Two-thirds of people living with HIV/AIDS reside in sub-Saharan Africa, 60% of whom are women.1 This pattern is reflected in Kenya, where 60% of the estimated 1.4 million HIV-infected adults are women.2 Before the era of combination antiretroviral therapy, lack of treatment increased morbidity and early mortality among HIV-infected patients.3,4 HIV has now become a chronic illness due to availability of combination antiretroviral therapy.5–7 With HIV-infected women living longer, there is increased need to address their family planning (FP) needs. The World Health Organization (WHO) and the United Nations regards FP as a cost-effective HIV prevention strategy and an important prong to Prevention of mother-to-child transmission of HIV/AIDS, hence their recommendation for the integration of HIV and FP services.8–11 This recommendation was based on a cochrane review of literature on reproductive health and HIV linkages, which found that, integrating FP and HIV services was feasible.12 The problem facing the majority of HIV care programs, in resource-poor settings however, is how to successfully provide reproductive health services including FP in a feasible, sustainable, and cost-effective manner.
The WHO defines modern contraception to encompass the use of any of the following methods for pregnancy prevention: female sterilization, oral contraceptive pill, intrauterine contraceptive device, implants, injectable depo provera, or condoms.13 The uptake of modern contraceptives in Kenya among married women is 39%14 and the unmet need for FP is estimated at 24%2 in the general population and approaches 30% among HIV-infected women.15 It is believed that the greater unmet need in HIV-infected women is related to provider perception that infected women are unable to use many of the modern contraceptives available. Contrary to this perception, it is now known that with individualized care, HIV-infected patients can use any contraceptive method.16,17 Studies have found that when provided with access and information about FP, HIV-infected patients increase their use of contraception.18–23
The WHO, World Bank, and the European Union support the integration of FP services and HIV care.8–11 Despite this support, most HIV programs focus on HIV treatment, with little or no emphasis on FP services. Such integration is further impeded by funding restrictions24; separate funding for these 2 programs result in vertical organization of the 2 services. This undermines coordination between departments and limits providers' ability to address the contraceptive needs of HIV-infected patients.25
To address the reproductive health needs of HIV-infected women, the United States Agency for International Development—Academic Model Providing Access To Healthcare (USAID-AMPATH) Partnership (known as AMPATH hereafter) started a pilot program integrating FP services into one of its HIV clinics. We carried out an operations research study among HIV-infected women cared for within USAID-AMPATH partnership accessing integrated family planning services (IFP) and routine care (RC), to determine the impact of integrated FP services and HIV care on use of new contraceptive methods and pregnancy rates.
Study Site and Setting
The AMPATH program26–29opened its initial HIV care program in 2001. At the end of February 2009, the program was caring for more than 78,000 HIV-infected adult patients (70% female) in 23 Kenya Ministry of Health facilities across Western Kenya. This study was conducted at the AMPATH Centre (Main Clinic), Moi Teaching and Referral Hospital, Eldoret, Kenya. At the end of February 2009, there were more than 17,000 adult patients enrolled in the AMPATH Centre with 65% of them being female. Within the AMPATH Centre, there are 3 clinical teams caring for adult patients. These teams function as independent practices with patients enrolled for care in 1 of the 3 teams (I, II, and III) and consistently returning for care within the same clinical team.
Nonintegrated Care Model (RC)
In the original AMPATH care model, the only FP service offered routinely in the HIV clinic is condom counseling with the primary goal of reduction of HIV transmission and not family planning. Condoms are strategically placed in the waiting bay, check in/out rooms, and consultation rooms for patients to access. Patients who need FP services beyond condom access are referred to maternal child health and family planning (MCH/FP) clinic after their appointment in the HIV clinic. The MCH/FP clinic is administrated by the departments of reproductive health and pediatrics and the HIV clinic administrated by the department of internal medicine. These departments are all independently run.
