Sexually Transmitted Diseases, November 2007, Vol. 34, No. 11, p.000 – 00 DOI: 10.1097/OLQ.0b013e31813e0c15 Copyright © 2007, American Sexually Transmitted Diseases Association All rights reserved.

Education Attainment and the Risk of HIV-1 Infections in Rural Kilimanjaro Region of Tanzania, 1991–2005. A Reversed Association ELIA J. MMBAGA, MD, MPH,*† GERMANA H. LEYNA, MD, MPH,*‡ KAGOMA S. MNYIKA, MD, PHD,* AKTHAR HUSSAIN, MD, PHD,† AND KNUT-INGE KLEPP, MPH, PHD‡

From the *Department of Epidemiology and Biostatistics, Muhimbili University of Health Sciences, Tanzania; and Departments of †General Practice and Community Medicine and ‡Nutrition, University of Oslo, Oslo, Norway

Objective: Previous studies found educated individuals to have higher risk of human immunodeficiency virus (HIV)-1 infection in Africa. A reverse in this association was predicted. We investigated the change in this association from 1991 to 2005 in a rural population in Tanzania. Study Design: Two cross-sectional surveys were conducted in 1991(N ⴝ 1152, response rate 76.4%) and 2005 (N ⴝ 1528, 73.0%). Consenting individuals aged 15 to 44 years living in Oria village were interviewed and gave blood for HIV-1 testing. Results: Primary [adjusted odds ratio (AOR), 2.7; 95% confidence interval (CI): 1.3–20.0] and secondary/higher education (AOR, 4.5; 95% CI: 1.4 –24.9) were associated with increased risk of HIV-1 infection in 1991. A reversed association was observed in 2005 where reduced odds of infection were associated with primary (AOR, 0.5; 95% CI: 0.2– 0.8) and secondary/higher education (AOR, 0.4; 95% CI: 0.3– 0.9). This was most pronounced among educated men. Corresponding reduction in risk behaviors were observed. Increased odds of reporting ever use of condom (AOR, 2.8; 95% CI: 1.1–7.3) and 70% reduction in reporting >2 sexual partners in the past year was observed among educated women. Educated men reported 60% reduction in the odds of reporting >2 sexual partners in the past month preceding the last survey. Conclusions: A reversed association between education attainment and HIV-1 infection was observed in this population between 1991 and 2005. Education seems to have an empowering role in women. Decreased risk among educated men may have an impact on HIV-1 transmission. Improving education sector in rural areas might be instrumental in the fight against the HIV epidemic.

income, and ability to exchange money for sex. Infectious diseases have been linked with poverty and illiteracy. As the HIV epidemic evolves and becomes generalized, it was believed that the pattern of increased risk among educated individuals will change adopting the usual pattern of other infectious diseases.10 The negative correlation between education attainment and HIV-1 infection has been explained to be due to the role of education as a “social vaccine.”11,12 The education vaccine may work in 2 ways: exposure to acquired immune deficiency syndrome (AIDS) information and sex education in school and the role of basic education in equipping and empowering people to understand and internalize relevant information and translate knowledge into behavioral change. Education attainment also helps to change the community environment in which behavioral change becomes socially acceptable.12,13 Studies have shown that, even attending and completing primary education might have an effect in the spread of HIV-1 infection compared to not attending school at all. This is supported by findings from a number of studies that staying in school even if it did not advance literacy skills was helpful in impacting correctional behavioral outcomes.14 As was the case in most African studies, earlier studies in Tanzania showed an increased risk of HIV infection among educated individuals.2,6,7,15–17 This relationship was more significant in rural areas than in urban areas.15,16,18 Evidence is now mounting on the reversed association between education attainment and HIV-1 infection in Zambia, Ethiopia, and Uganda.5,11,19 –24 Moreover, a review by Hargreaves and Glynn concluded that pattern of new HIV infections may be changing toward a greater burden among less educated groups.3 In Tanzania, hard data are lacking and recent studies have failed to show the reported reversed association.18 Local descriptions of the reversed association are important in redefining the package of current interventions and the group at risk. We surveyed a rural population in Kilimanjaro Tanzania, in 1991; the beginning of the epidemic in this population, and again in 2005 when the epidemic was mature.25,26 Crude analyses have shown increasing HIV-1 prevalence in this rural population with less educated individuals at an increased risk.26 To investigate

