|Year : 2016 | Volume
| Issue : 2 | Page : 576-581
Cancer incidence and all-cause mortality in HIV-positive patients in Northeastern Algeria before and during the era of highly active antiretroviral therapy
Karima Chaabna1, Robert Newton2, Philippe Vanhems3, Maamar Laouar4, David Forman5, Zahira Boudiaf4, Isabelle Soerjomataram5
1 Section of Cancer Information, International Agency for Research on Cancer, Lyon; Section of Cancer Information, University of Lyon, University of Lyon, France
2 Section of Cancer Information, International Agency for Research on Cancer, Lyon, France; MRC/UVRI Research Unit on AIDS, Entebbe, Uganda; Department of Health Sciences, University of York, York, United Kingdom
3 Section of Cancer Information, University of Lyon, University of Lyon 1; Department of Hygiene, Epidemiology and Prevention, Edouard Herriot Hospital, Civil Hospital of Lyon; Laboratory of Epidemiology and Public Health, CNRS, UMR 5558, Lyon, France
4 Infectious Disease Service, Durban Hospital, University Teaching Hospital, 23000 Annaba, Algeria
5 Section of Cancer Information, International Agency for Research on Cancer, Lyon, France
|Date of Web Publication||25-Jul-2016|
Section of Cancer Information, International Agency for Research on Cancer, 150 Cours Albert Thomas, 69372 Lyon
Source of Support: None, Conflict of Interest: None
Aims: To assess cancer incidence and all-cause mortality trends in HIV-positive patients in Algeria before and during the highly active antiretroviral therapy (HAART) era.
Settings and Design: Cross-sectional study.
Subjects and Methods: We used hospital-based data of patients with HIV/AIDS between January 1988 and December 2010.
Statistical Analysis Used: Cancer incidence, standardized mortality ratios (SMRs), risk of death, and proportion of HIV-positive patients treated before and during the HAART era were calculated. The joinpoint model was used to assess the magnitude of changes in SMRs.
Results: In 1988–2010, 156 patients were diagnosed as HIV-positive. During pre-HAART era, Kaposi sarcoma (KS) incidence was 5%. After the introduction of HAART, KS incidence decreased to 2%. No other AIDS-related cancer was diagnosed during the study. One-third died (52/156), of which 83.6% died in the same year as or in the year after HIV diagnosis; median age at death (interquartile range) was 34.5 (11.8) years. Yearly risk of death declined from 100% in 1998 to 8% in 2010; percentage of patients treated with HAART increased from 13% in 1998 to >80% after 2002. Overall SMR decreased from 200.2 (95% confidence interval [95% CI], 123.2–325.2) before the HAART era to 91.4 (95% CI, 66.0–126.6) thereafter. From 2003, yearly SMRs decreased significantly by 66.1% (P < 0.05) until 2006 but not thereafter.
Conclusions: Since 1998, the proportion of HIV-positive patients treated with HAART increased, reaching 84% in 2010, all-cause mortality decreased, and cancer remained rare. However, almost all patients who died during the study seemed to be diagnosed at a late stage of the disease, emphasizing the need for earlier diagnosis of HIV in Algeria.
