World Health Organization (WHO) promotes prevention of HIV infection, unintended pregnancies, perinatal HIV infection and providing care and support for HIV positive mothers and their families [1]. Adherence to antiretroviraldrugs is necessary to prevent drug resistance and achieve the prevention of HIV transmission to children through Mother to Child Transmission (MTCT). Important strides have been made in recent years in the PMTCT of the HIV. Yet, despite these advances, approximately 15% of all new cases of HIV infection have been diagnosed in children in developing countries [2]. Between one half and two thirds of children who become infected with HIV die before their second birthday [3].

In 2010, only 48% of HIV-positive pregnant mothers received antiretroviral medicines to prevent MTCT and around 390,000 children aged less than 15 became infected with HIV [4]. Almost all of these infections occur in low- and middle-income countries, and more than 90% are the result of MTCT during pregnancy, labor and delivery, or breastfeeding [5]. Beside this, Ethiopia is one of the top 20 countries affected by HIV in the world [6]. Only 26% of the pregnant women were tested for HIV. Fifty three percent of known HIV-positive mothers and 48% of known HIV - exposed infants have received ARV prophylaxis and the estimated ARV among HIV positive mother coverage was only 11.6% and 8.4% of their babies [7]. One of the reasons could be low male partner involvement in PMTCT care and support.

Furthermore, WHO’s 2010- 2015 PMTCT strategic vision emphasizes the need to involve male partners in scaling up PMTCT services in Sub-Saharan Africa (SSA) [6]. Studies have also shown that the utilization and adherence of PMTCT services by the pregnant women is influenced by both factors related to the health system such as accessibility of Voluntary Counseling and Testing (VCT) services and individual factors such as fear of disclosure of HIV results and lack of male partner support [8,9]. Therefore, the objective of this study was to examine the role of male partner’s involvement in improving adherence of HIV positive women to PMTCT care and support.

Study setting and population

Cross-sectional study, which contains quantitative methods of data collection, was conducted from May to June 2013. The study was carried out in health centers of Tigray region, northern Ethiopia, among HIV positive pregnant women attending ART. There are about 16 governmental, four private hospitals and 256 health centers in Tigray region. The PMTCT program is operational in hospitals and health centers [10]. Mothers receive PMTCT packages free of charge on a monthly basis.

Sampling

The sample size was estimated a single population proportion formula with the assumption of 95% confidence level, 5% margin of error and male involvement of 26% in the Uganda study [11]. To compensate the non-response rate, 10% of the estimated sample was added up to make the total sample of 326 HIV positive pregnant women.

Systematic random sampling was used to select mothers participating in the study. There were a total of 60 public health centers supported by ENHAT CS in the different zone of the region, from 20 health centers were selected by simple random sampling technique. Mothers were selected at equal interval. Mothers who didn’t satisfy the inclusion criteria were excluded and were replaced by next mothers in the list. However, mothers who refused to participate in the study were considered as non-respondent.

Data collection procedure

Data were collected using structured and pre-tested questionnaire by face-to-face interview. It was first prepared in English, translated to Tigrigna- the local language- and then translated back to English to check consistency of the questions. The questionnaire constituted questions that provide information on socio-demographic and economic characteristics of respondents, adherence of mothers to PMTCT drugs, health and health care related factors, male partner factors, mother’s behavior, knowledge and attitude of mothers towards PMTCT drugs. The questionnaire was adapted from literature reviews and considering the local situation of the study areas and subjects [7, 11,12]. Before the actual data collection, the questionnaire was pretested on 16 HIV positive women in Quiha health center which was excluded from the study. Based on the pre-test, modifications on the translation, time needed to complete interview and a number of data collectors needed were estimated.

Twenty clinical nurses for data collectors and six degree holder nurses for supervision were recruited and trained for two consecutive days on purpose, data collection tools and interview techniques. Each interview was conducted in a place where the woman feels free to express her feelings and ideas. The filled out questionnaires were checked for completeness and errors by the supervisors on a daily basis.

The composite measure of male partner involvement towards PMTCT care and support was measured by the total number of 10 questions with a minimum score of 0 and maximum of 10. Finally, it was categorized into three “better”, “moderate” and “less” male partner involvements by calculating and categorizing of the responses of the respondents out of 10. With correct answers for at least seven questions, male involvement was considered better. Male involvement was considered moderate if the correct answers were 4-6 and correct answers less than 4 were designated as less male involvement. Adherence to PMTCT care and support was measured by three items: adherence to PMTCT prophylactic drugs, ANC visit and their plan of delivery. Then it was dichotomized to adherent and non-adherent. The respondent was taken as adherent to PMTCT care and support, when they take greater than or equal to 95% of PMTCT prophylactic drugs, misses only one appointment day for ANC follows up and has a plan to give birth at the health facility.

