Background: Despite the scale-up of prevention of mother-to-child transmission (PMTCT) programs worldwide, the translation from research studies into public health policy has been slow. This report details the experiences of a city-driven PMTCT program in China using existing health resources.
Methods: The PMTCT program was devised to hospital based and city-wide. It achieves full use of available resources: the local Centers for Disease Control and Prevention, the Infectious Disease Hospital, Maternal and Child Health Hospitals, and all qualified comprehensive hospitals.
Results: From 2000 to 2010, 1,843,122 pregnant women attended prenatal care or labor and delivery services. Overall, 97.4% received pretest HIV counseling, and 96.2% were tested for HIV. Among the 81.1% (1,495,122) of women who attended prenatal clinics, 97.2% (1,452,753) received pretest counseling and 95.7% (1,430,799) were tested for HIV. Among the 18.9% (348,000) of women with an undocumented HIV status at labor and delivery, 98.6% (343,038) received pretest counseling, and 98.1% (341,371) were tested for HIV. In total, 229 women were determined HIV positive for a prevalence of 1.3 per 10,000 pregnant women. Among the 107 HIV-infected women who carried to delivery, 87.9% received antiretroviral prophylaxis for themselves and their infants. Among the 58 women who were identified HIV positive at labor, 10.3% of mothers and 72.4% of infants received antiretroviral prophylaxis. The estimated mother-to-child transmission rate was 5.3% (95% confidence interval, 2.2%–10.7%).
Conclusions: With appropriate integration, existing health care resources are adequate for a comprehensive city-driven PMTCT program in an area with a low HIV prevalence.
A city-driven prevention of mother-to-child transmission program in Shenzhen, China, showed that with appropriate integration, existing health care resources were adequate for a hospital-based comprehensive prevention of mother-to-child transmission program.
From the *Shenzhen Center for Disease Control and Prevention, Shenzhen, China; †Shenzhen Center for Chronic Disease Control, Shenzhen, China; ‡Global AIDS Program, US Centers for Disease Control and Prevention, Beijing, China; and §The Health, Population and Family Planning Commission of Shenzhen Municipality, Shenzhen, China.
The authors thank all participants from Shenzhen Center for Disease Control and Prevention (CDC), Maternal and Child Health Hospitals, the Infectious Disease Hospital, and all comprehensive hospitals in Shenzhen city. They also thank the National Center for Women and Children’s Health, China CDC, Beijing; The Guangdong Women and Children’s Hospital and Health Institute; and the Health, Population and Family Planning Commission of Shenzhen Municipality for guidance and support. The US CDC Global AIDS Program has provided technical support and guidance for program development in collaboration with Shenzhen City CDC since 2008.
Funding: This work received financial support from both the local government and transfer payments from the central government of China.
Competing interests: None of the authors report any financial relationships or other conflicts of interest.
Disclaimer: The findings and conclusions in this article are those of the authors and do not necessarily represent the views of the US Centers for Disease Control and Prevention.
Correspondence: Jinquan Cheng, PhD, Shenzhen Center for Disease Control and Prevention, Shenzhen, China. E-mail: email@example.com; Lin Chen, MD, Shenzhen Center for Disease Control and Prevention, Shenzhen, 8 Longyuan Rd, Nanshan District, Shenzhen, Guangdong Province, China. E-mail: firstname.lastname@example.org.
Received for publication September 14, 2012, and accepted November 28, 2012.
After the successful outcome of the Pediatric AIDS Clinical Trial Group Protocol 076 in 1994, many countries have implemented prevention of mother-to-child transmission (PMTCT) programs for HIV. It is clear that PMTCT programs should be integrated into existing maternal and child health (MCH) services.1–3 However, in areas or countries with low HIV seroprevalence and resource limitations, logistical and workload issues have been reported such as inadequate numbers of lay counselors, missed opportunities for antiretroviral (ARV) prophylaxis interventions, insufficient coverage, and lack of follow-up for mothers and children.4–6 Furthermore, reproductive health goals including providing family planning services and condom-based contraception to women after delivery have been difficult to achieve.7 Thus, an innovative and easy-to-implement model that addresses the goals of PMTCT and reproductive health programs for wide-scale practice is crucial for the scale-up of PMTCT programs.2,8–11
This report presents a city-driven PMTCT program in Shenzhen, a special economic zone in southern China with more than 80% of the total urban population ( >12 million people) migrant laborers.12 Increased sexual transmission of HIV is a particular concern because many women come from provinces with a higher HIV prevalence such as Yunnan, Guangxi, Sichuan, Xinjiang, and Henan.13
In July 2000, 10 state-owned hospitals were chosen as pilot sites to provide PMTCT services. In 2002, the PMTCT program formally began in 60 hospitals. In 2007, the national PMTCT program began providing support, and services gradually expanded to 85 sites by 2010. The program covered most of the city’s deliveries. Only 0.3% of deliveries occurred outside of hospitals and some of these mothers also received HIV screening at hospitals later after delivery. Through the use of existing health care resources, the PMTCT program has provided professional services for HIV-infected pregnant and postpartum women. This report describes the experience and the constraints encountered during the program implementation.
