HIV in Vietnam: The Evolving Epidemic and the Prevention Response, 1996 Through 1999

continues to evolve, with rapid increases in infection in the north, a region previously little affected. In this paper, we summarize the recent changes through 1999 and describe the national response to the epidemic. Vietnam is a country of 76.7 million people, according to the Vietnamese Ministry of Health. It is divided into 61 administrative units that comprise 57 provinces and four major cities (Hanoi, Hai Phong, Da Nang, and Ho Chi Minh City HCMC), expanded from a total of 53 such units in 1996. A summary of the country’s demographic characteristics relevant to HIV has been published (1)

JAIDS Journal of Acquired Immune Deficiency Syndromes 25:360–369 © 2000 Lippincott Williams & Wilkins, Inc., Philadelphia HIV in Vietnam: The Evolving Epidemic and the Prevention Response, 1996 Through 1999 *‡Vu Minh Quan, *A Chung, †Hoang Thuy Long, and ‡Timothy J Dondero *National AIDS Committee of Vietnam, Hanoi; †National Institute of Hygiene and Epidemiology, Hanoi, Vietnam; and ‡Division of HIV/AIDS Prevention—Surveillance and Epidemiology, National Center for HIV, STD, and TB Prevention, U.S Centers for Disease Control and Prevention, Atlanta, Georgia, U.S.A Objectives: To describe epidemiologic patterns and trends in HIV infection in Vietnam from 1996 through 1999, and to summarize the national response to the epidemic Methods: We reviewed nationwide HIV case reports, and we analyzed annual seroprevalence among different sentinel populations in 21 provinces, using the ␹2 test for linear trend to assess trends in HIV prevalence HIV prevention efforts were also reviewed Results: Through 1999, 17,046 HIV infections, including 2947 AIDS cases and 1523 deaths had been reported in Vietnam The cumulative incidence rate for the country was 22.5 per 100,000 population Injection drug users (IDUs) represented 89.0% of all those for whom risk was reported before 1997 and 88.0% in the period 1997 to 1999 In 1999, HIV prevalence rates among IDUs ranged by province from 0% to 89.4% Significantly increasing HIV trends among IDUs (p < 05) were found in 14 of the 21 sentinel provinces during 1996 to 1999 HIV prevalence among commercial sex workers (CSWs) ranged from 0% to 13.2%, increased significantly in 6 of 21 provinces In 1999, prevalence among pregnant women, blood donors, and military recruits were 0.12%, 0.20% and 0.61%, respectively Major prevention activities in- clude mass information; peer education and outreach among groups at increased risk; availability of low-cost syringes and condoms through pharmacies; needle exchange pilot projects; widely available treatment for sexually transmitted diseases; antibody screening of blood for transfusion; and free medical treatment at government hospitals Discussion: The HIV epidemic continues to evolve rapidly, intensifying among IDUs and increasing among CSWs Serosurveillance indicators of HIV in the popu- lation at large continue to indicate the relatively slow extension beyond those at highest risk Immediate, intensive preventions in high-risk groups may decelerate expansion to the broader population Key Words: HIV—AIDS—Vietnam—Asia—Prevalence—Surveillance—Epi- demiology—Trend—Prevention The HIV epidemic has emerged rapidly in Vietnam continues to evolve, with rapid increases in infection in since the first infection was detected in 1990 The epi- the north, a region previously little affected In this pa- demiologic situation through the end of 1996 has been per, we summarize the recent changes through 1999 and summarized elsewhere (1) Since that time, the epidemic describe the national response to the epidemic Address correspondence and reprint requests to Vu Minh Quan, Vietnam is a country of 76.7 million people, according Division of HIV/AIDS Prevention—Surveillance and Epidemiology, to the Vietnamese Ministry of Health It is divided into National Center for HIV, STD, and TB Prevention, Centers for Disease 61 administrative units that comprise 57 provinces and Control and Prevention, 1600 Clifton Road, Mailstop E-46, Atlanta, four major cities (Hanoi, Hai Phong, Da Nang, and Ho GA 30333, U.S.A.; e-mail: vaq1@cdc.gov Chi Minh City [HCMC]), expanded from a total of 53 such units in 1996 A summary of the country’s demo- Vu Minh Quan is currently an officer in the Epidemic Intelligence graphic characteristics relevant to HIV has been pub- Service [EIS] Program and Division of HIV/AIDS Prevention at the lished (1) Centers for Disease Control and Prevention Manuscript received March 28, 2000; accepted August 2, 2000 360 HIV IN VIETNAM: 1996–1999 361 Social changes in Vietnam continue to occur as the for Hygiene and Epidemiology, which is part of the pro- economy advances Many of these changes have been vincial department of health beneficial, but there are unwanted consequences as well Drug use continues to be a concern, as it is throughout Reports of HIV infection as well as reports of AIDS much of Asia During the past decade, heroin has cases and deaths are provided weekly by the provincial emerged as one of the drugs most commonly used, and center for hygiene and epidemiology to the Pasteur In- injection drug use remains the principal mode of HIV stitute serving the region or (for the north) to the Na- transmission in Vietnam Increases in investment and tional Institute for Hygiene and Epidemiology (NIHE) business, along with