Contraceptive use among HIV-positive women in Quang Ninh province, Vietnam

Authors


Corresponding Author Vibeke Rasch, Department of Obstetric and Gynaecology, Odense University Hospital, Odense, Denmark. E-mail: vrasch@health.sdu.dk

Abstract

Objective  To investigate contraceptive use among HIV-positive women in Ha long city and Cam Pha town of Quang Ninh, a Northern province of Vietnam.

Methods  Cross-sectional questionnaire study among HIV-positive women identified through the district HIV/AIDS register. Information on socioeconomic characteristics and contraceptive use before and after HIV diagnosis was obtained through structured questionnaires. Chi-square testing was used to assess whether the included women differed from the target population in terms of age. Crude Odds ratios (ORs) were calculated to show the association between contraceptive use and the women’s socioeconomic characteristics. Logistic regression analyses were applied to adjust for possible confounding. The women’s contraceptive use before HIV testing and after HIV testing was described and compared by Chi-square testing, and the association between post-test counselling and the women’s use of condom was assessed by ORs.

Results  Of the 351 participants, 63% stated they had used contraception before HIV diagnosis and 89% stated they had used contraception after HIV diagnosis. Forty six per cent of the women had been using either the pill or an intrauterine device (IUD) before the diagnosis whereas the same applied for only 8% of the women after diagnosis. Thirty-nine per cent stated they had been using condom before HIV diagnosis whereas 87% stated condom use after HIV diagnosis. Condom use was more common among women who had received post-test counselling (OR 3.03, 95% CI 1.03–8.90).

Conclusions  A change of contraceptive methods from IUD and oral contraception before HIV diagnosis to condom use after HIV diagnosis was observed. The women’s use of condoms after HIV diagnosis was associated with having received post-test counselling.

Abstract

Objectif:  Investiguer l’utilisation des contraceptifs chez les femmes VIH-positives dans les villes de Ha Long et de Cam Pha à Quang Ninh, une province du Nord du Vietnam.

Méthodes:  Etude transversale par questionnaire chez les femmes VIH-positives identifiées grâce au registre VIH/SIDA du district. L’information sur les caractéristiques socio-économiques et l’utilisation de contraceptifs avant et après le diagnostic du VIH a été obtenue au moyen de questionnaires structurés. Le test Chi-carré a été utilisé pour évaluer si les femmes incluses différaient de celles de la population cible en termes d’âge. Les rapports de côte bruts (OR) ont été calculés pour montrer l’association entre l’utilisation de contraceptifs et les caractéristiques socio-économiques des femmes. Des analyses de régression logistique ont été appliquées pour ajuster pour d’éventuels facteurs confusionnels. L’utilisation de la contraception par les femmes avant et après le dépistage du VIH a été décrite et comparée par le test de Chi-carré et l’association entre le conseil post-test et l’utilisation du préservatif par les femmes a étéévaluée par les OR.

Résultats:  Sur les 351 participantes, 63% ont déclaré avoir utilisé la contraception avant le diagnostic du VIH et 89% ont déclaré avoir utilisé la contraception après le diagnostic du VIH. 46% des femmes utilisaient soit la pilule ou un dispositif intra-utérin (DIU) avant le diagnostic, alors que cela s’appliquait à seulement 8% des femmes après le diagnostic. 39% des participantes ont déclaré avoir utilisé des préservatifs avant le diagnostic du VIH, tandis que 87% ont déclaré l’utilisation du préservatif après le diagnostic du VIH. L’utilisation du préservatif était plus fréquente chez les femmes qui avaient reçu des conseils post-test (OR: 3,03; IC95%: 1,03 à 8,90).

Conclusions:  Un changement dans les méthodes de contraception passant de l’utilisation de DIU et de la pilule orale avant le diagnostic du VIH à l’utilisation du préservatif après le diagnostic du VIH a été observé. L’utilisation du préservatif après le diagnostic du VIH était associée au conseil post-test reçu.

