Pregnant mother’s intention to use mobile phone-based messaging interventions for improving maternal and newborn health practices in Jimma Zone, Ethiopia

Background Mobile phone-based messaging for maternal education has shown promising outcomes in promoting maternal and child healthcare in low-and middle-income countries, where there is poor utilization of healthcare services. However, the success of a mobile phone messaging-based intervention depends on identifying and addressing the underlying factors that determine its utilization before implementation. Objective To assess pregnant mothers’ intention to use a mobile phone-based messaging intervention to improve maternal and newborn health in Jimma Zone, Ethiopia. Method This study employed a cross-sectional study design among randomly selected pregnant mothers residing in six primary health care units that were selected from three districts of Jimma Zone, Ethiopia. Data were collected using an interviewer-administered structured questionnaire. Descriptive statistics such as frequencies, percentages, mean, and standard deviation were calculated and presented using tables. Multivariable linear regression analysis was used to identify predictors of intention. Standardized regression coefficients were used to understand the effect of the independent variables and outcome variable. A 95% confidence interval and a p -value of < 0.05 were considered significant. Result Two hundred seventy-nine pregnant mothers participated in this study, resulting in a 98.9% response rate. Almost all (98.9%) respondents preferred the Afan Oromo language, 41.9% of respondents preferred the afternoon and 48.4% preferred receiving health information as text messages once daily. Having family members who can share SMS information (β = 0.098, 95% CI: (0.279—1.867), perceived usefulness (β = 0.283, 95% CI: (0.143—0.341)


Background
Globally, maternal and newborn remains an urgent priority due to the high death toll among these population groups [1].Maternal and newborn deaths primarily occur due to complications during pregnancy, labor, childbirth, and the postpartum period [2].The majority of these deaths takes place in low-and middle-income countries (LMICs), where access to and utilization of healthcare services are limited [3,4].Although the government of Ethiopia has implemented several initiatives and strategies to improve maternal and newborn health outcomes [5,6], the country still has the highest maternal and neonatal mortality ratios, with a maternal mortality ratio of 267 per 100,000 live births [7] and a neonatal mortality rate of 33 per 1,000 live births [8].
Utilization of essential maternal health care services such as antenatal care (ANC), skilled delivery, postnatal care (PNC), and newborn care practices plays a significant role in promoting maternal and newborn health [9].However, these services are underutilized in LMICs due to various factors [10,11].In addition to health systemrelated factors like infrastructure, finance, and human resources, individual factors such as low health literacy [12], knowledge [13], attitude [14], and other behavioral factors [15] also contribute to inadequate utilization of maternal and newborn care services in LMICs.Evidence indicates that educational interventions targeting mothers have a positive impact on the utilization of maternal health care services and newborn care practices, including colostrum administration and early initiation of breastfeeding in LMICs [16].
The global growth of mobile connectivity has created an opportunity for the emergence of a subset of eHealth known as mHealth [17].mHealth is defined as the use of mobile wireless technologies for health [18].It covers using mobile phones and tablets for voice calls, texting, audio, images, video, voice short message service (SMS), multimedia messaging, and internet access [19].According to the International Telecommunication Union, in 2022, mobile phone coverage reached 97.4% of the world's population, with 72.7% owning mobile phones [20].This high coverage and penetration of mobile phones have provided an opportunity to utilize mHealth for monitoring health services, facilitating interaction among healthcare providers, managing human resources, providing educational information, and enhancing behavior change [21,22].Mobile phone messaging, particularly through SMS, is a commonly used component of mHealth, employed for improving self-management of conditions such as diabetes, weight loss, physical activity, quitting smoking, medication adherence in antiretroviral therapy, and promoting the utilization of maternal and newborn health care services [23][24][25][26].
However, the success of mobile phone messagingbased interventions depends on the identification and addressing of underlying factors prior to implementation [27].Factors such as acceptance, end-users' preferences regarding the receipt of text messages, experience with using mobile phones and text messages, perception, and intention to use text messages affect the implementation of mobile phone messaging-based intervention [28,29].Additionally, factors such as culture, accessibility, and acceptance in the local context play a significant role in determining the design and success of an mHealth intervention [30,31].Therefore, it is crucial to identify these factors when developing mobile phone messaging-based interventions to ensure their alignment with the local context and to avoid implementing interventions that fail due to low user adoption.
This study is part of the formative phase of an intervention aimed at evaluating the effectiveness of mobile phone messaging-based message framing to improve maternal and newborn care practices among mothers in Jimma Zone, Ethiopia.The current study aims to assess the intention, perceptions, and preferences of pregnant mothers regarding the use of mobile phone-based messaging for maternal and newborn health information (MNHI) in the Jimma Zone, Ethiopia.