increased numbers of visitors, have Compilation of all data are done at NIHE, and the results helped the economy but have also encouraged commer- are provided to the National AIDS Committee cial sex Cases of sexually transmitted diseases (STD), primar- AIDS-related activities, including surveillance, pre- ily those seen at public-sector provincial STD centers vention, and treatment, are coordinated and funded na- and clinics, are reported to the National Institute for Der- tionally by the National AIDS Committee of Vietnam, matology and Venereology, which analyzes the data and which is made up of numerous government ministries provides them to the Ministry of Health and the National and mass organizations At the provincial and city levels, AIDS Committee AIDS activities are coordinated by the provincial or city AIDS committee, which includes the local counterparts HIV AND AIDS REPORTING of the same groups that comprise the National AIDS Committee Through 1999, 17,046 HIV infections, including 2947 AIDS cases and 1523 deaths, were reported The number AIDS cases are reported to the respective provincial of reported HIV infections per year continues to increase AIDS committee in two ways: hospitals report AIDS rapidly, whereas AIDS cases and deaths climb more cases to the provincial Department of Health (which par- slowly (Fig 1) The 1999 cumulative incidence rate for ticipates in the provincial AIDS committee); additional the country as a whole was 22.5 per 100,000 population, cases are detected by the provincial center for hygiene compared with only 7.0 per 100,000 population in 1996 and epidemiology in the course of public health follow- (1) up of persons who have tested HIV positive Infections have been reported from all of the country’s HIV infection is detected through several means, in- 61 provinces and major cities Most of the recent in- cluding serologic surveys, counseling and testing, diag- crease in reported HIV infections has taken place in the nostic procedures, evaluation of partners of HIV-infected north, a region where HIV was very limited even 3 years persons, and occasional epidemiologic field investiga- before Indeed, by the end of 1999, of the 10 areas with tions Within a province or city, most of these public the largest cumulative numbers of cases, five were in the health activities are conducted by the provincial Center FIG 1 Reported HIV infections, AIDS cases, and AIDS deaths, Viet- nam, by year of report, 1990 through 1999 JAIDS Journal of Acquired Immune Deficiency Syndromes, Vol 25, No 4, December 1, 2000 362 V M QUAN ET AL north; 3 years earlier, no northern province or city was infections associated with injection drug use far exceed among the top 10 areas The three areas with the highest the infections associated with all other risks except in the cumulative incidence rates of reported HIV (Fig 2) are southern provinces of An Giang and Can Tho, and this in the north: Quang Ninh (222 per 100,000 population), predominance has not changed over time; injection drug Lang Son (81 per 100,000 population) and Hai Phong users (IDUs) represented 89.0% of all persons for whom (70 per 100,000 population) risk was reported before 1997 and 88.0% in the period from 1997 to 1999 The most dramatic recent increase in HIV infections is associated with injection drug use and has occurred pri- Although currently much less frequent than drug- marily in the northeastern corner of the country In the related transmission, the second most common risk re- area that includes the provinces of Lang Son, Thai mains sexual transmission, principally heterosexual Nguyen, Bac Giang, Quang Ninh, Bac Ninh, Hanoi, transmission Of all infections reported through 1999, Hung Yen, Hai Duong, and Hai Phong, the number of 5.0% of cases with risk information were in commercial newly detected infections has increased from 8 in 1995, sex workers (CSWs); an additional 3.6% of cases were in to 126 in 1996, to 969 in 1997, to 2043 in 1998, to 2885 patients being seen clinically for an STD (for whom in 1999 During the same 5 years, the proportion of sexual risk is assumed to be the source of transmission), newly detected infections rose from 1% of the country’s and another 2.3% were in workers in the entertainment total in 1995, to 9% in 1996, to 32% in 1997, to 55% in industry or others thought to be at increased sexual risk 1998, and leveled to 52% in 1999 Thus, at least 10.9% of cases with known or reasonably presumed risk seem to have been sexually acquired The principal risk for HIV in Vietnam remains injec- tion drug use, which accounts for 88.3% of all cases for No case of HIV has been determined to be associated which risk has been reported or can be inferred Al- with sexual transmission between men Although this though there is some geographic variation in frequency, risk may be underascertained, the frequency of male-to- FIG 2 Cumulative HIV incidence per 100,000 population by prov- ince, Vietnam, 1996 and 1999 Map does not illustrate small offshore is- lands JAIDS Journal of Acquired Immune Deficiency Syndromes, Vol 25, No 4, December 1, 2000 HIV IN VIETNAM: 1996–1999 363 FIG 3 Reported HIV infections by age group, Vietnam, by year, 1990 through 1999 male HIV transmission still seems to be very low Chil- provinces in 1994; by 1999, they were being conducted dren thought to have acquired HIV infection from their in all 20 sentinel areas In addition, Quang Ninh