Effect of mHealth intervention on maternal health service utilization and birth outcomes in Northwest Ethiopia: a two-site non-randomized controlled trial

Abstract

Background

See More

Globally, mobile phone technologies, including short message service (SMS), have emerged as an essential tool for supporting the healthcare system and improving health outcomes. However, evidence on their effectiveness in improving maternal healthcare service utilization and birth outcomes remains limited. This study therefore aimed to examine whether a mobile phone-based SMS reminders with follow-up calls to pregnant women increased the uptake of focused antenatal care (FANC) and skilled birth attendance (SBA), and reduced the incidence of low birth weight (LBW) in neonates.

Methods

We conducted a non-randomized controlled trial (nRCT) between July 2019 and December 2020 among pregnant women attending antenatal care (ANC) at Debre Markos and Finoteselam public hospitals in northwest Ethiopia. The hospitals were purposively assigned to either the intervention (Debre Markos) or control group (Finoteselam) to avoid cross-contamination. Eligible women attending ANC at each hospital were enrolled accordingly. The intervention group received weekly SMS reminders and biweekly phone calls throughout their pregnancy, starting from their first visit until delivery. The control group received standard routine care. Analyses followed an intention-to-treat approach. Binary and multivariable logistic regressions were used, and any variable with a p-value less than 0.05 was declared statistically significant.

Results

We enrolled 386 pregnant women–194 in the intervention group at Debre Markos hospital and 192 in the control group at Finoteselam hospital. Among participants who completed our evaluation, the proportion of FANC uptake was higher in the intervention group (69.6%) compared to the control group (55.7%) (adjusted odds ratio (AOR): 1.75; 95% confidence interval (CI): 1.10, 2.80) during the follow-up. SBA was 51% in the intervention group and 41.1% in the control group (AOR: 1.49; 95% CI: 0.95, 2.34). The proportion of LBW was lower in the intervention group (17.5%) than in the control group (25.5%) (AOR: 0.55; 95%CI: 0.32, 0.90).

Conclusion

The mHealth intervention was associated with FANC services uptake and LBW. Policy- and decision-makers should consider integrating mHealth intervention into routine maternal, newborn, and child healthcare delivery systems.

Trial registration number

This trial is registered with the Pan African Clinical Trials Registry (PACTR202205597300585), date: 04/05/2022.

Introduction

To mitigate maternal, newborn, and child health-related complications, the World Health Organization (WHO) recommends access to and use of essential services throughout pregnancy, childbirth, and the postpartum period for every pregnant woman [1]. However, the availability and quality of these services vary widely across different settings, particularly in low-income countries, where women continue to die from preventable causes [2,3,4]. The Maternal Mortality Estimation Interagency Group predicted a total of 295,000 maternal deaths worldwide in 2017, where sub-Saharan Africa (SSA) accounted for 196,000 (66%) [5]. In addition, the maternal mortality ratio (MMR) in SSA was 542 deaths per 100,000 live births, compared to 216 deaths per 100,000 live births globally [6, 7]. This indicates that maternal health interventions must be prioritized, and greater efforts are needed in SSA countries to achieve the Sustainable Development Goals (SDGs).

Antenatal care (ANC) is critical to improving health-seeking behaviors and use of preventive measures to reduce maternal morbidity and mortality, while enhancing newborn survival [8]. Access to high-quality ANC has been identified as one of the pillars to lower maternal and newborn morbidity and mortality. However, this intervention remains insufficient, and its benefits are often missed, particularly in SSA, where ANC attendance rates are low [9]. For instance, only half (52%) of pregnant women in low- and middle-income countries (LMICs), including SSA, receive the recommended four ANC visits (focused ANC or FANC) by WHO [10]. Moreover, the proportion of skilled birth attendance (SBA) remains low in southern Asia (45%) and SSA (46%)—regions with the highest burden of maternal deaths [11, 12]. In addition, the number of live births with low birth weight (LBW) in SSA is estimated to have increased from 4.4 million in 2000 to 5 million in 2015 [13].