Keywords: Algeria, cancer, HIV, late presentation at diagnosis, mortality
|How to cite this article:|
Chaabna K, Newton R, Vanhems P, Laouar M, Forman D, Boudiaf Z, Soerjomataram I. Cancer incidence and all-cause mortality in HIV-positive patients in Northeastern Algeria before and during the era of highly active antiretroviral therapy. J Can Res Ther 2016;12:576-81
|How to cite this URL:|
Chaabna K, Newton R, Vanhems P, Laouar M, Forman D, Boudiaf Z, Soerjomataram I. Cancer incidence and all-cause mortality in HIV-positive patients in Northeastern Algeria before and during the era of highly active antiretroviral therapy. J Can Res Ther [serial online] 2016 [cited 2019 Dec 6];12:576-81. Available from: http://www.cancerjournal.net/text.asp?2016/12/2/576/179521
| > Introduction|| |
The estimated HIV/AIDS prevalence in the general population of the Middle East and North Africa (MENA) was very low (<0.2%). In 2010, while antiretroviral therapy coverage was 23% in Europe and Central Asia, 8% of HIV-positive patients in MENA in need of antiretroviral therapy received prescribed treatments such as single drug, dual-drug therapy, and/or highly active antiretroviral therapy (HAART). This proportion is higher in some MENA countries, i.e., the percentage of adults and children living with HIV who receive prescribed treatment is estimated at 45% in Oman, 37% in Lebanon, and 30% in Morocco. HAART has been available free of charge in Algeria since 1998, 14 years after the diagnosis of the first HIV-positive case. Compared to uninfected people, HIV-positive patients have a substantially higher risk of some types of cancer such as Kaposi sarcoma (KS), non-Hodgkin lymphoma, and cervical cancer.
This study aimed to assess cancer incidence among HIV-positive patients. In addition, this work aimed to compare all-cause mortality of HIV-positive patients in Northeastern Algeria with that of the general population in periods before and after the introduction of HAART. To the best of our knowledge, no previous study has assessed the evolution of all-cause mortality in HIV-positive patients over a long period to examine the effect of the introduction of HAART into MENA. In Sub-Saharan Africa, an excess mortality continues to be observed in HIV-positive patients treated with HAART; however, it might be prevented by timely initiation of HAART.
| > Subjects and Methods|| |
Study population and data collection
We used hospital-based data for all patients with HIV/AIDS who visited the infectious disease service in a University Hospital Center (CHU) between January 1, 1988, and December 31, 2010. All such patients were included in the current study, independent of the stage of disease or the degree of immunosuppression. The infectious disease service covers six Northeastern wilayas (regions) such as Annaba, El Taref, Skikda, Tebessa, Souk Ahras, and Guelma. It was founded in 1988 and, since 1998, it has been one of the seven STI/HIV/AIDS reference centers in Algeria. Individuals with HIV/AIDS diagnosed in one of the six wilayas were almost invariably referred to the center for diagnosis, confirmation, treatment, and (active) follow-up and were thus captured by the registration network. These cases were not excluded from the analysis. In 2008, the population coverage of this infectious disease service was 3,485,841 people – 10.2% of the Algerian national population.
From the 156 patients' medical records collected, we retrieved demographic information (sex, age at diagnosis, city of origin, and date of death, where relevant), HAART use (yes or no), occurrence of cancer (site-specific), vital status in December 31, 2010 (alive, dead, or lost to follow-up), and date of death – date of last visit for lost to follow-up patients were not available. Data were extracted by completion of questionnaires. Patients were lost to follow-up when they missed all appointments for at least 6 months. Other variables were not considered for the analysis because of data incompleteness, for example, information on transmission route (heterosexual mode, intravenous drug use, mother-to-child …) was not completed in all patients or viral load and CD4+ count were not tested systematically in all patients. HAART is defined as a combination of three antiretroviral treatments containing nucleoside reverse transcriptase, proteinase inhibitor, or nonnucleoside reverse transcriptase inhibitor. HAART is given to the following HIV-infected patients: (a) symptomatic (e.g., weight loss >10 kg, continuous diarrhea or fever, “Stage C” according to the 1993 centers for disease control and prevention classification, or oropharyngeal candidiasis); (b) asymptomatic but with CD4+ T-lymphocyte counts <350 cells/mm 3 ( 2006 guidelines) and then <500 cells/mm 3 (2010 guidelines) during exams performed at least 1 month apart; and (c) some specific patients, for example, pregnant women, infected members of serodiscordant heterosexual couples, children <2 years old, and patients with hepatitis B and/or C coinfections, with cardiovascular risks, and >50 years old.