In this model, it is the patients' responsibility to ensure that they attend their FP appointment after referral from the HIV clinic. Unlike HIV care, which is provided free of charge, services in the MCH/FP clinic require a patient's co-pay. Based on this model, 2 challenges were anecdotally observed. Patients who require FP at MCH/FP had an increased burden of hospital visits. Second, there was a relative underutilization of FP services by HIV-infected patients due to the fact that getting a FP appointment is patient dependent not provider initiated.
Family Planning Services and HIV Care Integrated Model
Integration of FP services into HIV care pilot study started in October 1, 2007, in the AMPATH center clinical team I. Teams II and III continue to offer original care model type of care described above. In the integrated model, nurses experienced in offering FP services were relocated to team I HIV clinic. The Reproductive Health room is integrated into the patient flow in the HIV clinic. Some degree of independence for both FP and HIV care is maintained to ensure that the focus and specific nature of these 2 service provisions is not weaken by a complete integration. The link between the 2 services forms the integrated nature of the 2 services run under the same in charge. Services that are integrated are as follows: routine offer of same-day “one-stop-shopping” appointments with a central check in/out, the use of same patient charts, and consistent messaging. Patients are counseled in a consistent and structured manner and are allowed to make informed choices on which modern FP methods to use. The study used the WHO definition of modern FP, which is the use of any of the following methods for pregnancy prevention: female sterilization, oral contraceptive pill, intrauterine contraceptive device, implants, injectable depo provera, or condoms.13 All FP methods except surgical sterilization are offered through the HIV clinic. Patients who request surgical sterilization are referred to MCH/FP clinic.
This is a retrospective cohort study approved by the Moi University/Moi Teaching and Referral Hospital Institutional Research and Ethics Committee and the Indiana University School of Medicine Institutional Review Board. For the purpose of this study, exposure to IFP was defined as care within the AMPATH clinic providing integrated FP and HIV services (team I). The RC group consisted of patients receiving care in modules without integrated FP services (teams II and III).
Adult HIV-infected women being cared for in the AMPATH Centre HIV clinic formed the study population. Patients were included in this analysis if they were aged 15–49 years at enrollment, enrolled on or after October 1, 2005, and attended 1 of the 3 clinical teams on or after October 1, 2007. Patients were excluded from this analysis if they had no follow-up visits after initial exposure or if they only had 1 visit on or after October 1, 2007.
Patients were also excluded if they were missing 1 or more of the following covariates collected at enrollment into AMPATH: years of school completed, sexual activity in the last 6 months, number of pregnancies, number of live births, number of children in household, and status of HIV disclosure. We have a relatively large cohort of several thousand patients; hence the main concern is possible bias than with precision in our inference of the effect of IFP service on contraceptive methods and pregnancy events. Therefore, we consider matching on baseline characteristics key in bias reduction. To match a patient under intervention and a control patient, we need to have complete data on baseline variables. This is the reason that we excluded patients with incomplete data.
Data Collection and Management
Data for this study was collected between October 1, 2005, and February 28, 2009. All data used in this study were derived from existing clinical data collected prospectively on encounter forms during routine patient care. All the data from paper forms were subsequently entered by trained data entry clerks into the AMPATH Medical Records System.
Continuous variables are summarized as mean (standard deviation) or median (interquartile range). Categorical variables are summarized as frequency and percentages. Primary outcomes include the following: corrected estimate of differences, incidence rate per 100 person-years, and the attributable risk between those accessing IFP and HIV care service and RC in new use of FP and new cases of pregnancies.
A match-with-replacement strategy30 was used to infer the effect of exposure (integrated FP and HIV services) by matching each exposed subject to 1 unexposed subject and each unexposed subject to 1 exposed subject on key baseline variables. We chose the method of matching to avoid parametric assumptions in conventional regression, so that bias induced by model misspecification can be reduced. The reduction in bias is at the cost of inflated variance. However, since we have a large cohort with several thousands of patients, the inflation in variance is not a big issue.
The Poisson model was used to estimate incidence rate of the primary outcomes. A P value (2-sided) of less than 0.05 is considered statistically significant. All analyses were conducted using SAS 9.2 (SAS Inc, Cary, NC) and R.