The 2016 Ethiopian Demographic and Health Survey (EDHS) reported 412 maternal deaths per 100,000 live births [14], although the Ethiopian reproductive health program is set to reach 99 maternal deaths per 100,000 live births by 2020 [15]. EDHS reports showed that four in ten women (43%) reported at least four ANC visits, 28% SBA, and 16% very small and 10% smaller than average deliveries for the most recent live births [14]. To reduce these undesired outcomes, the Ethiopian government has deployed 40,000 trained Health Extension Workers (HEWs) at health posts to provide essential preventive and curative maternal, newborn, and child health (MNCH) services at the community level [16]. Despite ongoing efforts, access to and equity in MNCH services in Ethiopia remain significantly constrained by a shortage of skilled workforce, a low level of societal awareness, and solid bureaucratic procedures, all of which continue to hinder the implementation and sustainability of effective interventions [17].

The rapid expansion of mobile phone use in low-income settings has positioned mHealth technologies as valuable tools for strengthening the healthcare system [18]. These technologies enhance access to MNCH services and improve treatment outcomes in a cost-effective manner [19]. In SSA, more than 60% of individuals have access to a mobile phone [20], while in Ethiopia, mobile subscriptions reached approximately 46.75 million in 2020, with a 41% penetration rate. The multifunctionality of mobile phones, including short message service (SMS), presents significant opportunities to address workforce shortages and improve health education and promotion for basic healthcare services [21].

SMS has been effectively utilized in healthcare to promote patient education [22,23,24,25], support chronic disease self-management [26,27,28,29], and improve clinic attendance [30,31,32,33,34]. Our meta-analysis indicates that pregnancy-focused mHealth applications are user-friendly and significantly improve FANC and SBA in LMICs [35]. Moreover, the WHO recognizes SMS as a widely used and effective tool for delivering appointment reminders [36].

The other advantage of mHealth is as a data collection and reporting tool to ensure data recording and completeness [37, 38]. However, we found limited evidence on the implementation of mHealth platforms specifically targeting MNCH service utilization in Ethiopia, highlighting the need for context-specific implementation and evaluation of such interventions. To address these gaps, we designed a non-randomized controlled trial (nRCT) to evaluate whether a mobile phone-based educational intervention can improve FANC, SBA, and birth weight outcomes in Ethiopia.

Methods

Ethics statement

This study was conducted in accordance with the ethical principles of the Declaration of Helsinki. This trial was approved by the Institutional Research Ethical Review Committee (IRERC) of Debre Markos University, College of Health Sciences (reference number: HSC/R/C/Ser/Co/145/11/11). Verbal informed consent was also approved by the IRERC. This committee had no role in the trial’s design, conduct of the study, analyses, or reporting. All procedures were performed in accordance with the organization’s established guidelines and regulations. Although the trial protocol was not prospectively published, it was retrospectively registered (trial registration number: PACTR202205597300585; date: 04/05/2022). A systematic review and meta-analysis in LMICs was also employed and published in the BMC Reproductive Health journal [35]. The reporting of this nRCT adheres to the Transparent Reporting of Evaluations with Nonrandomized Designs (TREND) statement, which provides guidelines for transparent reporting of nonrandomized evaluations of behavioral and public health interventions (Supplementary file: Table S1).

Study setting, design, and period

The study was conducted in Debre Markos and Finoteselam towns, located 300 and 375 kms northwest of Addis Ababa, Ethiopia’s capital. Debre Markos town is the administrative center of the East Gojjam zone and has one public hospital, three health centers and 14 health posts, and seven private clinics. Finoteselam town is the administrative center of the West Gojjam zone and has one public hospital, one health center and five health posts, and four private clinics [39, 40]. Both towns had reasonably good access to electricity and internet services. Women in both zones had similar characteristics in terms of socioeconomic status, educational background, and cultural practices related to pregnancy and childbirth.

This was a nRCT conducted at Debre Markos (intervention group) and Finoteselam (control group) hospitals between July 2019 and December 2020. The hospitals were intentionally selected and assigned to either the intervention or control arms based on logistical and operational feasibility, as well as to prevent contamination of the intervention.

Participants

Source population

All pregnant women attending ANC services at Debre Markos and Finoteselam public hospitals in Northwest Ethiopia during the study period.

Study population

Pregnant women who met the eligibility criteria, including those in their first or second trimester, with mobile phone access, able to read and write in Amharic and/or English, and who provided informed consent to participate.