IN THE BEGINNING OF the human immunodeficiency virus (HIV) epidemic in Africa, it was shown that HIV infection increased with increased level of education and socioeconomic position.1–9 High education and socioeconomic position were associated with increased opportunity for travel, disposable

The authors thank the study population for their participation. Supported by a grant from the Norwegian Council for Higher Education’s Program for Development Research and Education and facilitated by the collaborating institutions: Muhimbili University of Health Sciences, Kilimanjaro Christian Medical College, Centre for Education in Health Development Arusha in Tanzania, and the Universities of Oslo and Bergen, Norway. Additional financial support was provided through the Centre for Prevention of Global Infections (GLOBINF) at the University of Oslo. Correspondence: Elia John Mmbaga, MD, MPH, Department of General Practice and Community Medicine, University of Oslo, P.O. Box 1130, Blindern, N-0318 Oslo, Norway. E-mail: [email protected]. Received for publication March 13, 2007, and accepted June 12, 2007.

1

2

MMBAGA ET AL

these dynamics further between 1991 and 2005, we compared prevalence rates across education level from the 2 surveys. To explain the observed changes in HIV prevalence over time, we described changes in risk sexual behaviors across education levels. Materials and Methods Study Area and Population The data in this article comes from studies conducted in Oria, a rural village located 30 km south of Moshi town at the foot of Mount Kilimanjaro in Kilimanjaro region, Tanzania. Details of this study area have been described elsewhere.25,26 In brief, the village is located in Moshi rural district, 1 of the 6 districts in the region. Most of the villagers are peasant farmers, laborers, craftsmen, and some engage in seasonal work at the nearby rice plantations. The village has 2 primary schools and 1 secondary school that provide ordinary (4 years) secondary education. Data Collection Two cross-sectional surveys were conducted in the village in 1991 and 2005 (N ⫽ 1152 and 1528). The response rates achieved in these surveys were 76.4% (1152 of 1507) for 1991 and 73.0% (1528 of 2093) for 2005. The overall age and sex adjusted prevalence of HIV-1 infection was 3.2% and 5.9% in 1991 and 2005 survey, respectively (Mmbaga et al., unpublished data, 2007). Eligibility in all the surveys was based on having a residential address in the village and age between 15 and 44 years at the time of the study. All eligible individuals were registered by the research team and then invited to participate at all surveys. In the first survey in 1991, a temporary survey center was established in the village and those eligible were invited to participate. During the follow-up survey in 2005, the research team did house-tohouse visits and conducted the interviews and sampling at the participants homes. Information on study aims and procedure was given to the participants before participation in both surveys, and those who agreed to participate provided an informed consent. A structured questionnaire was administered where sociodemographic data and behavioral risk factors for HIV-1 infection were collected. Data on education status was based on the Tanzanian system of education. In this system, individuals attend 7 years of primary education, 4 years of ordinary secondary education, and 2 years of advanced secondary education. After ordinary secondary education or advance secondary education, individuals qualify for college education. Although fees are required for other levels of education, primary education is free and compulsory for all Tanzanians. Failure to enroll an eligible child could lead into jail time for the parent or guardian. There is also a nationwide encouragement for parents to enroll their children for primary education. The interviews were then followed by a pretest counseling and venous blood collection for HIV-1 testing. Appointments were made with each participant for posttest counseling within 2 weeks from the date of sample collection. Test results were issued to respondents in person and those who were HIV-1 positive, were offered medical follow-up. Ethical Consideration Both surveys were approved by the Ethical Committee of the Ministry of Health in Tanzania and the Norwegian Committee for Medical Research Ethics. The village government of Oria gave permission for the study. As mentioned above, participants gave their consent before participation. Free access to treatment of various medical conditions was provided by the research team during the survey period. The treatments were provided equally for

