Fertility Desire and Associated Factors among Sexually Active HIV Positive Women at Felege Hiwot Referral Hospital, Bahir Dar, Ethiopia: Institution-Based Cross-Sectional Study

Background: Many HIV-positive women continue to want to have children in spite of knowledge of their HIV status and the risk it imposes. This study aimed to assess the level of fertility desire and associated factors human immune-deciency-positive women at Felege Hiwot referral Hospital. Methods: An institution based cross-sectional study was conducted among randomly 308 HIV positive reproductive age women at Felege Hiwot referral Hospital from June 01 st -30 th , 2018. Data were collected using structured interviewer-administered questionnaire. Data were analyzed using SPSS version20.Descriptive, bivariate and multivariate logistic regression statistics were carried out. Statistical signicance was declared at P -value <0.05. Results: Out of 308 of participants, 45.5% reported a desire to have children in the near future. Being sexually activity (aOR=2.62, 95%CI; 1.58- 6.73), family planning use (aOR=7.62, 95%CI; 3.23-17.98), consistent condom use (aOR=2.6, 95%CI; 1.37- 4.94), and absence of children (aOR=4.43, 95%CI; 1.85-10.64) were predictors of fertility desire. Conclusion: A considerable proportion HIV-positive women in the study area reported to have children in the future. Integration of family planning and ART services, attention to sexually active HIV positive women and supporting HIV positive women with unmet need of family planning could have signicant impact.

2017.However,progress is so far slower than what is required to reach in the 2020 milestone of less than 500,00new infections [2].
Sub-Saharan African (SSA) contributed 76% of the total HIV-infected people, 76% of the total new infections, and 75% of the total HIV/AIDS related deaths in 2015 [1].As one of the Sub-Saharan country the case in Ethiopia is not different. Ethiopia has a generalized epidemic with an estimated adult (15 years and above) prevalence of 0.9 percent.HIV prevalence ranges from 4.7% in Gambella regional state to below 0.1% in Somali regional state [3,4]. In 2017, there were about 615,000 people living with HIV, of whom 65% were female, in the country. Of these, almost one -third (30%) were from Amhara Regional State. Prevalence of HIV among women of reproductive age is 1.2% which is higher than men (0.6%) in the same age [4].Accordingly, to end HIV/AIDS epidemic in 2030, intervention to meet the need of reproductive health; speci cally the need related to fertility desire of these population group must be prioritized as many HIV-positive women continue to want to have children despite knowledge of their HIV status [5][6][7][8][9].Consequently, the desire of HIV-positive women to have children has important implication for transmission of HIV to sexual partners and newborns [6].
Evidences from different studies indicate that HIV-positive women continue to desire more children in the future. In studies conducted in Rwanda, Cameroon, and Democratic Republic Congo, for example, the proportion of HIV-positive women who desired to have children in the future were 45.8%, 65% and 83% respectively [10][11][12] .In Ethiopia, a few studies have revealed different level of fertility desire among women living with HIV.Studies in Addis Ababa, West Ethiopia, and Jimma, Ethiopia indicated that 40.8% ,42% and 46.8% of HIV positive women had fertility desire respectively [8,9,13].Studies in different settings have pointed out different factors which determine the fertility desire of HIV-positive women. For example, family planning use in HIV positive women was found to be signi cantly associated with fertility desire [6,13,14].An increased fertility desire is found to be associated with absence of living children in several studies across the world [5,6,[14][15][16][17]. However, a study in the United States showed a decreased fertility desire among women of HIV-positive with absence of living children [18].In Ethiopia, although few studies are conducted in different health care institutions regarding fertility desire of HIVpositive women and associated factors with it, the information is inadequate. Furthermore, the extent of fertility desire among HIV-positive women receiving HIV care and how reproductive issue are affected by sociodemographic characteristics and reproductive related factors is not well understood in the study area. This research is therefore aimed to assess level of fertility desire and factors associated with it among HIV-positive women at Felege Hiwot Referral Hospital, Northwest Ethiopia. Accordingly, the ndings of this study could be used to improve reproductive health of HIV-positive women, decrease vertical transmission of HIV to children, decrease maternal and child mortality, and ultimately eliminate Mother-to-Child Transmission of HIV.

Study design, setting and period
A cross-sectional study was conducted at Felege Hiwot referral Hospital ART clinic which is located in Bahir Dar city, capital city of Amhara regional State, from June 01 th -30 th , 2018. The city is located 564 km northwest from Addis Ababa, the capital city of the Ethiopia. The hospital provides different inpatient and outpatient services to the population of the region including ART and family planning services for clients referred from all district and zonal health care facilities in the Regional State.

Source population
All people living with HIV/AIDS who were attending Felege Hiwot referral Hospital, ART clinic.

Study population
All randomly selected sexually active HIV positive between 15 and 49 years old available at the time of study and having at least one visit.