Because no ethical committee existed in the CHU involved in this study, the head of the infectious disease service was responsible for ethical review, and he fully approved the project. The examination of patient medical records and completion of the questionnaires were performed by those personnel who would normally have access to these records (physicians and nurses). The initials and record number of each patient were noted on the questionnaires to allow the medical professionals to find the patient's records again if necessary. The participants did not receive any payments, reimbursement of expenses, or any other benefits or incentives for taking part in this research, and they did not sign any type of consent.
Statistical analyses included descriptive statistics to compare patient demographic characteristics (sex, age at diagnosis, age at death, and city of origin), clinical characteristics (time to death, follow-up time, and proportion treated), and risk of death before (1988–1997) and after (1998–2010) the introduction of HAART. Comparisons were performed using Fisher's exact test for categorical variables and Student's t-test for continuous variables. Analysis comparing the characteristics of patients who were lost to follow-up with those with known status did not show significant differences in sex, age at diagnosis, city of origin, or HAART distribution (results not shown). Therefore, analyses were performed after excluding those cases lost to follow-up. For 1988–2010, yearly risk of death was calculated by dividing the yearly number of deaths by the number of prevalent patients in the specific year. To illustrate the relationship with changes in treatment, we calculated the annual proportion of prevalent patients receiving HAART since 1998. To present annual estimates, we used 3-year moving average rates to smooth out short-term fluctuations and highlight longer-term trends.
To compare mortality rates of HIV-positive patients with those of the general population, we calculated standardized mortality ratios (SMRs) between 1988 and 2010 as the ratio of the observed number of deaths in the HIV/AIDS patient cohort to the expected number obtained by applying the standard rates (all-cause mortality in the general population) to the cohort age structure. The National Algerian sex- and age-specific mortality rates in 1990, 2000, and 2009 were retrieved from the Global Health Observatory and used as the reference population. We performed linear interpolation of the mortality rates by sex and age to estimate the missing annual sex- and age-specific mortality rates between 1991 and 1999 and between 2001 and 2008. For instance, we used sex- and age-specific mortality rates in 1990 and 2000 to estimate the annual mortality rates during 1991–1999. Data for 2010 were calculated from the life table produced by the Algerian National Office of Statistics. Expected numbers of deaths were calculated by multiplying the number of prevalent HIV/AIDS cases by the corresponding sex-, 5-year age group-, and calendar year-specific mortality rate. 95% confidence intervals (95% CIs) for SMRs were derived from the standard Chi-square test.
A log-linear regression model with SMR as the outcome was used to determine joinpoints in the trend. Joinpoints are breakpoints in time for which significant changes in the linear trends are detected. A maximum number of four joinpoints was specified. Standard errors of SMR  were input into the joinpoint regression program to take into account random errors in the regression model. Using joinpoints, we estimated the magnitude of the trend expressed by the annual percentage change (APC) in SMR. Because multiple tests were performed, the significance level of each test was adjusted with the Bonferroni correction by the joinpoint software to maintain the overall type I error at the specified level (0.05).
To assess the validity of SMRs, we performed sensitivity analyses, mainly to evaluate the impact of patients lost to follow-up on the estimates. In the main analysis, we excluded patients lost to follow-up, and in these sensitivity analyses, we performed analysis for two additional scenarios. In Scenario 1, patients lost to follow-up were assumed to have died within the year of the HIV/AIDS diagnosis, and in Scenario 2, patients lost to follow-up were assumed to be alive at the end of the study period. These two scenarios represent upper (Scenario 1) and lower (Scenario 2) bounds of possible directions of SMR trends as a consequence of including the patients lost to follow-up.
| > Results|| |
Between January 1, 1988, and December 31, 2010, 156 patients from the six Northeastern wilayas covered by the STI/HIV/AIDS center of Annaba were diagnosed with HIV [Table 1]. The center accepted patients living outside its coverage area, i.e., 5.8% of all prevalent HIV/AIDS cases in this period came from areas outside the CHU coverage area. The median age at diagnosis (interquartile range [IQR]) of all patients was 34.5 (11.8) years. During the study, 55 (35%) patients died (median age at death [IQR] was 34.5 [11.8] years) and 54 (35%) were lost to follow-up. At the end of the study, 47 (30%) of the patients were alive. The majority of patients was male (73%), was diagnosed between 1998 and 2010 (74%), and was treated with HAART (89% of followed up patients). We did not observe any significant differences between the periods before and during HAART era in patient age at diagnosis, age at death, sex distribution, city of origin, time to death, or follow-up time of patients alive at the end of the study.
