At the end of 2014, 77% of HIV-infected pregnant women in the 21 Global Plan priority countries in sub-Saharan Africa were receiving antiretrovirals for their own health or for the prevention of mother-to-child transmission (PMTCT).1 Moreover, there has been a rapid shift to the 2013 WHO recommendations of lifelong antiretroviral therapy (ART) for all HIV-infected pregnant and breastfeeding women regardless of their clinical or immune status (Option B+).2 If implemented optimally, these guidelines could lead to mother-to-child transmission (MTCT) rates below 5% in a breastfeeding population and below 2% in a nonbreastfeeding population.3 Recent data from the Promise Study showed that with Option B+, a transmission rate as low as 0.5%–0.6% can be achieved by age 14 days.4
To ensure maximum benefit from ART for themselves and for PMTCT, HIV-infected pregnant women have to adhere to a continuum of services often referred to as the “PMTCT cascade.”5 The cascade includes attendance of regular clinic visits (at a minimum for ART refills), delivery in a health facility, and testing of the HIV-exposed infant at 6 weeks, at 9 months, and at the end of breastfeeding (between 18 and 24 months). Yet, data from early implementation of Option B+ in Malawi showed that about one in 6 pregnant women initiated on ART at antenatal care (ANC) registration do not return to the clinic after their initial visit.6 In addition, a recent meta-analysis of women lost to follow-up (LTFU) across the “PMTCT cascade” found that 49% of HIV-infected pregnant women are LTFU between ANC registration and delivery, and 34% of mother–infant pairs are LTFU within 3 months of delivery.7 Although the MTCT rate was 5% at 6 weeks in the 21 priority countries in sub-Saharan Africa in 2014, as a result of high rate of LTFU, it rose to 14% by the end of breastfeeding when many of the mothers are no longer on ART.8
Identification of effective and scalable strategies to improve retention along the PMTCT cascade and improve uptake of HIV services is at the forefront of PMTCT optimization efforts. In a recent randomized controlled trial, we found that providing newly diagnosed HIV-infected women with small and increasing cash incentives, on the condition that they attended scheduled clinic visits and received available PMTCT services, significantly increased the proportion of women who were retained in care and received available PMTCT services through 6 weeks postpartum.9,10 The intervention choice was informed by an important finding in the behavioral economics literature that many individuals have present-biased preferences, whereby they place disproportionate weight on present costs and benefits relative to those in the future.11 Thus, consistent with the Health Belief Model's12 elements of “perceived risk,” “perceived benefit,” and “perceived barriers,” it was hypothesized that although there may be strong health incentives (eg, desire to ovoid MTCT) for HIV-infected pregnant women to engage in HIV care and treatment, women may place substantial weight on the immediate costs of PMTCT, which include the financial and opportunity costs of attending regular clinic visits, and heavily discount the delayed benefits.13,14 Verification of the mechanisms by which cash incentive interventions increase desired PMTCT outcomes would (1) enhance our understanding of how these interventions work and (2) improve our ability to design and scale up similar interventions for PMTCT and other health and social issues.
To elucidate the mechanisms by which a cash incentive intervention increases desired PMTCT outcomes using a modified Health Belief Model as conceptual framework (Fig. 1), we examined the associations between perceptual factors (perceived severity of HIV, perceived MTCT susceptibility/risk, perceived PMTCT benefit, perceived barriers, and self-efficacy to access and use PMTCT services) and retention in care. We also assessed the potential interactions between (1) perceptual factors found to be associated with retention in care and (2) the effect of a conditional cash incentive intervention on the same outcome.
Study Design and Study Population
This is an analysis of data from a randomized controlled trial that was conducted in Kinshasa, the Democratic Republic of Congo. Between April 2013 and August 2014, 433 newly diagnosed HIV-infected women, pregnant for ≤32 weeks, registering for ANC at 1 of 89 clinics during the enrollment period were recruited and randomized to receive either the standard of care or the standard of care plus small and increasing cash payments. These payments started at $5 and increased by $1 each month on the condition that the woman attended scheduled clinic visits and provided a blood sample for a CD4 count, accepted referral for ART if referred, delivered in a health facility, and at 6 weeks postpartum provided a blood sample for early infant HIV diagnosis. A detailed description of the intervention, and results of the analysis of primary outcomes showing that the intervention resulted in significantly higher retention along the PMTCT cascade and better uptake of available services, has been reported.