Study design and setting
A Cross-sectional study was conducted from October 3 to 28, 2022, in the Manna, Shebe Sombo, and Dedo districts of the Jimma Zone, Oromia, Ethiopia.Jimma Zone is characterized by poor health infrastructure, including a high health worker-to-population ratio (1:329) and a limited number of health facilities providing comprehensive emergency obstetric and newborn care.The zone also has a high maternal mortality ratio (412 per 100,000 live births).The three districts were selected from a total of 21 districts in Jimma Zone, based on the confirmation that each of them had no ongoing initiatives focused on digital health aimed at improving maternal and newborn health.This study was reported in accordance with the STROBE checklist for cross sectional studies (Additional file 1).

Study participants
Pregnant mothers living in the catchment area of the selected primary health care unit (PHCU) were selected to participate in the study.The study included pregnant mothers who had a follow-up for ANC service, owned a mobile phone or had a family member who owned one, and provided written informed consent to participate in the study.

Sample size calculation and sampling procedure
The sample size for this study was calculated using the single population proportion formula.The following assumptions were used: a proportion of mother who have the intention to use SMS reminders, which is 78.9% from previous study [31], a 95% confidence interval, and a 5% margin of error.Based on these, the initial sample size was calculated to be 256.Considering a 10% nonresponse, the final sample size was determined to be 282.The study participants were recruited using a simple random sampling method.First, two PHCUs were randomly selected from each of the three districts.The calculated sample was then proportionally allocated to each of the PHCU based on client flow.All pregnant mothers who met the enrollment criteria in the selected PHCU were registered, and this list served as sampling frame.The list was then entered into Microsoft Excel, and a random number generator was run to select participants.With the assistance of local volunteers, data collectors identified and visited the homes of randomly selected pregnant mothers.They personally informed the mothers about the study, provided a clear explanation of its objectives, and invited them to participate in the study.

Data collection tool and procedure
Data were collected using an interviewer-administered structured questionnaire, employing a face-to-face interview technique.According to the technology acceptance model, intention is determined by perceived easiness and perceived usefulness [32], and the questionnaire included these variables.The items for the intention to use text messages for MNH were adapted from the literature [33,34], as were the perceived usefulness and easiness of using text messages for MNH [32][33][34].The theoretical framework of acceptability suggests the possibility of assessing acceptability before exposure to the intervention, and the items used to measure perceived acceptability were adapted from this model [35] and another study that validated implementation outcomes [36].The questionnaire also included socio-demographic characteristics, text message usage experience, and preference (language, timing, and frequency), self-efficacy-adapted from the literature [34,37], and perceived feasibilityadapted from the literature [36] of mobile phone-based messaging interventions for MNH.The questionnaire was translated into Afan Oromo from English and the back-translated into English to ensure consistency.The additional file shows the questionnaire in more detail (see Additional file 2).
As this is a pre-implementation study, a detailed description of the planned intervention was presented to each participant.The description included a definition of the planned mobile phone messaging-based intervention, how it is delivered, and its aims to encourage participants to utilize maternal health services and practice the recommended newborn care.Following this, an explanation was provided on what a mobile phone messaging-based intervention entails, namely sending text messages to mothers' mobile phone, the topic areas of the messages, the duration of the intervention, and ensuring participants receive key messages that are specific to their gestational age, infant age and reminders for their appointment.
Six data collectors and three supervisors were recruited for data collection.The data collectors and supervisors were trained for two days on the objectives of the study, data collection tools and procedures, providing informed consent, maintaining confidentiality, and respecting respondents' rights.Continuous follow-up was conducted by the supervisors throughout the data collection period to ensure that the data were collected according to the plan.

