Family Planning and HIV in Postpartum Women in Mbarara

HIV status and availability of ART have been found to widely influence the desirability and expectations to have children among childbearing women living with HIV (Nieves et al., 2015; Kaida et al., 2011). Up to 85% of the pregnancies within 3 years following ART are often unwanted, often leading to illegal abortions and maternal deaths (Desgrees-Du-Lou et al., 2002).

Uganda's HIV epidemic has declined over time to about a sero prevalence of 7.2% and this decline has been attributed in part to improved prevention behavior like condom use, reduction in multiple concurrent partnerships and other strategies to prevent unwanted pregnancies. Specifically, HIV prevalence is estimated at 7% among women attending antenatal clinics in Uganda (UN, 2010). Whereas ART availability has improved general health outcomes through viral load suppression and immune reconstitution, there is still a high unmet need for family planning among HIV positive women in Uganda (Snow et al., 2011), where modern contraceptive prevalence is at 33% in the general population despite the widespread promotional messages across the country. This continues to expose HIV positive women to increased risks of unwanted/unplanned pregnancies, perinatal HIV transmission, pregnancy complications, and increased economic burden of care for self and others among others.

The high average fertility rate of 5.9 children per woman in Uganda may further complicate the immunity and wellbeing of the already burdened HIV positive women and their families even amongst those already accessing HIV care (Muyindike et al., 2012; Homsy et al., 2009).

Supporting HIV positive women to delay or prevent an unwanted pregnancy may improve women's health through family planning choices, regular health reviews, and support. A cohort study that was done in Mbarara amongst HIV-infected women documented 50% of the enrolled women having no desire for a (another) child during the 4 years since ART initiation, 51% had a serodiscordant partner, with only 45% using effective contraceptive method. Key predictors of contraceptive use were family planning goals within the DYAD. One of the key predictors of effective contraceptive use was male partner involvement and CD4 count. The WHO recommends dual contraception (use of condoms and a hormonal or permanent method) to prevent both HIV transmission and unwanted pregnancies (WHO,2014).This prospective intervention study aims at providing information on whether continuous family planning accessibility and support has a measurable impact on pregnancy intentions and contraceptive usage among postpartum HIV positive mothers delivering at MRRH, SW Uganda.

General objective To test the effect of a family planning availability and support intervention on pregnancy intentions and use of effective contraception among recently postpartum women living with HIV who delivered at MRRH.

Specific objectives

1. To assess the influence of the intervention on pregnancy intentions among recently postpartum HIV positive mothers delivering at MRRH, SW Uganda one year after the intervention.
2. To assess the impact of the intervention (voucher) on contraceptive uptake among recently postpartum HIV positive mothers delivering at MRRH, SW Uganda
3. To assess the influences of the intervention (SMS +Partner involvement) on contraceptive uptake among recently postpartum HIV positive mothers delivering at MRRH, SW Uganda
4. To identify factors that influence pregnancy intentions and contraceptive use
5. To assess the influence of the intervention on the incidence of unintended pregnancy among recently postpartum HIV positive mothers delivering at MRRH, SW Uganda
6. To explore decision making challenges and perceptions of family planning among HIV positive women.
7. To assess the role of sexual partners on pregnancy intentions and effective contraceptive use

Methods This will be a prospective randomized controlled trial of 2 arms. I. The investigators intend to enroll a total of 320 HIV positive women delivering at MRRH. The participants will be equally randomized into the intervention arm (Family planning support) and control group (no family planning support, except the routine counseling following discharge). These women will be followed for 1 year. Their male partners will also voluntarily choose to participate in the study or not. Eligibility will also require access to a mobile phone.

II. Intervention group. Following delivery, the participants will be given a voucher to freely access any of the five family planning options of their choice. A voucher will also be given to the spouse/sexual partner due to its identified effect on family planning utilization (Ashraf et al., 2014). Although family planning is already free, there are usually stock outs. The voucher will therefore be an incentive to access the family planning timely (maximum of 1 hour waiting time) and continuously from the study in case of stock outs. These include; condoms (40 condoms), injectables, contraceptive pills, copper IUD and contraceptive implants. The women will be contacted at 6 weeks postpartum for initiation of family planning by a study nurse well trained in specialized family planning care and support for HIV positive women if they have not done so yet. The choice of family planning will be entirely up to the participants. For those women who select oral contraceptive pills, women will be sent daily [adherence support] reminders for the first 4 months, then weekly reminders (via SMS) for the next 4 month and then monthly for the next 4 months. This level of SMS support has been found to have a positive impact on adherence (Haberer et al., 2016). Sexual partners/regular spouses of women in intervention arm will also receive these reminders weekly [but not daily or monthly] throughout the study. The reminders will also be sent monthly if one chooses an injectable contraception [consider whether it's the 2-month or 3-month injectable] for the 12 study months. Daily reminders will also be sent for women who choose male or female condoms. Routine reviews on family planning will be done for women in the intervention arm alongside their routine visits at the HIV clinic or post-natal PMTCT visits. Interviews with study participants will be done at baseline, 6 and 12 months postpartum. The spouses/ sexual partners will be contacted, enrolled and interviewed at baseline, 6 months and at 1 year.