Inclusion criteria

Pregnant women were eligible if they owned a mobile phone and were literate, defined as the ability to read and write a simple sentence in Amharic and/or English, as assessed verbally by data collectors during recruitment. These women were also requested to confirm whether they had access to a phone number of a husband/friend/mother who could relay the messages. Women were enrolled based on gestational age or pregnancy trimester to test the hypothesis that longer exposure to the messages would lead to improved outcomes (dose-response), as the messages were designed to coincide with specific stages of pregnancy. Accordingly, pregnancies in the first and second trimesters were included in the study. Moreover, a pregnant woman attending either of the selected hospitals had to reside in the study zone.

Exclusion criteria

Pregnant women with critical health conditions who register at the hospital, or who were temporary residents (less than six months) in the study zones, or those who spent a significant amount of time outside their regular residence, were excluded from the study.

Intervention

The intervention consisted of a manual SMS system that enabled bidirectional communication through text messaging and phone calls. While only women with registered phone numbers received text messages, all pregnant women in the intervention group received both texts and calls regularly. Participants were also encouraged to text/call their providers for advice and counselling as needed. The SMS and phone call components provided detailed health and nutrition education, counseling on regular moderate exercise, information on danger signs during pregnancy, programmed appointment reminders, encouragement to attend routine ANC, and guidance on the importance of SBA. Messages were sent weekly, and phone calls were made fortnightly. The content was developed by a team of experts, including pediatricians, gynecologists, and local partners (see details in the supplementary file: Table S2).

Control

Women in the control group received routine ANC and were followed up as per hospital protocol. Finoteselam hospital provided ANC services according to the national FANC guideline, which consisted of three core packages of interventions. The first package focused on maternal and fetal assessment (essential blood tests, early ultrasound for all pregnant women, promoting early initiation, adherence to the schedule, etc.). The second package focused on health promotion, prevention, and treatment during pregnancy (counselling women for adherence to iron supplements, birth preparedness, providing appropriate treatment, etc.). Finally, the third package focused on strengthening the healthcare system for ANC (improving infrastructure, integrating reproductive health services, enhancing documentation practices, etc.).

Outcomes

We evaluated the effect of mobile phone-based intervention on three outcomes. The primary endpoint was the proportion of women receiving four or more ANC visits–FANC; the secondary endpoint was SBA; and the tertiary endpoint was LBW. We used the normal course of pregnancy and childbirth to determine the outcomes of interest (prenatal, intranatal, and postnatal) as this provides the opportunity for a more meaningful description and discussion. We also conducted and published a meta-analysis of literature using the same outcomes of interest [35], while developing a proposal for this study.

Masking

The hospitals were purposively allocated to either the intervention or control group in a 1:1 ratio. There was no individual-level randomization. Due to the nature of the intervention and site-level allocation, neither participants nor researchers were blinded.

Sample size determination

This study was designed as an nRCT involving two hospitals—one intervention and one control site. The sample size was estimated using proportions of home delivery, which serve as a proxy inverse for SBA, one of our primary outcomes. Assuming 33.7% home delivery in the intervention group and 58% in the control group [17], with a 5% significance level, 90% power, and a 1:1 group ratio, the calculated sample size was 468 (234 per group). No adjustment for clustering or intra-cluster correlation (ICC) was applied, as the study included only two clusters and such an adjustment is statistically inappropriate in this context. The sample size should therefore be interpreted as a pragmatic estimate, and the study was considered exploratory in nature.

Sampling procedure

Debre Markos and Finoteselam public hospitals were purposively selected to minimize the risk of information contamination between groups. Within each hospital, eligible pregnant women were enrolled using a consecutive sampling technique as they attended their first ANC visit during the study period. Women were screened based on inclusion and exclusion criteria, and sampling continued until the target sample size was reached.

Data collection procedure

A semi-structured questionnaire was developed in English and Amharic versions, which was adapted from previous studies [17, 22, 41, 42]. The survey was collected twice, at baseline and after the intervention. Four trained research assistants, with a midwifery qualification, were recruited to support the data collection. The principal investigator was responsible for responding to and advising pregnant women who sought guidance as needed during the study period. Data collection was conducted using a semi-structured questionnaire administered by research assistants during initial enrollment. At the end of the intervention, participants were asked to complete a follow-up questionnaire that collected information on pregnancy progress and outcomes. In some instances, clients who defaulted in hospital attendance were followed via a mobile phone. When a phone number was inactive, a review of medical records was performed to gather the necessary information (e.g., to obtain baby weight at birth or within seven days postpartum). Occasionally, data collectors made home visits or called mothers to remind them to attend postpartum care services at the hospital within seven days, and where appropriate, measured birth weight. The classification of birth weight was done in accordance with the Ethiopian Integrated Management of Childhood and Neonates Illness (IMNCI) standard [43], where needed, within seven days after delivery.