Abstract

Objetivo:  Investigar el uso de anticonceptivos entre mujeres VIH positivas en la ciudad de Ha long y el poblado de Cam Pha de Quang Ninh, una provincia en el norte de Vietnam.

Métodos:  Estudio crosseccional en mujeres VIH positivas identificadas a través del registro de VIH/SIDA del distrito. Mediante el uso de cuestionarios estructurados, se obtuvo información sobre las características socioeconómicas y el uso de anticonceptivos antes y después del diagnóstico de VIH. Se utilizó una prueba de Chi-cuadrado para evaluar si las mujeres incluidas diferían de la población objetivo en términos de edad. Se calcularon los odds ratios (ORs) crudos para mostrar la asociación entre el uso de anticonceptivos y las características socioeconómicas de las mujeres. Se realizaron análisis de regresión logística ajustando para posibles factores de confusión. Se describe el uso de anticonceptivos antes y después de la prueba de VIH y se comparan mediante la prueba de Chi cuadrado. La asociación entre el tener una sesión de aconsejamiento posterior a la prueba y el uso de preservativos se evaluó mediante ORs.

Resultados:  De las 351 participantes, un 63% afirmó que habían utilizado anticonceptivos antes del diagnóstico de VIH y un 89% afirmó que había utilizado anticoncepción después del diagnóstico de VIH. Un 46% de las mujeres había estado utilizando la píldora o un DIU (dispositivo intrauterino) antes del diagnóstico, mientras que esto era cierto solo a un 8% de las mujeres después del diagnóstico. Un 39% afirmó haber utilizado preservativos antes del diagnóstico del VIH, mientras que un 87% afirmó utilizar preservativos tras el diagnóstico de VIH. El uso de preservativos era más común entre mujeres que habían recibido aconsejamiento tras el diagnóstico (OR 3.03, 95% CI 1.03-8.90).

Conclusiones:  Se observó un cambio en los métodos anticonceptivos utilizados tras el diagnóstico de VIH: de DIU y anticonceptivos orales antes del diagnóstico, al preservativo después del mismo. El uso de preservativos por parte de las mujeres estaba asociado con haber recibido aconsejamiento posterior al diagnóstico.

Introduction

Vietnam has during the past two decades experienced a rapid transition from high to low fertility. In 2007, the country had a fertility rate of 2.1 children, down from 3.8 children in 1989 (VGSO 1997). The family planning program has played a critical role in the decline by making modern contraception more available for women in reproductive age. This trend is illustrated by an increase in contraceptive prevalence from 38% in 1988 to 68% in 2007 (VGSO 1997, 2008). At present, 55% are using an (IUD) intrauterine device, which makes it the most predominant of all contraceptive methods (VGSO 2008).

In Vietnam, the HIV/AIDS epidemic is in the stage of concentration, and the main risk factors are injecting drug and unsafe sex. In 2009, the number of people living with HIV/AIDS was 160 000 (UNGASS 2010). Over the past decade, there have been an increase in the number of women living with HIV and in 2009 women comprised 29% of the infected population (AVERT 2012). Although some women acquire HIV through injecting drugs and others while selling sex, the majority are exposed while having sex with a husband or partner who contracted HIV through injecting drugs, unprotected sex with a sex worker or unprotected sex with a male partner (UNAIDS 2010).

Since 2008, the strategy of the Vietnamese Government against HIV/AIDS has focused on three core preventive programs: Harm Reduction, Volunteer Counseling and Testing (VCT) and Prevention of Mother-to-child transmission (PMTCT). The Harm Reduction Program includes two major components: a Needle and Syringe Program and a Condom Use Program (UNGASS 2010). The aim of the Condom Use Program is to increase the availability of condoms, and in 2009, nearly 25 million condoms were distributed by the program. The existing VCT program was implemented in 2002 by the Vietnam Ministry of Health with support from international donors. The aim was to expand the national HIV testing capacity and make VCT freely available, especially to high-risk populations. The program seems to be successful in targeting high-risk populations and in changing their risky behaviour. Of a total of 158 888 records, 81% reported high-risk behaviour; condom use in this risk population ranged from 34% to 71%– after having attended VCT (Hong et al. 2011). Focusing on the PMTCT program, which has also received support from international donors, the number of antenatal care facilities offering PMTCT service has expanded from 21% in 2006 to 44% in 2009 (UNGASS 2010).