Perceptual Factors and Measurements
At enrollment into the study, which occurred between 2 and 4 weeks after registration for ANC and HIV diagnosis, participants were interviewed using a structured questionnaire to collect information on perceived HIV severity, perceived MTCT susceptibility/risk, perceived PMTCT benefits, perceived difficulties and barriers accessing PMTCT services, and perceived self-efficacy to access and adhere to clinic visits and PMTCT services (Fig. 1 for the conceptual model). All survey questions for the perceived factors had Likert-style scale response options (eg, “very likely,” “somewhat likely,” “somewhat unlikely,” “very unlikely,” “don't know,” and “refuse to answer”). The questions, response options, and the final coding of those responses are summarized in Table 1.
The main outcome considered in this analysis was LTFU—not in care at 6 weeks postpartum with unknown whereabouts (participants who were dead, or who did not return to the clinic because they experienced a poor pregnancy outcome but could still be reached, were classified as not in care but not as LTFU).
Identification of Perceptual Factors Associated With LTFU
Using data from participants randomized to the control group (to avoid any mixing of effect), we examined the associations of the measured perceptual factors with LTFU by comparing the proportion of participants LTFU by 6 weeks postpartum across categories of perceptual variables. Multivariate logistic regression models including all of the perceptual factors were used to identify factors that were independently predictive of LTFU and to estimate odds ratios (OR) and 95% confidence intervals (CI) assessing the strength of their associations with the outcome. Perceptual variables found to be significantly associated (alpha = 0.20) with LTFU after adjustment for all other factors considered and the wealth index were retained for the evaluation of interaction with the cash incentive intervention. The wealth index was included in the model because subgroup analysis of the primary trial outcomes showed that the effect of the cash intervention was particularly strong for poorer women.9 The wealth index was determined using principal component analysis. The model for this factor analysis included years of education, average number of household members per room (an indication of crowding), number of sleeping beds in the household, types of household water source (communal or private pipe), cooking fuel type (electrical stove or wood/charcoal), and ownership status for several durable assets. The first component explained 27% of the variability in the data. The factor score was categorized into upper (2 richest quintiles), middle (third quintile), and lower (2 poorest quintiles) categories.
Modification of the Effects of Identified Perceptual Factors on LTFU by the Cash Incentive Intervention
To assess potential synergism or antagonism between perceptual factors found to be significantly and independently associated with LTFU and the cash incentive intervention, log-linear risk models were used to estimate the risk difference (RD) measuring the individual and joint effects of the conditional cash incentive intervention and the perceptual factor on the outcome. The likelihood ratio test was used to assess RD modification, and excess risk due to interaction [interaction contrast (IC) and 95% CI] was used to assess for synergism or antagonism. IC is the difference between the observed RD comparing the doubly exposed category with the doubly unexposed category, and the expected RD comparing the same 2 groups assuming additivity (without the interaction term) should be equal to zero in the absence of RD modification. All analyses were completed using SAS 9.3 (SAS Institute Inc., Cary, NC). All tests were 2-sided and used a 0.20-significance level unless otherwise indicated.
The study is approved by the Institutional Review Board of the Ohio State University and the Ethical Committee of the Kinshasa School of Public Health. This parent trial is registered at ClinicalTrials.gov, number NCT01838005.
Overall, most participants perceived HIV as a “very serious” health problem for themselves (78.5%, 340/433) and for their infants (84.3%, 365/433) (Table 2). The perception of MTCT susceptibility/risk was also high; 83.6% (362/433) of participants responded that it was “very likely” that they will pass HIV to their baby if they did not take any HIV drug. Similarly, the perceived benefit of PMTCT medication was also very high; 84.8% (367/433) of participants believed that it would be “very unlikely” that they will pass HIV to their baby if they take all their HIV drugs. Over 90% (391/433) of participants stated that they were “very confident” to be able to come to the clinic every month to collect their HIV drugs and get checked; accordingly, 83.8% (363/433) anticipated that it would be “very easy/not at all difficult” for them to come to the clinic. The anticipated difficulties included “transportation” for 15.9% (69/433) and “not having money” for 11.6% (50/433) of respondents. Virtually all participants, 98.4% (426/433), stated that it was “very likely” that they would take all of their medication every day without missing any doses.
