Influences of Mobile Phones-Based Short-Message-Services Communications on the HIV Sexual Risk Behaviors among Pregnant Women’s Male Partners, Case Control Study Tanzania

Introduction: In the twenty-first century, Human Immunodeficiency Virus (HIV) 
continues to be among the major public health problem globally. The disease is 
implicated to the result of the over 32 million deaths, and in 2018 the disease 
caused over 770,000 
deaths globally. The disease can be transmitted from an HIV-positive pregnant 
woman to her child during pregnancy and childbirth about 15% - 20%, and 5% - 15% in 
breastfeeding. Awareness of HIV infection sexual risk behaviors and practices 
among male partners of pregnant mothers, therefore, plays a greater role in reducing 
new infection to them and the prevention of mother-to-child transmission of 
HIV. Methods: We conducted a retrospective case-control study involving randomly 
selected 314 pregnant women’s male partners attending PMTCT services in the 
Tabora Municipality for from September 2018 to August 2019. The enrolled study 
participants were randomly allocated to cases and controls and, SMS 
communication was sent twice weekly with 5 standard key messages from the PMTCT 
community guide and followed up. Results: The awareness that HIV infected mothers can transmit HIV infection in 
labor and during breastfeeding among their male partners was raised to 73.3% 
and 61.8% from 55.4% and 47.5% to study and control groups respectively 
following intervention. Similarly, men aged below 35 years from the study group 
showed an 18.3% change and a 35.4% change for those who were over 35 years old 
compared to the control. The difference was tested and found to be 
statistically significant with a p-value 
of 0.001 (relative risk of 0.11) and less than 0.001 (relative risk of 0.07) respectively 
at a 95% confidence interval. This indicates that following SMS communication 
intervention men from the study group were less likely to practice HIV 
infection sexual risk behavior compared to the control group.


