Increasing trend of HIV/AIDS among Arab and Jewish male persons in Israel, 1986–2010

Authors


Correspondence: Dr Zohar Mor, Ramla Department of Health, 3 Danny Mass St., Ramla 72100, Israel. Tel: +972 89181258; fax: +972 89251607; e-mail: zohar.mor@rml.health.gov.il

Abstract

Objectives

The aim of the study was to compare the HIV/AIDS burdens in Jewish and Arab Israeli males, as HIV/AIDS affects different population groups disproportionally.

Methods

The National HIV/AIDS Registry (NHAR) was used as the source of HIV/AIDS infection records, while the Israeli Central Bureau of Statistics was used to determine group-specific disease rates.

Results

Between 1986 and 2010, 3499 HIV/AIDS-infected male Israelis were reported to the NHAR: 3369 (96.3%) Jews and 130 (3.7%) Arabs, with an average annual incidence of 5.5 and 0.8 per 100 000 of the population, respectively (P = 0.05). Of the Jews, 1018 (29.9%) were born in Ethiopia, while 2389 were Jews who were not Ethiopian-born (JNE). Most of the Arabs (n = 99; 74.8%) were Muslims, followed by Christians (21; 16.2%) and Druze (13; 10%). AIDS rather than HIV infection at the time of reporting was diagnosed in 568 (23.8%) of the JNE and 31 (23.8%) of the Arabs (p = 1). The most affected age group was those aged 25–34 years among the JNE and those aged 20–24 years among the Arabs, and the respective cumulative death rates were 24.9% (n = 594) and 32.5% (n = 40) (P = 0.1). The point prevalences in 2010 were 58.4 and 11.4 per 100 000 for JNE and Arabs, and in adults aged 15–59 years they were 71.5 and 26.3 per 100 000, respectively. In Muslims, Christians and Druze, the point prevalences were 4.2, 11.2 and 7.1 per 100 000, and in adults aged 15–59 years they were 22.6, 42.9 and 29.4, respectively. The most common risk group among JNE was men who have sex with men (MSM; n = 1223; 51.2%), followed by injecting drug users (n = 661; 27.7%), while among Arabs it was MSM (n = 63; 48.1%), followed by heterosexuals (n = 36; 27.3%).

Conclusions

The HIV/AIDS burden in Israeli Arab males was significantly lower than that in Jews, and in both populations the most common risk group was MSM, with the proportion of MSM increasing with time.

Introduction

The HIV/AIDS burden in Israel is generally low in comparison to other countries [1], although subpopulations characterized by high-risk behaviours are affected disproportionally [2]. These include men who have sex with men (MSM), injecting drug users (IDUs), and migrants from countries with generalized HIV epidemics.

HIV tests in Israel are offered widely by both community and hospital physicians. An individual who is interested in HIV testing can either ask his/her doctor to refer him/her to the laboratory or go to any of the HIV-testing clinics. These clinics are situated in the AIDS centres of hospitals country-wide. Several other testing clinics throughout Israel are operated by nongovernmental organizations or the Ministry of Health [3]. The first HIV screening test can be performed anonymously, but those who are found to be positive for HIV in the enzyme-linked immunosorbent assay (ELISA) and who undergo a confirmatory western blot test are expected to disclose their identity. More than 200 000 voluntary tests are performed annually in Israel, in addition to 250 000 tests that are performed at blood banks while screening blood donations.

Although Israel is situated in the Middle East, it is a member of the World Health Organization (WHO) European region. It is a developed country with a gross domestic product (GDP) of $29 800 per person [4], and had a population of nearly 7.7 million people in 2010. Of the total population of Israeli citizens in 2010, Jews comprised 6.1 million (79.2%), and 1.6 million (19.7%) were Arabs. The Arab minority is defined culturally and linguistically, and includes Muslims (1.3 million; 82.6%), Christians (151 700; 9.5%) and Druze (127 600; 7.9%) [5]. Although the Arab communities are considered to be conservative, the incidence of HIV/AIDS in Arab countries neighbouring Israel is increasing [6], especially among men [7]. Knowledge about trends in the incidence of infection and the characteristics of HIV-affected communities is essential for planning and evaluating prevention efforts and for allocating budgets appropriately [8]. This study compares the HIV/AIDS burdens in Jewish and Arab male Israelis, and assesses the associations between demographic and behavioural determinants practised in Israel, a mixed, pluralistic Jewish and Arab country in which mutual cultural influences prevail.

Methods

The diagnosis of HIV/AIDS in Israel requires two positive ELISA results and confirmation with a western blot blood test. Both HIV infection and AIDS have been reportable nominally in Israel at the national level and recorded in the National HIV/AIDS Registry (NHAR) since 1981. The reports contain individual demographic, behavioural and clinical data for each notification. In order to address the possibility of reporting bias caused by newly diagnosed patients not disclosing complete personal information or not reporting their risk behaviour at diagnosis, the NHAR records are actively updated annually with reports from all the AIDS treatment centres and regional health departments. Annual cross-matching with all four medical insurers allows the updating of information regarding antiretroviral treatment, and cross-matching with the National Civil Registry at the Ministry of Interior allows the updating of information regarding deaths or emigration from Israel. Thus, the NHAR is both valid and accurate.