For the study, the composite measure of knowledge towards PMTCT was measured by the total number of correct answers to 6 items of knowledge with a minimum score of 0 and maximum of 6. Finally, it was categorized into three “high”, “moderate” and “low knowledgeable” by calculating and categorizing of the responses of the respondents out of 100%. Women were said “highly Knowledgeable” if their answers 80% and above distinct characteristics of PMTCT care and support from knowledge questions. In addition, “moderate knowledgeable” those who knows 60 up to 79% distinct characteristics of PMTCT care and support from knowledge questions. They were labeled as “low knowledgeable” if they answer less than 60% characteristics of any of the PMTCT care and support.

For analysis items the attitude of the respondents towards PMTCT was grouped into three “strongly agrees” and “agree”, as “agree”, “neutral” as “neutral” and, “disagree” and “strongly disagree” as “disagree”. Finally by adding the overall variables the total score of HIV positive pregnant women’s attitude towards PMTCT was done based on mean. Then, they were said to have Positive Attitude, if their scores above mean to the correct answers from attitude measuring PMTCT care and support questions and having a negative attitude if they score mean and below the mean.

Data analysis

Data were entered into and cleaned in Epi-DATA version 3.1. Further analyses were done using SPSS 20 for windows (SPSS Inc. Version 20, Chicago, Illinois). Descriptive analyses were computed for the level of male involvement in PMTCT care and support and other important characteristics of the study participants. We computed crude odds ratio (COR) using bivariate logistic regression. To identify adjusted effect of predictors on adherence to PMTCT care and support, a multiple logistic regression model was employed. The effect of each predictor was adjusted for the rest of the variables to get net effect. The sample effect sizes were depicted using odds ratio (OR) and effects of the factors in the community was estimated using 95% confidence interval of adjusted OR. Multi collinearity among independent factors was checked using VIF. A P-value of less than 0.05 was set for the significance of estimates.

Ethical consideration

The study protocol was approved by Research and ethical review committees in Mekelle University, College of health sciences. Written consent was obtained from each participant to ensure their voluntariness to participate in the study and they were told that all have a right to withdraw at any time or to put an end for single question, segment of questions or refuse to participate at all with no negative repercussions.

Characteristics of the study subjects

A total of 299 HIV positive pregnant mothers were included in the study making the response rate to 91.7%. Slightly lower than half of the mothers were in the age range of 25-29 years old. A higher proportion of the mothers (66.4%) were resident of urban and 86.3% of the respondents were followers of Orthodox Christianity Pertaining to educational status, 120(40.1%) of respondents were unable to read and write. Nearly to 39% of the respondents didn’t have job. They were limited to indoor activities. Of the total respondents, 198(66.2%) and 96(32.1%) had radio and television, respectively. One hundred sixty five (55.2%) of the respondents had household monthly income of 51.3-102.56 $ with a median amount of 120$ (Table 1).

Table 1: Socio demographic and economic characteristics of respondents and their partner, Tigray, northern Ethiopia, 2013.

Clinical characteristics of respondents

The majority, 181(60.5%), of the respondents was using ART. Of those, 101(33.8%) of them were currently on the ART regimen of D4TSampling 3TC-EFP. A majority, 190 (63.5%) of the participants were in WHO stage of III during they start treatment. While 65.6% of them had CD4 count less than 200/mm3. During the study period, however, 65.6% of the women were at WHO stage T1 and 97.3% had CD4 count over 200/mm3. Pertaining to their anthropometric composition during the study period, 44.6% of the women had a Body Mass Index(BMI) of less than 18.5 kg/m2 (Table 2).

Table 2: Clinical characteristics of respondents, Tigray, northern Ethiopia, 2013.

Reproductive history of the respondents

Average age at marriage was 18.45 yrs (± 3.2 yrs) and age at first birth was 20 yrs (+ 2.5). While 64.2% of the women got married below the age of 18, 85.1% gave birth to their first child at 18 or beyond. On average, each woman had 2.1 children (+ 1.8). The birth spacing was less than 3 years among 61.6% of the women and over half of the women (57.5%) intend to have more children in the future, the majority of them within three years. Eleven percent of the women had child death in which 89% occurred within one year of birth. In addition, 11.1% of the women experienced at least one abortion (Table 3).