prov- mothers (24 cases) remain a small part (0.2%) of the ince, not a formal part of the sentinel surveillance pro- epidemic thus far HIV infection resulting from transfu- gram, has also conducted surveys since 1996 Serosur- sion of blood or blood products has not yet been de- veillance data are more reliable than case reporting for tected evaluating trends, because the methods for survey sam- pling remain essentially consistent, whereas the com- Of those reported with HIV infection, 86.7% have pleteness of case reporting, which is affected by factors been men Distribution by sex has not changed apprecia- aside from the epidemiologic situation, can differ from bly: 84.4% men during 1990 to 1996 and 87.9% during year to year We used the ␹2 test for linear trend to assess 1997 to 1999 However, infections are occurring increas- trends in HIV prevalence in the 20 sentinel provinces and ingly in younger persons, especially in those 20 to 29 Quang Ninh province, from 1996 through 1999 years old (Fig 3) Sentinel surveillance among IDUs continues to indi- Although reported HIV infections and AIDS cases il- cate high levels of infection associated with drug injec- lustrate the geographic, demographic, and risk character- tion (Table 1) The epidemic among IDUs has intensified istics of the epidemic, the number of detected, reported in 14 of the 21 provinces surveyed (Fig 4) In the central infections and AIDS cases will always be a fraction of region, where HIV prevalence among IDUs was moder- the total in the country The National AIDS Committee ately high by 1996, the levels by 1999 had uniformly, estimates that in reality 120,000 to 140,000 infections and in some places dramatically, increased (e.g., Binh existed in Vietnam by the beginning of 1999 Dinh, from 23.3% to 71.1%; Khanh Hoa, from 50.8% to 89.4%) In the south, the HIV prevalence among the SEROSURVEILLANCE sentinel IDU populations increased significantly in four provinces, An Giang, Can Tho, Kien Giang and Vung An important part of monitoring HIV in Vietnam is Tau-Ba Ria, but stabilized in three areas, of which the sentinel serosurveillance program (2) As part of this HCMC and Dong Nai already had high prevalences program, in 1999, 20 provinces and major cities con- However, the greatest overall change in HIV prevalence ducted annual or semiannual HIV surveys in sentinel has been the explosive increase from 1996 through 1999 populations (IDUs, CSWs, STD patients, tuberculosis in 7 of 9 provinces surveyed in the north (Table 1) The [TB] patients, pregnant women, and military recruits) most dramatic increases have occurred in Quang Ninh The annual target sample size (not always reached in and Hai Phong some provinces) is 400 each for IDUs, CSWs, STD pa- tients, and TB patients, and 800 each for pregnant The two sentinel indicators of sexual transmission women and military recruits, respectively Data from the risk, surveys among CSWs and surveys among STD screening of blood donors are also available in many clinic patients, continue to indicate that sexual trans- sentinel provinces Serosurveys were begun in some mission is a distant second to transmission associated JAIDS Journal of Acquired Immune Deficiency Syndromes, Vol 25, No 4, December 1, 2000 364 V M QUAN ET AL FIG 4 Significant increases* in HIV prevalence among injection drug users, commercial sex work- ers, and sexually transmitted dis- ease patients in sentinel prov- inces, Vietnam, 1996 through 1999 Map does not illustrate small offshore islands with injection drug use, again, a pattern that is consistent prevalence among CSWs is still very low and stable (10 with HIV case reports Nonetheless, the prevalence of of 14 northern and southern provinces have rates less HIV among CSWs has increased in several areas (Table than 1% in 1999) The exceptions are in the north: Hanoi 1), although to levels well below those in some other (from 0.1% in 1996 to 6.0% in 1999) and Hai Phong southeast Asian countries (3–5) The HIV prevalence (from 0.8% to 2.6%) HIV infection also seems to be among sentinel groups at increased sexual risk has been increasing among STD patients (who are mostly men) in highest in the south, where sex-related transmission was some of the southern sentinel provinces, although the already evident by 1996 In 1999, HIV prevalence increase is significant in only two (An Giang and Dong among CSWs reached 5.0% to 13.2% in some areas, Nai; Table 1) In 1999, the prevalence among STD pa- compared with 4.1%, the highest level in 1996 The tients in the central region continued to be as low as it prevalence in CSWs increased significantly in four of had been in 1996 In the north, the prevalence among seven Mekong delta provinces from 1996 through 1999 STD patients in 1999 is generally low, except in Hanoi (Table 1, Fig 4) Outside the southern region, HIV (2.0%), Hai Phong (5.5%), and Quang Ninh (3.9%) JAIDS Journal of Acquired Immune Deficiency Syndromes, Vol 25, No 4, December 1, 2000 HIV IN VIETNAM: 1996–1999 365 TABLE 1 Changes in HIV seroprevalence among injection drug users, commercial sex workers and sexually transmitted disease (STD) patients in Vietnam, 1996 through 1999,a by province Injection drug users Commercial sex workers STD patients percentage positive percentage positive percentage positive (n) (n) (n) 1996 1997 1998 1999 p Valueb 1996 1997 1998 1999 p Value 1996 1997 1998 1999 p Value Northern region 0.6 2.4 3.3 13.5