Framework and Implementation of the Shenzhen City PMTCT Program
There are several organizations involved at different levels of the program.
The Health, Population, and Family Planning Commission of Shenzhen Municipality is an administrative leader for all Shenzhen health care institutions and undertakes supervisory duty of the program. The Shenzhen municipal Center for Disease Control and Prevention (CDC) is the primary manager for the program providing technical assistance, reagent supplies, and laboratory testing including confirmatory HIV tests (Western blots), CD4 cell counts, and viral load tests. The Shenzhen municipal MCH Hospital provides supplemental management including program monitoring, training, and technical assistance on maternal-child health care. Shenzhen City Infectious Disease Hospital (Third People’s Hospital) provides ARV prophylaxis and treatment.
In addition to the municipal organizations listed previously, each of the 6 districts under the Shenzhen municipality has their own CDC station and MCH hospital. Antenatal care (ANC) services including delivery and HIV testing are provided in 85 hospitals in the city: 7 MCH hospitals (6 districts and 1 municipal), 47 public non-MCH hospitals, and 31 private hospitals.
Yearly PMTCT training was conducted by the CDCs and MCH hospitals at the municipal and district levels, emphasizing the nationally adapted training modules and problem-solving solutions. Health care personnel in prenatal clinics, obstetric wards, laboratories, and disease prevention and care departments were trained. The municipal CDC also held annual laboratory trainings.
Pretest Counseling and Information Registration
Free HIV tests were first provided using an opt-in strategy. In 2007, an opt-out strategy was adopted under the instruction of the National Center for Women and Children’s Health. Obstetricians notified pregnant women about the availability of free HIV tests and provided a brochure with PMTCT service and support information. Women were encouraged to consult with a doctor if they had questions. Syphilis testing was simultaneously offered.14 Each woman attending prenatal care was registered, and the result of the HIV test was recorded in a city handbook to avoid double counting. Women who were not tested initially were provided the opportunity of free testing at a subsequent visit. Monthly reports with numbers of HIV counseling sessions and screening tests performed were collected from each hospital clinic and inpatient ward.
Preliminary maternal HIV status was determined by enzyme-linked immunosorbent assay. Positive specimens were sent to the municipal CDC for confirmatory testing and CD4 enumeration, usually reported in 2 to 3 working days. Women with undocumented HIV status at labor and delivery were tested with a rapid test (Kehua HIV-1/2 Colloidal Gold, Shanghai, China) completed in 30 to 60 minutes.15 If the initial rapid test was positive, a second rapid test (Abbott HIV-1/2 Determine; Abbott Japan, Chiba, Japan) was used. Expedited informed consent and ARV intervention were only performed with 2 positive results. Expedited Western Blot (WB) confirmation testing at the municipal CDC was completed in 12 to 24 hours. Viral load testing of HIV-exposed infants was performed at 6 weeks and/or 3 months of age using Roche Amplicor HIV-1 Monitor test v1.5 viral load kit (Branchburg, NJ).16 HIV antibody tests were performed on exposed children at 12 and 18 months of age.
After testing, HIV-positive women were referred to CDC specialists for additional posttest counseling including steps for positive living, prevention of HIV transmission, and guidance on contraception and family planning. Two documents were provided at that time. The Informed Consent of PMTCT contained information on HIV/AIDS issues, safer sex, mother-to-child transmission, and information on replacement feeding. The Informed Consent of Intervention in Antepartum, Intrapartum and Postpartum Periods contained information on the ARV drug prophylaxis regimen, the recommended vaccine schedule, follow-up, and early HIV diagnosis of exposed children.