Intention
Intention is defined as an individual's readiness to engage in a given behavior [38].For this study, a scale consisting of seven 5-point Likert scale items was adapted from the literature [33,34] to measure of pregnant mothers' readiness to receive mobile phone-based messages on MNH.Item scores were summed after reverse coding negatively worded items and higher composite scores indicate a higher intention.

Perceived usefulness
Perceived usefulness is defined as the degree to which a person believes that using a particular system would enhance his or her job performance [32].Nine items designed on a 5-point Likert scale, adapted from the literature [32][33][34], were used to measure the extent to which pregnant mothers believe that using mobile phone-based messaging for MNH would enhance their maternal and newborn health practices.Item scores were summed after reverse coding negatively worded items, and higher composite scores indicate a higher perceived usefulness.

Perceived easiness
Perceived easiness is defined as the degree to which a person believes that using a particular system would be free of effort [32].In this study, seven items designed on a 5-point Likert scale, adapted from previous studies [32][33][34], were used to measure mothers' belief that using mobile phone-based messaging for MNH would be effortless.Item scores were summed after reverse coding negatively worded items, and higher composite scores indicate a higher perceived easiness.

Self-efficacy
Self-efficacy was defined as the extent to which a pregnant mother believes that she can use mobile phonebased messaging for MNH [39].Seven items designed on a 5-point Likert scale, adapted from the literature [34,37], were used to measure this belief.Item scores were summed after reverse coding negatively worded items, and higher composite scores indicate a higher self-efficacy.

Perceived acceptability
Perceived acceptability is defined as the belief that a given innovation is agreeable, appropriate, suitable, convenient, effective and satisfactory [35,40].In this study, ten 5-point Likert scale items, adapted from the literature [36,41], were used to measure pregnant mothers' belief about appropriateness, suitability, convenience, and effectiveness of mobile phone-based messaging for improving MNH.Items score was summed after reverse coding negatively worded items, and higher composite scores indicate a higher perceived acceptability.

Perceive feasibility
Perceived feasibility is defined as an individual's belief in the extent to which an innovation can be successfully used or carried out within a given setting [40].For this study, it was used to measure the degree to which pregnant mothers believe that mobile phone-based messaging can be successfully used for MNH.This was assessed using seven 5-point Likert scale items adapted from the literature [36].Item scores were summed after reverse coding negatively worded items, and higher composite scores indicate a higher perceived feasibility.

Household wealth level
Household wealth level was calculated using principal component analysis by considering 15 household asset properties.it was then categorized into wealth quantiles; lowest, second, middle, fourth, and highest [42].

Quality assurance
The face and content validity of the data collection instrument were ensured by adapting the tool from previous related studies.Before the actual data collection, the tool was pretested on 5% of the sample size.Modifications were made to the questionnaire content based on the findings.The internal consistency of the key constructs was checked using Cronbach's alpha, and all the constructs showed acceptable values of Cronbach alpha coefficient.