Contraceptive use by HIV-positive women is an area of particular interest. They use condoms or other types of family planning for various reasons: some women fear giving birth to an infected child, while others want to avoid infecting their partner and yet others want to focus their resources on maintaining their own health and the well-being of their family (De Bruyn 2005; Feldman & Maposhere 2003; Heard et al. 2004; Delvaux & Nostlinger 2007). Some HIV-positive persons may want to avoid HIV superinfection. In a study conducted among Vietnamese injecting drug users, the risk of superinfection was described to be a pronounced problem (Thanh et al. 2009). However, it may also be a problem in the general population, because some HIV-positive persons may have become involved in a new sexual relationship with a partner who may carry a new HIV type. From a Vietnamese perspective, there is a lack of data and studies on contraceptive use and experience among HIV-positive women. Thus, the aim of this study is to explore contraceptive practices among HIV-positive women in Ha Long city and Cam Pha town of Quang Ninh province, an area where massive investments have been made in Harm Reduction, VCT and PMTCT programs. The study also aims at describing factors that affect change of contraceptive use over time. A greater understanding of contraceptive attitudes among HIV-positive women may be helpful for the future scale-up of HIV preventive programs.

Methods

This cross-sectional study was conducted in Quang Ninh province during the period of April to October, 2007. The cumulative number of HIV-positive cases in the province was 17 917 in 2009, equivalent to a prevalence rate of 1% while the rate of the country was only 0.4% (UNGASS 2010). Data were collected in all communes of Ha Long city (20 communes) and Cam Pha Town (16 communes).

Study population

The women were identified through the HIV/AIDS register, a documentation of HIV-positive people maintained by the local health centres. The register contains information on names, sex, age, address and AIDS stage and includes all HIV-positive women who have been registered as HIV-positive since 2002. The list is updated weekly, to add new and remove deceased cases. In March 2007, 518 women in Ha Long city and Cam Pha town were listed in the HIV/AIDS register. When attempts were made to contact these women, it was found that 108 women had provided a wrong address and 24 were not at home at the time of the study. We thus managed to invite 386 HIV-positive women to participate in the study, of whom 351 (91%) accepted.

Data collection

A structured questionnaire was developed with the main topics: (i) socio-demographic characteristics; (ii) pregnancy experiences; (iii) contraceptive use; and (iv) time of HIV infection. The main outcomes of this study are ‘contraceptive use before HIV diagnosis’ and ‘contraceptive use at the time of the study’. ‘Contraceptive use before HIV diagnosis’ implied contraceptive use during 6 months before the women were diagnosed HIV-positive and ‘Contraceptive use at the time of the study’ implied contraceptive use 1 month prior to the interview. The contraceptive method used was classified into five categories: oral contraception, IUD, condom, natural family planning and others. Condom use was defined as using a condom at each act of sexual intercourse before HIV diagnosis and at the time of the study. A few of the questions such as motives for not using contraception were open-ended and the answers were subsequently translated to relevant categories.

Three teachers from the Provincial Secondary Medical College of Quang Ninh were recruited as interviewers. The first author carefully instructed the interviewers about how to find the women and how to perform the interviews. One health worker from each commune health centre accompanied each interviewer as a guide and contact person in the field. The health worker contacted all the HIV-positive women from the HIV/AIDS register on the phone or by visiting them. They introduced the study to the women and invited them to participate and if the women agreed an interview appointment was made.

The questionnaire was pilot-tested in a group of 20 women where the appropriateness of the content and the interview technique was assessed. The questionnaire was revised and then validated by the researchers, interviewers, supervisors and a specialist on HIV prevention from Viet Nam Administration of HIV/AIDS Control (VAAC).

The interviewers cross-checked the collected questionnaires daily and the first author supervised the process of data collection, the quality of data, entered data, and cleaned and checked the data set before analysis.