Exclusion criteria
Women with surgical removal of uterus and unable to communicate for any reason Women who were pregnant and infertile.
Women below age of 18 without guardian or family (consent issue).
Sample size and sampling procedure A total sample size of 308 was estimated using single population formula; assuming a p-value p=76% from a previous reported prevalence in Addis Ababa Ethiopia [19]; [19]; a 95% con dence interval (CI); a tolerable margin of error (d)=5% ;Zα⁄ =1.96 and 10% Non-response rate.
A total of 11,120 people living with HIV were enrolled in ART of the hospital during the study period. Out of this, a total of 5644 sexually active HIV positive women aged 15-49 years old in ART clinic were enumerated and sampling frame was developed. Finally, study participants were selected using simple random sampling method i.e., lottery method.

Variables
The dependent variable was self-reported fertility desire. Independent variables were age of woman, religion, educational level, number of living children, drinking habit, partner HIV status, HIV status disclosure to partner, having stable sexual partner, contraceptive use experience, condom use, and having sexual activity.

Operational de nition
Fertility desire: This refers to when women have a desire to have a child in the near future [20]. Consistent condom use: Use of female or male condoms in all vaginal sexual relationships with casual and/or steady partners [20].
Family planning use: Respondents who responded positively being asked whether they are currently using any method of contraception to delay or avoid pregnancy.
Sexually active: A client who had sexual intercourse in the last six months.
Data collection and data quality assurance Trained data collectors collected the data using structured interviewer administered questionnaire which we adopted from earlier studies [19,21]. The questionnaire had the following contents; Sociodemographic characteristics, reproductive and sexual characteristics, HIV and fertility related factors.
First, we prepared English version of the questionnaire then language experts translated it to local language (Amharic) and back to English to check consistency and accuracy. We recruited four diploma nurses and two Bachelor of Science nurses as data collectors and supervisors. Assigned supervisors closely managed the data collection process. In addition, we gave training for data collectors and supervisors on the overall content of questionnaire and data collection process for two consecutive days.
We have carried out pre-test study on 5% of the calculated sample size of women in Adiss Alem Hospital ATR clinic which is out of study area and readjusted the questionnaire. We also reviewed medical records of participants to get clinical information on anti-retroviral treatment and CD4 count.

Data processing and analysis
The collected data were checked manually for completeness and consistency. Then, coded and entered into EPI Info version3.5.3 and transferred to SPSS version 20 for analysis. We used descriptive statistics to summarize socio-demographic characteristics of participants and proportion of fertility desire. To show factors associated with fertility desire, we carried out binary logistic regression analysis at two levels. We performed bivariate logistic regression to each independent variable with the outcome variable. Then, variables with p-value < 0.05 were included in multivariate analysis. We measured strength of association using odds ratio, and 95% con dence intervals. Finally, we declare statistical signi cance at p-value <0.05.

Result
Socio-demographic characteristics of participants A total of 308 sexually active reproductive age HIV positive women participated in the current study making a response rate of 100%. The mean age of women was 26 years (SD±4.08). Majority of participants were aged between 25 and 34 years. Regarding to religion, more than three fourth of participants 241(78.2%) were Orthodox Christians. Only one -fth 66(21.4%) participants were married while more than one-third 114(37%) were single. All participants were urban dwellers (not mentioned in table) .Less than half of participants 144(46.7%) attained primary school and above; about one fth of participants (18.1%) were commercial sex workers and majority of participants (84.4%) had drinking habit (Table1).

HIV, Reproductive and Fertility Related Pro le of Participants
Prevalence of fertility desire was 45.5 %( 95%CI: 39.6-51.0).Out of 308 participants, recent CD4 count of 142 participants was accessed. Out of this, 47.2% had CD4 count of more than 500cells/m 3 . More than half (55.8%) had sex in the last 6 months, 118 (38.3%) had used any method of contraception, and 146(44.2%) were practicing consistent condom use. More than one-fourth 80(26%) of participants had used dual contraceptive method. The most commonly used contraceptive was injectable 97 (52.2%) followed by implant 35 (18.85) and pills 30 (16.1%).Majority 224(72.7%) had known their partners' HIV status and 182(81.3%) reported that their partners were sero-positive. Almost one-fth (21.5%) had disclosed their status to their sex partner. About one-fth participants 67(21.8%) reported that they faced child death. Most of participants 251(81.8%) had one and above children. Almost one-third of participants75 (24.4%) had changed their sexual partner since their diagnosis (Table2).

Factors Associated With Fertility Desire
After adjusting for potential confounders, family planning use, consistent condom use, sexual activity, sexual partner change, living children were found to be signi cantly associated with fertility desire in multivariate analysis at p-value <0.05.The study revealed that participants who were using family were seven times less likely to desire children compared to women who were not using family planning (aOR=7.62, 95%CI; 3.23-17.98). Similarly, participants who were using condom consistently were two times less likely to desire children compared to their counterparts (aOR=2.6, 95%CI; 1.37-4.94).Furthermore, participants who did not have children were four times less odds of fertility desire (aOR=4.43, 95%CI; 1.85-10.64) compared with participants who had one or more children unexpectedly.
However, participants who had sexual activity six months prior to the study were three times more likely to desire fertility (AOR=2.62,95%CI; 1.58-6.73) compared to their counter parts(see Table3).

