min of exercise per week), and ‘active’ (reported 150 min or more of physical activity) based on the combined moderate–vigorous minutes per week [49]. We applied bivariate and multivariate analyses to assess the factors affecting physical activity behaviour during pregnancy. The Chi-square was used to determine the associations between the physical activity levels and socio-demographic, lifestyle and obstetric characteristics. All the covariates associated with physical activity (that is, age, area of residence, marital status, educational level, employment status, parity, family support, mode of pregnancy delivery, antepartum haemorrhage, pre-pregnancy BMI, employment status, lifestyle behaviours, and physical activity before and during pregnancy) were included in the models. The odds ratio (OR) and corresponding confidence interval (CI) of 95% were calculated. A multiple logistic regression, using automatic variable selection procedure was applied to determine the factors that predict physical activity levels. Automatic variable selection procedures are statistical tools for choosing the best subset of predictor variables for a given response variable. The significance level was set at p = 0.05. The Statistical Package for Social Sciences (SPSS) (Version 24.0, IBM SPSS, Chicago, IL, USA) was used to perform all statistical analyses.