|Table 1: Patient characteristics for the periods before (1988-1997) and during (1998-2010) the era of highly active antiretroviral therapy|
Click here to view
During the pre-HAART era, two cases of KS were diagnosed among the 41 HIV-positive patients (5%). After the introduction of HAART, two cases of KS and one case of female breast cancer were diagnosed among the HIV-positive patients (2% and 1% of HIV-positive patients, respectively). For the KS cases, the skin lesions were localized to the extremities and the face and were responsible for the diagnosis of both cancer and HIV. The breast cancer was diagnosed 1-year after HIV diagnosis.
During the HAART era, the majority of patients who died did so in the same year as or in the year after their HIV diagnosis (84%). Their demographic characteristics were similar to those of the overall population studied: 79% were male, median age at diagnosis (IQR) was 36 (10.3) years, and 46% were from the city of Annaba. However, the only difference observed between these patients and the rest of the study population is the lower proportion of patients receiving HAART (63% vs. 94%). [Figure 1] shows the number of diagnosed cases and the number of deaths, by 5-year age group, for the periods before and during the HAART era. The curves showing the number of diagnosed cases and deaths were similar in shape, with a peak in the 35–39 years age group during the pre-HAART period; thereafter, for the number of diagnosed cases only, the peak was in the 30–34 years age group.
|Figure 1: Newly diagnosed HIV/AIDS cases and number of deaths by age group before (1988–1997) and during the highly active antiretroviral therapy era (1998–2010) in Northeastern Algeria. Solid lines, number of HIV-diagnosed patients in 1988–1998 (squares) and 1999–2010 (diamonds); dashed lines, number of deaths in 1988–1998 (squares) and 1999–2010 (diamonds). Patients lost to follow-up were excluded|
Click here to view
Overall risk of death and standardized mortality ratios
[Figure 2] shows the yearly risk of death over time, together with the proportion of patients receiving HAART. The risk of death increased from 33% in 1988 to 100% in 2001 and then decreased to 8% in 2010; this marked decline coincided with an increase in the proportion of patients receiving HAART, which increased from 13% in 1998 to 84% in 2010 [Figure 2]. The overall SMR declined from 200.2 (95% CI: 123.2–325.2) during the pre-HAART era to 91.4 (95% CI: 66.0–126.6) during the HAART era. The annual SMR trend derived from the joinpoint model is shown in [Figure 3]; it consists of 5-line segments joined at the joinpoints of 1997, 2000, 2003, and 2006. SMRs increased significantly until 1997 (APC = +12.1; 95% CI = 5.5–19.0), remained steady until 2003, and then decreased significantly (APC = −66.1; 95% CI = −78.0 to–47.6) until 2006, with only a modest increase thereafter.
|Figure 2: Trends of the risk of death and use of highly active antiretroviral therapy among HIV/AIDS patients in the University Hospital Center of Annaba, 3-year moving average rates, 1988–2010. Solid line, risk of death; dashed line, percentage of prevalent patients treated by highly active antiretroviral therapy; gray area and gray dashed line around line, 95% confidence interval|
Click here to view
|Figure 3: Standardized mortality ratios and annual percentage change in standardized mortality ratio between 1988 and 2010. Dashed line, 3-year moving average standardized mortality ratio; solid line, modeled curve; gray area around line, 95% confidence interval of standardized mortality ratio. *Annual percentage change was significant with P< 0.05|
Click here to view
The sensitivity analysis shows that the SMRs of the upper-bound scenario [Figure 4]a, in which patients lost to follow-up were assumed to have died within a year of the HIV diagnosis, were considerably higher than the estimates in the main approach whereas the SMRs of the lower-bound scenario [Figure 4]b, in which patients lost to follow-up were assumed to be alive at the end of the study, were considerably lower than the estimates in the main approach. The estimated APC before the HAART era also shows contradictory results, with a significant increase in SMRs before the HAART era in the lower-bound scenario and stable SMRs in that period in the upper-bound scenario. However, in spite of this, these two extreme assumptions demonstrate a significant decrease in SMRs starting in the early 2000s.
