Factors associated with high-risk mortality in HIV patients receiving highly active antiretroviral therapy in southwestern Ethiopia: a retrospective cohort study
- (0) Download
https://www.eduzhai.net Public Health Research 2013, 3(6): 169-177 DOI: 10.5923/j.phr.20130306.03 Factors Associated with High Risk Mortality of HIV Patients Treated with Highly Active Anti Retroviral Therapy in South Western Ethiopia: A Retrospective Cohort Study Birtukan T Adamu1, Eshetu Wencheko2, Hailay A Gesesew1,*, Amanuel T Gebremedhin3 1Department of Epidemiology, Jimma University, Jimma, Ethiopia 2Department of Statistics, Addis Ababa University, Addis Ababa, Ethiopia 3Department of Population and Family Health, Jimma University, Jimma, Ethiopia Abstract The introduction of HAART has brought about a significant reduction in the morbidity and mortality of patients with HIV infection. However, the mortality rate of patients treated with HAART is still very high in resource-poor settings. Factors contributing to this high mortality rate are poorly understood. Therefore, the objective of this study was to describe survival status and to identify the determinant factors of HIV associated mortality in a cohort of HIV infected patients treated with HAART. The study has reviewed patient forms and follow up cards of 832 patients treated with HAART in Jimma University Specialized Hospital from 2003-2007. Kaplan-Meier survival curves and Log-Rank test were used to compare the survival experience of different groups of patients and proportional hazards Cox model was used to explore the factors associated with increased risk of mortality. Some 144 patients died during the follow up time of which 48.6% and 68.8% deaths occurred within three and six months of HAART initiation, respectively. The overall mean estimated survival time of patients was 63.7 months. Factors/covariates associated with increased risk of mortality were older age (HR=1.03, 95% CI: 1.01-1.051), low CD4 count at baseline (HR=0.994, 95% CI: 0.992-0.996), low weight at baseline (HR for a 5kg change=0.902, 95% CI: 0.816-0.996), bedridden and ambulatory functional status (HR=6.904, 95% CI: 4.005-11.902) and (HR=2.877, 95% CI: 1.899-4.360), respectively, co-infection with TB (HR=1.906, 95% CI: 1.305-2.784) and substance use (HR=1.42, 95% CI: 1.016-1.985). Thus, early diagnosis and treatment should be encouraged and focus should be given in these predictors. Keywords HAART, HIV mortality, Factors, Survival analysis, Ethiopia 1. Introduction AIDS is an acronym for acquired immune deficiency syndrome, the final disease manifestation from infection by human immunodeficiency virus (HIV). Today it is one of the largest public health crises endangering the human race. In almost three decades since its first cases were recognised, it has claimed the lives of millions of people making it one of the most devastating epidemics. In the countries most heavily affected, HIV has reduced life expectancy by more than 20 years, slowed economic growth, and deepened household poverty. It has already caused an estimated 25 million deaths worldwide and has generated profound demographic changes. In Sub-Saharan Africa alone, the * Corresponding author: email@example.com (Hailay A Gesesew) Published online at https://www.eduzhai.net Copyright © 2013 Scientific & Academic Publishing. All Rights Reserved epidemic has orphaned nearly 12 million children aged less than 18 years. The earliest evidence of HIV infection in Ethiopia was found in 1984, with the first case reported in 1986. Adult HIV prevalence in 2009 was estimated to be between 1.4% and 2.8% in the country. Highly active antiretroviral therapy (HAART) was a breakthrough in care and treatment of people living with HIV, leading to a reduction in mortality and an improvement in the quality of life. Antiretroviral drugs significantly lowered the rate of HIV transmission from mother to child, and antiretroviral therapy (ART) has become an integral part of the continuum of HIV care. A study in the US conducted from January 1994 through June 1997 which evaluated 1255 patients who were using antiretroviral revealed that mortality among the patients declined from 29.4 per 100 person-years in 1995 to 8.8 per 100 person-years