III. Control group: These will be counseled on family planning at discharge and enrolled. They will be interviewed at baseline, 6 months and 1 year. Data on reproductive history, HIV/AIDS experience, socio-sexual relationships, fertility aspirations, use and knowledge of family planning, decision making, and perceptions on Family planning, alcohol use, food security, education, socio-economic, side effects or undesirable effects, economic and social support will be documented.

Natural family planning methods like lactation amenorrhea, withdrawal methods and others have been presented with mixed literature, painting a grim picture on their effectiveness as standalone contraceptive methods especially among HIV positive women. All women are advised to exclusively breast feed for 6 months while on ART (MoH, 2015). Because hormonal contraception inclusive of estrogen interferes with breast milk production among lactating mothers, progestin-only methods will be recommended to the women.

IV. Randomization:

Consecutive numbers ranging from 001 to 320 will be computer generated to indicate the 2 main groups into which mothers can be randomly assigned and enrolled. These consecutive numbers will be labeled on top of the opaque envelopes. To be eligible, mothers must be HIV positive, have had a delivery within the last 48 hours, above 18 years of age, anticipating to begin family planning at 6 weeks postpartum, must be mentally stable and qualify for any family planning method available. Once eligibility of a mother is confirmed, they will be invited to participate in the study.

The aim of the study and details of the procedures to be involved in the trial, potential side effects and therapeutic benefits will be explained before randomization occurs. Once the mothers consent to participate in the study, a study number will be allocated by the nurse research assistant (who will be recruited to work on the ward/post natal clinic for this study) by taking the next in a series of similar opaque envelopes provided to conceal allocation of groups. These opaque envelopes will be labeled with computer-generated list of numbers with group allocation (either Intervention group or Control Group) of the mothers randomized in blocks of 20. Nurse research assistants will be blinded to the group allocation until eligibility and study participation is confirmed. The research assistants will also be blinded to hypothesis of the study.

A blood sample will be drawn at baseline to confirm the HIV status and or do a CD4 cell count in case it is unknown at the time of enrollment. This will be done to support women enroll for the recommended Option B+ in case the women are not yet lifelong ART].

A different study nurse will be enrolled and trained to specifically collect data from participants. The socio-demographic data and obstetric characteristics will be documented. Other documented potential covariates of contraceptive use like participant age, primary partner's age, primary partner's HIV status, personal and partner fertility desires, number of children, most recent CD4 cell count, Efavirenz-containing ART Regimen (due to its association in decreasing levels of hormonal contraception thus recommending dual methods and its teratogenicity concerns- although these concerns have been found to be exaggerated), socio-economic status, body mass index, education level, time on ART . Prior contraceptive use experiences, fertility desire/ pregnancy intentions or aspiration will be measured using the CDC pregnancy Risk Assessment Monitoring System Instrument (Ahluwalia et al., 1999)

Effective contraception use will be defined as consistent use [both self report and observational chart review from study nurse at the family planning clinic] of a family planning method (including consistent condom use) , including dual contraception (use of condoms and another contraceptive method-hormonal or permanent) for HIV-sero discordant couples/partners (for participants who know partner's status). A primary partner will be defined either as a regular spouse, who is also a regular sexual partner or the most recent sexual partner if no main partner is named. These sexual partners will also be enrolled and interviewed at baseline and at 1 year. In case a consented mother ends up with a serious adverse event, the time and indication will be documented and reported. In case a woman changes the contraceptive method, the reasons for the change will also be documented.

Data Preparation All data will be cross checked for completeness before entry. Data will be coded and entered independently into the excel sheet and exported into STATA Version 12 for statistical analysis. For each variable, a reference category will be identified.

Data Analysis Plan Data analysis will be by intention-to-treat (ITT) consisting of all mothers randomized and thus supposed to be treated (Lesaffre, 2008). Descriptive statistics will be used to describe key characteristics of study participants. Different variables will be explored for normal distribution. Continuous variables will be compared with binary outcomes of contraception effectiveness and pregnancy intentions using a t- test. Selected demographic, clinical, post-natal factors will be compared for differences and similarities in the three groups. Crude relative risks will be derived. Relative risks with 95% confidence Interval (two tailed) will be calculated and used to compare contraception effectiveness and pregnancy intentions in all groups. Factors associated with effective contraception use and pregnancy intentions will be explored using univariate analysis and for results whose p value ≤ 0.10, a multivariate analysis will be done to rule out confounding factors.

Data Reporting The means will be presented with their standard deviations. Relative risks and odds ratios will be presented with their 95% confidence Intervals. Two-tailed statistical analysis will be used and statistical significance will be defined at level of p < 0.05.