During the 16-month pilot period between October 1, 2007 (commencement of integrated model) and February 28, 2009, 4031 patients met the eligibility criteria for study; 1453 accessed integrated FP and HIV services (IFP) and 2578 accessed RC (Fig. 1). The mean (SD) length of follow-up was 342 (155) and 361 (147) days P value <0.001 for the IFP and RC groups, respectively. At enrollment, there were statistically significant differences between the IFP and RC groups in the following: age, years of school, sexual activity in the previous 6 months, number of pregnancies, number of live births, number of children in household, HIV disclosure to partner, HIV disclosure to family, and HIV disclosure to others. At the start of follow-up, there were statistical differences between the IFP and RC groups in the following: ARV status and those pregnant (Table 1). Though statistically significant, most of these differences are small and are unlikely to be clinically relevant.
After adjusting for potential confounders, among the IFP group, there was a 16.7% increase P < 0.001 [95% confidence interval (CI): 13.2% to 20.2%] in the incidence of new condom use, 12.9% increase P < 0.001; (95% CI: 9.4% to 16.4%) in the incidence of new FP use including condoms, 3.8% reduction P < 0.001 (95% CI: 1.9% to 5.6%) in the incidence of new FP use excluding condoms, and no statistically significant increase in new cases of pregnancies of 0.1% increase P = 0.9; (95% CI: −1.9% to 2.1%) (Table 2).
The incidence rate per 100 person-years of: new condom use was 53.4 (95% CI: 50.7 to 56.0) versus 36.9 (95% CI: 35.0 to 38.8) for the IFP and RC groups, respectively, with an attributable risk of 16.4 (95% CI: 11.9 to 21.0), new FP use including condoms was 58.1 (95% CI: 55.5 to 60.8) versus 44.7 (95% CI: 42.7 to 46.6) for the IFP and RC groups, respectively, with an attributable risk of 13.5 (95% CI: 8.7 to 18.3), new FP use excluding condoms was 4.8 (95% CI: 3.7 to 6.0) versus 7.8 (95% CI: 6.8 to 8.9) for the IFP and RC groups, respectively, with an attributable risk of −3.0 (95% CI: −4.6 to −1.4), and new cases of pregnancies was 8.2 (95% CI: 6.8 to 9.8) versus 7.0 (95% CI: 6.1 to 8.1) between IFP and RC groups with an attributable risk of 1.2 (95% CI: −0.6 to 3.0) (Table 3).
The proportions of ever: used condoms, used FP including condoms, and got pregnant are 41.4%, 47.9% and 7.2% for the 4031 among study participants with complete covariates included in this analysis. Corresponding proportions for the 1101 subjects with incomplete covariates values are 34.9%, 39.7%, and 5.7%. The difference is significant for ever-used condoms and ever-used FP including condoms.
Integration of FP into HIV care model as described in this article is in accordance with recommendations by WHO, World Bank, and the European Union.8–11 Previous studies demonstrated presence of policy commitment to such integration.31 Few care programs have integrated FP into HIV care due to funding restrictions, separate funding, independent running of both care programs and reluctance to integrate FP into HIV/AIDS funding.24,25,32,33 The AMPATH program has been able to demonstrate the feasibility of integrating FP services into HIV care, by providing same-day “one-stop shop care” service provision of both FP services and HIV care in its integrated model.
This study demonstrates that integration of FP services into HIV care is associated with an increase in new use of modern FP methods among HIV-infected women. Similar findings have been reported in the past.18–22 The following factors may have led to this increase as follows: sensitivity of health care providers to FP needs of this group of patients; “one-stop-shopping” same visit service provision; accessibility and reproductive health room being part of the patient flow.
Ultimately, the goal with increased FP is a decrease in unwanted pregnancy and improved pregnancy spacing. However, we found no reduction in the incidence of pregnancy in the group accessing integrated FP services. This finding may be a result of the following 2 reasons. First, the increase in FP use in this study was driven by an increase in condom use, which is a less effective contraceptive method and prone to reporting bias. The study did not evaluate adherence to the contraceptive methods. Second, patient desires regarding whether the pregnancy was planned or unplanned was not determined due to the retrospective nature of the study.