Intervention and Follow-up of HIV-positive Women and Infants: Before, During, and After Delivery
Antenatal visits and delivery for HIV-infected women were performed in the 85 program hospitals. HIV-infected women obtained zidovudine starting at 28 weeks of pregnancy from the CDC. Women were referred to Shenzhen City Infectious Disease Hospital for ARV prophylaxis and treatment starting in July 2007, according to the guidelines of the National Center for Women and Children’s Health (Table 1). Early initiation of highly active antiretroviral therapy (HAART) during pregnancy was adopted regardless of CD4 count since late 2009.17–19 Prepartum ARV drugs were available in the obstetric ward for women diagnosed during labor along with medical steps to restrain milk secretion. Women in labor were immediately given ARV prophylaxis after 2 positive HIV rapid test results.
Follow-up of infants was performed by hospital staff 42 days and 3, 6, 9, 12, and 18 months after birth. Blood was drawn at the initial follow-up visit and/or the 3-month visit for viral load testing and at 12 and 18 months for antibody testing. HIV-infected mothers were referred to the HIV follow-up services network to receive care and treatment.
Monitoring of the Shenzhen PMTCT program was conducted at the district level twice a year. The monitoring group consisted of experts on HIV/AIDS, laboratory testing, and maternal-child health from local CDC and MCH hospitals. Four elements were reviewed: organizational management, capacity development, implementation process, and accomplishments. A quantitative and detailed index was set up to assure the comparability of monitoring conducted by different groups. Questionnaires for pregnant women and health care workers were used to assess the quality of PMTCT services. Deficiencies in registration and data collection were identified through personal interviews in gynecologic clinics, obstetric wards, laboratories, and the Disease Prevention and Care Department. The annual quality control and evaluation test scores for the HIV screening laboratories were also assessed.
Analysis was conducted using SPSS software (version 16.0).20 Demographic data for HIV-infected women enrolled in PMTCT services were described. The χ2 test and unadjusted and adjusted odds ratios were used to evaluate associations between HIV-infected and HIV-noninfected pregnant women enrolled in services for selected variables. A P value less than 0.05 was considered statistically significant.
Counseling and Screening for HIV
From 1 July 2000 to 31 December 2010, 1,843,122 pregnant women visited 85 hospitals in Shenzhen for prenatal care or labor and delivery services. Overall, 97.4% (1,795,791) received pretest HIV counseling, and of these, 96.2% (1,772,170) were tested for HIV. Among the 81.1% (1,495,122) women who attended prenatal clinics, 97.2% (1,452,753) received pretest counseling and 95.7% (1,430,799) were tested for HIV. Among the 348,000 (18.9%) women with an undocumented HIV status at delivery, 98.6% (343,038) received pretest counseling and 98.1% (341,371) were tested for HIV. In total, 229 women were found to be HIV infected with a prevalence of 1.3 per 10,000 pregnant women (Table 2). The prevalence among women tested at labor and delivery was marginally higher (1.7 per 10,000) than that among women tested in ANC (1.2 per 10,000; P = 0.02). Staff were unable to notify 8 women of their HIV test results. Therefore, 221 women received HIV test results and posttest counseling (Fig. 1).
Characteristics of HIV-Infected Pregnant Women
The mean age of the 221 women diagnosed with HIV was 27.0 years (interquartile range, 23–30). Women with less education and who were non-residents residing in Shenzhen for 12 months or more were more likely to be diagnosed with HIV (Table 3). Most (87.8%) infections were reported to be acquired through sexual partners. Seventy-six percent (167) of male partners accepted HIV testing, and 38.3% (64) were HIV positive.
The 221 HIV-positive women carried 229 single-fetus pregnancies during the study period. Among the 229 pregnancies, 158 (69.0%) women were identified as HIV positive at ANC visit, 13 (5.7%) already knew their HIV-positive status and asked for PMTCT services, and 58 (25.3%) were identified at labor and delivery. Among the 158 pregnant women diagnosed as having HIV at ANC, 94 (59.5%) chose to deliver and 64 (40.5%) terminated their pregnancies. In the early phases of Shenzhen PMTCT program, there was a higher proportion of terminations because of a lack of comprehension of the HIV disease process and fear of discrimination. Abortion is legally available in China with given indications. From 2000 to 2005, 26 (61.9%) of 42 women opted for an abortion. Since then, this proportion has declined to 34.8% in 2006 to 2007 and 31.4% in 2008 to 2010 (test for trend, P = 0.003).