in the second quarter of 1997. The incidence of major opportunistic infections (OI) also declined from 21.9 per 100 person-years in 1994 to 3.7 per 100 person-years by 170 Birtukan T Adamu et al.: Factors Associated with High Risk Mortality of HIV Patients Treated with Highly Active Anti Retroviral Therapy in South Western Ethiopia: A Retrospective Cohort Study mid-1997. Another multinational study conducted in Europe also demonstrated that mortality rates for HIV-infected persons have become much closer to general mortality rates since the introduction of HAART. According to this study, persons infected sexually with HIV now appear to experience mortality rates similar to those of the general population in the first five years following infection though a mortality excess remains as duration of HIV infection lengthens. A similar retrospective cohort study in northern Thailand, which used the same method of analysis, also revealed that sex, age group, registered year, clinical status, CD4 group, and ARV drug group were all significantly related to death in the univariate analysis. In Ethiopia, according to the 2007 single point estimate, there were an estimated 1,116,216 people living with HIV in 2009, of which 336,160 were eligible for ART. There were an estimated 131,145 new HIV infections and 44,751 AIDS-related deaths of which females accounted for 57% of the total infections and deaths. The total estimated number of HIV-positive pregnant women and annual HIV positive births in the same year were 84,189 and 14,140, respectively . There were an estimated 72,945 children less than 15 years old living with HIV, out of which 20,522 needed ART. Due to the combined effect of poverty and AIDS, more than 5.4 million children under the age of 18 years were orphaned out of which 855,720 (16%) lost at least one parent due to AIDS. Age, religion, education level and active TB were some of the factors associated with High Risk of Mortality of HIV Patients Treated with HAART. Numerous researches have also been conducted that tried to address many of the issues that arise in connection with the HIV epidemic. But most studies focused on the assessment of the prevalence, adherence and prevention measures. Given this as a back drop, this study focused on the consideration of possible factors that may possibly influence the survival status of people who are following ART in South West Ethiopia. The study of these factors will provide information to put in place efficient ART system that will take into account the huge cost that is involved in the system. baseline and follow up clinical and laboratory measurement information, and treatment outcomes. It was collected by ART trained nurses. A standard data abstraction format was used for recording information extracted from patients’ cards which was developed using the standardized ART entry and follow up used by the ART clinic. Four experienced ART nurses who were trained on comprehensive HIV care and involved in patient follow ups collected the data. To assure the quality of the collected data, training was given for the data collectors and supervisor and the principal investigators were supervising the data collection. The data were entered and cleaned by the investigators before analysis and were coded and analyzed using the statistical packages SPSS 13, SAS 9.2 and STATA 9.2. Descriptive survival analyses including graph of the estimate of overall Kaplan-Meier survivor function and Cox regression were conducted. To investigate the significance of the observed difference in the Kaplan-Meier estimates of the survivor functions among different categories of the factors, log-rank test was performed. Hazard ratios (HR), as well as 95% confidence intervals were used as effect measures with p-value of 0.05. The final model was built with back ward elimination. Cox-Snell residuals are used to asses the overall goodness of fit of the model and the deviance residuals from the final Cox model are used to check if each observation is well fitted by the Cox model. The Log partial likelihood was also used. To check the adequacy of the proportional hazards assumption, extended Cox model was used and graphical check was also used to provide any additional insight into any departure from proportionality. The assumption of proportionality was also assessed graphically by plotting the scaled Schoenfeld residuals of each covariate against log time. Moreover, linearity of the continuous variables in the model was checked by using martingale residuals. Ethical clearance was obtained from the health research and post graduate coordinating office of College of public health and medical sciences of Jimma University. Permission letter was obtained from JUSH and all information was confidentially used only for research purpose. 