Statistical analysis

Statistical analyses were performed using the Statistical Package for the Social Sciences (SPSS), version 15.0. Chi-square testing was used to assess whether the included women differed from the target population in terms of age. Odd ratios (ORs) were calculated by a case–control approach, with the group of women using contraception as the dependent variable in comparison with women who were not using contraception. The association between contraceptive use and no contraceptive use is presented as OR with 95% confidence intervals (CI). Possible confounding by age, marital status and number of children alive was controlled for by including these factors in logistic regression analyses. Contraceptive use before HIV testing and at the time of study was compared by Chi-square testing and the association between post-test counselling and condom use was assessed through ORs, where condom use comprised the dependant variable and post-test counselling the independent variable.

Ethical considerations

Through collaboration with the local health system, an ethically acceptable strategy was developed to approach women living with HIV/AIDS. The main priority in the strategy in approaching the women was to assure them of conditions of strict confidentiality and anonymity. The participants were informed that their personal details would not be attached to the questionnaire responses. Informed consent was obtained and the study was ethically approved by the Scientific Committee of the Vietnamese Commission for Population, Family and Children and by the Danish National Committee on Biomedical Research Ethics.

Results

Table 1 summarises the age distribution of the HIV-positive women registered in the HIV/AIDS register and the HIV-positive women who accepted to participate in the study. The mean age of the study population was 31.5, compared with 31.6 of the women registered in the HIV/AIDS register. 62% of the women in the study population as well as the women who were registered in the HIV/AIDS register were 25–34 years old. The women in the study population tended to be slightly younger, 11% were 24 years or younger and 34% were 25–29 years old. The corresponding figures from the HIV/AIDS register were 9% and 32%. The difference, however, was not statistically significant.

Table 1.   Age categories of HIV-positive women
Age categoriesHIV/AIDS register N = 518Study population N = 351 P
n % N %
<24448.53710.50.67
25–2916431.712034.2 
30–3415930.79627.4 
35–3910119.56418.2 
40+509.6349.7 
Mean age31.6 31.5  

The women’s background characteristics in terms of socioeconomic factors and HIV situation are summarised in Table 2. More than half (54%) were widowed, 37% were married and 9% were single. The vast majority (93%) had been infected through sexual contact with their husbands and 14% had an HIV-infected child. 65% had been diagnosed for <3 years, 59% participated in clubs for people living with HIV/AIDS and 46% reported being on anti-retroviral treatment.

Table 2.   Description of participants in the study
Characteristics N = 351Per cent
  1. *Missing 7 cases.

Marital status
 Single/divorced318.8
 Married12936.8
 Widowed19154.4
Education
 No education51.4
 Some/completed primary school5515.7
 Completed lower secondary school14641.6
 Completed upper secondary school12936.8
 Higher164.5
Employment
 Unemployment7822.2
 Petty trade10329.3
 Independent labour7922.5
 Government staff72.0
 Farmer/worker/handicraft and other8424.0
Mode of transmission
 Drug use10.3
 Transmission from having sexual relation with husband32592.6
 Transmission from having sexual relation with partner92.6
 No identifiable risk164.5
How long ago HIV was diagnosed
 <3 years22965.2
 3–5 years7922.5
 >5 years4312.3
Number of children alive* (= 344)
 No children3710.8
 1 child29986.9
 2 + children82.3
Number of dead children (n = 351329.1
Number of HIV-positive children5014.2
Unmarried women having a sexual partner (n = 222)5123.0
Participation in club for people living with HIV/AIDS20859.3
Under ARV treatment16045.6

Table 3 describes the association between the women’s socioeconomic situation and their use of contraception. The crude ORs show a significant association between contraceptive use and the type of relationship the women were in. More specifically, women who were married had 11 times increased OR for using contraception in comparison with women who were single or divorced. In contrast, women aged 40 or older and women who had two or more children had lower ORs of 0.4 and 0.6 for using contraception in comparison with women who were aged 30–34 and women who had only one child, respectively. After adjusting for age, marital situation, cohabitation status and number of children alive, the only factor that remained significantly associated with contraceptive use was whether the women were married or not. Hence, the multivariable analysis showed that women who were married had a seven times higher OR for using contraception in comparison with women who were single/divorced, whereas no significant association was found between age, number of children and contraceptive use.

