Trends and projection of incidence, mortality, and disability-adjusted life years of HIV in the Middle East and North Africa (1990–2030)

Evidence shows a growing trend of the HIV epidemic in the Middle East and North Africa (MENA). We aimed to project the incidence, mortality, and disability-adjusted life years (DALY) in the region from 1990 to 2019 and assess its trend by 2025, and 2030. We extracted the HIV incidence, mortality, and DALY data from the Global Burden of Disease (GBD) and UNAIDS databases. The joinpoint regression model was used to examine changes in HIV trends. The trend changes were estimated by average annual percent change (AAPC). In most countries, an increasing trend was observed in HIV incidence, mortality, and DALY. Specifically, the highest growth in the annual incidence rate was related to Egypt (AAPC = 14.4, GBD) and Iran (AAPC = 9.6, UNAIDS). Notably, Qatar (AAPC = − 5.6, GBD), Bahrain (AAPC = − 3.3, GBD), and Somalia (AAPC = − 4.2, UNAIDS) demonstrated a significant reduction in incidence. Regarding mortality rates, Djibouti (AAPC = 24.2, GBD) and Iran (AAPC = 16.2, UNAIDS) exhibited a significant increasing pattern. Furthermore, the estimated increase in incidence by 2030 was most marked in Djibouti (985%) and Iran (174%). Iran (422%) and Egypt (339%) showed a prominent rise in mortality rates. GBD data showed 16 countries had an increasing pattern in DALY in both genders. According to age and period effects, there was a significant upward trend in incidence, mortality rates, and DALY. Findings highlighted the urgent need for improved prevention and treatment services, including expanding access to HIV testing, promoting safe practices, increasing antiretroviral therapy coverage, and supporting targeted interventions for high-risk populations.

Data from 1994 to 2014 were used to predict HIV incidence in 2015.Data from 1995 to 2015 were used to predict HIV incidence in 2016.Data from 1996 to 2016 were used to predict HIV incidence in 2017.Data from 1997 to 2017 were used to predict HIV incidence in 2018.Second, we fitted the Joinpoint Regression model for each of the datasets.Third, we estimated the prediction in incidence, mortality, and DALY using the betas displayed on the APC in the final segment.
To measure the accuracy of the prediction counts, the prediction error is estimated for each year by computing the predicted minus the observed value.We then find the ratio of the prediction error to the observed count, adding 0.5 to the denominator to avoid numerical error when the observed count is zero.The ratio is the percentage error of the prediction, compared to the observed count: Fourth, the average absolute relative deviation (AARD) was computed as the average of the error ratios over all the cases considered in the data.More specifically, assume θ S is the predicted mortality for specific scenario s, s = 1,…, S, and that θ S is the true observed count.
Smaller values in AARD indicate closer estimates to the true values 15,16 .The results of the prediction model of the incidence rate of Algeria country between 1993 and 2019 were shown in Supplementary Tables S1 and S2.
Ethics approval.Ethical approval was obtained (IR.KMU.REC.1401.263)from the Ethics Committee of Kerman University of Medical Sciences.
Figure 1 presents the trends of the age-standardized incidence rates (ASIR) for HIV in MENA countries from 1990 to 2019.The ASIR of men was higher than women, except in Djibouti, Iraq, Somalia, and Sudan (Fig. 1).Table S3 presents the AAPC of HIV incidence in the region from 1990 to 2019.Among women (GBD data), seventeen countries had increases in ASIR and 4 countries showed declining trends.The highest AAPC was related to Iran (AAPC = 9.6, 95% CI 8.8, 10.5; P < 0.05) (12.3, and 8.3 in women and men, respectively).In the UNIAIDS dataset, the majority of incidence rises occurred in Egypt and Iran (P<0.05)(Table S3).
A total of 17 and 18 out of 21 reported increasing mortality trends in men and women, respectively.Among them, Djibouti (AAPC men = 23.3,95% CI 21.5, 25, AAPC women = 24.6,95%CI 23.4,25.9; P < 0.05) showed the most marked increase in mortality rates.In UNIAIDS data, Iran had the greatest mortality increase (AAPC = 16.2, 95% CI 16, 16.4; P < 0.05) (Table S5).  3 and Fig. 3, and the corresponding findings from the joinpoint regression analysis were presented in Tables S7  and S8.According to GBD datasets, 16 countries had an increasing pattern in DALY and five countries showed declining trends in both genders between 1990 and 2019 (Table S7).www.nature.com/scientificreports/Projection of incidence, mortality rates, and DALY.Tables 4, 5 and 6 summarize the projected agestandardized incidence, mortality rates, and DALY of HIV by 2025 and 2030 per 100,000 in the countries in the region.The estimated increase in the incidence by 2030 was most marked in Egypt (339%), based on UNAIDS estimates.However, the estimated higher age-standardized incidence rate by 2030 was found in Iran (174%), according to estimates from GBD. Countries with substantial growth rates of age-standardized incidence rates and incidence rates between 2019 and 2030 include Lebanon (59%) and Qatar (478%), respectively (Table 4 and Fig. 1).
APC model analysis results of HIV incidence, mortality rates, and DALY.Table 7 shows the estimated impact of age, period, and birth cohort of the whole population on the age-standardized incidence rate, mortality rates, and DALY of HIV in the region.The IE algorithm was used for quantitative analysis among different age groups and periods.The age effect of HIV incidence, mortality rates, and DALY showed a net decrease of 0.83, 0.29, and 0.01 from the age of 35-74 years (the lowest value 35-44 years age group as a reference).According to period effects, there is a significant upward trend in the risk of incidence, mortality rates, and DALY from HIV among the total population.According to the analysis of cohort effects, the incidence, mortality rates, and DALY from HIV have no clear or consistent trend.The 1972-1975 period had the lowest cohort effect on incidence risk from HIV.Also, the 1990-1994 period had the lowest cohort effect on mortality and DALY risk from HIV in the region (Table 7).