ARV Prophylaxis Among HIV-Infected Women and Infants
Among the 107 women who knew their HIV infection status during pregnancy and carried to delivery, 94 (87.9%) mother-child pairs received ARV prophylaxis. Among them, 26 mothers accepted HAART. Five women spontaneously aborted. Eight women moved away from Shenzen, and their ARV prophylaxis status was unknown. Among the 58 women identified as HIV positive during labor, 6 (10.3%) mothers and 42 (72.4%) infants received ARV prophylaxis (Fig. 1). Only 14.3% (2/14) of infants whose mothers were diagnosed during labor received ARV prophylaxis between 2000–2005 while this proportion reached 90.9% (40/44) from 2006–2010. Sixteen infants did not receive ARV prophylaxis; 12 (75%) due to the late identification of the mother’s HIV status since nevirapine must be provided in the first 72 hours after delivery.
Follow-up and Care for HIV-Infected Mothers and HIV-Exposed Infants
From 2000 to 2010, 113 (74.3%) of 152 HIV-exposed infants received follow-up HIV virological tests. Among them, 75 (79.8%) of the 94 infants of mothers who had prenatal care accepted virological tests, and 38 (65.5%) of the 58 infants whose mothers were diagnosed at labor and delivery accepted tests (Fig. 1). HIV status of 39 infants (19 mothers diagnosed at ANC and 20 during labor) was not available, primarily due to relocation to their hometown after delivery. All 75 tested infants whose mothers had prenatal care accepted ARV prophylaxis. Among the 38 tested infants whose mothers were identified during labor, 22 received ARV prophylaxis. In all, 6 infants were diagnosed as being HIV infected with a transmission rate of 5.3% (95% confidence interval [CI], 2.2%–10.7%). The period-specific transmission rates for each ART regimen are shown in Table 1.
Among the mothers of the 6 infants diagnosed with HIV, 1 mother was diagnosed at ANC for a transmission rate of 1.3% (95% CI, 0.1%–6.4%). Five were diagnosed at delivery and missed ARV prophylaxis for a transmission rate of 13.2% (95% CI, 5.0%–26.8%). HIV-infected women presenting at labor and delivery comprised only 33.6% of the total women enrolled in the program but resulted in 83.3% of the cases of MTCT. Among the 6 HIV-infected infants, 2 did not receive ARV prophylaxis (in 2002 and 2003) due to late identification. The other 4 (2 in 2007 and 2 in 2009) received single-dose NVP (either with or without zidovudine for 1 month). No vertical transmission occurred in the 26 pregnant women who received HAART during pregnancy since 2007. Free infant formula has been provided to all HIV-infected mothers through the PMTCT program since 2004. Replacement feeding was adopted by nearly all mothers since safe water and sterilized bottles can be easily accessed in an industrial city like Shenzhen.
Expansion of PMTCT Sites
Prevention of mother-to-child transmission services in private hospitals expanded from 6 sites in 2000 to 31 in 2010, so that 36.5% of the 85 PMTCT sites were in private hospitals.
In a large city-driven perinatal HIV prevention program involving hospitals, disease control institutions, and the main referral Infectious Disease Hospital, the prevalence of MTCT was 5.3%. This decade-long intervention program screened more than 1.8 million pregnant women and enrolled 221 women and infants into comprehensive PMTCT services. Services were provided for the HIV-infected women and their infants including early infant diagnosis using RNA viral load. In recent years, HAART was provided to all HIV-infected pregnant women. None of the 26 women who received HAART since late 2009 transmitted HIV to their infants. Because HAART during pregnancy has now become routine, the MTCT rate in HIV-infected women identified early in antenatal care and started on timely antiretroviral treatment is expected to remain very low.21,22
Shenzhen’s large migrant population is a huge challenge for the city PMTCT program. It was decided to expand the program to as many hospitals as possible, including the smaller private hospitals where many migrant women get their health care. It turned out that 16.9% of HIV-positive women attended private hospitals even though only 9.6% of women received HIV screening there.
