Estimation of life-time survival and predictors of mortality among the people living with HIV/AIDS: a case study in Andhra Pradesh, India

INTRODUCTION

HIV/AIDS illness remains global public health challenge and main cause of morbidity and mortality worldwide since the first case was reported in the early 1980’s. After 30 years, roughly 34 million people (31.6-35.2 million) were living with HIV (PLHIV) worldwide in 2010 that increased 17% from 2001.¹ Demographically, India is the second largest country in the world has also the third largest number of people living with HIV/AIDS. The national adult HIV prevalence was estimated as 0.31% (C.I. 0.25-0.39%) in 2009 with persons living with HIV/AIDS were 2.40 million (C.I. 1.93-3.04 million).² Andhra Pradesh has second highest HIV prevalence (0.90%) after Manipur among six high prevalent states with more than 1, 00,000 PLHIV at one point of time.³ The early initiation of antiretroviral therapy (ART) improved survival of HIV infected individuals worldwide since its inception as in the absence of cure for HIV.^4,5 The main purpose of ART is to prolong lives, making HIV/AIDS as a chronic disease rather than death sentence.⁶ The low- and middle-income countries expended the access to ART in early 2000, particularly with the World Health Organisations’ ‘‘3 by 5’’ initiative, which aimed to have three million people on ART by 2005.⁷ In 2004, the ART program in India was initiated under national AIDS control programme (NACP) with the aim to provide free services and across the country more than 350 ART centers serving the nation at present.⁸ As of January 2014 in Andhra Pradesh, 55 ART centers are functional.⁹ The scale-up of ART access took place in 2007 with new criteria for ART initiation that results more than seven million PLHIVs had access to ART in 2012.¹⁰

The AIDS-related mortality decreases due to introduction of highly active antiretroviral therapy in developed countries and the life-time survival after infection was estimated in the range from 12-25 years.¹¹ However, in developing and low-income countries the lower survival was reported after HIV-infection.^12-14 The survival of PLHIV individuals and underlying associated factors with mortality depend on demographic characteristics, laboratory variables and clinical indicators.^15-19 A nationwide large scale household survey suggested that only 17% and 25% currently married women have heard about STIs and HIV, respectively.²⁰ Moreover, empirical research in India have reported conflicting trends in mortality rates; few studies reported increasing mortality trends among HIV infected individuals while another set of studies reported a decline in mortality rates over the years.^18,19,21 This has prompted researchers and policy makers to questions on contributions of ART on survival status of HIV infected individuals.

There is a little evidence regarding life-time survival and correlates of mortality in HIV-infected individuals in Andhra Pradesh and India by large due to the lack of prospective studies. This analysis aimed to investigate the survival and associated determinants of survival probability in HIV/AIDS-patients registered at various ART centers in Andhra Pradesh, India.

METHODS

Study settings and participants

The records of HIV-positive individuals who registered at different ART centers (around 55) across the Andhra Pradesh state, India during Jan 2007 to Dec 2011 with the follow-up till Dec 2013 was extracted from ART surveillance database. A retrospective cohort of 3, 13, 175 HIV-positive patients were registered at different ART centers during 2007-11. Out of this, 2,93,560 were adult (age ≥15 years; 1,51,878 male, 1,81,514 female and 451 were TS/TG) and remaining 19,615 were registered as paediatric cases. Out of these, 63,882 records were found with complete follow-up information (Figure 1).

All adult male and female were included in the present study. The baseline information on different demographic covariate includes: age, education, gender, marital status, occupation. It also includes important programme variables: year of HIV detection, year of ART registration, ART status, mode of transmission and the follow up status at the end of the study. Out of 33, 060 males 22,270 were alive, 5,024 died, 3,301 were transferred out, 162 on medical advice and 2,303 were lost to follow up. Similarly among 30,822 females 23,205 were alive, 2,597 were died, 3,158 were transferred out, 232 on medical advice and 1,630 were lost to follow up during the period. Death of a HIV patient was taken as the primary end point.

Permission to use ART programme data for this analysis has been obtained from National AIDS Control Organization (NACO), Ministry of Health and Family Welfare, Government of India, New Delhi.

Data description and analysis

Demographic information on age, education, gender, marital status, occupation, mode of HIV transmission and initial clinical indicators: clinical stage, CD4 count and functional status of patients was categorized to assess any association with the survival probability. Age was categorized into four groups as 15-30, 31-49 and ≥50 years. Education was categorized into illiterate, primary, secondary and college or above. Occupation was categorized as unemployed and employed. Marital status was categorized as single (unmarried, widow/widower, divorced/separate) and married/live-in. The mode of transmission was categorized as sexual (heterosexual+MSM) and others (injecting drug users/blood transfusion/mother to child). HIV clinical stage was categorized as I, II, III and IV. CD4 count was labeled into <200, 200-349, 350-499 and ≥500. The functional status of patients was assessed whether the patient was ambulatory, bedridden or working at the time of registration. At the end, individual case in the database can be classified into the following four categories: alive, death, LFU/transferred and don’t know based on the status as on 31^st Dec 2013.