Discussion
Our study showed that the incidence and mortality of HIV have increased in most countries in the MENA region from 1990 to 2019.Iran and Egypt showed the highest annual growth rates in incidence, while Qatar and Somalia experienced substantial reductions in incidence based on the modeling results using GBD and UNAIDS estimates, respectively.Djibouti and Iran had the highest mortality rates according to the modeling results using GBD and UNAIDS estimates, respectively, whereas Qatar had the lowest mortality rate according to both datasets.Moreover, incidence trends from 21 countries in a period of 30 years reported that a total of 14 and 11 countries experienced increases in the annual growth rate according to the GBD and UNAIDS estimates, respectively.Up to five out of 21 countries showed a reduction in mortality trends according to GBD information, with Qatar being the only country to demonstrate a significant decrease in mortality rate according to both estimates.Regarding the incidence rate, Iran, Egypt, Tunisia, Libya, and Djibouti exhibited the highest percentage of change from 1990 to 2019, while Lebanon, Morocco, Syria, and Somalia reported the lowest percentage of change in Table 6.Disability-adjusted life years and prediction of disability-adjusted life years from HIV in 2025 and 2030 per 100000 in MENA countries.UAE United Arab Emirates.diagnosis in these countries.In any case, these countries are likely experiencing a combination of different situations, and all aspects of each country need to be examined to obtain an accurate result.While the incidence trend decreased in Qatar and Bahrain during the whole period, it increased in Iran, Egypt, Iraq, Libya, Palestine, Tunisia, and Yemen.In the rest of the countries, the trend was unsteady.The significant decreasing trend we observed in Qatar was consistent with findings from a cohort study during a 17-year period, which demonstrated the lower incidence rate of HIV in comparison to the world average 28 .Shakiba, et al., reported Qatar as the country with the lowest HIV burden in 2017 in the region 20 .The declining trend of the HIV incidence rate in Qatar coincided with the implementation of diagnostic activity and the consumption of antiretroviral therapy 29 .This reduction trend may be explained by earlier diagnosis and implementation of the curative intervention.The HIV screening programs, prophylaxis, and early use of antiretroviral therapy may lead to a decrease in the rate of HIV-related morbidity and mortality 28 .The possible reasons for the reduction of morbidity and mortality of HIV in Qatar maybe the regular implementation of HIV early screening programs for immigrants, especially if they stay for more than one month, as well as premarital, pre-employment, and antenatal screening.Furthermore, HIV treatment is available for all patients 29,30 .In the MENA region, Egypt has the fastest-increasing newly discovered HIV cases (25-30% annually in the past 10 years) 24 .Whereas, based on modelling results from the WHO mortality database, 2001-2018, Egypt had the lowest age-standardized death rate (ASDR) for males (0.2/100,000) and modelling results showed a positive percent change in ASDR for females (114.38% between 2001 and 2015).In 2018, the disparity in ASDR rates in males, measured using rate ratios was the lowest rate observed in Egypt 31 .
In this study, utilizing GBD data, we observed that the trend of HIV-related mortality was increasing in 15 out of 21 countries, stable in two countries, and decreasing in four countries.Similarly, according to UNAIDS data, there was an increase in HIV-related mortality in 13 out of 14 countries and a decrease in one country.The total death trend in most MENA countries was increasing.But looking at the last years of the graphs, the death rate due to AIDS has decreased in most countries.This suggests that in recent years, these countries have made efforts to implement programs proposed by WHO and UNAIDS aimed at reducing HIV incidence and mortality, although the degree of success achieved thus far may be limited 32,33 .Only in Iran, Iraq, and Libya, the death trend has always been increasing, even in the last years of the study.Similar to the results presented here, Hasankhani, et al., in separate modelling revealed an ascending trend for the incidence and mortality of this disease in Iran over the past three decades 34 .According to the results of another study carried out in Iran, the raw number of mortality, incidence, and burden of HIV increased from 2008 to 2016 in Iran 35 .Studies about the trends of HIV incidence in Iran have shown that a change in the pattern of HIV transmission from injecting the drug to unsafe sexual contact has resulted in an increased number of HIV-positive cases among women 36 .Another explanation for the higher incidence could be attributed to the shortage of educational and preventive programs, lack of knowledge, social stigma, low access to counselling and diagnostic service, and prognostic diagnostic technique in the Iranian population 34 .The increase in HIV infection notifications is difficult to interpret because of changes in the availability of tests, and the willingness of both health professionals and individuals to be tested 37 .