Jews who have migrated from Ethiopia are routinely screened for HIV during the first few weeks following their arrival in Israel. As members of the Jewish Ethiopian community in Israel have a unique and distinct lifestyle, they were excluded from most of the calculations. Jews immigrating to Israel who were born in countries other than Ethiopia assimilate into Israeli society with Israeli-born Jews. Therefore, most of the analyses related to behavioural factors compared Jews who were not Ethiopian-born (JNE) with Arabs.

The denominators used to determine age-specific disease rates among male Israelis of different religions for the years 1986–2010 and other calculations were based on the official census of the Israeli Central Bureau of Statistics [9]. The year-specific HIV transmission rates were calculated by dividing the annual incidence for that year by the prevalence for that year [10], while the number of newly reported cases of HIV infection in male Israelis was used as a proxy for the incidence.

Categorical variables were compared using the χ2 test, and continuous variables were compared using Student's t-test. As this was a routine epidemiological analysis using non-nominal data within a standard surveillance reporting system, institutional review board approval was not required.

Results

Between 1986 and 2010, 3499 HIV-infected Israeli male citizens were reported to the Ministry of Health. Of those, 3369 (96.3%) were Jews and 130 (3.7%) Arabs, with an average annual incidence rate of 5.5 (annual range 2.7–8.0) and 0.8 (annual range 0.2–1.7) per 100 000 of the population, respectively (P = 0.05); incidence rates in both Jews and Arab males showed an increasing trend with time since the mid-1990s (Fig. 1).

Figure 1.

Annual rate of HIV/AIDS in male Israelis divided into Jews and Arabs, for the period 1986–2010. JNE, Jews who were not Ethiopian-born.

Of the HIV-infected male Jewish Israelis, 1018 (29.9%) were born in Ethiopia and the remainder (2389; 70.1%) were JNE; of these, 1253 (36.8%) were Israeli-born and 1136 (33.3%) were born in other countries and were naturalized following their immigration to Israel. Overall, most male Jews (n = 2097; 62.2%) were infected in Israel, while the majority (n = 1018; 80.0%) of the 1272 Jews who were infected overseas were born and infected in Ethiopia.

Most HIV-infected Arabs (n = 125; 94.7%) were born in Israel, and the majority (n = 99; 74.8%) were Muslims, followed by Christians (21; 16.2%) and Druze (13; 10%). Of the male Arab Israelis, 117 (90%) were infected in Israel.

At reporting, AIDS rather than HIV infection was diagnosed in 568 (23.8%) of the JNE and 31 (23.8%) of the Arabs (P = 1). The most common (16.1%) AIDS-defining illness in JNE was Pneumocystis carinii pneumonia (PCP) followed by Kaposi's sarcoma (9.6%), while in Arabs the most common AIDS-defining illness was PCP (12.8%) followed by brain toxoplasmosis (10.1%). Progression from HIV diagnosis to AIDS disease was detected within 5.0 ± 3.7 years (mean and standard error) in JNE, and within 6.6 ± 4.6 years in Arabs (P = 0.14).

The age group with the highest average detection rate throughout the study period was those aged 25–34 years for Jews and those aged 20–24 years for Arabs (10.4 vs. 3.1 in Jews and Arabs per 100 000 male persons aged 25–34 years, and 5.5 vs. 6.1 in Jews and Arabs per 100 000 male persons aged 20–24 years, respectively). The respective average ages at reporting were 34.3 ± 11.3 (mean and standard error) and 33.0 ± 12.1 years (P = 0.27), and the cumulative death rates at the end of the follow-up period in December 2010 were 24.9% (n = 594) and 32.5% (n = 40), respectively (P = 0.1). Deaths in those cases were recorded 4.5 ± 0.3 (mean and standard error) and 4.3 ± 0.3 years following initial notification, respectively (P = 0.9). After exclusion of those who died or left Israel, the point prevalence at the end of 2010 was 58.4 and 11.4 per 100 000 in JNE and Arab males respectively, for all age groups, and 71.5 and 26.3 per 100 000, respectively, for those aged 15–59 years. The point prevalences in Christians, Druze and Muslims were 11.2, 7.1 and 4.2 per 100 000 for male persons of all ages and 42.9, 29.4 and 22.6 for those aged 15–59 years, respectively.

The most common risk group in male Jews, including those who were born in Ethiopia, was heterosexuals (n = 1282; 37.7%), followed by MSM (n = 1224; 35.9%) and IDUs (n = 664; 19.5%), while in JNE the most common risk group was MSM (n = 1223; 51.2%), followed by IDUs (n = 661; 27.7%) and heterosexuals (n = 313; 13.1%). Among male Arabs, the most common risk group was MSM (n = 63; 48.1%), followed by heterosexuals (n = 36; 27.3%) and IDUs (n = 21; 15.9%) (P = 0.01). The proportion of MSM in the HIV-infected male population has increased from 2000 to 2010 among both JNE and Arabs, with reciprocal decrease in the infection rate among heterosexuals (Fig. 2).