There are 2 major limitations to this analysis: (1) enrollment to the intervention group was not randomized hence the outcomes may have been affected by possible confounding due to unmeasured factors (retrospective nature of the study) such as number of times the patients had vaginal intercourse during follow-up, adherence to contraceptive methods, differences in HIV treatment modalities, and patient desires regarding whether the pregnancy was planned or unplanned; and (2) multiple filtering steps to exclude subjects could have introduced selection bias. Examination of the outcomes between those included and excluded shows that the 4031 subjects included in the analysis tend to have higher rates of condom and FP use. Therefore, the intervention effect found for this group of patients may represent the effect of those who are more engaged in family planning instead of the total patient population.
The strength of this analysis is that, it took place in the real-world setting of a busy HIV care and treatment program. Our pilot study clearly demonstrates the feasibility of integrating FP into routine HIV care within a busy program.
Integrating FP services into HIV care is feasible and significantly increases the uptake of modern FP methods. No reduction in the incidence of pregnancy occurred. HIV programs need to consider integrating primary care services such as FP into their program structure.
We wish to thank all the staff and patients of AMPATH Eldoret clinical teams I, II, and III, and Prof Susan Cu-Uvin her invaluable input.
1. UNAIDS. Report on the Global AIDS Epidemic. 2008 Geneva, Switzerland Joint UN Program on HIV/AIDS (UNAIDS) Available at: http://data.unaids.org/pub/GlobalReport/2008/jc1511_gr08_executivesummary_en.pdf
. Accessed March 22, 2011
2. National AIDS and STI Control Programme, Ministry of Health, Kenya. . Kenya AIDS indicator survey 2007 2008 Available at: http://www.aidskenya.org
. Accessed March 22, 2011
3. Maier M, Andia I, Emenyonu N, et al. Antiretroviral therapy is associated with increased fertility desire, but not pregnancy or live birth, among HIV+ women in an early HIV treatment program in rural Uganda. AIDS Behav. 2009;13(suppl 1):28–37
4. Gray RH, Wawer MJ, Serwadda D, et al. Population-based study of fertility in women with HIV-1 infection in Uganda. Lancet. 1998;351:98–103
5. Wools-Kaloustian K, Kimaiyo S, Diero L, et al. Viability and effectiveness of large-scale HIV treatment initiatives in sub-Saharan Africa: experience from western Kenya. AIDS. 2006;20:41–48
6. Marins JR, Jamal LF, Chen SY, et al. Dramatic improvement in survival among adult Brazilian AIDS patients. AIDS. 2003;17:1675–1682
7. Seyler C, Anglaret X, Dakoury-Dogbo N, et al. Medium-term survival, morbidity and immunovirological evolution in HIV-infected adults receiving antiretroviral therapy, Abidjan, Cote d'Ivoire. Antivir Ther. 2003;8:385–393
8. World Health Organization. . Reproductive health website: linkages between sexual and reproductive health and HIV Available at: http://www.who.int/reproductive-health/index.htm
. Accessed December 5, 2009
9. Reynolds HW, Janowitz B, Homan R, et al. The value of contraception to prevent perinatal HIV transmission. Sex Transm Dis. 2006;33:350–356
10. Reynolds HW, Janowitz B, Wilcher R, et al. Contraception to prevent HIV-positive births: current contribution and potential cost savings in PEPFAR countries. Sex Transm Infect. 2008;84(suppl 2):ii49–ii53
11. Sweat MD, O'Reilly KR, Schmid GP, et al. Cost-effectiveness of nevirapine to prevent mother-to-child HIV transmission in eight African countries. AIDS. 2004;18:1661–1671
12. World Health Organization (WHO), United Nations Population Fund (UNFPA), International Planned Parenthood Fed-eration (IPPF), Joint United Nations Programme on HIV/AIDS (UNAIDS), University of California San Francisco (UCSF). Sexual and Reproductive Health and HIV Linkages: Evidence Review and Recommendations. 2008 Geneva, Switzerland Available at: http://data.unaids.org/pub/Agenda/2009/2009_linkages_evidence_review_en.pdf