Sexually Transmitted Diseases

●

November 2007

participants, nonparticipants, and their families, and medical conditions treated were not a function of participation. Testing Strategies All blood samples in all surveys were tested for HIV-1 antibodies using a similar testing algorithm. Details of the blood testing strategies have been presented in detail elsewhere.26 In summary, HIV-1 antibodies were detected using 2 independent recombinant enzyme-linked immunosorbent assay systems. All discordant or doubtful sera were confirmed by Western blot (Organon, Epitope, Beaverton, OR). Quality control of the HIV-1 testing was done at the Centre for International Health, University of Bergen by retesting 11.2% of the sera in 1991 where excellent agreement was found with the testing performed at Kilimanjaro Christian Medical Centre.25 Statistical Methods Data from the interviews and laboratory tests were checked for consistency, entered, and analyzed using Statistical Package for Social Scientists versions 13.0 (SPSS, Chicago, IL). Education levels were defined as ⬍7 years primary education (no or ⬍7 years of primary education), primary education completed (7 years of education), and secondary education or higher (⬎7 years of education). Only a small fraction of participants had postsecondary education [7 (0.6%) in 1991 and 15 (1.0%) in 2005]. These were, therefore, included in the last category of secondary education or higher. HIV-1 prevalence was computed using cross-sectional measures. The numerator was the number of adults with an HIV-1 seropositive status and the denominator was the total number of adult residents with a definitive HIV-1 serostatus at that particular survey. Comparison of HIV-1 prevalence by various sociodemographic characteristics between 1991 and 2005 was done where ␹2 test for difference in proportion was used. For each survey, overall trends in HIV-1 prevalence across education level were computed and P values for trends in proportions are presented. To follow the changes in the association between HIV-1 prevalence and education level between 1991 and 2005, crude odds ratios by education level were estimated. We subsequently presented adjusted odds ratios (AOR) for sex, marital status, and age as continuous variable. To be able to explain the changes in HIV-1 prevalence, we compared the sexual risk behaviors (condom use and number of sexual partners) by education level. These analyses were stratified by sex. All the reported confidence intervals (CIs) were at 95% level and the P values were 2-tailed. Results The comparison of HIV-1 prevalence by sociodemographic characteristics between 1991 and 2005 surveys are depicted in Table 1. Significant increase in HIV-1 prevalence were observed among women, individuals aged 35 to 44 years, married/cohabiting, divorced/separated individuals, those with primary education (completed) or less, Chagga and other mixed tribes, farmers, and Catholics. The distribution of participants by education levels in 1991 were ⬍7 years of primary 93 (8.1%), primary (completed) 985 (85.5%), and secondary or higher 74 (6.4%). The corresponding distribution for 2005 were 161 (10.5%), 1250 (81.8%), and 117 (7.7%), respectively (Table 1) (Mmbaga et al., unpublished data, 2007). In the 1991 survey, the overall prevalence of HIV-1 infection by education level was ⬍7 years of primary 1.2%, primary (completed) 3.2%, and secondary or above 5.3% (Ptrends ⬍0.001). The corresponding prevalence’s in the 2005 survey were 11.2%, 5.8%, and 5.7%, respectively (Ptrends 0.028). The sex specific patterns of

EDUCATION ATTAINMENT AND THE RISK OF HIV-1 INFECTIONS

Vol. 34 ● No. 11

TABLE 1. Comparison of HIV-1 Prevalence by Sociodemographic Characteristics Between 1991 and 2005 Survey in Rural Kilimanjaro, Tanzania 1991 Variable Sex Male Female Age groups 15–24 25–34 35–44 Marital status Single Married/cohabiting Divorced/separated Education level ⬍7 yr primary Primary (completed) Secondary⫹ Tribe Chagga Pare/Kahe Other Occupation Farmer Employed Business Religion Catholic Protestant Muslim

2005

N

HIV (%)

N

HIV (%)

P*

517 635

1.8 4.4

614 914

3.5 8.4

0.058 0.002

523 319 310

2.0 5.0 2.4

575 534 419

2.2 8.4 9.8

0.858 0.061 ⬍0.001

447 575 130

2.7 3.3 8.0

508 875 145

1.9 7.3 15.9

0.557 0.002 0.043

93 985 74

1.2 3.2 5.3

161 1250 117

11.2 5.8 5.7

0.008 0.002 0.917

321 459 372

3.3 3.5 3.1

438 563 527

7.6 5.4 6.6

0.017 0.125 0.023

1037 68 47

3.2 3.4 6.3

1327 49 152

6.3 2.1 10.8

⬍0.001 0.486 0.496

322 279 551

2.3 5.2 3.0

404 471 653

7.0 6.9 5.8

0.005 0.567 0.329

*P value from ␹2 test for difference in proportions.