This study predicted incidence and mortality results based on both GBD and UNAIDS databases.Based on GBD data, both incidence and mortality rates were decreasing in Qatar, Bahrain, Kuwait, and Lebanon.However, based on UNAIDS data, it was estimated that the incidence rate will decrease in Somalia, Djibouti, and Morocco, and the mortality rate will decrease only in Qatar.Based on both databases, it was estimated that the incidence and mortality rate will increase for most of the MENA countries, with the largest increase predicted for Iran.In 2019, Khalifa et al. conducted a study to predict the number of people living with HIV and the number of new HIV infections worldwide using UNAIDS data and Spectrum software.The findings indicated that the number of people living with HIV in the MENA region would increase until 2030 and gradually decrease thereafter, with a similar trend observed for the number of new HIV infections.However, the study did not report these indices by country, providing regional-level projections instead 38 .
In this study, different epidemiological patterns are observed in the five periods identified by the Joinpoint Regression analysis during the studied period.In most countries in the MENA region, the rising tendency in the incidence and mortality rate was the most pronounced in trends 1 and 2 (1990-2004).The drop in these rates can be explained by increased access to prevention, early diagnosis, and care services for HIV infection 37 .Significant progress in HIV response in the region has been witnessed in recent years.For instance, Algeria and Morocco have made remarkable progress in growing access to HIV care services.However, other countries, including Somalia and Sudan, have encountered considerable challenges 1,39 .A notable gap between available care services and required facilities existed in the region; the existing resources in 2020 were less than 20% of what is needed to boost HIV programs and achieve the 2025 targets 1 Progress towards the 2030 targets will depend on cooperation between organizations, linking the HIV service to efforts to reach universal health coverage and improved access to reproductive health and social protection systems.These programs will be achieved with a stronger commitment from governments and more emphasis on the recognition of the social and economic impacts of HIV on the Sustainable Development Goals 1 .
Despite the expectation of similar results, some countries showed completely different trends in the data obtained from the two different datasets (GBD and UNAIDS).For example, the death rate in Egypt and Lebanon decreased based on GBD data but increased based on UNAIDS data.Similarly, the incidence rate for Qatar was decreasing based on GBD data, but increasing based on UNAIDS data.Except for some inconsistencies mentioned in the results obtained from the two different data, the incidence and mortality rates in the rest of the countries in the two data although not the same had almost similar results.One of the reasons for this discrepancy is that the GBD data calculated age-adjusted incidence and mortality rates, while the UNAIDS data calculated incidence and mortality rates.Another reason may be the different sources of data collection in some countries, which led to different results.In general, in countries where both incidence and mortality rates are decreasing, it can be concluded as a sign of correct policy making and planning and the accurate implementation of these programs.In countries where both the incidence and mortality rates are increasing, it is probably a sign of the lack of a correct program or proper implementation of existing programs.In these countries, it is difficult to conclude that the increase in incidence is due to the improvement in disease diagnosis.In countries where one of the incidence or mortality rates is increasing and the other is decreasing, it will be a very complicated task to interpret the results and more investigations will be needed.