Association Between Perceptual Factors and LTFU
In bivariate analysis, perceived MTCT susceptibility/risk was the strongest predictor of LTFU; participants who perceived passing HIV to their baby was “very likely” if they did not take their HIV drugs were strongly and statistically less likely to be LTFU at 6 weeks postpartum compared with those who did not perceive the risk of transmission as “very likely” (OR, 0.44; 95% CI: 0.19 to 0.99) (Table 3). Participants who perceived HIV to be a serious health problem for their infants (OR, 0.50; 95% CI: 0.21 to 1.14) were also less likely to be LTFU at 6 weeks postpartum compared with women who did not perceive HIV as a serious health problem. In contrast, compared with participants who responded “nothing” to the question “What makes it difficult for you to come to the clinic to receive your treatment?,” those who responded “not having money” were about twice as likely to be LTFU (OR, 1.90; 95% CI: 0.75 to 4.84).
In a multivariate logistic model, women who perceived it as “very likely” that they will transmit HIV to their baby if they did not take any HIV drug remained statistically less likely to be LTFU [adjusted odds ratio (aOR), 0.39; 95% CI: 0.14 to 1.08]. Although not statistically significant, HIV-infected women who perceived HIV to be a serious health problem for their baby were also less likely to be LTFU at 6 weeks postpartum (aOR, 0.55; 95% CI: 0.18 to 1.65). Accordingly, perceived lack of benefit of PMTCT medication was associated with LTFU. Women who responded that it was “very unlikely” that they will pass HIV to their baby if they take all their HIV drugs were twice as likely to be LTFU (aOR, 2.01; 95% CI: 0.64 to 6.30). Believing that “not having money” will make it difficult to come to the clinic was the only anticipated barrier associated with an increased tendency of LTFU (aOR, 1.83; 95% CI: 0.67 to 5.02).
Modification of the Effects of Identified Perceptual Factors on LTFU by the Cash Incentive Intervention
Table 4 presents the RDs and 95% CIs for LTFU for each exposure (perceptual factors and cash incentives) individually and jointly, as well as the excess RDs due to interaction (IC) and their 95% CIs. Perceiving HIV as a “very serious” health problem for their baby vs. not was associated with a 13% decrease in the proportion of women LTFU by 6 weeks postpartum (RD, −0.13; 95% CI: −0.30 to 0.04) among women who did not receive the cash incentive intervention. Among women who did not perceive HIV as a “very serious” health problem for their baby, receiving the cash incentive was associated with a 23% (RD, −0.23; 95% CI: −0.41 to −0.04) reduction in the proportion of participants LTFU. The effects of the 2 exposures were not additive for participants who perceived HIV as a “very serious” health problem and received the cash incentive (RD, −0.20; 95% CI: −0.37 to −0.03). The antagonistic excess RD due to interaction (IC, 0.16; 95% CI: −0.04 to 0.35). Similar antagonism was observed for the perception that it will be “very likely” to pass HIV to the baby if one did not take any HIV drug.
Anticipating that “not having money” will make it difficult to come to the clinic to receive treatment, although not statistically significant, was associated with a 12% (95% CI: −7% to 30%) increase in the proportion LTFU at 6 weeks postpartum among participants who did not receive the cash incentive (compared with participants who reported “nothing” when asked about barriers to clinic attendance). Providing women in this group with a cash incentive perfectly antagonized the negative effect of “not having money” on LTFU (IC, −0.12; 95% CI: −0.35 to 0.10).
If ART is started at an early stage of pregnancy and continued uninterrupted throughout breastfeeding, the risk of HIV transmission from mother to child decreases to almost zero.4 But without novel strategies to increase retention in and adherence to PMTCT services, the goal of an AIDS-free generation is unlikely to be achieved. In this study, we examined the mechanisms through which a cash incentive intervention recently tested by our group in a randomized controlled study increases retention in care.