HIV-1 prevalence by education level from 1991 to 2005 are illustrated in Figures 1 and 2. In 1991, HIV-1 prevalence increased with education both for women and men. In contrast to the above, a reversed pattern was observed in the 2005 survey (Figs. 1 and 2).

Fig 1. Pattern of HIV-1 prevalence by education level from 1991 to 2005 among women in rural Kilimanjaro Tanzania.

3

From the 1991 overall analyses, individuals with primary education (AOR, 2.7; 95% CI: 1.3–20.0) and secondary education or higher (AOR, 4.5; 95% CI: 1.4 –24.9) were at increased risk of infection when compared with those with ⬍7 years of primary education. These associations were reversed in 2005, where individuals with primary education (AOR, 0.5; 95% CI: 0.2– 0.8) and those with secondary education or higher (AOR, 0.4; 95% CI: 0.3– 0.9) had a reduced odds of infection when compared with those with ⬍7 years primary education. The associations between education level and HIV-1 prevalence by sex in 1991 and 2005 are shown in Table 2. Although women with completed primary education had almost 3 times the likelihood of been infected with HIV infection, the corresponding likelihood of infection was 6 times among those with secondary education or higher. In the same year (1991), no men was infected in the ⬍7 years primary education group, but HIV-1 prevalence was higher among those with higher education. In the 2005 survey, the data indicate that higher education level was associated with lower HIV-1 prevalence. Women who completed primary education and above had 30% to 50% less risk of HIV-1 infection compared with those with ⬍7 years of primary education. The corresponding group of men had 60% to 70% reduced risk of infection (Table 2). Associations between education level and HIV-1 risk sexual behaviors are depicted in Table 3. The likelihood of having ever used a condom increased with increased education level among women (AOR, 2.8; 95% CI: 1.1–7.3). Women with secondary education or higher also recorded a 70% decreased odds of reporting multiple sexual partners in the past year preceding the survey (AOR, 0.3; 95% CI: 0.1– 0.7). This was in contrast to what was observed in 1991 where the increased likelihood of reporting ⱖ2 sexual partners among those with secondary education or higher in the past year was recorded (AOR, 2.1; 95% CI: 1.2–7.4). The relationship between education level and reported sexual risk behaviors was similarly reversed among educated men. When compared with noneducated men (⬍7 years primary education), educated men had almost 3 times higher odds (AOR, 2.7; 95% CI: 1.3–9.9) of reporting ⱖ2 sexual partners in the past month preceding the 1991 survey, this was reversed in 2005 where educated men had 60% lower

4

MMBAGA ET AL

Sexually Transmitted Diseases

●

November 2007

Fig 2. Pattern of HIV-1 prevalence by education level from 1991 to 2005 among male in rural Kilimanjaro Tanzania.

risk of reporting ⱖ2 sexual partners when compared with noneducated men (Table 3). Discussion This article reports a reversed association between HIV-1 infection and education attainment in rural Kilimanjaro, Tanzania. The overall change in the risk of HIV-1 infection from educated to noneducated individuals was observed between 1991 and 2005. The observed reversed association was highly significant among men. The risk of HIV-1 infection was reduced by 70% among men who attained secondary school or higher. This risk reduction among educated men in rural areas has recently been reported in Zambia24 and Ethiopia.23 As in the Zambian study, the modest decline in the risk of infection was also recorded among educated women. A decline in HIV-1 prevalence among educated women was also seen in this population. Despite large number of studies in Africa with information on education level and HIV-1 infection, the association between education attainment and HIV-1 infection itself

TABLE 2.

has not been the focus of analysis in those studies.15,18,27 This has therefore shielded thorough assessment of the changing role of education in HIV-1 transmission in Africa. This is the first study in Tanzania to focus in its analysis and show in hard data a reversed association between education attainment and HIV-1 infection. The patterns of behavioral change corroborated the findings of reversed association between education attainment and HIV-1 prevalence in this population. Decreased prevalence of HIV-1 infection among educated individuals was associated with increased practice of safer sexual behaviors. Educated women were 3 times more likely to report ever use of condom in 2005 than noneducated women. With the increased condom use, educated women also had 70% decreased odds of reporting ⱖ2 sexual partners in the past year. These findings are of paramount importance as education seems to empower women, promote self-efficacy, and increase their negotiating power during sexual acts.28,29 In rural sub-Saharan Africa, traditional normative behaviors such as dominance against women and