Table 3.   Demographic characteristics of women using contraception and women not using contraception at the time of the study
 Using contraception = 160Not using contraception = 191Using vs. not using OR (95% CI)Using vs. not using Adj OR (95% CI)
  1. *Missing 7 cases.

Age   = 0.0016 
 <2420 (12.5)17 (8.9)1.33 (0.62–2.85)0.65 (0.23–1.85)
 25–2963 (39.4)57 (29.8)1.25 (0.73–2.14)1.50 (0.72–3.1)
 30–3445 (28.1)51 (26.7)11
 35–3923 (14.4)41 (21.5)0.64 (0.33–1.22)0.68 (0.29–1.6)
 40+9 (5.6)25 (13.1)0.41 (0.17–0.97)0.39 (0.12–1.28)
Marital status
 Single/divorced10 (6.3)21 (11.0)11
 Married108 (67.5)21 (11.0)10.8 (4.45–26.2)7.33 (2.68–20.02)
 Widowed42 (26.3)149 (78.0)0.59 (0.26–1.35)0.69 (0.44–1.1)
Number of children alive*
 No children18 (11.5)19 (10.2)0.97 (0.48–1.95)0.48 (0.18–1.28)
 1 child102 (65.0)104 (55.6)11
 2 + children37 (23.6)64 (34.2)0.59 (0.36–0.96)0.57 (0.27–1.23)

Of the 351 women, 180 (51%) stated they were sexually active with either their husband or a sexual partner. In the group of sexually active women, 63% stated they had been using contraception before HIV diagnosis and 89% stated they were using contraception at the time of the study. The contraceptive methods described by the women included IUD, pill, condom and natural methods (periodic abstinence and withdrawal). The contraceptive method used before HIV diagnosis and at the time of the study differed, 47% stated they had been using either the pill or an IUD before the diagnosis whereas only 8% of the women stated use of the pill or IUD at the time of the study. When focusing on condom use, 39% of the women stated they had been using condom before HIV diagnosis, whereas 85% stated condom use when interviewed (Table 4). The proportion of women who were using dual protection, for example, IUD for contraception and condoms for HIV prevention/contraception, was low and post-test counselling was not associated with an increase in number of women using dual protection. More specifically, 5.6% of the women stated they were using dual protection before HIV diagnosis and 2.3% stated the same after HIV diagnosis.

Table 4.   Contraceptive use before HIV diagnosis and at the study time among women who were sexually active*
Kinds of contraceptionContraceptive use before HIV diagnosis (N = 180)Contraceptive use at the time of the study (N = 180) P
  1. *Sexually active women were defined as married women and single, divorced and widowed women who stated they had a sexual partner.

  2. †Total of the women who used different methods may exceed the total of the women who used contraception because some women used the mixed method of contraception.

Contraceptive use
 Yes114 (63.0)160 (88.9)<0.0001
 No66 (37.0)20 (11.1)
Type of contraception† (n = 114/160)
 Pill/IUD54 (47.)13 (8.1)<0.0001
 Condom44 (38.6)136 (85.0)
 Natural52 (45.6)13 (8.1)

Among sexually active women who were using contraception at the time of the study, the main reason for using contraception was to avoid becoming infected with a new type of HIV through sexual contact (44%). The majority of the women (69%) who used condoms claimed that they used them to avoid both HIV super infection and unwanted pregnancies, whereas one-third stated that they used condoms only to prevent HIV super infection. Another prevalent reason for contraceptive use was fear of giving birth to an infected child, which 25% of the women stated was their main reason for using contraception.

Condoms were the main contraceptive method used and women who had received post-test counselling had a three times increased OR for using condom in comparison with women who had not received post-test counselling (Table 5).