The life-time survival was measured as a duration between date of HIV confirmation (as the date of infection is unknown) to date of death for patients who died or the date of the last recorded visit for patients who were censored (e.g. either alive till the end of the study period or lost to follow-up). For the comparison of unadjusted survival rates, Kaplan-Meier (K-M) methods were used and patients were stratified by different population sub-groups and it was plotted among different categories of demographic variables. We used the Wilcoxon log rank tests to assess statistical differences between the groups. To visually present the product limit estimates of the cumulative survival rate, Kaplan-Meier curves were generated to show the life time survival of the patients. Student’s’t’ test was used for statistical difference between mean ages of male and female and Spearman’s rank correlation was applied for correlation between age and median survival.

The widely accepted Cox proportional hazard regression model was applied to identify the associated factors.^11,18,19 The proportional hazards assumption was tested using Schoenfeld residuals. Unadjusted hazard ratios (HR) and adjusted hazard ratios were estimated with 95% confidence intervals (95% C.I.). The data were analyzed using IBM SPSS statistics for windows version 21.0 (Armonk, NY, USA). The two tailed ‘p’ value less than 0.05 was considered as statistical significance.

RESULTS

A total of 63,882 adult HIV-positive cases were retrospectively extracted with complete follow-up information from ART surveillance database. The demographic and initial clinical profile of the PLHIV’s presented in Table 1.

The average age (±SD) at registration of PLHIV was recorded as 36.5±9.1 years among male and 32.1±8.8 years among female (t = 62.1; p <0.001). More than 90% of HIV patients were acquired infection at their younger age (below 50 years), 52.9% were illiterate, 97.4% employed, 68.8% reported as married/Live-in, 93.9% of them were acquired HIV through sexual (hetero+MSM) mode of transmission. Around 40% of them were belong to clinical stage III and IV, 53.1% had CD4 count ≤200, 84.9% were functionally well at the time of registration and 71.2% were alive at the end of the follow-up.

The mortality distribution within the subgroup of predictors during the study period, life-time survival estimation (with 95% C.I.) and their comparison were presented in Table 2.

Around 11.9% PLHIV (n = 7,621) died during the follow-up period. 27.9% (n = 2,128) of them were died within one year of their registration with similar distribution over male (28.0%) and female (29.6%). Life-time survival was estimated to 12.2 (95% C.I. 12.0-12.3) years that was significantly higher among women 13.0 (95% C.I. 12.8-13.2) years than men 11.3 (95% C.I. 11.1-11.6) years. Further, the stratified Kaplan-Meier survival plots illustrated the survival probabilities in different subgroups of in dependent variables (Figure 2) and the log-rank test were used for statistical significance (Table 2).

The mean survival after ART registration was estimated to 5.3 (95% C.I. 4.9-5.7) years and it was also negatively correlated with age at registration (r = -0.13; p <0.001).

Univariate and multivariate Cox proportional hazard models were used to examine the association between different predictor variables and hazard of death showed in Table 3.

The univariable regression analysis demonstrated age, gender, clinical stage CD4 count and functional status were associated with the survival of a patient. Further, all these variables were used in multivariable regression analysis to estimate the adjusted hazard of death. Adjusted hazard ratio (HR) was significantly higher in older age groups (1.2 for 31-49 years and 1.5 for ≥50 years) with reference to the ≤30 years. Males were higher at risk (1.6 times) then female. Multivariate HR was also higher in the higher clinical stage groups (1.2 for stage III, 1.5 for stage IV) as compared with the combination of stage I+II. The risk of death was significantly increase as CD4 decreases (6.3 for <200, 3.6 for 200-349 and 1.4 for 350-499) with respect to CD4 count ≥500. The PHIV’s with bedridden functional status at the time of ART registration was significantly associated with risk of death (1.8 times) with reference to well (or working).