Discussion
Reproductive health issue including fertility desire among women enrolled in HIV care and treatment programs in the study area has important implications for the health of women, their partners, and their infants. This study revealed that the prevalence of fertility desire among HIV positive reproductive age women in the study area was 45% .The nding was exactly consistent with the study done in Tigray Region, Ethiopia which has reported that 45% of women living with HIV/AIDS had desire to have children in the future [20].The nding was also consistent with studies done in Rwanda (45.8%), Jimma Town, South West Ethiopia (46.8%) respectively [10,13]. On the other hand, the nding of the current study was lower than evidences from Addis Ababa, Ethiopia(54.6%) [22],Cameroon (83.3%) [12],Democratic Republic Congo(64%) [11]. However, prevalence of this study was higher than the study done in Hosanna Town ,southern Ethiopia(36.3%) [6], Ababa, Ethiopia(40.8%) [8],Central Brazil (25.9%) [23],Kenya (26.4%) [24].The difference could be explained in terms of dissimilarities in sociodemographic of study populations and cultural concerns about having family and fertility desire.
According to the result of this study, factors that associated with fertility desire in bivariate analysis were age of participant, drinking habit, current family planning use, consistent condom use, sexual activity at the time of study, sexual partner change, and number of living children. Even though the abovementioned factors have demonstrated signi cant association with fertility desire; age, sexual partner change, and drinking habit lost their signi cance in multivariate analysis (see Table3).
Participants who were using family planning during the study were found to have fertility desire about seven times less likely than their counterparts. The nding is in line with studies done in Jimma town, Ethiopia and Harari Regional State, Eastern Ethiopia but in contrast with the study done in Hosanna Town, Southern Ethiopia. [6,13,14]. This could be also due to the in uence of health care providers counselling and awareness of study participants about vertical transmission of HIV.Increasing awareness about vertical transmission of HIV among HIV-positive clients is powerful means of fostering family planning use to reduce HIV transmission [25].
Participants who were using condom consistently in the last six months prior to the study were almost two times less likely to desire children compared to their counterparts. The nding is supported by G. J.Wagner et.al. which has reported that consistent condom use was greater among those with no fertility desire [26]. Actually, consistent condom use is one of the contraceptive method to avoid pregnancy. Use of male and female condoms could provide dual protection. It could protect acquisition or transmission of HIV and contribute to PMTCT of HIV [27].
Furthermore, participants who were sexually active in the last six months were three times more likely to desire children in the near future which is similar with studies done in Ethiopia and Tanzania [14,28]. This could be explained in terms of that being sexually active meant desiring pregnancy. When women has desire to become pregnant they engaged in sexual activity.
Unexpectedly, participants who had no child were four times less likely to desire children compared to participants who had one or more children. The nding is in line with study done in the United States [18] .Otherwise, it is in contrast with other several studies done in different areas of the world [5,6,[14][15][16][17]. This could be explained in terms of differences in sociodemographic, socioeconomic status, health condition, and culture between study populations. The level of fertility is partly in uenced by demographic, health status, prevailing social-economic conditions and cultural factor [29]. It could be also due to in this study, majority (81.8%) of participants had at least one child (see Table2).

Conclusion
The results of this study showed that a considerable proportion of women living with HIV/AIDS in the study area reported the desire to have children in the near future. Family planning use, consistent condom use, being sexually active, and absence living children were predictors of fertility desire. Therefore, Health care providers working in HIV care centers should discuss the issue of reproductive desire with HIVpositive women. Supporting HIV positive women with unmet need of family planning could have signi cant impact. In addition, addressing the issue of health concerns related to fertility desire and risky sexual behavior among HIV infected reproductive age women is essential as it could play a signi cant role in struggling to eliminate mother-to-child HIV transmission. Furthermore, it is important to integrate family planning services and HIV/AIDS care and treatment service.

Limitations
The rst limitation of this study is men were not included in the study in despite they have important role to play in deciding about fertility issues and family size. Critically ill women and pregnant and women were not also included in this study. Fertility desire among this women might be very different than women participated in this study. Since it is a single institution study we can't generalize it for the general population. As fertility issue is a sensitive topic, social desirability bias cannot be avoided. Another limitation for this study was reasons for fertility desire were not investigated.

Declarations
Ethics approval and consent to participate Ethical clearance obtained from Bahir Dar University, department of nursing research committee and college of health science institutional review board. Each study participant was adequately informed about the aim of the study and anticipated bene t and risk of the study by their data collector. Written consent sought from all study participants for protecting autonomy and ensuring con dentiality.

Consent for publication
Not applicable.

Availability of data and materials
The data of this study cannot be shared publically due to presence of sensitive (con dential) participants' information.

Competing interests
There are no competing interests among authors