. Accessed April 13, 2011
13. WHO. . World Health Organization Statistical Information System (WHOSIS) 2011 Available at: http://www.who.int/whosis/indicators/compendium/2008/3pcf/en/index.html
. Accessed March 22, 2011
14. Kenya National Bureau of Statistics (KNBS) and ICF Macro. . Kenya demographic and health survey 2008–2009 2010 Available at: http://apps.who.int/medicinedocs/documents/s17116e/s17116e.pdf
. Accessed February 28, 2011
15. Mutiso SM, Kinuthia J, Qureshi Z. Contraceptive use among HIV infected women attending Comprehensive Care Centre. East Afr Med J. 2008;85:171–177
16. Chu JH, Gange SJ, Anastos K, et al. Hormonal contraceptive use and the effectiveness of highly active antiretroviral therapy. Am J Epidemiol. 2005;161:881–890
17. Cohn SE, Park JG, Watts DH, et al. Depo-medroxyprogesterone in women on antiretroviral therapy: effective contraception and lack of clinically significant interactions. Clin Pharmacol Ther. 2007;81:222–227
18. King R, Estey J, Allen S, et al. A family planning intervention to reduce vertical transmission of HIV in Rwanda. AIDS. 1995;9(suppl 1):S45–S51
19. Dawson SG, Callander N, Roche C, et al. Integrated sexual healthcare: the development and review of one model of service delivery. Int J STD AIDS. 2000;11:428–434
20. De Vincenzi I, Jadand C, Couturier E, et al. Pregnancy and contraception in a French cohort of HIV-infected women. SEROCO Study Group. AIDS. 1997;11:333–338
21. Desgrées-Du-Loû A, Msellati P, Viho I, et al. Contraceptive use, protected sexual intercourse and incidence of pregnancies among African HIV-infected women. DITRAME ANRS 049 Project, Abidjan 1995–2000. Int J STD AIDS. 2002;13:462–468
22. Magalhães JA, Amaral E, Giraldo PC, et al. HIV infection in women: impact on contraception. Contraception. 2002;66:87–91
23. Peck R, Fitzgerald DW, Liautaud B, et al. The feasibility, demand, and effect of integrating primary care services with HIV voluntary counseling and testing: evaluation of a 15-year experience in Haiti, 1985–2000. J Acquir Immune Defic Syndr. 2003;33:470–475
24. Forbes A, Engle N. Re-building distribution networks to assure future microbicide access. AIDS Public Policy J. 2005;20:92–101
25. Wilcher R, Petruney T, Reynolds HW, et al. From effectiveness to impact: contraception as an HIV prevention intervention. Sex Transm Infect. 2008;84(suppl 2):ii54–ii60
26. Einterz RM, Kimaiyo S, Mengech HN, et al. Responding to the HIV pandemic: the power of an academic medical partnership. Acad Med. 2007;82:812–818
27. World Health Organization. . Perspectives and practice in antiretroviral treatment, academic institutions linking access to treatment and prevention: case study 2004 Available at: http://www.who.int/hiv/pub/prev_care/en/ampath.pdf
. Accessed October 11, 2011
28. Inui TS, Nyandiko WM, Kimaiyo SN, et al. AMPATH: living proof that no one has to die from HIV. J Gen Intern Med. 2007;22:1745–1750
29. Voelker R. Conquering HIV and stigma in Kenya. JAMA. 2004;292:157–159
30. Abadie A, Imbens G. Large sample properties of matching estimators for average treatment effects. Econometrica. 2006;74:235–267
31. French RS, Coope CM, Graham A, et al. One stop shop versus collaborative integration: what is the best way of delivering sexual health services? Sex Transm Infect. 2006;82:202–206
32. Lush L, Walt G, Cleland J, et al. The role of MCH and family planning services in HIV/STD control: is integration the answer? Afr J Reprod Health. 2001;5:29–46
33. Pachauri S. Relationship between AIDS and family planning programmes: a rationale for developing integrated reproductive health services. Health Transit Rev. 1994;4(suppl):321–347