Changes in the Association Between Education Level and HIV-1 Prevalence, 1991–2005, by Sex in Rural Kilimanjaro, Tanzania 1991

Education Level Female ⬍7 yr primary education Primary education Secondary education⫹ Male ⬍7 yr primary education Primary education Secondary education⫹

2005

% HIV⫹ (n)

OR (95% CI)*

AOR (95% CI)†

% HIV⫹ (n)

OR (95% CI)*

AOR (95% CI)†

1.6 (61) 4.5 (552) 9.5 (21)

1.0 2.7 (1.3–41.0) 6.3 (1.5–73.5)

1.0 2.6 (1.3–42.0) 6.0 (1.5–73.2)

13.4 (104) 7.6 (738) 8.6 (72)

1.0 0.5 (0.2–1.2) 0.6 (0.2–1.6)

1.0 0.5 (0.3–1.9) 0.7 (0.3–2.5)

0.0 (32) 1.6 (433) 2.8 (53)

— — —

— — —

6.8 (57) 3.2 (512) 2.8 (45)

1.0 0.4 (0.1–0.8) 0.3 (0.1–1.4)

1.0 0.4 (0.1–0.9) 0.3 (0.1–1.7)

⬍7 yr primary education category was used as a reference category. *Unadjusted for marital status and age as continues variable. † Adjusted for marital status and age as continues variable. OR shown in bold are significant at the 5% level. OR indicates odds ratio; AOR, adjusted odds ratio; CI, confidence interval.

EDUCATION ATTAINMENT AND THE RISK OF HIV-1 INFECTIONS

Vol. 34 ● No. 11

5

TABLE 3. Changes in the Association Between Education Level and Behavioral Risk Factors From 1991 to 2005 in Rural Kilimanjaro, Tanzania 1991 Variable Female Ever used condom ⬍7 yr education Primary Secondary⫹ ⱖ2 Partners past mo ⬍7 yr education Primary Secondary⫹ ⱖ2 Partners past yr ⬍7 yr education Primary Secondary⫹ ⱖ2 Partners past 5 yr ⬍7 yr education Primary Secondary⫹ Male Ever used condom ⬍7 yr education Primary Secondary⫹ ⱖ2 Partners past mo ⬍7 yr education Primary Secondary⫹ ⱖ2 Partners past yr ⬍7 yr education Primary Secondary⫹ ⱖ2 Partners past 5 yr ⬍7 yr education Primary Secondary⫹

% (n)

2005

AOR (95% CI)*

% (n)

AOR (95% CI)*

22.2 (61) 24.7 (475) 21.4 (13)

1.0 1.2 (0.3–4.1) 0.9 (0.1–5.1)

20.2 (36) 30.5 (717) 47.5 (59)

1.0 1.5 (0.6–3.4) 2.8 (1.1–7.3)

3.3 (61) 3.4 (475) 4.3 (13)

1.0 1.0 (0.2–4.5) 1.3 (0.7–6.9)

5.6 (36) 3.7 (707) 3.5 (57)

1.0 0.6 (0.1–2.8) 0.5 (0.1–4.1)

14.8 (61) 11.8 (475) 35.4 (13)

1.0 0.9 (0.2–1.3) 2.1 (1.2–7.4)

16.2 (36) 11.2 (704) 7.0 (57)

1.0 0.6 (0.2–1.6) 0.3 (0.1–0.7)

29.5 (61) 29.4 (475) 45.4 (13)

1.0 1.0 (0.5–1.7) 1.9 (0.9–4.6)

25.0 (36) 21.5 (711) 24.1 (56)

1.0 0.8 (0.3–1.7) 0.8 (0.3–2.2)

18.2 (20) 36.8 (410) 38.2 (35)

1.0 2.2 (0.4–10.9) 2.4 (0.5–13.0)

47.1 (17) 47.0 (470) 55.4 (56)

1.0 1.0 (0.3–2.7) 1.4 (0.4–4.4)

10.0 (20) 16.1 (410) 27.1 (35)

1.0 1.6 (0.3–7.4) 2.7 (1.3–9.9)

18.8 (16) 11.2 (460) 9.3 (54)

