Menu
Menu
Menu
Menu
Introduction: Globally, postnatal care (PNC) is fraught with challenges. Despite high PNC coverages in Ghana’s Greater Accra Region (GAR), maternal and newborn health outcomes are of great concern. In 2017, neonatal and post-neonatal mortality rates in GAR were 19 and 13 per 1000 live births respectively despite PNC coverages of 93% for at least one PNC and 87.5% for PNC within 48 hours post-delivery. Telephone follow-up has been used to improve health outcomes in some settings, however, its usefulness in improving maternal and infant health during the postnatal period is not well known in Ghana. We assessed effectiveness of telephone-based PNC on infant and maternal illness in selected hospitals in GAR. Methods: An open-label, assessor-blinded, parallel-group, two-arm superiority randomized controlled trial with 1:1 allocation ratio was conducted from September 2020 to March 2021. Mother-baby pairs in intervention arm, in addition to usual PNC, received midwife-led telephone counselling within 48 hours post-discharge plus telephone access to midwife during postnatal period. In control arm, only usual PNC was provided. Descriptive and inferential data analyses were conducted to generate frequencies, relative frequencies, risk ratios and 95% confidence intervals. Primary analysis was by intention-to-treat (ITT), complemented by per-protocol (PP) analysis. Results: Of 608 mother-baby pairs assessed for eligibility, 400 (65.8%) were enrolled. During 3 months follow-up, proportion of infants who fell ill was 62.5% in intervention arm and 77.5% in control arm (p = 0.001). Maternal illness occurred in 27.5% of intervention and 38.5% of control participants (p = 0.02). Risk of infant illness was 20% less in intervention than control arm in both ITT analysis [RR = 0.8 (95%CI = 0.71–0.92] and PP analysis [RR = 0.8 (95%CI = 0.67–0.89)]. Compared to controls, risk of maternal illness in intervention arm was 30% lower in both ITT [RR = 0.7 (95%CI = 0.54–95.00)] and PP analysis [RR = 0.7 (95%CI = 0.51–0.94)]. Conclusion: Telephone-based PNC significantly reduced risk of maternal and infant illness within first 3 months after delivery. This intervention merits consideration as a tool for adoption and scale up to improve infant and maternal health. Trial registration: This trial was retrospectively registered with the International Standard Randomized Controlled Trial Number (ISRCTN) Registry with number ISRCTN46905855 on 09/04/2021.
The study was a two-arm, parallel-group, open-label randomized controlled trial conducted across two hospitals, with blinding of the outcome assessors to compare the effectiveness of telephone-based PNC plus usual hospital-based PNC with usual hospital-based PNC alone on infant and maternal illness. The study was designed to answer the question: Among infants and their mothers, how does complementary telephone-based health education initiated within 48 hours after discharge plus telephone access of mother to a dedicated health worker throughout the postnatal period compared with usual PNC alone affect infant and maternal health within the first 3 months after delivery? The trial was conducted at the Greater Accra Regional and Tema General Hospitals in the Greater Accra Region located in the southeastern part of Ghana (Fig. 1). The Greater Accra Regional Hospital and the Tema General Hospital are two largest hospitals and referral centres in the Accra and Tema Metropolitan Areas. These two metropolises were chosen because they are the largest and most cosmopolitan of the 29 Metropolitan, Municipal and District Assemblies (MMDAs) in the Greater Accra Region. The Greater Accra regional capital city, Accra, serves as both the national and regional capital and is about 30 km from Tema, the capital city of Tema Metropolitan Area. Map of Greater Accra Region showing Accra and Tema Metropolitan Areas and the study sites The Greater Accra Regional Hospital has a 620-bed capacity and provides services to over 1000 outpatients and inpatients on daily basis. A total of over 150 midwives provide maternal and child health services at the hospital. The hospital had an average monthly delivery of approximately 600 babies in 2019. The Tema General Hospital is a 399-bed capacity hospital which accommodates approximately 500 patients daily. It has a midwifery staff strength of 90 and had an average of approximately 460 deliveries per month in 2019. Postnatal care at both health facilities is conducted according to national guidelines adapted from WHO recommendations for postnatal care. Mothers who delivered via spontaneous vaginal delivery were typically discharged after 24 hours in the absence of complications while mothers who delivered by caesarian section were discharge averagely between three to 5 days after surgery. Both hospitals have postnatal clinics which provide PNC to mothers and their babies at 1–2 weeks and 6 weeks. The study included mother-singleton baby pairs delivered in the Greater Accra Regional and Tema General hospitals during the study period. Mother-baby pairs were eligible for inclusion if mother was at least 18 years of age and have access to a functional mobile phone, discharged from hospital but not have left the hospital at time of recruitment, not involved in another study and had no intention of moving out of the Greater Accra Region within 3 months after delivery. Exclusion criteria were babies born preterm, babies with major congenital