|Figure 4: Sensitivity analysis for standardized mortality ratios between 1988 and 2010. (a) Scenario 1: Patients lost to follow-up were assumed to have died within the year of HIV/AIDS diagnosis. (b) Scenario 2: Patients lost to follow-up were assumed to be alive at the end of the study period. *Annual percentage change was significant with P< 0.05. Dashed line, 3-year moving average standardized mortality ratio; solid line, modeled curve; gray area around line, 95% confidence interval of standardized mortality ratio|
Click here to view
| > Discussion|| |
This study reports cancer incidence and changes in HIV/AIDS mortality rates in six Northeastern Algerian wilayas over a 23-year period. During the study, KS incidence remained rare at 5% during the pre-HAART era and at 2% after the introduction of HAART. No other AIDS-related cancer such as non-Hodgkin lymphoma or cervical cancer was diagnosed among the studied population. This low incidence of AIDS-related cancers may reflect the low prevalence of HIV/AIDS in Algeria. It may also suggest that Algeria is at a very beginning stage of the HIV/AIDS epidemiological transition. The risk of death among HIV-positive patients decreased from 100% to 8% whereas the proportion of patients treated with HAART increased from 13% to 84%. Decreasing mortality was also observed when HIV/AIDS patients were compared with the general population, in which relative mortality rates have decreased significantly since 2000. However, mortality rates among HIV/AIDS patients continued to be higher than that of general population in Northeastern Algeria (SMR = 91.4) after the introduction of HAART.
During the HAART era, one-third of HIV-diagnosed patients died, most of them within 2 years after diagnosis. The only difference observed between these patients and the rest of the study population is the lower proportion of patients receiving HAART (63% vs. 94%). During the HAART era, median time to death and median follow-up time of patients alive at the end of the study period who were diagnosed during the HAART era were less than the median survival time from enrollment to death for prevalent cases (4.5 years) in the HIV-1 natural history cohort 5-year prospective study in the rural Masaka District of Uganda which ended in 1995, where HIV prevalence in adults was 8% and no antiretroviral therapy was available. Late stage of disease at diagnosis could explain the relatively short time to death in the patients of the current study and lower HAART prevalence. In addition, lack of adequate therapy might be an additional explanation for the failure to limit disease progression for some of the patients. Triple therapy (HAART) has been used in Algeria as first- and second-line treatment, yet there is currently no validated therapeutic approach for third-line treatment. Unlike in Europe, fusion inhibitors, CCR5 inhibitors, and integrase inhibitors are not currently in use in Algeria.
Despite the reported low prevalence of HIV/AIDS in Algeria, the number of diagnosed patients has tripled during the HAART era compared with the pre-HAART era. In 2000, the first national HIV serosurveys were conducted among pregnant women, sex workers, and drug users, followed by surveys in 2004 and 2007.,, Across the country, 54 testing centers were opened in 2006, which referred patients identified as HIV-positive to the STI/HIV/AIDS centers. Part of the increase in diagnosed cases in our study was probably an artifact of increasing awareness among medical professionals, but this result likely also reflects a genuine increase in HIV/AIDS cases. Among the general population, knowledge of HIV/AIDS is very poor as shown in two surveys conducted among students in Algeria;, hence, this is unlikely to motivate a reduction in high-risk behaviors that lead to increased risk of HIV transmission. Furthermore, a recent study in two hospitals in Northeastern Algeria reported an HIV prevalence of 5.3/1000 among pregnant women.