1.0 0.5 (0.1–1.9) 0.4 (0.1–0.7)

45.0 (20) 46.6 (410) 57.1 (35)

1.0 1.1 (0.3–1.9) 1.1 (0.3–3.5)

23.5 (17) 28.0 (457) 26.4 (53)

1.0 1.1 (0.3–3.7) 1.1 (0.3–3.9)

65.0 (20) 75.4 (410) 85.7 (35)

1.0 1.9 (0.7–5.1) 3.8 (1.1–14.6)

52.9 (17) 46.2 (470) 43.6 (55)

1.0 0.7 (0.2–1.8) 0.6 (0.2–1.9)

⬍7 yr primary education category was used as a reference category. *Adjusted for marital status and age as continues variable. OR in bold are significant at the 5% level. AOR indicates adjusted odds ratio; CI, confidence interval.

men having concurrent partners are socially acceptable. The empowering role of education is an important tool for women who want to change their behaviors in ways that may conflict with this traditional norms and customs. A change toward safer sex among educated men was also observed. When compared with the noneducated, educated men had 60% decreased odds of reporting ⱖ2 sexual partners in the past month preceding the 2005 survey. This is the reverse of what was observed in 1991. Adoption of safer sex among educated men is encouraging because studies have shown that marriage occurs predominately between people of the same socioeconomic status and that a woman’s risk of infection is associated with the husband’s behavior and education level.1 All these findings of positive behavioral change associated with decreased HIV-1 infection among educated individuals have been reported in other studies in Africa.19,21,22,24,28 –30 The association between these behaviors and HIV-1 infection in this population have been presented elsewhere.26 All these behaviors were associated with HIV-1 transmission in this population hence change in these behaviors as observed among educated individuals may have a positive impact in reducing HIV-1 transmission.

The positive response of educated individuals to preventive messages could explain the current observation on reversed association between education attainment and HIV infection. Both AIDS-specific and basic education play an equally important role in behavioral change that reduces the risk of HIV infection. Staying in school is also associated with increased likelihood of exposure to health education including knowledge on HIV/AIDS and other sexually transmitted infection prevention.10,12,31–33 When AIDS-specific health education diffuses into the community, educated individuals are more likely to process, understand, adapt quickly, and act on health promotion messages.5,12,13,31 This may lead to behavioral change faster than among illiterate people in terms of delaying first sexual encounter and marriage, reducing number of sexual partners and negotiating, accepting and increasing condom use.32,34 This could thus mean that through adoption of safer sexual behaviors, educational attainment acts both in reducing the risk and vulnerability to HIV infection. HIV/AIDS education was not provided in Tanzanian public schools until recently. However, an in-school health education campaign called “Afya kwa vijana” (“Health for youth” in

6

MMBAGA ET AL

Swahili the national language used in Tanzania and most of east Africa) was introduced in the study area involving grade 6 and 7 pupils in late 1990s until 2005.33 The program goal was to reduce students’ risk of HIV infection by delaying initiation of sexual activity. The programs focused on increasing HIV/AIDS transmission knowledge, skills on condom use, improve perceived susceptibility, and foster communication on HIV/AIDS. This program was implemented in the area until 2005. Individuals who attended primary school up to at least grade 6, had the opportunity to be exposed to the program and this might explain the observed positive behavioral change among the educated group. In our study no significant increase in HIV-1 prevalence was noted among young age groups (15–34 years). This could be because this younger group who went to school more recently when compared with the older group (35– 44 years) may have benefited from the HIV/AIDS education campaigns hence adopting safer sexual behaviors (Mmbaga et al., unpublished data, 2007).26 Reproductive health education was recently introduced to all Tanzanian public primary and secondary education curriculums, meaning that all schools are now offering skills vital in mitigating the spread of HIV infection. In Africa, majority of the population live in rural areas (78% in the case of Tanzania) where the level of education is known to be low compared with urban population. Although primary education is still free in Tanzania, the rates of school enrolment are still low in rural areas because of poor motivation, and follow-up to ensure adequate enrolment. In this article, we have demonstrated that even attending 7 years of primary education could have an important impact on HIV-1 risk. These findings should encourage more involvement of parents and all education stakeholders in the fight against the epidemic by promoting increased educational attainment to the younger generation with vigorous follow-up and legal action for those failing to enroll their children. These measures may not only save their lives but open more doors for economical development. The results of this article may be affected by a number of biases. First, change in population structure over time might have affected comparability between the 2 cross-sectional studies. Moreover, we only achieved a response rate of about 75% in both studies. Although this response rate is acceptable, selection bias because of the 25% nonrespondents could have biased our findings. Second, we have assessed only the level of education in this study but the quality of the education received was not possible to assess. Finally, it is possible that education attainment may influence reported behavior rather than actual behavior. Despite this, the observed decrease in HIV-1 prevalence (a biologic marker out of participant’s control) among educated individuals suggests that the reported decrease in risk behavior might reflect the actual behavior. In conclusion, although HIV-1 infection increased in rural Kilimanjaro, the risk of infection seems to have shifted from educated to noneducated individuals. Reversed association between education attainment and HIV-1 infection was observed between 1991 and 2005. This reversed risk was corroborated by an increase in reported practice of safer sexual behaviors among educated individuals. Educated individuals might have benefited most from health education campaigns. Increased condom use among educated women suggests that education could be an important tool to empower women and increase their negotiating power. The role of the education sector in the fight against HIV/AIDS should be emphasized, as the more educated society is more responsive to health information.