Another major step in decreasing the MTCT rate is the acceleration of HIV detection during labor and delivery. Late diagnosis limits the effectiveness of PMTCT interventions. Mothers of 5 of the 6 infants who became HIV infected in this decade-long, urban PMTCT program were diagnosed at delivery. HIV rapid tests were promoted in 2006 in Shenzhen. Before that, the results of enzyme-linked immunosorbent assay tests were usually too late for the infants to receive nevirapine.23
Ongoing integration of health care resources contributed to the relatively high levels of follow-up for mother-child pairs. Municipal and district CDCs and participating hospitals were the first to perform counseling and follow-up. In 2003, community-based prevention and care units of the CDCs were formed. Staff in these units performed infectious disease surveillance and case management and worked to strengthen the social support system for the infected women. In 2007, HAART began to be prescribed, and doctors from the City Infectious Disease Hospital provided treatment-related follow-up. Personnel in the MCH hospitals performed data management and statistics starting in 2008, allowing for easier monthly reporting.
The integration of health resources also provides a framework for the prevention of transmission of perinatal syphilis and hepatitis B.24 These 2 prevention programs were added to the PMTCT program in Shenzhen in 2001 and 2012.14 The national PMTCT program in China was initiated in 2002 and expanded to 1156 counties by 2010, increasing counseling and testing yearly.25 Now, a combined PMTCT program addressing HIV, syphilis, and hepatitis B virus is expanding nationwide through MCH hospitals in 2011. However, the existing system for follow-up through local CDCs and infectious disease hospitals has been effective and should not be discounted.26 Committed local leadership plays a key role in the integrated program.27 In Shenzhen, the Health, Population and Family Planning Commission of Shenzhen Municipality, the administrative authority and policy maker for all hospitals and CDCs in the city, provided leadership for the PMTCT program to ensure effective implementation.
Despite substantial progress in reducing perinatal HIV transmission during the past decade, there are still critical challenges in the Shenzhen PMTCT program. First, as a low HIV prevalence area, some health care workers do not see the need for timely HIV testing for women presenting at delivery which can lead to a sizeable proportion of mothers missing the last chance of ARV prophylaxis. To address this issue, a protocol has been developed for managing HIV-positive women in labor wards. Second, some women opt to receive ANC in private outpatient clinics and community health centers which have not been integrated into the PMTCT program. Third, there are 500 to 700 deliveries outside the hospitals each year. A large portion of these mothers may have missed HIV screening and intervention opportunities.
Some women diagnosed as having HIV opted out of follow-up. This affects the MTCT rate. Most of the loss-to-follow-up occurred among women diagnosed as having HIV during labor and delivery. These women are especially vulnerable and require additional psychosocial support.28 Abortion rates were high during the first few years of the PMTCT program. This has decreased significantly as a result of ARV prophylaxis and an education campaign among health care providers.
Some women also face suspicion from their families when adopting formula feeding because breast-feeding is widely encouraged and accepted in China.29 In addition, psychosocial support for HIV-infected mothers and their families is still relatively weak. Some male partners who were HIV negative chose not to accept the baby or left the family. In conclusion, earlier HIV testing and enrollment, appropriate care and follow-up, and ensuring the availability of comprehensive social support for the families after PMTCT services end are all ongoing challenges. In addition, more prevention programs and provider education could be targeted toward the mobile population of women living in Shenzhen and those public and private health care providers who care for them.
1. Paintsil E, Andiman WA. Update on successes and challenges regarding mother-to-child transmission of HIV. Curr Opin Pediatr 2009; 21: 94–101.
3. Spensley A, Sripipatana T, Turner AN, et al.. Preventing mother-to-child transmission of HIV in resource-limited settings: The Elizabeth Glaser Pediatric AIDS Foundation experience. Am J Public Health 2009; 99: 631–637.
4. Peltzer K, Mosala T, Dana P, et al.. Follow-up survey of women who have undergone a prevention of mother-to-child transmission program in a resource-poor setting in South Africa. J Assoc Nurses AIDS Care 2008; 19: 450–460.
5. Dube S, Boily MC, Mugurungi O, et al.. Estimating vertically acquired HIV infections and the impact of the prevention of mother-to-child transmission program in Zimbabwe: Insights from decision analysis models. J Acquir Immune Defic Syndr 2008; 48: 72–81.