Table 3: Reproductive history of HIV positive pregnant women, Tigray, northern Ethiopia, 2013.

Health and health related characteristics of respondents

With regard to HIV testing, 186(62.2%) of the women got tested in VCT center. Fifty one percent of the women didn’t have any discussion with their partners whether they needed to get tested for HIV or not. As such, 222(73.6%) of them got tested alone. Pertaining to pregnancy care, 30.4% of the pregnantwomen attended ANC two times while 26.4% of them didn’t visit for pregnancy care at all. Even among the ANC users, the visit was not consistent and the frequently mentioned reason has been illness. As asked about their preference of place of delivery for their future births, 86.3% of them had an intention to delivery at health facility (Table 4).

Table 4: Health and health related characteristics of respondents towards PMTCT care and support, Tigray, northern Ethiopia, 2013.

Information regarding to composite measure of HIV positive married women’ knowledge towards PMTCT care and support show that, 100(33.4%), 116(38.8%) and 83(27.8%) had high, moderate and low knowledge about PMTCT care and support, respectively. Moreover, information related to composite measure of attitude, 152 (50.8%) of the respondent had positive attitude towards PMTCT care and support, whereas the remaining had a negative attitude, 147(49.2%).

Partner involvement in PMTCT care and support

Nearly eight out of ten women disclosed their HIV status to their partner (77.3%) in which majority of these (74.5%) disclosed their status before they started ARV drugs in PMTCT service. Their most reason for disclosure was the need of support from their partners. Not only 25.5% hadn’t disclosed their HIV status, 81% of them didn’t have any intention to, mainly with the fear of divorce. Out of the women who knew their partner’s HIV status, 91.1% of them were positive for HIV (Table 5).

Table 5: Partner related factor of HIV positive women towards PMTCT care and support, Tigray, Northern Ethiopia, 2013

Regarding to male partner involvement in PMTCT care and support, 139(59.9%) of their partners participated in decision making and 173(74.6%) of them supported their wives in domestic activities. Half of the women discussed on use of condom with their partner. As an indicator to the partners support, 17.3%, 32.2% and 41.4% of them knew the name of the ARV, dosage and types of regime, respectively. In addition, 104(44.8%) of the partners had accompanied their wives to ANC, 133(59.3%) knew the appointment date for ANC and 119(51.3%) used to discuss with their wives on the prophylaxis and advices they received in ANC. Over fifty percent (53.9%) of the women also received financial support from their partner (Figure 1).

Figure 1: Percentage respondents who had male involvement to adherence of HIV positive pregnant women to PMTCT care and support in Tigray Region, 2013.

The composite measure of male involvement showed that 196 (65.6%) of the respondents had less male involvement, while 28(9.4%) and 75(25 %) had moderate and better male involvement, respectively.

Adherence of respondents towards PMTCT care and support

The composite measure of adherence to PMTCT care and support showed that only 254 (84.9%) of HIV positive women were adherent to PMTCT care and support as gauged based on three attributesadherence to PMTCT prophylactic drugs, ANC visit and their plan of delivery

Among the adherent women, 81.3%, 93.5% and 82.1% were among the women who disclosed their HIV status and employed and who knew their partners’ HIV status, respectively. As partners increased their educational status, they were more likely to support their wives and thus contribute to adherence of ARV in PMTCT. This was supported with evidence that the adherence rate was 20.1% among the women partnered with illiterate husbands as compared to 27.6% in women whose partners completed their secondary school. As adherence is contrasted with the level of male involvement, 51% of the women were adherent to PMTCT care and support among those whose partners had better involvement in care and support. However, only 24% and 25% of the women whose partners’ involvement was moderate and less were adherent to PMTCT care and support.

Determinants of Adherence to PMTCT care and support

Multivariate logistic regression was done to assess the predictors of adherence towards PMTCT care and support among women attending PMTCT. Accordingly, attitude towards to and knowledge on PMTCT, discussion with husband on HIV test, male partner involvement, access to information through radio and sufficient birth space were significant predictors of adherence to PMTCT.