Data analysis plan
The collected data were entered into Epi-data V.3.1 and exported to SPSS V.25 software for analysis after ensuring completeness.Descriptive statistics, such as frequencies, means, standard deviations and percentages were computed, and presented using tables.Pearson's correlation analysis was done to identify associations between continuous variables and the intention to use mobile phonebased messaging for MNH.An independent sample t-test was performed to see association between dichotomous variables and intention.Multivariable linear regression analysis was employed to identify predictors of the intention to use mobile phone-based messaging for MNH.Variables that showed significant associations with the outcome variable in the bivariable analysis were included in the multivariable linear regression analysis to control for potential confounding effects.Standardized regression coefficients, along with their 95% CI, were used to understand the effect of the independent variables on outcome variable.A p-value < 0.05 was considered statistically significant.

Patient and public involvement
The technical committee for the project included maternal and child health experts from zonal and district health office, information technology experts, and academicians.This group evaluated the content of the questions and the overall research method.Three maternal and child health experts at the district level assisted with the data collector selection for this study.Six volunteers living in the catchment area of the selected primary health care unit contributed to the study by helping us review the Afan Oromo language version of the questions.We refined the questionnaire based on the feedback we received from the volunteers.The pregnant mothers provided their opinions, experiences, preferences, and perceptions about the mobile phone messaging-based intervention after receiving a brief description of the proposed intervention.

Sociodemographic characteristics
A total of 279 mothers participated in this study, resulting in a response rate of 98.9%.One hundred twenty (43%) of the study participants were in the age group of 25-29 years.One-third (66.7%) of the respondents lived in rural areas.A large number of the respondents were married (95%), identified as Oromo in ethnicity (96.1%), and were Muslim (94.3%).The majority of the participants (59.9%) and their husband (62.6%) had completed primary school.Regarding occupation, a large number (72.8%) of the respondents were housewives, followed by farmers (18.3%).The majority (63.1%) of the study participants had a family size of four and fewer.All of the study participants had family members who can read and write, and two-third (66.7%) of them had two family members who can read and write Table 1.

Mobile phone use and preference to receive message for maternal and newborn health
The majority (69.5%) of respondents had been using a mobile phone for 1 to 4 years.A Large number (91.8%) of respondents were using the standard mobile type.The majority of the participants can read (79.9%), and around half (54.1%) of them were able to send mobile phone text messages.Only 16.8% of the respondents had ever received health information from a healthcare provider through their mobile phone.More than three-fourths (77.4%) of the respondents stated that a family member shares information they receive through their phone.For almost all (98.9%) of respondents, Afan Oromo was the preferred language to receive health information as text messages.Regarding the preferred time of day to receive health information as text messages, 41.9% of respondents preferred the afternoon, and 35.8% preferred the evening.Around half (48.4%) of respondents preferred receiving one health information text message per day (Table 2).