In other countries such as Zimbabwe and Uganda, the decline of the HIV/AIDS epidemic has been attributed to increased access to antiretroviral treatment coupled with active prevention campaigns to increase population awareness and change risk behaviors.,, The success of these prevention strategy encourage the expansion of these experiences and the inclusion HIV-infected population from resource-limited settings in global campaigns and research programs. In Sub-Saharan African countries, the mortality of HIV-infected patients treated with HAART continues to be higher than that in the general population, but for patients with CD4 count ≥200 cells/ml and starting HAART with the best prognosis, excess mortality was moderate and mortality reached that of the general population in the 2nd year of receiving HAART. Much of the excess mortality might be prevented by timely initiation of HAART. Algerian public health policy should continue to encourage early detection programs, especially among young males who were the most affected group because of use of sex worker services. Furthermore, because heterosexual intercourse is the primary mode of HIV transmission in Algeria (98%), the spread of HIV in families is of great concern. Therefore, pregnant women might be another potential target group for early prevention, to avoid mother-to-child transmission  and improve the prognosis of the unborn children.
The retrospective study on HIV/AIDS burden in Northeastern Algeria was affected by some missing information at various levels. Incompleteness of data for salient etiological, predictive, and/or prognostic factors in HIV/AIDS patient records was often encountered during data collection. For example, some important variables such as CD4 count were not found in all records. In addition to economic limitations hampering complete blood work, this can be explained by the fact that many patients were diagnosed at the late stage of HIV/AIDS and therefore died before all exams had been completed.
Another limitation of our study is the large number of patients lost to follow-up. It is very probable that some of these patients had died, and thus the number of deaths may be underestimated in the main analysis. However, the sensitivity analysis, including extreme scenarios, confirmed the decrease in mortality among HIV-diagnosed patients after the introduction of HAART. In addition, the small number of cases in the current study means that we have limited power to examine some associations. In addition, for some patients, the long distance from their residence to the STI/HIV/AIDS reference center may cause these patients to go closer reference centers. However, based on a report of a neighboring center (Setif), no cases from the Annaba center were followed up there. Social stigma and physician reputation may also influence patient preferences in selecting a care provider. Of the study patients, 5% came from outside the center's coverage area, and we expect that a similar proportion of patients went outside the Annaba coverage area for care. Finally, because we observed a noticeable proportion of patients diagnosed in the late stage of the disease, a number of patients who died of HIV without diagnosed and referred to the STI/HIV/AIDS reference center were probably missed in the current study.
This study showed that cancer remained rare among this high-risk population suggesting that Algeria is at a very beginning stage of the HIV/AIDS epidemiological transition. Our study also showed a decrease in the all-cause mortality rate after the introduction of HAART in 1998 for HIV-diagnosed patients, for which the proportion of patients receiving HAART reached 84% in 2010. However, the mortality rate remained higher than that of the general population. Almost all patients who died during the study period seemed to be diagnosed in the late stage of the disease. Public health programs encouraging early detection of HIV in Algeria should be coupled with the increasing proportion of patients treated with HAART to reduce risk of death among patients with late-stage disease.
| > References|| |
UNAIDS/WHO. UNAIDS World AIDS Day Report 2011. Geneva, Switzerland: UNAIDS/WHO; 2011.
UNAIDS. Middle East and North Africa. Regional report on AIDS. 2011. Geneva, Switzerland: UNAIDS/WHO; 2011.
Touaibia R, Merbout G, Ait-Oubelli K, Benmakhlouf M. UNGASS Report Algeria 2010: Report on the national situation on on follow-up to the Declaration of Commitment on HIV/AIDS (2008-2009); 2011.
Bouvard V, Baan R, Straif K, Grosse Y, Secretan B, El Ghissassi F, et al.
A review of human carcinogens – Part B: Biological agents. Lancet Oncol 2009;10:321-2.
Brinkhof MW, Boulle A, Weigel R, Messou E, Mathers C, Orrell C, et al.
Mortality of HIV-infected patients starting antiretroviral therapy in Sub-Saharan Africa: Comparison with HIV-unrelated mortality. PLoS Med 2009;6:e1000066.