Table 5.   Association between post-test counselling and condom use
 Condom use (= 136)No condom use (= 44) OR (95% CI)
 n (%) n (%)
Being counselled128 (94.1)37 (84.1)3.03 (0.87–10.02)
Not being counselled8 (5.9)7 (15.9)1

Discussion

This study is the first study examining contraceptive use among the general population of HIV-infected women in Vietnam. The findings seem to indicate that women are more likely to use contraception after being diagnosed HIV positive and also more likely to use a condom. Women who had received post-test counselling had a three times higher OR for using a condom than women who had not.

A limitation of the study is that it only covered 64% of the women who were listed in the HIV/AIDS register in Ha Long and Cam Pha communes. The women’s age was the only parameter by which we were able to judge how representative the study population was. A comparison of this variable showed no statistical difference in age distribution between women registered in the HIV register and in our study sample, a finding that indicates that the study population age wise may be considered a representative subsample of the population of registered HIV-positive women in Ha Long and Cam Pha. However it seems that women of reproductive age are overrepresented in our study sample as well as in the HIV/AIDS register. In a study focusing on contraceptive use, the lack of inclusion of women beyond reproductive age is not a major problem because they have little rationale for using contraception. On the other hand, the absence of older women in the present study may be considered problematic when it comes to HIV prevention, because older women may have concerns regarding how to avoid HIV transmission.

Structured questionnaire interviews were used to obtain information about the women’s contraceptive use. Such reliance on self reported contraceptive use may be criticised because of the questionable reliability of the subjectively reported use. Hence, it may not be considered appropriate for a single, divorced or widowed woman to discuss contraception and sexual activity and information bias could thus be a problem. However, the study was conducted in an area with many peer groups for HIV-positive women. Most of the women interviewed were therefore used to discussing safe sex and contraceptive use. In addition to get as valid information as possible, the interviewers were trained in performing the interviews as a friendly dialogue, free of moral judgment and the women were assured confidentiality and anonymity. These conditions made it possible to have an open-minded interview and allowed the women to be frank in the interview situation.

Recall bias may be a problem in the present study. Sixty five per cent of the women had received the HIV diagnosis <3 years before the study and 12% had received the diagnosis 5 or more years before. The time span the women were asked to recall may have led to differences in the accuracy of the reported contraceptive use, especially among those women who were asked to remember a pre-diagnosis period of five or more years. If some women understated past contraceptive use because of recall bias, the impact of post-test counselling may be overestimated in the present study.

The contraceptive use among the HIV-positive women reflects their relationship. Women who were married had a higher contraceptive prevalence rate than women who were single or divorced. This finding is consistent with a number of studies. For instance, in a US based study, 43% of the HIV-positive women who were married used contraception in comparison with 14% of the women who were not (Standwood et al. 2007). Another study from the US has also shown that contraceptive use is more frequent among HIV-positive women if they are in a steady relationship (Massad et al. 2007). A similar observation was made in Kenya, where 35% of the married used contraception in comparison with 9% of the single women (Mutiso et al. 2008).

Before HIV diagnosis, the women did not use contraception because they ‘wanted to have children’ while after HIV diagnosis the most prevalent answer was no contraceptive use because of ‘being sexually inactive’. This finding is in contrast to the motives for contraceptive use found in the National Population Change and Family Planning (NPCFP) survey, where 42% of the general population of Vietnamese women reported ‘child desire’ to be the main reason for not using contraception (VGSO 2008). Our findings are in line with other studies which have explored contraceptive use among HIV-positive women (Desgrées-du-Lou et al. 2002).