Descriptive statistics of items and internal consistency measure of constructs
Table 3 presents the descriptive statistics for individual items and the internal consistency of key constructs.Among the items related to intention, using SMS in the future for information about pregnancy and newborn care had the highest mean score (µ = 4.15), followed by the willing to receive SMS about service provided at health facilities for mother and newborns (µ = 4).The reversely coded item, which was about unwillingness to be contacted by one's own or a family member's mobile phone for important health message, had a lower mean score (µ = 3.18) within this construct.Items of perceived usefulness related to respondents' beliefs about convenience of SMS for maternal and newborn health information and its capability to improving their communication with healthcare providers had similar mean score (µ = 4.10).The item concerning the role of SMS-based interventions in reducing the time pregnant mother spend receiving routine education at health facilities had a lower mean score (µ = 3.75) compared to other items in perceived usefulness construct.All five items used to measure perceived easiness had relatively similar mean score, ranging from 3.86 to 3.99, indicating that a large number of the respondents agreed on easy to use of SMS-based intervention.All items of self-efficacy also had nearly similar mean score within the 3.8 -3.93 range.The item concerning the capability to differentiate health information from other SMS messages received on their phone scored the highest mean (µ=3.93), while items on participants' confidence in their ability to read SMS and belief in their capability to manipulate their own mobile phone to read health information had a mean score of 3.8.Most items of perceived acceptability showed relatively higher mean scores.The item stating that using texts is a good way to teach about maternal and newborn health scored the higher mean (µ=4.09).Items related to respondent's belief in other mothers' interest in receiving maternal and newborn  related education by text message (µ=3.87) and belief that the SMS-based maternal education approach can effectively meet need and fit the target issue and setting (µ=3.95) had relatively lower score compared with other items in this construct.All items of the perceived feasibility had similar mean score between 3.95 and 3.99.The Cronbach's alpha coefficient was used to check the internal consistency of key constructs, and all constructs showed acceptable alpha coefficient.The alpha score of intention was (α = 0.72) after deleting one item due to a lower item-total correlation.The alpha score for perceived usefulness and perceived easiness was 0.86 and 0.87, respectively, after deleting two items from each dimension for lower item-total correlation.The alpha score for self-efficacy and perceived acceptability was 0.93 and 0.91 respectively.Perceived feasibility scored an alpha of 0.9 after removing one item (Table 3).
Correlations between Intention, perceived usefulness, perceived easy to use, self-efficacy, perceived acceptability, and perceived feasibility were moderate to strongly positive with r values between 0.54 and 0.84, (p < 0.01).The highest positive significant correlation was observed between self-efficacy and perceived easy to use (r = 84, p < 0.01).All the perceptions and the self-efficacy constructs have shown positive correlation with intention with r value in the 0.64 -0.72 range, p < 0.01) (Table 4).

Factors associated with intention to use mobile phone-based messaging for MNH
Bivariable and multivariable linear regression analyses were conducted to identify factors associated with the intention to use mobile phone-based messaging for MNH.Variables that were significant during the bivariable analysis were included in the multivariable linear regression analysis.Accordingly, age, education, husband education, the number of literate family members, ability to read a text message, having a family member who can share SMS information, perceived usefulness, perceived easiness, self-efficacy, perceived acceptability, and perceived feasibility of mobile phone-based messaging for MNH, were included in the multivariable regression.However, only having family members who can share SMS information, perceived usefulness, perceived easiness, perceived acceptability, and perceived feasibility showed a statistically significant association with pregnant mothers' intention to use mobile phone-based messaging for MNH with (p < 0.05).