Limitation
There are several limitations to the present analysis that need to be considered.First, we used two databases (GBD and UNIAIDS) to compute the AAPC.These databases can recognise only in-care patients diagnosed by physicians; therefore, there is a possibility of underestimating the results in modelling.Hence, our findings of modelling may underestimate the true incidence and mortality rate of HIV in the region.Although the GBD 2019 made various modifications to the source and evaluation of the HIV incidence to improve data accuracy, it was obvious that some deviations in the precision and completeness of the GBD data were unavoidable.To come up with this issue, we used two databases for modelling to cross-validate the results.Thus, both databases to estimate the trend of HIV infection in MENA countries have their particular advantage and restriction.The results of both of them showed a significant increase in incidence and mortality rates in most countries in the region, so there are not enough claims to identify a better database for our modelling.The outputs of the two sources supported each other in the majority of countries.
Second, it is essential to acknowledge the potential risk of the ecological fallacy, which arises from making individual-level inferences based on aggregate-level data.Despite the limitations, our study offers a comprehensive depiction of HIV status in MENA countries.This analysis provides an extensive overview of the trend of HIV infection and mortality in MENA countries.

Conclusion
Our study found increasing incidence and mortality rates, as well as DALYs, associated with HIV in several countries across the MENA region.Future studies are needed to explore the underlying mechanisms for these epidemiological trends with potential risk factors incorporated into further analysis.To effectively address this rising trend and control the epidemic, policymakers must prioritize preventive measures in each country, including educating the pathways of transmission, informing and using the early detection methods, and timely treatment.We suggest that governments should strengthen HIV prevention and treatment programs, especially in countries with high incidence and mortality of HIV.Special attention should be directed toward improving HIV testing scale-up approaches, designing targeted interventions, expanding harm reduction services, and enhancing HIV surveillance systems.There is still plenty of room for improvement to achieve public health goals, such as HIV elimination in the MENA.Some structural interventions like addressing stigma and discrimination, and implementing supportive laws for key populations and people living with HIV are necessary to manage the epidemic in this region.

Table 1 .
Table S6 presents the AAPC of mortality rates from HIV in MENA countries by age groups, 1990-2019.Joinpoint regression analysis of age-standardized incidence rates, and incidence rates from HIV in MENA countries, 1990-2019.APC annual percentage change, UAE United Arab Emirates.

Table 3 .
Joinpoint regression analysis of disability-adjusted life years from HIV in MENA countries, 1990-2019.APC annual percentage change, UAE United Arab Emirates.*P < 0.05 versus 0 (output from joinpoint regression analysis).

Table 5 .
Age-standardized mortality rates and prediction of age-standardized mortality rates from HIV in 2025 and 2030 per 100000 in MENA countries.UAE United Arab Emirates.