Half of the HIV-positive women who participated in the current study were sexually active. This is a relatively low rate in comparison with other studies which have documented that approximately 70% of women with HIV infection are sexually active (Mitchel & Stephens 2004) and may be explained by the fact that more than half of the women were widows. The contraceptive prevalence rate among women who were sexually active in the current study was slightly higher than among the general population; 87% of the women were using contraception whereas 79% of women from the NPCFP survey stated they were using contraception in 2007 (VGSO 2008). Among sexually active women who were using contraception at the time of the study, the main reason was to avoid becoming infected with a new type of HIV through sexual contact (44%). Among women who were sexually active, the rate of contraceptive use at the time of the study was higher than before HIV diagnosis (87%vs. 63%). The high contraception prevalence rate found in this study may reflect that many women did not want to give birth after they were diagnosed with HIV. A similar result has been reported in a US based study on contraception and fertility plans of HIV-positive women where only 7% of the women stated they were not using any contraception after HIV diagnosis (Standwood et al. 2007). In contrast, studies from Côte d’Ivoire and Kenya have documented that only 39% and 44%, respectively, of HIV-infected women stated they were using contraception (Desgrées-du-Lou et al. 2002; Mutiso et al. 2008).

The women were more likely to use condoms after having received a HIV-positive diagnosis, more specifically 39% of the women stated they had been using condoms before HIV diagnosis whereas 87% reported they were using condoms at the time of the study. During the same time period, the use of IUD and oral contraception decreased significantly. A similar change in contraceptive use before and after HIV diagnosis has been observed in US where a permutation between the condom and other contraceptive methods has been documented. In the study from US, 81% of the HIV-positive women reported use of the male condom and 25% reported use of the female condom after HIV diagnosis, whereas the use of other contraceptive methods decreased significantly (Standwood et al. 2007). In the present study, avoiding infection with a new type of HIV and preventing unwanted pregnancy were the main purposes of women’s condom use. Studies on contraceptive use among HIV-infected women in Europe, the United States, Africa and Asia have also proven that the condom is the preferred contraceptive method among HIV- positive women and women mainly used condom for HIV/STD prevention rather than fertility control (Heard et al. 2004; Mutiso et al. 2008; Mark et al. 2007; Raiford et al. 2007). For example, in a study from the US, 70% of HIV-positive Afro American women stated they had used the condom in the past 30 days (Raiford et al. 2007) while in a Kenyan study conducted among HIV- positive women, 82% of the women, who stated contraceptive use, were using it (Mutiso et al. 2008). The high rate of condom use in the current study was associated with post-test counselling. This may reflect a positive effect of the HIV counselling and testing program in the setting studied and is in line with a recent study reporting condom use rates of 34% to 71% in a Vietnamese HIV high-risk population who had participated in a country wide VCT program (Hong et al. 2011). The assumption is further supported by a recent study reporting that women who had participated in the VCT program in Quang Ninh province were satisfied with the service and found that the health staff provided good emotional support, counselling and care through post-test counselling (Hanh et al. 2008).

Dual protection was not common in the present study, a finding which is in contrast to WHO’s current recommendations suggesting that condoms are used in combination with hormonal contraceptives or IUD to prevent both HIV spread and unwanted pregnancies (WHO 2012). The structured questionnaire approach did not allow for detailed information about the women’s notion of condoms as a means of HIV prevention. To better understand how HIV-positive women perceive the role of contraception and condom in relation to HIV prevention as well as a birth control more qualitative studies are needed.

Conclusion

This study has shown a high contraceptive prevalence rate among sexual active women with HIV infection. It further documents that women are more likely to use condom after having received a HIV-positive diagnosis and this tendency was especially pronounced among women who had received post-test counselling. Apparently post-test counselling is effective in providing knowledge on HIV transmission to an uninfected partner or on how to avoid becoming infected with a new type of HIV. This was a cross-sectional study, which explored HIV-positive women’s contraceptive use. However, to get more detailed information about HIV-positive women’s contraceptive decision-making, qualitative studies are needed. Well-designed prospective cohort studies are warranted to better describe factors associated with change in contraceptive use over time among HIV-positive women. Such information will help healthcare providers to better direct their counselling to meet and address HIV-positive women’s contraceptive priorities and concerns.

Acknowledgements

We would like to thank all the women who participated in the study for their information and time and the authorities in the study site for their support during data collection. This work was supported by the project ‘Strengthening Population and Reproductive Health Research in Vietnam’, funded by the Danish International Development Assistance (Danida).

Ancillary