anomaly or admitted to the neonatal intensive care unit (NICU) after delivery for severe illness and babies with either or both parents being a health worker. Based on prior data from an initial pilot study conducted at the two hospitals that showed that the proportion of infant illness among mother-baby pairs in the usual PNC alone arm was 25% and our hypothesis that the risk of childhood illness in the telephone-based PNC relative to the usual PNC alone is reduced by 50%, with a two-sided type I error rate of 5, 80% power and 1:1 allocation ratio, a sample size of 186 was calculated for each arm given an anticipated loss to follow-up rate of 27% observed in a neonatal follow-up study [22]. A total of 200 mother-baby pairs per study arm were studied. The sample size was proportionately allocated to the hospitals based on the number of deliveries per month. The intervention tested was telephone-based PNC delivered to mother-baby pairs in the intervention arm in two parts: This intervention was given to mother-baby pairs in addition to the usual PNC services provided by the health facility. Content of telephone call to the mother was based on a client education protocol validated and used in Ecuador [18] and adapted to reflect the National Safe Motherhood Service Protocol as well as WHO and UNICEF guidance on care of the newborn [8, 23] (Table 1). The telephone-based education focused on educating mothers on key maternal and newborn care issues. Mother-baby pairs in the control arm received only the usual health facility-based PNC services. Content of postnatal education session Adapted from Maslowsky et al., 2018 The primary outcome of the study was proportion of infants who fell ill within the first 3 months after delivery. Proportion of mothers who fell ill within the first 3 months after delivery was co-primary outcome. Secondary outcomes of proportion of mothers practicing exclusive breastfeeding, using contraceptives and those who had postpartum depression were also measured but not reported in this paper. The outcome measures were assessed verbally through interview of the mothers at 3 months after delivery. Infant illness was defined as any acute health problem of the child since birth requiring a visit to a clinic, chemical shop, pharmacy or for which medical help was sought for the child. Maternal illness was any acute health problem of the mother since discharge from the hospital requiring a visit to a clinic, chemical shop, pharmacy or for which mother sought medical help. Each mother-baby pair was randomized into either of two study arms using block randomization technique. The randomization scheme was made up of computer-generated random numbers of 100 blocks and 4 participants per block by using web-based randomization software [24]. The random allocation sequence generated was printed and concealed by masking tape from the research team enrolling the participants with access restricted only to independent assistants at the two hospitals who were not directly involved in the data collection. The sequence code was only revealed serially by these independent assistants by peeling off the masking tape on a case-by-case basis after a participant consented and was enrolled in the study. The study was conducted as an open-label trial. Owing to the nature of the study, it was not possible to blind study participants (mothers) and investigators to the intervention received. However, research assistants who assessed the outcomes at the end of the follow-up period were not aware of the study groups to which the participants were allocated, and the intervention given to each group. The study participants were enrolled soon after delivery or within few days after delivery whilst still in the hospital. The mother-baby pairs were provided with basic information on the study design and objectives for their interest. Mothers willing to be part of the study were screened for eligibility using the inclusion and exclusion criteria. Eligible participants were taken through informed written consent process. Mothers who consented, were enrolled, interviewed to obtain data on their sociodemographic characteristics, obstetric history and household assets. The allocation of the mother-baby pair to either arm of the study was done after consent, enrollment and obtaining all the baseline information. The independent assistant at each hospital who had access to the concealed allocation sequence peeled of the masking tape sequentially to reveal the next group assignment for each participant enrolled. Participants in the intervention arm were asked to provide information on what time of the day they preferred to be called. Mother-baby pairs in both arms were followed up through the hospitals’ routine follow-up. In both groups, mother-baby pairs received routine PNC and education provided by health facilities without interference from the research team. Outcomes of maternal and child illness were assessed at 3 months. The mother-baby pairs were followed up while at home at 3 months and interviewed face-to-face on baby and maternal health using a structured questionnaire. They were called on phone to schedule the follow-up interviews. The interviews were arranged at a time and place convenient for the mothers. In instances where telephone interviews were preferred to the face-to-face interviews, the interviews were arranged at time convenient for the mothers. The interviews were initially