7. Agadjanian V, Hayford SR. PMTCT, HAART, and childbearing in Mozambique: An institutional perspective. AIDS Behav 2009; 13 (suppl 1): 103–112.
8. Wilcher R, Cates W. Reproductive choices for women with HIV. Bull World Health Organ 2009; 87: 833–839.
9. Fowler MG, Lampe MA, Jamieson DJ, et al.. Reducing the risk of mother-to-child human immunodeficiency virus transmission: Past successes, current progress and challenges, and future directions. Am J Obstet Gynecol 2007; 197 (3 suppl): S3–S9.
10. Becquet R, Ekouevi DK, Arrive E, et al.. Universal antiretroviral therapy for pregnant and breast-feeding HIV-1–infected women: Towards the elimination of mother-to-child transmission of HIV-1 in resource-limited settings. Clin Infect Dis 2009; 49: 1936–1945.
12. Cai WD, Zhao J, Zhao JK, et al.. HIV prevalence and related risk factors among male sex workers in Shenzhen, China: Results from a time-location sampling survey. Sex Transm Infect 2010; 86: 15–20.
13. Wang N, Wang L, Wu Z, et al.. Estimating the number of people living with HIV/AIDS in China: 2003–09. Int J Epidemiol 2010; 39 (suppl 2): ii21–ii28.
14. Cheng JQ, Zhou H, Hong FC, et al.. Syphilis screening and intervention in 500,000 pregnant women in Shenzhen, the People’s Republic of China. Sex Transm Infect 2007; 83: 347–350.
15. Bulterys M, Jamieson DJ, O’Sullivan MJ, et al.. Rapid HIV-1 testing during labor: A multicenter study. JAMA 2004; 292: 219–223.
16. Donovan M, Palumbo P. Diagnosis of HIV: Challenges and strategies for HIV prevention and detection among pregnant women and their infants. Clin Perinatol 2010; 37: 751–763, viii.
18. Stringer JS, McConnell MS, Kiarie J, et al.. Effectiveness of non-nucleoside reverse-transcriptase inhibitor–based antiretroviral therapy in women previously exposed to a single intrapartum dose of nevirapine: A multi-country, prospective cohort study. PLoS Med 2010; 7: e1000233.
19. Kourtis AP, Bulterys M. Mother-to-child transmission of HIV: Pathogenesis, mechanisms and pathways. Clin Perinatol 2010; 37: 721–737, vii.
20. SPSS software version 16.0. Chicago, IL; 2007.
21. Zhou Z, Meyers K, Li X, et al.. Prevention of mother-to-child transmission of HIV-1 using highly active antiretroviral therapy in rural Yunnan, China. J Acquir Immune Defic Syndr 2010; 53 (suppl 1): S15–S22.
22. Bulterys M, Wilfert CM. HAART during pregnancy and during breastfeeding among HIV-infected women in the developing world: Has the time come? AIDS 2009; 23: 2473–2477.
23. Wettsteina C, Mugglina C, Egger M, et al.. Missed Opportunities to Prevent Mother-to-Child-Transmission in sub-Saharan Africa: Systematic Review and Meta-Analysis. AIDS 2012; 26: 2361–2373.
24. Kourtis AP, Bulterys M, Hu DJ, et al.. HIV-HBV coinfection–a global challenge. N Engl J Med 2012; 366: 1749–1752.
25. Fang LW, Wang LH, Wang XY, et al.. Analysis on the intervention services status of prevention of mother to child transmission of HIV/AIDS in China. Zhonghua Yu Fang Yi Xue Za Zhi 2010; 44: 1003–1006.
26. Wang JJ, Reilly KH, Han H, et al.. Dynamic characteristic analysis of HIV mother to child transmission in China. Biomed Environ Sci 2010; 23: 402–408.
27. Malyuta R, Newell ML, Ostergren M, et al.. Prevention of mother-to-child transmission of HIV infection: Ukraine experience to date. Eur J Public Health 2006; 16: 123–127.
28. Cohen MH, Olszewski Y, Webber MP, et al.. Women identified with HIV at labor and delivery: Testing, disclosing and linking to care challenges. Matern Child Health J 2008; 12: 568–576.
© Copyright 2013 American Sexually Transmitted Diseases Association
29. Korhonen C, Wang L, Wang LH, et al.. Breastfeeding and HIV infection in China. Adv Exp Med Biol 2012; 743: 237–246.