Study variables

Dependent variables included FANC (‘no’ versus ‘yes’), SBA (‘no’ versus ‘yes’), and LBW (‘no’ versus ‘yes’). Independent variables included sociodemographic and socioeconomic characteristics (maternal age, residence, education, occupation, monthly income, etc.), obstetric history (parity, gravidity, outcome of pregnancy and pregnancy intention (planned or not planned)), and others (e.g., husband’s education).

Operational definitions

Mobile health (mHealth)

A component of electronic health and has been defined as a medical and public health practice supported by mobile devices, such as mobile phones, patient-monitoring devices or other wireless devices [44].

Focused antenatal care

A minimum of four scheduled comprehensive ANC visits during pregnancy [45].

Skilled birth attendance

Delivery at a health facility attended by midwives, nurses, or doctors [46].

Low birth weight

A neonatal weight of less than 2,500 g measured within the first seven days after birth (early neonatal weight) [43].

Urban

An area with a high population density (more than 100,000 people per square kilometer of land area) [47].

Rural

All population, housing, and territory not included within an urban area.

Data quality control

Supervisors and data collectors received orientation on the specific data to be collected and the appropriate procedures for obtaining it from designated sources. Data extraction forms were reviewed before data collection to ensure accuracy and relevance. The completeness and consistency of the collected data were checked daily during the data collection period. Prompt feedback was provided by the primary investigator and supervisors when necessary. Information formats were also cross-checked on-site to identify and address any incompleteness, errors, and ambiguities in the records.

Data analysis

All statistical analyses were performed using STATA/se version 14. Descriptive statistics were presented as numbers and percentages, using tables or graphs. Analytical analyses were performed for all binary outcome variables: FANC (‘‘yes’ versus ‘no’), SBA (‘yes’ versus ‘no’), and LBW (‘yes’ versus ‘no’) to evaluate the impact of the intervention. We employed a binary logistic regression model. The models included fixed-effects for relevant covariates such as maternal age, residence, marital status, and education level. By adjusting for these key sociodemographic factors, the model aims to provide an unbiased estimate of the intervention effect across the two study sites (Debre Markos and Finoteselam). The analyses also applied an intention–to–treat (ITT) approach to assess the effect of the intervention. Any statistical significance was set as p-value < 0.05. All analyses were designed to assess both direct intervention effects and potential confounding factors. In addition, we conducted a site-level sensitivity analysis by comparing outcome proportions (FANC, SBA, and LBW) between the two sites. Summary measures, including risk differences, odds ratios, and risk ratios, were calculated. The Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) were also presented in the models to assess the model’s goodness of fit.

Results

The trial-group assignments, loss-to-follow-up, and reasons for withdrawal are illustrated in Fig. 1. Of the 468 pregnant women screened for eligibility, 392 were enrolled in the study—194 were assigned to the mHealth intervention group, and 192 to the control group (Fig. 1).

Fig. 1

Flow diagram of participant recruitment and retention

Sociodemographic characteristics

Due to an nRCT, the baseline characteristics in both groups were not homogeneous. As presented in Table 1, a total of 386 pregnant women participated in the trial at baseline and were followed up. Of these, most (75.8% in the intervention and 71.9% in the control group) participants were between the ages of 15 and 29 years, with a mean age of 26.7 (SD ± 4.8) years. Regarding educational level, about three-fifth (59.8%) of women in the intervention group had a formal education, including college and above, compared to only 20.3% in the control group. Moreover, 47% of participants in the intervention and 28.2% in the control group had a monthly income of ETB 4000.00 and above. Almost all participants (363 pregnant women) had their own mobile phones: 186 in the mHealth intervention and 177 in routine care (Table 1).

Table 1 Baseline characteristics of participants enrolled in the trial in Debre Markos and Finoteselam hospitals, 2020

Proportion of FANC, SBA, and LBW

The intervention group demonstrated higher percentages compared to the control group across key outcomes: a 14-percetage-point increase in FANC utilization (69.6% versus 55.7%), 10-percetage-point increase in SBA (51% versus 41.1%), and an 8-percetage-point increase in LBW (17.5% versus 25.5%) (Fig. 2).