planned to be conducted face-to-face but were not possible for all the participants. This was because the study period coincided with the initial peak phase of COVID-19 in the region and there were restrictions on movement and stay at home orders coupled with high level of uncertainty among the populace. Some participants were reluctant to have face-to-face meetings with the research assistants. Outcomes were assessed for mother-baby pairs in the intervention and the control groups at 3 months after delivery. At each contact, the mother was asked if she or her baby had been ill since the last physical contact with the research team, and if so, when the illness started and ended. Other data collected were whether care was sought for illness and from where. The mothers were provided counselling and support when needed during the outcome assessment period. Participants were defined as lost-to-follow-up when they were unable to be reached within 1 week after the 3 months follow-up. The research assistants were trained on the data collection process and the delivery of the intervention. The questionnaires were pretested to ensure clarity and ambiguities resolved. The data were entered directly into the Open Data Kit (ODK) software downloaded onto portable mobile devices to prevent errors. The ODK database was reviewed daily to ensure that the questionnaires were properly filled. In order to ensure that the outcome assessors were not aware of the arm of the study to which participants randomized, different set of research assistants made up of community health nurses were used as outcome assessors. The telephone call conversations during the intervention delivery were audio recorded if permission was granted by the participants and reviewed to ensure the intervention was delivered as planned. Data from the field were directly entered into ODK, exported to Microsoft Excel and then to STATA Version 16.0 [25] for analysis. Data was cleaned by running frequencies of all variables that were used in the analysis. Incorrect data points were resolved by contacting the study participants and appropriate corrections done. Descriptive statistics were used to describe baseline socio-demographic and obstetrics characteristics as well as distribution of infant and maternal illness among participants in the study arms. Wealth index was constructed using the principal component analysis (PCA), as a proxy of the SES of the household of each mother-baby pair using housing characteristics and utility variables. The items included in the wealth index calculation were: household ownership of assets (clock, mosquito net, bed, livestock, blender, table, room divider, fridge, fan, sewing machine, washing machine, computer, air conditioner, mobile phone, radio set, television, bicycle, motor bike, car, land,), characteristics of materials used for housing (material used for roof, floor and wall of house, number of rooms) and access to basic services (electricity supply, source of water, access to toilet facility, type of fuel used). Based on the wealth scores, households were put into five categories of wealth: quintile 1, quintile 2, quintile 3, quintile 4 and quintile 5. In order to cater for the effects of missing data, minimize bias, preserve the sample size and representativeness of the data, multiple imputation (MI) was used to impute missing outcome data for analysis by ITT. The ITT method of analysis was used for all participants based on how they were originally randomized. The imputation model predicting missing outcome values was implemented by the multiple chain equations method with the assumption that missing outcome data were missing at random. The multiple imputed data was used for the ITT analysis. Proportions of mothers and infants who fell ill were compared between the study arms using the Chi-squared-test with its resultant p-values. Risk ratios (RR) were calculated to determine the strength of association between exposures and outcomes. Statistical evidence of associations between the intervention and each outcome was assessed using 95% confidence interval (CI). In order to examine whether there was effect heterogeneity across subgroups, post hoc subgroup analyses were done for the infant illness and maternal illness to explore any differences in magnitude of intervention effect among different categories of participants. The categories considered were highest level of education, antenatal care attendance, parity, and socioeconomic status. With an initial significance level of 0.05 and four post hoc subgroup analyses done, adjustments for multiplicity were made using the Bonferroni correction. The significance level was set at 0.01 for all the subgroup analyses. Subgroup analyses was done using modified Poisson regression model with robust error variances. The interaction p-values were used to determine any significant subgroup effects. To investigate the effect of protocol violation on the results obtained in the primary analysis, PP analysis was done by restricting analysis to only participants who complied with the protocol (were reached on phone within 48 hours after discharge and taken through the postnatal education according to the checklist). The sensitivity analysis was done by comparing the ITT and PP analyses. This trial was retrospectively registered on 09/04/2021 with the International Standard Randomized Controlled Trial Number (ISRCTN) Registry with number ISRCTN46905855.
N/A