HIV positive pregnant women with better partner involvement in PMTCT care and support had 8 times more odds to adhere to PMTCT (AOR=8.4; 95% CI:4.2, 12.9). Having better knowledge on PMTCT was associated with better adherence to PMTCT care and support (AOR=6. 2; 95% CI: 3.1, 9.3). Likewise, women with positive attitude were at 8 times higher odds to adhere to PMTCT as compare to their counterparts (AOR=8.2; 95% CI: 4.3, 12.6). The odd of adhering to PMTCT was 8 times higher among women who had 3 years or beyond as compared to women with insufficient birth spacing AOR=8.2; 95% CI: 3.8, 14.1). Women with exposure to radio messages were more likely to adhere to PMTCT (AOR=12.0; 95% CI: 5.8, 15.3). The odds of adhering to PMTCT was 12 times higher among women who had an open discussion with their partner on HIV testing (AOR=12.0, 95% CI: 6.2, 15.3) (Table 6).

Table 6: Predictors of adherence to PMTCT care and support among Respondents, Tigray, northern Ethiopia, 2013

The adherence rate of mothers towards PMTCT care and support was 84.9%. Attitude towards to and knowledge on PMTCT, discussion with husband prior an HIV test, male partner involvement, access to information through radio and sufficient birth space were significant predictors of adherence to PMTCT care and support.

Adherence to PMTCT is low in the current study, though it is comparatively higher than estimate reported in Zambia [13]. This study and other studies show that the adherence is influenced both by factors related to the health system such as accessibility of VCT services, and by individual factors such as fear of disclosure of HIV results, lack of male partner support, fear of domestic violence, abandonment and stigmatization [8,9].

Education is good intervention to improve health seeking behavior and health status of one’s population. As such, women with secondary education were more likely to adhere to PMTCT care and support. This result was consistent with findings in Uganda [11] but different as compared to a study in Zimbabwe [14]. The discrepancy could be accountable for the differences in the nature of the cross sectional which doesn’t sample the study participants based on the different characteristics.

Unlike to study in Nigeria [15], male involvement in PMTCT care and support in this study was higher. This could be attribuTable to the integrated community interventions including PMTCT services in Ethiopia through the aid of Health Extension Program. Male partner involvement was predictor for adherence to PMTCT and this was consistent with a study done in Tanzania which indicates that women who disclosed their HIV status were significantly more adherent to prophylaxis in the pre-delivery period than women who did not [16]. This association shows that male partners who get involved in PMTCT care and support could have better understanding and awareness towards the treatment. Hence, women how get psychological and financial support would be more adherent to PMTCT.

In the current study, 65.6% of the women received less male involvement in their PMTCT care and support. Similar findings were also reported in studies done in Tanzania, Nepal, South Africa and Malawi [16-19]. The low level of involvement could be attributable to the less awareness and education, culture and tradition of the society in which males don’t care on their partners. Similarly, the reasons mentioned for the less involvement of male partners in the other studies were unawareness to the importance of PMTCT care and support, environmental factors and feeling of shy. Moreover, as the educational level of male partners increases, the level of involvement in PMTCT care and support of their wives increases. However, the educational level of the population in Ethiopia is very low which could affect the male involvement negatively.

Knowledge is powerful to utilize health care service. Women with better knowledge and understanding that PMTCT can prevent HIV transmission to their baby are more likely to adhere to the PMTCT care and support. That’s, utilizing and comply with the treatment protocols of PMTCT without the knowledge of its advantages is unlikely. This finding was consistent with other studies, which showed that women with inadequate knowledge were 3.5 times more likely to non-adhered as compared to those with adequate knowledge [20].

Open discussion with partners on the need of HIV testing is associated with good adherence of mothers to PMTCT. Similar result is reported in a study conducted in Tanzania [21]. Discussion creates favorable condition for understanding and supporting each other. If the couples discuss on the need of HIV testing, that means they are ready to accept the test result and they are more likely to adhere to the PMTCT treatment.

Radio is one of the channels that health promotion is broadcasted through. Particularly, radio programs transmit messages through dramas and best experiences which have positive influence on the mothers to seek for the treatment. Thus, women who have exposure to radio messages are more likely to retain messages about the need of HIV testing, PMTCT care to prevent HIV transmission to baby, and adhere to the care.

The adherence rate of mothers towards PMTCT care and support was fair. Attitude towards to and knowledge on PMTCT, discussion with husband on an HIV test, male partner involvement, access to information through radio and sufficient birth space were significant predictors of adherence to PMTCT.

The authors would like to thank ENAT-CS/MSH Ethiopia for funding the research. We are also grateful to Mekelle University and Tigray Regional Health Bureau for facilitating the study.