Fig. 2

Proportion of FANC, SBA, and LBW in the intervention and control groups among pregnant women attending Debre Markos and Finote Selam Hospitals. Point estimates are presented without 95% CI

Effect of mobile phone text message reminders with calling on FANC

Multivariable analysis illustrated that women who received the intervention had significantly higher FANC visits compared to the control group: (AOR; 1.75 (95%CI: 1.1, 2.8)) (Table 2).

Table 2 Effect of mobile phone text message reminders with calling on FANC at Debre Markos and Finoteselam hospitals, 2020

Effect of mobile phone text message reminders with calling on SBA

Although the intervention did not show a statistically significant effect on SBA, women in the intervention group were more likely to deliver in health institutions compared to those in the control group (AOR: 1.49 (95%CI: 0.95, 2.34)) (Table 3).

Table 3 Effect of mobile phone text message reminders with calling on SBA at Debre Markos and Finoteselam hospitals, 2020

Effect of mobile phone text message reminders with calling on LBW

Women who received SMS and phone call reminders in addition to routine prenatal care experienced slightly different birth outcomes compared to those who received routine care alone. The risk of LBW (OR for the intervention group compared to the control group) was AOR: 0.55(95%CI: 0.32, 0.90). In contrast, women who did not attend at least 4 ANC visits were more likely to have LBW compared to their counterparts (AOR: 1.77 (95%CI 1.1, 2.94) (Table 4).

Table 4 Effect of mobile phone text message reminders with calling on LBW at Debre Markos and Finoteselam hospitals, 2020

The site-level sensitivity analysis showed results consistent with the main individual-level models. FANC uptake was higher in the intervention site, and SBA and LBW also showed favorable trends with comparable effect directions. Full summary measures are presented in the supplementary file Table S3.

Discussion

In this study, after adjusting for relevant baseline covariates, FANC utilization remained significantly higher among women in the intervention group than in the control group. A significant variation was also detected between the intervention and control groups regarding LBW. However, no significant difference was observed in SBA between the intervention and control groups.

To our knowledge, this trial demonstrates the effect of an innovative mHealth intervention on MNCH care services utilization and birth outcomes in low-resource settings.

Indeed, there is evidence supporting the use of mHealth interventions for managing various health conditions in developed countries [48, 49]. These interventions are accessible, cost-effective, practical, and capable of automatically sending pertinent information or guidance directly to targeted clients. It is, therefore, not surprising that numerous studies use SMS reminders and/or phone calls to provide and disseminate relevant health information [50]. Moreover, some hospitals have reported substantial budgetary savings after implementing the SMS reminder system and observing reduced rates of missed appointments [50,51,52]. This implies that mHealth interventions should be integrated into initiatives aimed at supporting Ethiopia’s effort to achieve SDG of ending preventable maternal, neonatal, and child deaths by 2030.

In fact, the Ethiopian government has initiated plans to digitize its healthcare industry [53, 54]. For instance, despite existing limitations in the country’s healthcare infrastructure, the Ministry of Health is expanding its eHealth strategy to address healthcare gaps by adopting mHealth technology as part of a progressive approach to tackle health disparities. The findings of this study indicate that a mobile phone-based SMS combined with phone calls can be integrated into other interventions. These findings may therefore inform the future adoption and implementation of mHealth interventions in the country.

In this study, we determined that the intervention, regular SMS communications combined with counselling via phone calls, had positive effects on FANC visits during pregnancy. There is an improvement in ANC visits by pregnant women. For instance, according to the recent EDHS, the proportion of women with four or more ANC visits is reported to be 43% [14]. This finding is consistent with previous studies in developed and some developing countries [32, 55,56,57,58], demonstrating SMS and call reminders improve ANC attendance. In addition, a systematic review and meta-analysis of seven trials reported that SMS targeting pregnant women has a positive impact on maternal healthcare-seeking behaviors [35]. This could be due to the fact that these reminders prompted pregnant women to attend their appointments and can play a critical role in influencing their health-seeking behavior. This means that reminders can serve as coaching or encouragement to help prospective memory, remembering to perform planned actions in the future, thereby leading to behavior change [59, 60].