As hypothesized during design of the study, our analysis showed that HIV-infected pregnant women who perceived HIV as a “very serious” health problem for their baby (perceived severity of HIV), and perceived that they will “very likely” pass HIV to their baby if they did not take their drugs (perceived benefit of PMTCT), were more likely be have been retained in care through 6 weeks postpartum. Our results are in agreement with previous studies showing that better PMTCT knowledge15 and greater understanding and belief in the effectiveness of ART16 are positively associated with higher adherence to PMTCT services. Interestingly, the observed antagonism between the cash incentive intervention and (1) the perception of HIV as a very serious health problem for the baby and (2) believing that it will be very likely that HIV will be transmitted to the baby if one does not take any HIV drugs suggests a competitive mechanism of action as represented in the conceptual framework. Consistent with the prospect theory,17 this observation suggests that interventions that combine (1) relatively short-term cash incentives to keep asymptomatic pregnant women who are likely to be LTFU and (2) an effective education program to improve women's understanding of HIV risk for themselves and for the baby may achieve meaningful and sustained retention.
Lack of financial support is a known risk factor for poor ART adherence among HIV-infected women.18 Our finding that provision of a small cash incentive to those women completely reverses the negative effect of “not having money” on LTFU is a confirmation that there are real financial barriers to accessing HIV care, at least in maternal and child health clinics where PMTCT programs are implemented despite HIV care itself being provided free of charge. In fact, in most countries in sub-Sahara Africa, hospitals and clinics are funded on a fee-per-service basis, and HIV-infected women have to be able to pay for their registration for antenatal visits and delivery costs, and register for the well-baby clinic visit in the postpartum period to be able to continue to have access to the clinic at each of those steps.19 The magnitude and structure of those fees vary by clinic and might explain why some women switch clinics during and after pregnancy.9
The observation of a negative effect of “not having money” on retention also has implications for determination of the magnitude of the cash incentive. It has been suggested that incentive magnitude be dependent on the opportunity cost incurred by the behavior change. A systematic review of the use of financial incentives in treatments for obese and overweight persons found evidence of very weak trends in favor of using amounts greater than 1.2% of personal disposable income.20 However, although the optimal size and structure of incentive payments remain important questions to be addressed empirically, this finding suggests that in areas where fee for service is in place, they must be taken into account when determining the opportunity costs of attending clinic visits.
This is the first study to examine the mechanisms by which a cash incentive intervention achieves the desired outcomes of retention in care among newly diagnosed HIV-infected pregnant women. The study used data from a randomized controlled trial to test hypotheses that informed the design of the intervention. Despite the potential risk of misreporting the perceptual factors considered in this analysis, the fact that interviews were conducted before participants were assigned to randomization groups eliminates any potential for differential misreporting by study group or outcome level. However, because the study was not powered for this specific analysis, some estimates are unstable with wide CIs.
In conclusion, our analyses show that cash transfers improve retention in PMTCT services mainly by mitigating the negative effect of not having money to come to the clinic. In designing financial incentive interventions to improve retention in PMTC services, the minimum amount proposed should be at least equal to the cost of attending clinic visits.
The authors are grateful for the participation and time of the mothers and infants in the study, the time and efforts of the personnel of the participating clinics, the technical support of Dr. Landry Kiketa and Mrs. M.T., the data collection and data entry contributions of Josée Nlandu Babela, Valerie B. Chalachala, Fanny Matadi, Espérance Mindia, and Georges Kihuma Nganguli, and the support of the administrative teams of the Ohio State University, the University of North Carolina at Chapel Hill, and the Kinshasa School of Public Health.
1. Joint United Nations Programme on HIV/AIDS (UNAIDS). 2014 Progress Report on the Global Plan Towards the Elimination of New HIV Infections Among Children by 2015 and Keeping Their Mothers Alive. Geneva, Switzerland: UNAIDS; 2014.
