Effectiveness of pregnant women’s active participation in their antenatal care for the control of malaria and anaemia in pregnancy in Ghana: a cluster randomized controlled trial

Malaria and anemia [hemoglobin (Hb) < 110 g/l] during pregnancy in sub-Saharan Africa remain a concern despite proven efficacious antenatal care (ANC) interventions. Recommended ANC interventions include Intermittent preventive treatment of malaria during pregnancy (IPTp) using Sulfadoxine–pyrimethamine (SP), the use of insecticide-treated bed nets (ITN) and ensuring effective case management of malaria illness and anaemia [1]. ANC coverage in Ghana has improved over the last two and a half decades with more than 95% of pregnant women attending ANC clinics at least once during their pregnancy and 87% four or more times [2]. It is expected that as pregnant women visit the clinics they receive all these efficacious ANC interventions but their pregnancy outcomes show otherwise. In Ghana, anaemia and asymptomatic malaria prevalence and their detrimental effects during pregnancy continue to be of public health concern. Almost 50% of pregnant women at term (36–40 weeks gestation) and 12–36% of women at 32 weeks gestation and above, have asymptomatic malaria parasitaemia [3‐5]. Eleven percent of babies born are of low birth weight [3] and ITN use during pregnancy is only 32%. Current national statistics show that 68% of pregnant women received two or more doses of IPTp. Fifty-nine percent of them reported taking their iron supplements for 3 months or more [2]. Factors that contribute to low intervention effectiveness include inaccessibility or unavailability of ANC clinics, unavailability of resources needed for intervention implementation and lack of adherence of ANC staff and pregnant women to these interventions [6]. Health system issues such as unclear policy and guidance, stock outs, user fees, poor organization leading to poor quality of care, poor healthcare provider performance as well as factors concerning pregnant women such as poor timing of ANC clinic visits, education, age, marital status, knowledge about malaria and IPTp, socio-economic status and parity have been identified as key determinants to the uptake of IPTp and ITN during antenatal care [7].

The focus of efforts to enhance effectiveness of ANC interventions has mainly been on improving health services for better health outcomes. While this occurs mainly in the clinical settings, evidence has also shown that community-based health intervention packages involving health personnel such as community health workers and community health nurses can improve maternal and child survival [8, 9]. Encouraging patients to actively participate in their healthcare leads to better health outcomes [10, 11] through improved adherence to or uptake of health interventions [12, 13]. Encouraging patients to take part in their health care decisions is considered ‘a sign of valuing humanity and the individuality of the patient,’ a legal right of the patient as well as an international gold standard for healthcare systems’ [14]. Although this concept has been applied to some aspects of patient care including decision-making during medical consultation, patient safety and the management of some chronic diseases [15‐17], little is known about its usefulness in antenatal care. Only a few studies have targeted strategies aimed at improving adherence of pregnant women to ANC interventions. Mbonye and colleagues have showed that by creating awareness about IPTp and its importance and improving health worker-pregnant woman trust, pregnant women can be influenced to improve their adherence to scheduled ANC visits, 2 doses of IPTp and health facility-based delivery [18] albeit not through active participation.

This study hypothesized that when pregnant women actively participate in their ANC, this would improve their adherence and hence uptake of ANC interventions translating eventually into better health outcomes (Fig. 1). This concept was tested in this study in relation to the control of malaria and anaemia in pregnancy. Active participation in this study means allowing pregnant women to see the conduct and results of their point-of-care tests for malaria and anaemia and be involved with the interpretation of the test results so as to influence their decision to adhere to ANC treatments and recommendations given at the ANC.

This was a pragmatic cluster randomized controlled trial designed to investigate the effectiveness of pregnant women’s active participation in their antenatal anaemia and malaria point-of-care testing in improving pregnancy outcomes. The study showed a potential of improving maternal and pregnancy outcomes since the risk of malaria parasitaemia, anaemia, low birth weight and sub-optimal pregnancies appeared reduced in the intervention arm compared to the control arm by the end of pregnancy although not significantly.

A limitation identified in this study is more than 30% loss to follow-up at each time point in both arms (Fig. 4). Also, an assumption of a 30% intervention effect based on documented effects of patient participation in previous studies [23‐25] may not have been applicable to our study and may have also contributed to reduced power to robustly detect significant benefit.