Sexually Transmitted Diseases

●

November 2007

References 1. Dallabetta GA, Miotti PG, Chiphangwi JD, et al. High socioeconomic status is a risk factor for human immunodeficiency virus type-1 (HIV-1) infection but not for sexually transmitted diseases in women in Malawi: Implication for HIV-1 control. J infect Dis 1993; 167: 36 – 42. 2. Grosskurth H, Mosha F, Todd J, et al. A community trial of the impact of improved sexually transmitted disease treatment on the HIV epidemic in rural Tanzania: Baseline survey results. AIDS 1995; 9:927–934. 3. Hargreaves JR, Glynn JR. Educational attainment and HIV-1 infection in developing countries: A systematic review. Trop Med Int Health 2002; 7:489 – 498. 4. Killewo J, Nyamuryekunge K, Sandstrom A, et al. Prevalence of HIV-1 infection in the Kagera region of Tanzania: A populationbased study. AIDS 1990; 4:1081–1085. 5. Killian AH, Gregson S, Ndyanabangi B, et al. Reduction in risk behaviours provides the most consistent explanation for the declining HIV-1 prevalence in Uganda. AIDS 1999; 13:391–398. 6. Kwesigabo G, Killewo J, Godoy C, et al. Decline in the prevalence of HIV-1 infection in young women in the Kagera region of Tanzania. J Acquir Immune Defic Syndr 1998; 17:262–268. 7. Senkoro KP, Boerma JT, Klokke AH, et al. HIV incidence and HIV associated mortality in a cohort of factory workers and their spouses in Tanzania 1991 through 1996. J Acquir Immune Defic Syndr 2001; 23:194 –202. 8. Smith J, Nalagoda F, Wawer MJ, et al. Education attainment as a predictor of HIV risk in rural Uganda: Results from a populationbased study. Int J STD AIDS 1999; 10:452– 459. 9. Kirunga CT, Ntozi JP. Socio-economic determinants of HIV serostatus: A study of Rakai district, Uganda. Health Transit Rev 1997; 7:175–188. 10. Over M, Piot P. Disease control priorities in developing countries. In: Jamson DT, Mosley WH, Minsham AR, et al., eds. HIV Infection and Sexually Transmitted Diseases. Oxford: Oxford University Press, 1993; 455–527. 11. Kelly R, Kwanuka N, Wawer MJ, et al. Age of male circumcision and risk of prevalent HIV infection in rural Uganda. AIDS 1999; 13: 399 – 405. 12. Vandermoortele J, Delamonica E. The “education vaccine” against HIV. Curr Issues Comparative Educ 2002; 3:6 –13. 13. Kelly MJ, Coombe C. Education as a vehicle for combating HIV/ AIDS. Prospects 2001; XXXI:438 – 445. 14. Paasche-Orlow NK, Clarke JG, Hebert MR, et al. Education attainment but not literacy is associated with HIV risk behaviours among incarcerated women. J Womens Health 2005; 14:852– 859. 15. Barongo LR, Borgdorff MW, Mosha FF, et al. The epidemiology of HIV-1 infection in urban areas, roadside settlements and rural villages in Mwanza region, Tanzania. AIDS 1992; 6:1521–1528. 16. Killewo J, Sandstrom A, Bredberg-raden U, et al. Incidence of HIV-1 infection among adults in the Kagera region of Tanzania. Int J Epidemiol 1993; 22:528 –536. 17. Quigley M, Munguti K, Grosskurth H, et al. Sexual behavior patterns and other risk factors for HIV infection in rural Tanzania: A casecontrol study. AIDS 1997; 11:237–248. 18. Kwesigabo G, Killewo J, Urassa W, et al. HIV-1 infection prevalence and incidence trends in areas of contrasting levels of infection in the Kagera region, Tanzania, 1987–2000. J Acquir Immune Defic Syndr 2005; 40:585–591. 19. Fontanet AL, Woldemichael T, Sahlu T, et al. Epidemiology of HIV and schistosoma mansoni infections among suger-estate residents in Ethiopia. Ann Trop Med Parasitol 2000; 94:145–155. 20. Fylkenes K, Musonda RM, Sichome M, et al. Declining HIV prevalence and risk behaviours in Zambia: Evidence from surveillance and population-based surveys. AIDS 2001; 15:907–916. 21. Glynn JR, Carael M, Buve A, et al. Does increased general schooling protect against HIV infection? A study in four African cities. Trop Med Int Health 2004; 9:4 –14. 22. Gregson S, Waddel H, Chandiwana S. School education and HIV control in sub-Saharan Africa: From discord to harmony. J Int Develop 2001; 13:467– 485.