Figure 2.

Risk groups of HIV-infected male Israelis, by population group, for the period 1986–2010. Hetero, heterosexual; JNE, Jews who were not Ethiopian-born; MCTC, mother to child transmission; MSM, men who have sex with men; UNK, unknown.

HIV transmission rates have been increasing among Arabs. The estimated transmission rates in 2000 were 6.3% for Arabs and 11.4% for JNE, while in 2010 they were 10.9% and 8.6%, respectively. That is, on average in 2010, each Arab or JNE living with HIV/AIDS transmitted the virus to about 0.11 and 0.09 previously uninfected persons per year, respectively.

Discussion

The HIV prevalence in male Arabs was significantly lower than that in Jews and also lower than that reported in most other countries in the Middle East and North Africa (MENA) WHO region [1]. Although 19.7% of the Israeli male population are Arabs, only 3.7% of HIV-infected male Israelis were Arabs. HIV prevalence rates in Israeli Arabs were different in different religious groups, with Christians having the highest rates, followed by Druze and Muslims. This generally low rate in Arabs, especially in Muslims, may be related to the influence of Islamic traditions on cultural and social customs conferring protection against HIV infection [11]. Muslims are allowed polygamic relationships (up to four concurrent wives), couples are prohibited from having premarital or extramarital sex, MSM activities are prohibited and alcohol consumption is banned [12]. Nevertheless, the vast majority of HIV infections in the MENA region are documented in men [7], and it might also be argued that Muslim culture, being more tolerant regarding male freedom and allowing men to have multiple female partners as part of the masculine ideal, places them at risk of HIV infection. It has been suggested that male circumcision [13, 14] and the ritual penile washing practised by Muslims [5] may limit the spread of HIV. However, Jewish and Druze men and many Christians are also circumcised. Thus, the advantage of circumcision in Muslims, if it exists, is irrelevant. A more likely explanation is that Jews and Christians are at higher risk of HIV infection than Druze or Muslims because of cultural differences.

Israeli society reflects an amalgam of Western and Middle Eastern influences [15], and is a relatively open society regarding sex and sexuality, being fairly similar in this respect to Western countries. Israel and, to some extent, Lebanon [16] are exceptional in the Middle East in that they have more liberal views on MSM activities and relatively tolerant views on sexual diversity and premarital and extramarital sex. Nevertheless, the HIV/AIDS burden in Arabs in our study was lower than that reported for most countries in the MENA region.

Some publications indicate that minorities have a greater vulnerability to HIV infection and a higher HIV/AIDS burden [17]. In our study, no differences were found between Arabs and JNE in age at detection of HIV infection, rate of late diagnosis or time for progression from HIV infection to AIDS disease. Therefore, it seems that there are no substantial barriers to accessing HIV testing points and that negative perceptions are not deterring Arabs from undergoing HIV tests. The high proportion of MSM in the HIV-infected Arab male population may be associated with positive utilization of HIV testing, but may also be related to unsafe sex practices. In the last 10 years, MSM have become the major risk group in Arab Israeli males, overtaking IDUs and heterosexuals, who comprised the majority of HIV-infected Arabs up to 2002 [18], similar to current epidemiological trends described among male Jews [19] and additional reports from other developed countries [1]. It may also be that Arab and Jewish men share similar sexual behaviours and take comparable risks. However, more extensive data need to be collected with regard to Arab MSM behaviour to better understand their knowledge, attitudes and sexual practices. These data could then be used to design targeted behavioural interventions, such as identifying venues frequented by Arab MSM, designing unique safe-sex messages, and establishing Arab-oriented health infrastructure for testing and treatment.

The validity of the Israeli NHAR is high. Although individuals undergoing confirmatory HIV tests in Israel are identified by name, less than 1% of individuals testing positive in the initial ELISA screening test decline to proceed to the western blot test (Z. Mor, personal observation). Additionally, the NHAR is constantly updated retrospectively. HIV-infected individuals may be embarrassed to disclose their risk behaviour on their first few visits to an AIDS centre. They usually feel more comfortable once they have established relationships of trust with the medical staff at the AIDS centre, and then disclose further details regarding their risk behaviour and sexual contacts. The NHAR is updated retrospectively in these cases. The low proportion of heterosexuals who are not IDUs or born in high-prevalence countries reflects the reliability of the NHAR, and the openness of HIV-infected individuals.

In conclusion, the HIV/AIDS burden in male Arabs was significantly lower than in Jews, but the HIV/AIDS burden differed among Christians, Druze and Muslims. The proportion of MSM in men diagnosed with HIV infection, regardless of their religion, is increasing. Interventions aimed at preventing further HIV transmission should address the cultural, linguistic and behavioural characteristics of this population.

Acknowledgements

The authors wish to thank Ms Zehuvit Wiexelbom for her exceptional maintenance of the National HIV/AIDS Registry and Ms Yana Roshal Levin for her assistance with the statistical analyses. The authors also thank Ms Judy Brandt for editing the manuscript.

Ancillary