2. World Health Organization. Consolidated Guidelines on the Use of Antiretroviral Drugs for Treating and Preventing HIV Infection: Recommendations for a Public Health Approach. Geneva, Switzerland: World Health Organization; 2013.
3. Ciaranello AL, Perez F, Keatinge J, et al. What will it take to eliminate pediatric HIV? Reaching WHO target rates of mother-to-child HIV transmission in Zimbabwe: a model-based analysis. PLoS Med. 2012;9:e1001156.
4. Fowler MG, Qin M, Shapiro D, et al. PROMISE: Efficacy and Safety of 2 Strategies to Prevent Perinatal HIV Transmission. Conference on Retroviruses and Opportunistic Infections (CROI), Seattle, WA, February 23–26, 2015.
5. Stringer EM, Chi BH, Chintu N, et al. Monitoring effectiveness of programmes to prevent mother-to-child HIV transmission in lower-income countries. Bull World Health Organ. 2008;86:57–62.
6. Tenthani L, Haas AD, Tweya H, et al. Retention in care
under universal antiretroviral therapy for HIV-infected pregnant and breastfeeding women (Option B+”) in Malawi. AIDS. 2014;28:589–598.
7. Sibanda E, Weller I, Hakim J, et al. The magnitude of loss to follow-up of HIV-exposed infants along the prevention of mother-to-child HIV transmission continuum of care: a systematic review and meta-analysis. AIDS. 2013;27:2787–2797.
8. Joint United Nations Programme on HIV/AIDS (UNAIDS). 2015 Progress Report on the Global Plan Towards the Elimination of New HIV Infections Among Children and Keeping Their Mothers Alive. Geneva, Switzerland: UNAIDS; 2015.
9. Yotebieng M, Thirumurthy H, Moracco KE, et al. Conditional cash transfers
and uptake of and retention in prevention of mother-to-child HIV transmission care: a randomised controlled trial. Lancet HIV. 2016;3:e85–e93.
10. Yotebieng M, Thirumurthy H, Moracco KE, et al. Conditional cash transfers
to increase retention in PMTCT
care, antiretroviral adherence, and postpartum virological suppression: a randomized controlled trial. J Acquir Immune Defic Syndr. 2016;72 (suppl 2):S124–S129.
11. Laibson D. Golden eggs and hyperbolic discounting. Q J Econ. 1997;112:443–478.
12. Rosentock IM. The health belief model: explaining health behavior through expectancies. In: Glanz KL, Rimmer FM, BK, eds. Health Behavior and Health Education. San Francisco, CA: Jossey-Bass;2002:39–62.
13. Shafir E. Decision Biases, Cognitive Psychology. In: Neil JS, Paul BB, eds. International Encyclopedia of the Social and Behavioral Sciences. Oxford, United Kingdom: Pergamon;2001:3296–3300.
14. Green L, Fry AF, Myerson J. Discounting of delayed rewards: a life-span comparison. Psychol Sci. 1994;5:33–36.
15. Mirkuzie AH, Hinderaker SG, Sisay MM, et al. Current status of medication adherence and infant follow up in the prevention of mother to child HIV transmission programme in Addis Ababa: a cohort study. J Int AIDS Soc. 2011;14:50.
16. Durante AJ, Bova CA, Fennie KP, et al. Home-based study of anti-HIV drug regimen adherence among HIV-infected women: feasibility and preliminary results. AIDS Care. 2003;15:103–115.
17. Kahneman D, Tversky A. Prospect theory: an analysis of decision under risk. Econometrica. 1979;47:263–291.
18. El-Khatib Z, Ekstrom A, Coovadia A, et al. Adherence and virologic suppression during the first 24 weeks on antiretroviral therapy among women in Johannesburg, South Africa—a prospective cohort study. BMC Public Health. 2011;11:1–13.
19. Manthalu G, Yi D, Farrar S, et al. The effect of user fee exemption on the utilization of maternal health care at mission health facilities in Malawi. Health Policy Plann. 2016;31:1184–1192.
20. Paul-Ebhohimhen V, Avenell A. Systematic review of the use of financial incentives in treatments for obesity and overweight. Obes Rev. 2008;9:355–367.