Introduction
In 2018, the Joint United Nations Programme on HIV/AIDS (UNADS) estimated that over 37.9 million people were living with Human Immunodeficiency Virus (HIV) and two-third of them were in the WHO African region indicating approximately 1 in every 25 people (UNAIDS, 2016(UNAIDS, , 2019. The HIV remained to be a global public health interest disease that contributed to over 770,000 deaths globally in 2018. The UNAIDS set a target of 90 -90 by 2030 referring that 90% of people should know their HIV status and 90% should be able to receive and use Antiretroviral therapy (ART). Studies indicate sub-Saharan Africa reached only 54%, due to several challenges towards reaching the target (Alary et al., 2017;Bain, Nkoke, & Noubiap, 2017). The Tanzania HIV survey in 2017 revealed that the HIV incidence for adults aged 15 -64 years in the country was at 0.29% (0.40% for women and 0.17% for males) which represents about 81,000 new infections (URT, 2018).
Subnational variation in HIV prevalence exists, as it was also reported that the HIV prevalence for Njombe was 11.4% while it was 0.3% for Lindi. In the Tabora region, the HIV prevalence was 5.1% [95% CI (4.2 -6)]. The viral load suppression among adults People Living With HIV/AIDs (PLWHA) is shown to be ranging from 66.8% in Kilimanjaro (high) to 29.1% in Arusha (low), and the viral suppression level in Tabora stands at 41.2% [0.6% -51.7%, 95% CI] (URT, 2018). Community surveys and studies are indicating high HIV prevalence compared to the National average, for example, female fisherfolks were found to have a prevalence of 14% (Kapesa et al., 2018).
The risk of people including male partners of pregnant women in getting HIV varies according to several risk factors including the type of exposure and behaviors. The Centre of Disease Control and Prevention (CDC) estimates in every 10,000 exposures, the risks of infection are high for blood transfusion (93%), receptive anal-intercourse (1.4%), and sharing needle-sharing for the drug addicts (0.6%). Matiko et al. conducted a study involving 408 people who inject drugs (PWID) in Tanzania and found out a prevalence of 11.3% for PWID in 2012 (CDC, 2014;Matiko et al., 2015). Nevertheless, a different similar study conducted in Mwanza indicated that the risks of HIV infection to PWID were mainly associated with their unsafe sex behaviors rather than expected parenteral transmission (CDC, 2014). Shayo  ported that female and male anal sex is less practiced in the 5 districts with high HIV prevalence in Tanzania and therefore maybe that behavior is less common in the country (Shayo et al., 2017 (Kashitala et al., 2015;Mmbaga, Kishimba, Mohammed, & Elias, 2017). A study by Moses et al. conducted in Nigeria involving 172 and revealed that among the participants, 44.2% were believing that HIV to be God's punishment and 3.5% believes that, the disease is of witchcraft origin (Moses, Chama, Udo, & Omotora, 2009). In that study, 12.8% of the study participants believed that HIV cannot be transmitted if withdrawal before ejaculation practices is being done during sexual intercourse, and 19.8% believed that, taking antibiotics after sexual intercourse prevents HIV infection. This study further indicates how appropriate awareness can reduce HIV infection to pregnant women and their expected babies.
Another qualitative study conducted in Malamulo Seventh Day Adventist hospital in Malawi alluded that, the high prevalence among women in the rural areas of Malawi was contributed by multiple sexual partners of either pregnant women or their male partners (Kasenga, 2010). In African communities, studies have shown that multiple partners have resulted from polygamy's cultural legacy and uncontrollable sexual urges among men (Agnarson, Strömdahl, Levira, Masanja, & Thorson, 2015;Shelton, 2009). The type of behavior and practices were also anticipated to exist in Tabora Municipality where the dominating tribes are Nyamwezi and Sukuma tribes exercising patriarchal values.
In South Africa, a study was conducted with 633 participants who had 1,051 partners, and it was revealed that there was a behavior for men who were introduced by their friends to young women who became their sexual partners regardless of their marital status (Fearon et al., 2019). In the same study, it was also shown that most of the old sexual partners out of their marriage were old friends. This study was a continuation of a study done before conducted by Jewkes et al. (2006) with an enrollment of 1275 volunteers which indicated that  (Ajzen, 2011;Hurling et al., 2007). In this study, the intervention group received the solutions to plan weekly exercise sessions with the mobile phone of perceived barriers and reminders through emails. The study revealed that, there were an increase in the higher level of moderate physical exercise to study group when compared to the controls (study group: −2.18, SD = 0.59; control group: −0.17, SD = 0.81; P = 0.04). It is therefore evidenced that, SMS communications to pregnant women's male partners may increase their skills, ideation as well as reinforced to change their HIV risk behaviors (Ajzen, 1985(Ajzen, , 2011. The behavioral decision-making process related to these behavioral theories depends on the perceived risk, probability of a loss, and the importance of a loss. These behavioral change theories are insisting the confidence in making an intention to change or in undergoing the stages of behavior change as key factors for changes to happen (Ajzen, 2011). In our study, we explored the importance of using technology (SMS messaging) to hasten the behavior change process.
There was low awareness of male partners of pregnant mothers to HIV risk behaviors in the Tabora Municipality this contributes to increased risk of mother-to-child transmission of HIV after being infected by male partners (URT, 2017(URT, , 2018. It was also viewed on the notion that, among African society perception that, child care is the primary role of a mother in contrary to the Convention on the Rights of the Child (UN, 1989).

Methods
A retrospective case-control study comparative study was conducted on know- The following were the selected five-key messages which were communicated twice weekly to enrolled men partners of pregnant women: 1) AIDS is caused by HIV. AIDS can be prevented and there are medicines to help PLWHA but there is no cure.
2) HIV can be transmitted in the following ways: • Having unprotected sex with an HIV infected person (not using condoms), this includes vaginal or anal sex.
• From an HIV-infected mother to a baby during pregnancy, childbirth or breastfeeding who are not taking ARVs 3) Attend with your spouse at ANC/PMTCT clinics to get information from health workers 4) Get HIV counseling and testing with your partner 5) Make sure your spouse/pregnant mother is delivering at a health facility Specific messages communicated (entire package in 5 SMS) in Swahili language as follows: 1) Ugonjwa wa UKIMWI husababishwa na Virusi vya Ukimwi (VVU), ugonjwa huu hauna tiba ila una kinga na kuna dawa za kusaidia wagojwa wa UKIMWI.