3. Results 2. Methods and Participants The study was conducted in Jimma University Specialized Hospital (JUSH). The hospital provides Voluntary Counselling and testing (VCT), Prevention of Mother to Child Transmission (PMTCT), ART and opportunistic infections (OIs) treatment service. Data were collected from March to April 2010. Retrospective cohort study was conducted among HIV positive participants. A total of 832 patients on HAART from October 2003 to August 2007 were included in the study. Data extraction format prepared in English and log books were used to collect the information from ART records about socio-demographic characteristics, 3.1. General Characteristics of the Cohort The study included 832 HIV patients, who started HAART in Jimma University Specialized Hospital (JUSH) between the years 2005 and 2007, about whom complete information related to the study covariates is available. Among the patients, 686 started their treatment on D4T-based drug regimens. Of the total of 832 patients included, 475 were females, 167 were never married, and 219 were living in Jimma town. When HAART was initiated, 197 patients had clinical AIDS (WHO stage IV) and 58 patients had bedridden functional status due to the severe progression of the disease. TB and opportunistic infections (OI) were prevalent among 428 and Public Health Research 2013, 3(6): 169-177 171 512 patients, respectively at baseline. Most of the patients of their family, 285 of the patients had casual sexual partners (728) disclosed their HIV sero-status at least for one member and 276 of the patients were substance users (Table 1). Table 1. Distribution of Socio-Demographic and Clinical Characteristics of HIV patients treated with HAART in JUSH, South west Ethiopia, 2010 Covariate / factor Sex Marital status Religion Level of education Employment status Place of residence No of Rooms Disclosure status Risk behaviour Substance use Functional status WHO clinical stage Past OI TB co-infection Category Female Male Never married Married Others Muslim Orthodox Others No education Primary Secondary or above Employed Non Employed Jimma Others One room Two or more rooms Disclosed Not disclosed Regular Casual or Both Yes No Working Ambulatory Bedridden Stage I or II Stage III Stage IV Yes No Yes No Censored 394 294 134 347 207 180 397 111 119 252 317 195 493 180 508 369 319 601 87 463 225 216 472 398 259 31 130 411 147 340 348 404 284 Dead (%) 81 (17.1) 63 (17.6) 33 (19.8) 72 (17.2) 39 (15.9) 42 (18.9) 85 (17.6) 17 (13.3) 27 (18.5) 43 (14.6) 74 (18.9) 39 (16.7) 105 (17.6) 39 (17.8) 105 (17.1) 73 (16.5) 71 (18.2) 127 (17.4) 17 (16.3) 84 (15.4) 60 (21.1) 60 (21.7) 84 (15.1) 33 (7.7) 84 (24.5) 27 (46.6) 19 (12.8) 75 (15.4) 50 (25.4) 88 (20.6) 56 (13.9) 108 (21.1) 36 (11.3) Total 475 357 167 419 246 222 482 128 146 295 391 234 598 219 613 442 390 728 104 547 285 276 556 431 343 58 149 486 197 428 404 512 320 3.2. Descriptive Survival Analyses The patients were followed up for a median of 40 months. The minimum follow up time was 1 month and the maximum was 78 months. Some 144 patients died during the follow up time of whom 70 (48.6%) and 99 (68.8%) deaths occurred within three months and six months of HAART initiation, respectively. The overall mean estimated survival time of patients under the study was 63.7 (95% CI: 61.1- 66.3) months. Females have relatively lower survival time (59.1 months) than males (63.8 months). Patients with younger age (40 years or less) had survived for about 64.4 months while the mean survival time for older patients was 57.6 months (Table 2). 