In the current trial, the intervention had a positive effect on LBW, highlighting women who received SMS-based pregnancy and delivery counseling had lower odds of giving a LBW baby compared to the control group [61]. Another study demonstrated that nutritional counseling during pregnancy has strong potential to improve birth weight and breastfeeding practices [62]. The possible reason might be that the intervention group was more frequently exposed to healthcare information and nutritional counseling through phone calls, which may have closed knowledge gaps and enhanced dietary diversity practices. This may then contribute to optimal weight gain in pregnant women, which is important for optimal fetal growth [63,64,65]. Additionally, pregnant women in the intervention group had greater autonomy to refuse harmful traditional practices, seek appropriate healthcare, and make financial decisions to invest in a healthier diet.

We also found that FANC utilization had a positive correlation with LBW. This is supported by a finding that more ANC visits by pregnant women are associated with improved neonatal nutritional status [66]. It is evident that essential ANC services serve as a platform for implementing several interventions that have been shown to positively impact MNCH outcomes. However, this trial was underpowered to examine the true effect of the intervention on SBA. Although there was no statistically significant difference between the intervention and control groups, a higher proportion of women in the intervention group used trained health personnel for delivery services (51% versus 41.5%). This finding is consistent with a previous study that reported no statistically significant difference between the intervention and control groups regarding institutional delivery service utilization [67]. In contrast, some existing literature [17, 42, 68, 69] suggests that mHealth applications can influence women’s choice of delivery place and have a positive impact on SBA. The possible explanation for the observed difference might be variations in educational level, socioeconomic status, and cultural factors, which influence the use of these technologies [70, 71]. Another potential explanation for this inconsistency might be a lack of transportation or financial resources during the sudden onset of labour, limited access to ambulance services, and women’s low decision-making power [72, 73].

Interestingly, in our study, nearly all pregnant women expressed a preference for receiving appointment reminders via text messages or phone calls. After delivery, some mothers contacted mHealth providers about postnatal care and immunization services. Within primary healthcare infrastructure, our findings suggest that substantial progress in health outcomes is both achievable and sustainable through mHealth interventions. The findings should inform policy decisions aimed at reducing maternal and neonatal morbidity and mortality in resource-poor settings. To maximize impact, the intervention should be scaled up and enhanced by including more tailored educational messages that align with women’s gestational age and pre-existing medical conditions, while also addressing the specific needs of rural women. Overall, this study offers valuable insights for policymakers and funders of MNCH programs targeting effective and scalable solutions.

Limitations of the study

Even though we attempted to control for as many limitations as possible, the internal validity of the study may have been compromised as the study design did not involve randomization. This trial was conducted among highly selected groups, which may limit the generalizability of the findings to the wider population. Particularly, the trial had very strict inclusion criteria and notably excluded women with serious illness, mental health disorders, those unable to read and write, and those who did not own mobile phones, which further limited its generalizability. Delayed (retrospective) trial registration is also acknowledged as a limitation that may affect perceived protocol compliance.

In addition, it was an nRCT with only two hospital clusters, and the sample size was estimated using individual-level assumptions based on home delivery rates as a proxy for SBA. While this approach was contextually reasonable, it does not reflect the clustering of participants within hospitals and may overestimate statistical power. Due to the small number of clusters, we did not adjust for ICC, and site-level confounding cannot be excluded. As such, the findings should be interpreted cautiously and considered exploratory. Moreover, the intervention for this trial was designed in accordance with the local context, which limits its reproducibility. While our analysis focused on sociodemographic factors and their association with maternal and neonatal outcomes, some relevant variables—such as maternal nutrition, access to health services, and psychosocial conditions—were not considered, potentially leading to residual confounding. Furthermore, our intervention lacked a digital platform, and due to financial restrictions, we were unable to use automated platform software for large-scale implementation. We did not conduct a cost-effectiveness analysis of the mHealth intervention, which limits our ability to assess its feasibility for large-scale implementation. Finally, birth weight was recorded at birth or within seven days, which may have introduced misclassification bias due to postnatal weight loss, particularly in home deliveries where weighing may have been delayed. Therefore, birth weight should be measured within 24 h to avoid misclassification caused by early neonatal weight loss.

Conclusion

The mHealth intervention package–comprising SMS as reminders and phone calls for disseminating health information–demonstrated a positive impact on FANC visits and a reduction in LBW, regardless of sociodemographic and economic characteristics. However, further large-scale investigations are needed to synthesize strong evidence on the effectiveness of mHealth interventions as part of MNCH initiatives in developing countries. In particular, the development of standardized, practical, and scalable automated software solutions is essential for addressing health challenges in primary healthcare settings to reduce preventable deaths in vulnerable populations.