Enrolment of Study Participants, Sample Size Estimation, and Study Participants
Probability sampling was used to enroll participants in this study using the list of PMTCT pregnant mothers' attendee's phone numbers in Tabora Municipality health facilities as a sampling frame. The cases and control were then randomly selected by the computer using numerical control. Both selected cases and con-

Study Procedures and Data Analysis
This study involved 314 male participants who were randomly enrolled and of which 157 men received SMS of 5 key messages for PMTCT/HIV obtained from the PMTCT and HIV community guidelines. We collected data from the online questionnaires administered to both cases and controls by 5 trained research assistants using the Open Data Kit (ODK) platform. The principle researcher monitored regularly through observing the geocoordinates of data collections and through physical supportive supervision. Also, a supervisor was deployed to monitor physically research assistants.
All collated data were validated and cleaned to remove data duplicates and or missing data for the important variables before analysis. High-quality data generated were subjected to the analysis to assess the impact of communicated SMS on the knowledge, awareness, and practices for the cases and the control groups.
Descriptive analysis of the enrolled participants was made on their demographic characteristics such including age, socioeconomic status, and education. Statistical analyses were performed using Stata software (StataCorp, College Station, TX, USA, version 12) and p-values < 0.05 were considered significant.
Human immunodeficiency infection sexual risks practices were measure from respondents' sexual acts. Sexual acts were rated on two response options: 1 score

Results
The study enrolled 314 men partners of pregnant women reflecting 109% of the expected 288 study participants who were randomly selected and distributed to cases and control groups. group was 62.4% and 26.1% respectively. In addition, the enrolled staff who were permanently employed by the Government was 29.3% for the study group compared to 9.6% in the control group.
The study found out that 90.2% of the enrolled pregnant women's male partners reported being informed on PMTCT services by health workers, 85.5% by informed by their friends and 59.4% through radio spots before the SMS communication intervention. Also, 54.3% reported to be informed by television spots, 23.2 through posters, and 14.7% through road shows as highlighted in Figure 1.
However, following SMS communication intervention 34% of all enrolled male partners-participants from the study group preferred SMS communication on PMTCT services/HIV information compared to 30.5% who reported continued to prefer to be communicated by health care workers. Furthermore, 25.2% of enrolled pregnant women male partners preferred PMTCT/HIV communication by friends and no one preferred posters (Figure 2).
Awareness of HIV infection prevention is very important for people and especially the pregnant women's male partners for not performing HIV infection risk behaviors. This study revealed that awareness to the enrolled study group of pregnant women's male partners that condoms prevent HIV infection increased from 75.9% to 80.8% and from 70.1% to 75.5% for the control group after SMS communication intervention as shown in Table 2.     The study found out that there was an increase in the awareness that there are ARV/PMTCT drugs in health facilities which when taken appropriately can prevent transmission of HIV infection from pregnant women to their babies increased from 53.6% to 66.7% after SMS communication only to the study group. For the importance of male testing during pregnancy, the understanding increased from 83.2% to 89% and from 80.4% to 86.6% for the study and control groups respectively. There was a statistical significance between the study and control group on the awareness that HIV infected mothers can transmit HIV infection to their babies after SMS communication with 13.2 percentage change, a p-value of 0.015 (95 C.I ranges 2.5% to 23.4%). Also, there was a significant change between study and control after SMS communication on the fact that HIV infected pregnant women can infect their babies with HIV (if not taking PMTCT drugs) during labor with 13.4 percentage change between the groups, a p-value of 0.012 (95% C.I ranges from 2.9 to 23.4%).
The study indicated that before SMS communication to the enrolled pregnant women's male partners, 56.5% of the study group aged less than 35 years conducted HIV infection risk behavior compared to 32.7% from the control groups of the same age group. There is a significant variation for the two groups with 23.8 percentage change and a p-value of 0.002 (Table 3(a)). Furthermore, the enrolled pregnant women of married male partners from the study group (42.8%) reported practicing more HIV infection risk behaviors compared to 30.9% of married men in the control group. This is not only an 11.9% variation but also with a significant change with a p-value of 0.045. The other parameters in this study indicated a significant variability between the study and control group regarding practice HIV infection risk behaviors before SMS communication was  the men with less than 2 live children, men completed with primary school or informal education, and men with 1 concurrent sexual partner within 6 months prior to the study (Table 3(a)). Following the implementation of SMS communication to the study group on 5 key messages as explained in the methodology, men in the study group for all age groups categories, were less likely to practice HIV infection risk behavior compared to control group with a significant change of p-value of 0.001 for the under 35 years and less than 0.001 for men above 35 years. The study also observed variation in all categories except married men categories for the study group, there was a significant reduction (a p-value of less than 0.05) for performing HIV infection risk behaviors, such categories of the study were the number of live children, education levels, occupational status, and partners in the six months before the study (Table 3(b)).
The study also identified in Table 3(c) that the risk of men aged less than 35 years in the study group was reduced with RR of 0.11 (95% C.I ranges from 0.02 to 0.76) compared to the controls. The risk of unmarried men was also lower with RR of 0.42 (95% C.I. ranges from 0.08 to 2.35) compared to the controls. The risk for the study group was also reduced to men with primary or informal education with RR of 0.26 (95% C.I. ranges from 0.04 to 1.50), with secondary education with RR of 0.10 (95% C.I. ranges from 0.02 to 0.72), self-employed 0.30 (95% C.I. ranges from 0.09 to 1.02) and with more than 1 concurrent partner with RR 0.16 (95% C.I. ranges from 0.04 to 0.61).  Figure 3 below.