172 Birtukan T Adamu et al.: Factors Associated with High Risk Mortality of HIV Patients Treated with Highly Active Anti Retroviral Therapy in South Western Ethiopia: A Retrospective Cohort Study Table 2. Kaplan-Meier analyses of survival times for patients on antiretroviral treatment according to important socio-demographic and clinical characteristics of HIV patients treated with HAART in JUSH, South west Ethiopia, 2010 Covariate / Factor Sex Age group Marital status Religion Level of education Employment status Place of residence No of Rooms Disclosure status Risk behaviour Substance use Functional status WHO clinical stage Past OI TB co-infection Category Female Male Below 40 40 and above Never married Married Others Muslim Orthodox Others No education Primary Sec. or above Employed Non Employed Jimma Others Only One Two or more Disclosed Not disclosed Regular Casual or Both Yes No Working Ambulatory Bed ridden Stage 1 or 2 Stage 3 Stage 4 Yes No Yes No Overall survival time Estimate 59.1 63.8 64.4 57.6 55.1 63.8 56.6 62.4 63.3 61.8 62.0 60.5 63.3 64.1 62.7 63.7 62.2 58.6 64.1 63.6 49.0 66.2 54.6 58.6 66.3 71.3 57.1 39.2 58.7 65.2 56.7 62.2 65.1 60.9 67.9 63.7 Std. Error 1.5 1.6 1.5 2.6 2.0 1.7 1.5 2.3 1.3 2.1 2.4 1.7 1.5 2.2 1.2 2.1 1.5 1.4 1.5 1.4 2.0 1.2 1.7 2.1 1.2 1.0 1.8 5.1 1.8 1.2 2.7 1.5 1.8 1.4 1.9 1.3 95% confidence interval 56.2 61.9 60.7 67.0 61.6 67.3 52.4 62.7 51.1 59.0 60.5 67.0 53.6 59.6 57.9 67.0 60.7 65.9 57.8 65.9 57.3 66.8 57.3 63.8 60.3 66.3 59.9 68.3 60.4 65.0 59.7 67.8 59.2 65.1 55.9 61.2 61.1 67.0 60.9 66.3 45.0 52.9 63.8 68.5 51.3 57.9 54.5 62.6 64.0 68.6 69.4 73.1 53.6 60.6 29.1 49.2 55.2 62.2 62.8 67.7 51.4 62.1 59.2 65.1 61.6 68.5 58.2 63.6 64.1 71.7 61.1 66.3 Similarly, HIV positive patients diagnosed with stages I and II have relatively higher survival time than those who were in stage IV and so does for patients diagnosed with out Tb co-infected than with Tb co-infected (figures 1 and 2). Public Health Research 2013, 3(6): 169-177 173 Kaplan-Meier survival estimates, by WHO 1=Stage I or II 2=Stage III 3=Stage IV 0.00 0.25 0.50 0.75 1.00 0 20 40 60 80 analysis time WHO = 1 WHO = 3 WHO = 2 Figure 1. Survival functions stratified according to WHO clinical staging in HIV infected patients in a cohort of patients on antiretroviral treatment in JUSH, South west Ethiopia Kaplan-Meier survival estimates, by TB 1=Co-infected 2=Not co-infected 0.00 0.25 0.50 0.75 1.00 0 20 40 60 80 analysis time TB = 1 TB = 2 Figure 2. Survival functions stratified according to Tb cp-infection in HIV infected patients in a cohort of patients on antiretroviral treatment in JUSH, South west Ethiopia 3.3. Predictors of Mortality Risky behaviour, substance use, functional status, WHO clinical stage, opportunistic Infections, HIV tuberculosis co-infection, age at HAART initiation, base line weight, base line CD4 count were significant variables in univariable proportional hazards Cox regression model. Finally, age, baseline CD4 count, baseline weight, substance use, functional status and HIV tuberculosis co-infection were independently associated with high mortality of the patients under HAART (Table 3). 174 Birtukan T Adamu et al.: Factors Associated with High Risk Mortality of HIV Patients Treated with Highly Active Anti Retroviral Therapy in South Western Ethiopia: A Retrospective Cohort Study Table 3. Independent predictors of mortality among a sample of HIV infected cohorts on HAART in JUSH, south Ethiopia, 2010 Covariates / Factors Age Baseline CD4 count Baseline weight Substance use No Yes Functional status Working Ambulatory Bedridden TB co-infection No Yes AHR (95% CI) 1.03 (1.010 -1.051) 0.994 (0.992 - 0.996) 0.979 (0.960 - 0.999) 1 1.42 (1.016 - 1.985) 1 2.877 (1.899 - 4.360) 6.904 (4.005 -11.902) 1 1.906 (1.305 - 2.784) P-Value 0.0031 <.0001 0.0426 0.0404 <.0001 <.0001 0.0009 commulative hazard of Cox-Snell residuals While checking the adequacy of the proportional hazards assumption using extended Cox model and graphical method, it was found as fit. Again, Cox-Snell residuals and Log partial likelihood were also used to assess the overall goodness of fit of the model and was found as fit (figure 3). 