Discussion
The purpose of the current study was to assess pregnant mothers' intention and perceptions regarding the use mobile phone-based messaging for MNH.The study also assessed respondents' mobile phone usage and preference for receiving mobile phone-based messages about MNH.The findings showed that having family members who can share SMS information, perceived easiness, perceived usefulness, perceived acceptability, and perceived feasibility were positively associated with the intention to use mobile phone-based messaging for MNH.This study also showed that 79.9% of the respondents can read mobile text messages, while only 54.1% of them can send them.This finding is comparable with a study conducted in India, where only 52.5% the participants could type and send text messages [43].However, it is lower than the finding of study conducted in northern Ethiopia, where 91% and 87.3% of participants could read and send mobile text messages, respectively [31].The difference might be due to variations in educational status of study participants.Although only about half of the participants in our study were able to send a text message, four out of five could read them, which would make a mobile phone-based messaging intervention feasible.Focusing on designing a one-way messaging intervention would likely maximize the provision of information on MNH to respondents in this study setting.
In this study, almost all participants preferred to receive maternal and newborn-related text messages in the Afan Oromo language.This high preference for the Afan Oromo could be because it is not only the official language but also the mother tongue for nearly all of the community members in the study setting.The preference for receiving message in the common language of their locality was also reported by the majority of participants in other studies [31,43].The afternoon was the preferred time for 41.9% of participants, followed by the evening, preferred by 35.8%.A study conducted in northern Ethiopia also revealed a comparable percent of mothers who preferred to receive text message reminders in the afternoon, although fewer preferred the evening [31].Around half (48.4%) of the participants preferred receiving a text message on MNH once a day.To be successful, any mobile phone-based messaging intervention in this setting should fully consider the language, timing, and frequency preferences of mothers during the development process before implementing the intervention [29].
Having a family member who shares text message information was significantly associated with the intention to use mobile phone-based messaging for MNH.The presence of text message information sharing within a family might have created awareness and increased the perceived importance of text messages, leading to the intention to receive the messages.Moreover, all participants have a family member who can read and write, and these family members might play more supportive role, and could influence pregnant mothers' intention to use mobile phone-based messaging [44].Therefore, the implementation of mobile phone-based messaging intervention could consider family members who can read and write to serve as facilitators and help in reading and explaining MNH information that mothers with lower literacy levels receive through their own or a family member's phone.Perceived usefulness was positively associated with the intention to use mobile phone-based messaging for MNH.This finding aligns with other studies where perceived usefulness positively predicted the intention to use mHealth-based services [31,[45][46][47].These findings imply the importance of providing the end users with clear and concise information about the purpose, how it works, and benefits of mobile phone-based messaging before its implementation.Additionally, this study found a significant positive association between perceived easiness and intention to use mobile phonebased messaging for MNH.Similar findings have been reported in other studies, where perceived easiness was a predictor of intention [31,46] and willingness to use mHealth services [48].Therefore, involving mothers in the design and development process of mobile phonebased messaging intervention-through interviews, message content testing, pilot testing, and feedback ensures its easiness.
In this study, perceived acceptability and perceived feasibility were also significantly associated with intention to use mobile phone-based messaging for MNH.It implies perceived acceptability and feasibility can be used as facilitators, increasing the belief in the mobile phone-based messaging service and thereby increasing the intention to use it.Making the intervention convenient, accessible, personalized, easiness and providing clear information about its benefits could enhance the perceived acceptability and feasibility of mobile phonebased messaging for MNH [49,50].This, in turn, could increase the intention to use and eventually lead to its adoption.

Limitation of the study
This study has some limitations, and its findings should be interpreted with consideration of these limitations.Since the study included only pregnant mothers who visited a health facility for ANC service, its findings may not be generalizable to all pregnant mothers, particularly those who do not have ANC follow-up.Focusing on pregnant mothers who own a mobile phone is a drawback of this study, as it has the potential to marginalize those who do not have a mobile phone.
Failing to publish the protocol of the study may limit its transparency, which could result in publication bias.This study addressed only individual level factors and did not assess systemic factors that affect the implementation of technology.Furthermore, social desirability bias might exist since the data was collected using intervieweradministered technique.

Conclusion
Most respondents preferred to receive mobile phonebased messaging on MNH in the Afan Oromo language, during the afternoon and once a day.A Pregnant mother's intention to use mobile phone-based messaging for MNH is determined by having family members who share text message information, perceived usefulness and perceived easiness, perceived acceptability and feasibility of mobile phone-based messaging for MNH.
In addition to these predictors of intention, engaging mothers through interview and message content testing, as well as involving stakeholders in the design process, is crucial.Considering the identified pregnant mothers' preferences and tailoring messages to gestational and newborn age are very important when designing mobile phone-based messaging intervention for MNH.Providing a clear description of the intervention is also important during its implementation to maximize its utilization.

Table 1
Sociodemographic characteristics of pregnant mother in Jimma Zone, Ethiopia (n = 279)

Table 2
Pregnant mothers mobile phone use and preferences to receive message for maternal and newborn health in Jimma Zone, Ethiopia (n = 279)

Table 3
Descriptive statistics of items and internal consistency measure of constructs

Table 4
Correlations of intention to use and perceptions of mobile phone-based messaging for MNHI among pregnant mothers in Jimma Zone, Ethiopia, (n = 279)

Table 5
Factors associated with intention to use mobile phone-based messaging intervention for MNHI Among pregnant mothers in Jimma Zone, Ethiopia, (n = 279)