*P<0.05; **P<0.01; ***P<0.0001

DISCUSSION

The present analysis is based on a retrospective cohort of HIV/AIDS patients registered at different ART centers during 2007-11 in Andhra Pradesh, India. The median CD4 count at the time of ART registration remained low, at 190 (IQR: 110-295). These data clearly indicate that people are typically being early diagnosed with HIV in India. At the end of follow-up, 71.2% of them were alive, 11.9% were dead, 10.1% were transferred out and 6.2% were LFU. Sexual root (heterosexual+MSM) (93.9%) remains highest vulnerable mode of HIV transmission. 27.9% deaths occurred within one year of HIV registration. The higher mortality in the first year estimated by this analysis corroborates the findings from several other studies.^{8,12,18,19,22} The higher mortality within the first year of HIV registration could be due to some other reasons like poor adherence to ART regimen combined with other co-morbid infections and low CD4 cell count.¹⁸ Furthermore, we found that the overall lifetime mean survival time was around 12.2 years from the date of HIV detection and around 5.3 years from the date of ART registration with higher average survival among women as reported by others.^12,19 The analysis showed a strong association between survival probability and demographic variables and clinical indicators. The increasing risk of death in HIV-patients has significantly been high in higher age groups as previously reported.^{12,15,17-19,23-25} This empirical analysis clearly revealed that higher mortality in older age. Therefore, survival of HIV-positive individuals was inversely correlated with age. This analysis also recorded the higher proportion of deaths among males (15.2%) as compared with females (8.4%). Around 7.0% in male and 5.3% in female were recorded as lost to follow up, however such differences were minimal in this study. The hazard of death among men was significantly higher than among women, as previously reported by others.^12,26 Higher mortality among male may be due to high reporting by males or biological differences or drug response to ART. However, in some studies it was not found statistically significant.^23-25 The clinical parameters like CD4 count, clinical stage and functional status play vital role in survival of an individual as reported by multivariate regression model. The findings are in agreement with other several studies.^12,26 Though the plenty of research is available worldwide regarding survival of HIV/AIDS patients, however it still required further research in Indian context.

The findings of the present study should be examined in light of certain limitations. The first limitation of this study was related with its study design. We analyzed a retrospective cohort with demographic characteristics and some of clinical variables as CD4 cell count, HIV staging and functional status. Hence we could not take into account all the potential confounders such as, viral load, co-morbid condition that could have influenced the survival of individuals due to non-availability in the data. Another limitation was associated with the cause of death. We considered all-cause mortality in the analysis. There is a possibility that a few of the deaths may have occurred due to causes not related to HIV/AIDS. The third limitation was related to the missing information in different covariates as well as date of death.

CONCLUSION

Despite these study limitations, the study findings have important policy implications for ART programme in India. ART services in India are freely available to all HIV infected persons. The estimates of the survival rate among ART individuals suggests that country’s ART program has been successful in increasing longevity of life of HIV infected individuals. However, there are certain concerns regarding missing information in the database. Efforts are required at the management as well as at implementation level for constant and regular follow-up with all patients. In addition, the program should expand to cover more remote areas to provide services to some of the underserved population groups like women. Our study also demonstrated higher mortality among individuals within one year of registration. This argues for the need to identify and treat eligible HIV infected individuals at an early stage. In addition, late appearance to ART centers is one of the main reasons for higher mortality in the first year after registration. Therefore, attention is needed to ensure early diagnosis and enrollment into ART for PLHIV’s.

ACKNOWLEDGEMENTS

The views expressed herein are those of the authors and do not necessarily reflect the official policy of AIDS control programme. Authors are thankful to the National AIDS Control Organization (NACO) for providing individual ART data of Andhra Pradesh, India.