1.9 1.8 1.7 1.6 1.5 1.4 1.3 1.2 1.1 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 Cox-Snell residual Figure 3. Cumulative hazard plot of the Cox-Snell residuals of the proportional hazards Cox regression model. The 450-striaght line through the origin is drawn for reference Age of patients at the start of HAART is found to have a significant effect on the mortality of patients (estimated HR=1.03, 95% CI: 1.01-1.051, p=0.0031). Since the estimated hazard ratio is greater than unity, other things being equal, the higher the age of a patient, the greater the hazard of death at any given time. In particular an increase of 1 year in age of patients increases the hazard rate of death by 3%. The 95% CI also indicates that the hazard rate could be as low as 1.01 or as large as 1.051. Baseline CD4 count is also another covariate which has a significant effect on the mortality of patients (estimated HR=0.994, 95% CI: 0.992-0.996, p=<0.0001). The estimated hazard ratio for a 50 cells/mm3 increase in the baseline CD4 count is exp(−0.006 * 50) = 0.741 , and the corresponding 95% CI for the hazard ratio is 0.659 to 0.833. The interpretation is that patients whose CD4 count is higher by 50 cells/ mm3 are dying at a hazard rate 26% lower than for patients with lower count. Also a 5 kilograms (kg) increase in the baseline value of weight of patients decreases the hazard rate of death by 10%. However, the decrease in Public Health Research 2013, 3(6): 169-177 175 the hazard rate could be as low as less than 1% or as high as 18% (the estimated HR for a 5kg change=0.902, 95% CI: 0.816-0.996). Substance use, functional status and TB co-infection are the three categorical variables that are found to be significantly associated with the survival of patients in the fitted Cox regression model. The hazard ratio for substance use is 1.42. Thus, patients who are abusing substance have a 42% higher risk rate of death than those that are not using substances. The confidence interval indicates that the risk of death for substance user patients could be higher by a quantity as large as 99% or as low as 1.6% than patients who are not using substance; p =0.0404. The estimated relative risk (hazard ratio) of dying for patients co-infected with TB as compared to those who are not infected with TB is 1.906 (95% CI: 1.305-2.784). This means being co-infected with TB almost doubles the hazard rate of death. The 95 % confidence interval also suggests that the risk of death for TB co-infected patients is 1.306 times as low and 2.784 times as large as compared to patients who are not co-infected with TB after adjusting for other variables in the model. The estimated risks of death for a patient with ambulatory and bedridden functional status compared to those patients in a working functional status are 2.877 (95% CI: 1.899-4.360) and 6.904 (95% CI: 4.005-11.902), respectively. This means that the hazard rate of death for ambulatory and bedridden patients is 2.8 times and 7 times the working patients, respectively. Moreover, the estimated hazard ratio of bedridden functional status compared to ambulatory functional status is exp(1.9321 − 1.0569) = 2.399 (95% CI: 1.53 to 3.75). Since the confidence interval does not contain 1, an individual whose functional status is bedridden has a significantly higher hazard rate, at any given time, than patients with ambulatory functional status. Thus, compared to patients with ambulatory functional status the hazard rate of bedridden patients is 2.4 times the hazard rate of ambulatory patients. 4. Discussion This historical cohort study found that the significant predictors of lower chance of survival in patients living with HIV/AIDS after initiation of ART were: older age, low CD4 count at baseline, low weight at baseline, substance use, being bedridden and ambulatory and co-infection with TB. WHO clinical stage, OI at baseline and risk behaviour are only significantly associated with mortality in the univariable analysis but not in the multivariable analysis. The mortality rate of patients in the earlier months of ART initiation was high and it declined in the later months of follow up. Studies conducted in southern Ethiopia and Tanzania also revealed that the mortality rates were high within the first three months of follow up[7, 8-10]. This may be attributable to the fact that most of the patients start HAART at the severe stage