The nature of the trial, being pragmatic, did not call for active follow up into the communities and this may have contributed to a reduced power to detect significant intervention effect. Pregnant women’s ANC seeking behaviour may be influenced by numerous factors including the educational level of the woman and her husband, distance to the clinic from their homes, direct (payment for some services at the clinic) and indirect costs (transportation, food to eat and time spent to seek care), complacency among elder women, and women’s previous experiences with ANC and the staff [26‐28]. Although these factors were not explored in this current study, they may have contributed to the loss to follow up. In order to achieve the maximum rate of data capture as the women attended their subsequent ANC visits, research assistants were stationed at the clinics every day of the week to be able to identify participants for follow up at the clinics. It is thus likely that this may be a true reflection of what pertains in the study area. However understanding what factors affect ANC seeking behaviour in subsequent visits in the study area may be necessary to help improve ANC re-attendance.

The intervention may have been responsible for the higher number of women re-attending the 4–8 weeks after recruitment follow-up visit in the intervention arm compared to the control arm (Fig. 4). Women in the intervention arm may have been curious to see whether there had been changes in their Hb and malaria parasitaemia levels after their first visit, hence motivated to visit the clinic at their scheduled date, accounting for the higher numbers. This may also have accounted for the higher risk of parasitaemia detected in the intervention arm compared to the control arm at this time point as a higher number of women were tested compared to the control arm. But this motivation was not enough to sustain attendance during pregnancy with, both arms recording similar reduced attendance at term and at delivery. Other factors like the age, parity, educational status, socio- economic situation and empowerment of the women [27, 29, 30] may be of equal importance in determining re-attendance to antenatal care.

The risk of parasitaemia recorded at term pregnancy in both arms was lower (Table 6) than has previously been reported [3, 4, 21]. The deployment of RDT to all ANC clinics by the Ghana Health Service during the time of the trial meant that RDT became available in the control ANC clinics as well. This may have reduced the effect of the intervention as it is possible that some ANC staff may have allowed the pregnant women in the control group to see their test results (as reported in a qualitative study yet to be published elsewhere). None the less, more widespread use of RDT may have helped identify more women with parasitaemia for treatment during pregnancy hence reducing their risk of malaria at term. The HCS and pictorial guides were at the intervention clinics only so these may have accounted for the slightly lower but not significant risk of anaemia observed in the intervention arm compared to the control arm.

Improved adherence of pregnant women to ANC treatments and recommendations was anticipated as a benefit of active participation of pregnant women in their care; however no significant difference was measured between intervention and control groups (Table 7). Patients’ adherence is a complex health behaviour, and has been found to be affected by a complex interplay of other factors including attitudes, beliefs and group norms; culture; psychological state, the patient’s involvement and participatory decision making level, type of illness and co-morbidity, socio-economic level, use of alternative medicine and stakes of proposed outcome [13, 17]. Understanding what factors influenced these pregnant women’s adherence to ANC recommendations and treatments might have helped explain the results in the context of complex inter-relations between patient adherence, improved health outcomes and intervention implementation. Health literacy, another factor that affects patient adherence [13] was slightly but not significantly increased in the intervention group compared to the control group. This was evidenced by the higher scores of adequate knowledge about malaria and anaemia in pregnancy (Table 7) in the intervention group compared with the control group however this alone may not have been enough to positively influence the pregnant woman’s level of adherence.

Although patient adherence to health interventions is important for health intervention effectiveness, health service factors including accessibility to healthcare, availability of diagnostics and health provider compliance with intervention implementation [6] are necessary for intervention effectiveness too. In this study, it was assumed that financial accessibility to antenatal care was assured by the free maternal health care and delivery policy that was underway in the country at time of the study [31] and geographical accessibility by the design of the study. There was also constant supply of the point-of-care tests (diagnostics) needed throughout the study since these were provided by the project. Health provider compliance with intervention implementation, measured during the study as implementation fidelity was however found to be below expectation (Table 8). ANC staff may have adopted the active participation concept with enthusiasm at the beginning of the trial but this was not sustained as evidenced by the lower than expected implementation fidelity. Health workers’ unwillingness to delegate power, their perceived lack of time, the type of health situation the health care worker is dealing with, their individual characteristics including their personal beliefs, perceptions and knowledge of the intervention, their specialization level, and their age and sex are known to influence their level of compliance [17, 32]. Although these factors were not explored in this study, some of them may have contributed to the sub-optimal implementation and hence reduced intervention effectiveness.

This study has demonstrated that encouraging pregnant women to see and be involved with the interpretation of their point-of-care test results for malaria and anaemia has the potential to improve maternal and pregnancy outcomes albeit not statistically significant. It appears there is a rather complex inter-relation between active patient participation, patient adherence, intervention implementation and expected health outcomes. Exploring what factors influence health worker compliance to health intervention implementation and patient adherence to health interventions within this context will contribute to understanding and improving intervention effectiveness and hence patient health outcomes.