Vol. 34 ● No. 11

EDUCATION ATTAINMENT AND THE RISK OF HIV-1 INFECTIONS

23. Bradley H, Bedada A, Brahmbhatt H, et al. Education attainment and HIV status among ethiopian voluntary counseling and testing clients. AIDS Behav 2006;2. Epub ahead of print. 24. Michello C, Sandoy IF, Fylkenes K. Marked HIV prevalence declines in higher educated young people: Evidence from population-based surveys (1995–2003) in Zambia. AIDS 2006; 20:1031–1038. 25. Klouman E, Masenga EJ, Klepp KI, et al. HIV and reproductive tract infections in a total village population in rural Kilimanjaro, Tanzania: Women at increased risk. J Acquir Immune Defic Syndr Hum Retrovirol 1997; 14:163–168. 26. Mmbaga EJ, Hussain A, Leyna G, et al. Prevalence and risk factors for HIV-1 infection in rural Kilimanjaro region of Tanzania: Implications for prevention and treatment. BMC Public Health 2007; 7:58. 27. Mnyika KS, Klepp KI, Kvale G, et al. Risk factors for HIV-1 infection among women in the Arusha region of Tanzania. J Acquir Immune Defic Syndr Hum Retrovirol 1996; 11:484 – 489. 28. de Walque D, Nakiyingi-Miiro JS, Busingye J, et al. Changing association between schooling levels and HIV-1 infection over 11 years in a rural

29. 30. 31. 32. 33. 34.

7

population cohort in south-west Uganda. Trop Med Int Health 2005; 10:1001. Lagarde E, Carael M, Glynn JR, et al. Education level is associated with condom use within non-spousal partnerships in four cities of sub-Saharan Africa. AIDS 2001; 15:1399 –1408. Crampin AC, Glynn JR, Ngwira BMM, et al. Trends and measurement of HIV prevalence in northern Malawi. AIDS 2003; 17:1817. Fylkenes K, Musonda RM, Kasumba K, et al. The HIV epidemic in Zambia: Socio-demographic prevalence patterns and indications of trends among childbearing women. AIDS 1997; 11:339 –345. Lugoe WL, Klepp KI, Skulle A. Sexual debut and predictors of condom use among secondary school students in Arusha, Tanzania. AIDS Care 1996; 8:443– 452. Klepp KI, Ndeki SS, Leshabari MT, et al. AIDS education in Tanzania: Promoting risk reduction among primary school children. Am J Public Health 1997; 87:1931–1936. Grunseit A. Impact of HIV and Sexual Health Education on the Sexual behaviors of Young People: A Review Update. Geneva: UNAIDS/ WHO. UNAIDS Report 1997.