of the disease. Besides, the development of IRIS may also contribute. CD4+ cell count was the most important marker of HIV disease progression and a strong predictor of survival which has been revealed by many studies. A similar study in Uganda also found that the risk of mortally of patients having CD4 count of less than 50 cells/mm3 is more than 4 times compared to those patients having a CD4 count greater than 50 cells/mm3. Studies from Malawi, South Africa, and Ivory Coast also identified lower CD4 count as a predictor for lower survival and/or increased rate of mortality[12-13, 14]. The use of CD4+ cell level as the primary means to evaluate survival among HIV-infected individuals is also supported by the observation that duration of infection has less prognostic value than the CD4+ cell count at a given point in time. Tuberculosis increased the rate of mortality. Patients co-infected with TB had nearly 2 times higher risk of dying while on ART compared to non-infected. Studies from Baltimore, USA and mainland China have also shown that TB is an independent predictor of mortality in patients on ART, after controlling for potential confounders, including CD4 cell count and viral load[15-16]. A cohort study in Abidjan, Ivory Coast also found out that TB is a risk factor for immunological and virological failure, which leads to severe morbidity and mortality in adult patients treated with ART. The probable reason could be its effect on compromising an immunity, drug interaction and exacerbation of side effects. The functional status of patients can be seen as an indicator of the severity of the progression of the disease. Those patients who are in ambulatory and working functional status have the strength to work and engage themselves in household activities which may help them generate additional income and improve their quality of life. Even though patients who are staying in bed in hospitals are accessing medical care and support, cannot do anything by their own to create a stress-free environment for them. Thus, as expected, patients with ambulatory and bedridden status are at a higher risk of mortality than working patients. Symptomatic disease (WHO stages III and IV) was associated with mortality in many studies[8-10]. However, our finding shows that WHO clinical stage is associated with mortality in the univariabe analysis only, possibly reflecting differences in the accuracy of clinical staging or homogeneity of cohorts with respect to this variable. Thus, in the univariable Cox regression, being WHO stage IV led to increased rate of mortality when compared to stage I and II. This may be due excess viral load which might not have the ability to defend the other opportunistic infections and off course the efficacy drug decreases as WHO stage increases. Substance abuse is another significant predictor of death in our study. This supported by many studies[17-18]. This explanation could be due to that substance abuse may impair judgment and the ability to adopt and maintain routine medication use. Therefore, screening for drug abuse and excessive alcohol use, and supportive counselling and treatment for drug abuse might help in promoting long term adherence to ART. 176 Birtukan T Adamu et al.: Factors Associated with High Risk Mortality of HIV Patients Treated with Highly Active Anti Retroviral Therapy in South Western Ethiopia: A Retrospective Cohort Study Even though different previous studies demonstrated body mass index (BMI) as one of the strong predictors of mortality [8-10], it was not possible to calculate BMI for patients since only the weight but not the height of the patients has been recorded at baseline. However, the study revealed that the weight of patients is significantly associated with mortality and those patients having lower weight are at higher risk of mortality. This study has its own strength and limitations. As being the study used was cohort study design, it allows showing the temporal relationship between the exposure variables and the outcome. Despite this, the study was conducted based on secondary data which might have incomplete and biased information. There might also underestimation of mortality due to lost to follow up patients included in