Discussion
The results from the study indicate age variability between the study and control group as shown and the median age was 35 years and 44 years for the study and control group respectively. In making consideration of HIV epidemiology in Tanzania (Ngilangwa, Ochako, Mboya, Noronha, & Mgomella, 2015). The findings from the study indicate that if the SMS communication could be implemented in the rest of districts in Tanzania especially the Tabora Region, it will accelerate both the milestones for the PMTCT elimination and HIV infection reduction goals. It was also indicated that there is the likelihood of unmarried male partners in the study group to practice HIV risk behavior compared to unmarried male partners in the control group the value of p was less than 0.001 (95% C.I) before implementing SMS communication. The study also indicated that there is a reduction of risk for both unmarried and married men in the study group following SMS communication in the study group compared to controls with a relative risk of 0.42 for unmarried and 0.05 and they are all statically significant. The finding is similar to a study conducted in Arusha and Kilimanjaro which picked up unmarried, alcohol consumption and living in urban areas (Arusha) to be among the risk for performing HIV infection sexual risk behavior (Ngilangwa et al., 2015).
In addition, the male partners with 2 or more live children in the study group were less likely to practice HIV sexual risk behaviors p-value < 0.001 (95% C.I).
This may be as people are becoming more responsible are also fearing to practice HIV infection sexual risk behavior. Similar findings were reported in South Africa when men with children were not avoiding performing HIV infection risk behavior but also if they were diagnosed HIV positive were likely to take ARV to prevent the sufferings of their children (Koo, Makin, & Forsyth, 2013). The study also indicates the variability of health workers believes and clients' needs in terms of where to receive health information. Even though a large-scale study may provide more information, however, the study highlighted how simple common interventions can be well accepted by community and change to healthy behaviors and practices. Sexual practices and behaviors contribute to HIV new infection and study opened eyes that interventions to prevent mothers and children HIV infection may be well addressed if male partners (formal husband and concurrent sexual partners) are also addressed similarly to studies results in Northern Tanzania and in Kenya (Kaiser et al., 2011;Ngilangwa et al., 2015).

Conclusion
The study results indicated a clear effect of SMS communication intervention in reducing HIV sexual risk behavior in pregnant women male partners especially in the study group which was tested and found out to be statistically significant.
The study also suggests an improved approach for awareness creation from traditional methods of mass media and posters. We also suggest that health promotion guidelines for PMTCT and HIV/AIDs should be reviewed to include adaptive innovation and strategies like SMS communications in the national PMTCT and HIV/AIDs programs.