Funding: No funding sources

Conflict of interest: None declared

Ethical approval: The study was approved by the Institutional Ethics Committee

REFERENCES

1. 1. World AIDS Day Report. UNAIDS, Geneva, Switzerland, 2012.
2. 2. Technical reports: India HIV estimates. National AIDS Control Organization. Ministry of Health & Family Welfare, India, 2010.
3. 3. Pandey A, Sahu D, Bakkali T, Reddy DCS, Venkatesh S, Kant S, et al. Estimate of HIV prevalence and number of people living with HIV in India 2008–2009. BMJ Open. 2012;2:e000926.
4. 4. Egger M, May M, Chene G, Phillips AN, Ledergerber B, and Dabis Fet al. Prognosis of HIV-I infected patients starting highly active antiretroviral therapy: a collaborative analysis of prospective studies. Lancet. 2002;360:119-29.
5. 5. Mocroft A, Vella S, Benfield TL, Chiesi A, Miller V, Gargalianos P et al. Changing patterns of mortality across European patients infected with HIV-I, EuroSIDA Study Group.Lancet. 1998;352:1725-30.
6. 6. Kim JY, Ammann A. Is the ‘‘3 by 5’’ initiative the best approach to tackling the HIV pandemic? PLoS Med. 2004;1:e37.
7. 7. Bachani D, Garg R, Rewari BB, Hegg L, Rajasekaran S, Deshpande A, et al. Two-year treatment outcomes of patients enrolled in India’s national first-line antiretroviral therapy programme. Natl Med J India, 2010;23:7-12.
8. 8. Annual reports: 2011-12. National AIDS Control Organization, Ministry of Health & Family Welfare, India.
9. 9. Vella S, Schwartlander B, Sow SP, Eholie SP, Murphy RL. The history of antiretroviral therapy and of its implementation in resource-limited areas of the world. AIDS. 2012;26:1231-41.
10. 10. Kee MK, Lee JH, Kim EJ, Lee J, Nam JG, Yoo BH et al. Improvement in survival among HIV-infected individuals in the Republic of Korea: Need for an early HIV diagnosis. BMC Infectious Diseases. 2009; 9:128.
11. 11. Bajpai RC, Raj PY, Jha UM, Chaturvedi HK, Pandey A. Demographic Correlates of Survival in Adult HIV Patients Registered at ART Centers in Andhra Pradesh, India: A Retrospective Cohort Study. PHR. 2014;4:31-8.
12. 12. WenchekoE AT. Survival analysis of patients under chronic HIV-care and antiretroviral treatment at Tikur Anbessa Specialized Hospital, Addis Ababa, Ethiopia. Ethiop J Health Dev. 2012;26:22-9.
13. 13. Shaboltas AV, Toussova OV, Hoffman IF et al. HIV prevalence, sociodemographic, and behavioral correlates and recruitment methods among injection drug users in St. Petersburg, Russian. J Acquir Immune Defic Syndr. 2006;41:5.
14. 14. Ghate M, Deshpande S, Tripathi S, Godbole S, Nene M, Thakar M et al. Mortality in HIV infected individuals in Pune, India. IJMR. 2011;133:414-20.
15. 15. District Level Household and Facility Survey (DLHS-3), 2007–08: India: Key Indicators: States and Districts, IIPS, Mumbai, India.
16. 16. Teja VD, Sudha T, Lakshmi V. Causes and pattern of mortality in HIV infected, hospitalized patients in a tertiary care hospital: a fourteen year study. Indian J Med Sci. 2007;61:555-61.
17. 17. Mermin J, Were W, Ekwaru JP, Moore D, Downing R, Behumbiize P, et al. Mortality in HIV-infected Ugandan adults receiving antiretroviral treatment and survival of their HIV-uninfected children: a prospective cohort study. Lancet. 2008;371:752-59.
18. 18. Hambisa MT, Ali A, Dessie Y. Determinants of mortality among HIV positives after initiating antiretroviral therapy in Western Ethiopia: a hospital-based retrospective cohort study. ISRN AIDS. 2013;1-7.
19. 19. Doherty MC, Garfein RS, Monterroso E, Brown D, and Vlahov D. Correlates of HIV infection among young adult shortterm injection drug users. AIDS 2000;717-26.
20. 20. Alonso-Gonzalez M, Martin L, Munoz S, Jacobson J. Patterns, trends and sex differences in HIV/AIDS reported mortality in Latin American countries: 1996–2007. BMC Public Health. 2011;11:605.
21. 21. Arrivillaga M, Ross M, Useche B, Alzate ML, Correa D. Social position, gender role, and treatment adherence among Colombian women living with HIV/AIDS: social determinants of health approach. Revista Panamaricana de Salud Publica. 2009;26:502-10.
22. 22. Mallitt KA, Grigoryan SR, Papoyan AS, Wand HC, Wilson DP. Access to antiretroviral therapy and survival in eastern Europe and central Asia: a case study in Armenia. J Int AIDS Soc. 2014;17:18795.
23. 23. Hambisa MT, Ali A, and Dessie Y. Determinants of mortality among HIV positives after initiating antiretroviral therapy in Western Ethiopia: a hospital-based retrospective cohort study. ISRN AIDS 2013;1-7.
24. 24. Dominique YMG. HIV/AIDS in Haiti: an analysis of demographics, lifestyle, STD awareness, HIV knowledge and perception that influence HIV infection among Haitians. Public Health Theses. 2011:172.
25. 25. Doherty MC, Garfein RS, Monterroso E, Brown D, Vlahov D. Correlates of HIV infection among young adult shortterm injection drug users. AIDS. 2000;717-26.
26. 26. Taylor-Smith K, Tweya H, Harries A, Schoutene E, and Jahn A. Gender differences in retention and survival on antiretroviral therapy of HIV-1 infected adults in Malawi. Malawi Med J. 2010;22:49-56.