Ouahdi M. Operational manual for national monitoring and evaluation of the national strategic Plan national 2008-2012. Geneva, Switzerland: UNAIDS/WHO; 2009.
CDC Staff Members. 1993 Revised Classification System for HIV Infection and Expanded Surveillance Case Definition for AIDS Among Adolescents and Adults. Atlanta, GA: Centers for Disease Control and Prevention; 1993.
Editorial Board of the Department of Prevention. National Guide of Therapeutic HIV/AIDS and opportunistic infection in adults and children. Algiers, Algeria: Ministry Health, Population and Hospital Reform; 2010.
Breslow NE, Day NE, editors. Statistical Methods in Cancer Research. The Analysis of Case – Control Studies. Vol. I. Lyon, France: International Agency for Research on Cancer (IARC Scientific Publication No. 32); 1980.
Statistical Research and Applications Branch. Joinpoint Regression Program. Bethesda, MD: National Cancer Institute; 2008.
Morgan D, Malamba SS, Maude GH, Okongo MJ, Wagner HU, Mulder DW, et al.
An HIV-1 natural history cohort and survival times in rural Uganda. AIDS 1997;11:633-40.
WHO. Provisional working estimates of adult HIV prevalence as of end 1994, by country. Wkly Epidemiol Rec 1995;70:355-7.
Yeni P. Medical management of HIV infected people. Expert group recommandation. The French documentation. France: Ministry of Health and Sports; 2010.
Fares EG, Mokhtari L. Report on epidemiological surveillance of HIV infection (sero-sentinel surveillance in 2000). Algiers, Algeria: Prevention Branch. Ministry of Health, Population and Hospital Reform; 2001.
Fares EG, Mokhtari L. Report of the sero-surveillance of HIV and syphilis in 2004. Algiers, Algeria: the Directorate Prevention. Ministry of Health, Population and Reform hospital; 2005
Fares EG, Mokhtari L. Report of the National sero-surveillance of HIV and syphilis in 2007. Algiers, Algeria: Prevention Branch. Ministry of Health, Population, and Hospital Reform; 2008.
Abdennour D. Level of knowledge and attitudes about HIV/AIDS among students. Algeria: 5th
National Days of Infectious Disease Prevention; 2012.
Abdennour D. Enquête sur le niveau de connaissances des jeunes de la Wilaya de Constantine sur l'infection VIH/SIDA. Algeria: 5th
National Days of Infectious Disease Prevention; 2012.
Aidaoui M, Bouzbid S, Laouar M. Seroprevalence of HIV infection in pregnant women in the Annaba region (Algeria). Rev Epidemiol Sante Publique 2008;56:261-6.
UNAIDS/WHO. Evidence for HIV decline in Zimbabwe: A comprehensive review of the epidemiological data. Geneva, Switzerland: UNAIDS/WHO; 2010.
Halperin DT, Mugurungi O, Hallett TB, Muchini B, Campbell B, Magure T, et al.
A surprising prevention success: Why did the HIV epidemic decline in Zimbabwe? PLoS Med 2011;8:e1000414.
IeDEA Network. International Epidemiologic Databases to Evaluate AIDS (IeDE). Available from: http://www.iedea.org/
[Last accessed on 2016 Feb].
IARC Working Group on the Evaluation of Carcinogenic Risks to Humans. IARC Monographs on the Evaluation of Carcinogenic Risk to Human. Vol. 67. Lyon, France: IARC/WHO; 1996.
Siegfried N, van der Merwe L, Brocklehurst P, Sint TT. Antiretrovirals for reducing the risk of mother-to-child transmission of HIV infection. Cochrane Database Syst Rev 2011;(7):CD003510.
Lacheheb A. Clinical management of HIV/AIDS in Algeria: Experience of Setif Reference Center. Algeria: 1st
Meeting Rennes-Sétif; 2007.
[Figure 1], [Figure 2], [Figure 3], [Figure 4]