the study. In addition to this all deaths are assumed to be caused by AIDS. The study includes only baseline values of the variables. i.e. CD4 cell count stability or improvement, weight lose or gain, treatment adherence, treatment switches or substitution, number of missed appointments, which are associated to mortality of AIDS patients, are not included in the study. Lastly, the study used CD4 cell count as indirect surrogate indicator instead of viral load. Also the study used weight, which might be affected by height, instead of BMI. 5. Conclusions In general, the mortality rate was very high in the earlier months of HAART initiation and tended to stabilize later. Majority of deaths occurred within three and six months of HAART initiation, respectively. Moreover, the results of the multivariable proportional hazards Cox regression model showed that lower CD4 count at the start of HAART, lower weight, older age, TB co-infection, substance use and being of bedridden or ambulatory functional status are associated with higher risk of mortality. Therefore, health workers and other ART clinic staff should plan for more frequent contacts with patients during the early phase of treatment in order to prevent the many deaths that occur during the early weeks of ART. Therefore, patients should be informed about the need for early diagnosis of HIV infection and starting treatment early is very important. Moreover, treatment of opportunistic infection parallel to the ART programme may reduce the risk of mortality. Further research should be conducted why lower weight and older patients associated with increased rate of mortality. A separate treatment programme for drug user patients is important and careful monitoring of drug adherence should be made available, as the effect of the treatment is highly dependent on adherence. Finally, health workers and peer educators and data clerks, working with patients under HAART, should be given special training to improve the quality of the data records of patients. Moreover, attempt should be made to investigate the causes of deaths that occurred out of hospitals, and mechanisms should be devised to trace patients lost to follow up. ACKNOWLEDGMENTS Our earnest gratitude goes to Department of Epidemiology of Jimma University, Jimma University Specialized Hospital, and data collectors of the institution for their cooperation and assistance. We also thank Jimma University for funding the study. REFERENCES  UNAIDS, Geneva. (2008) Report on the global AIDS epidemic.  Federal HIV/AIDS Prevention and Control Office, Ethiopia. (2010) Report on progress towards implementations of the UN declaration of commitment on HIV/AIDS.  Federal HIV/AIDS Prevention and Control Office, Ethiopia. (2007) Guidelines for implementation of the antiretroviral therapy programme in Ethiopia.  Palella FJ, Delaney KM, Moorman AC, Loveless MO, Fuhrer J, Satten GA, Aschman J, Holmberg SD. (1998) Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. N Eng. J Med 338:853-860.  Bhaskaran K, Hamouda O, Sannes M. (2008) Changes in the Risk of Death after HIV Sero-conversion Compared with Mortality in the General Population. JAMA 300 (1):51-59.  Pathipvanich P, Ariyoshi K, Rojanawiwat A, et al. (2003) Survival Benefit from Non-Highly Active Antiretroviral Therapy in a Resource-Constrained Setting. JAIDS Journal of Acquired Immune Deficiency Syndromes 32:157-160.  Johannessen A, Naman E, Ngowi BJ, et al. (2008) Predictors of mortality in HIV-1 infected patients starting antiretroviral therapy in a rural hospital in Tanzania. BMC Infectious Diseases 8:52.  Degu J, Lindtjørn B. (2005) Disease Progression among Untreated HIV-Infected Patients in South Ethiopia: Implications for Patient Care. Meds cape General Medicine 7(3):66.  Degu J, Næss A, Lindtjørn B. (2006) Antiretroviral therapy at a district hospital in Ethiopia prevents death and tuberculosis in a cohort of HIV patients. AIDS Research and Therapy 3:10.  Degu J, Endale A, Hailu Y, Lindtjørn B. (2006) Predictors of early death in a cohort of Ethiopian patients treated with HAART. BMC Infectious Diseases 6:136.  Abaasa AM, Todd J, Ekoru K, Kalyango JN, Levin J, Odeke E, Karamagi C. (2008) Good adherence to HAART and improved survival in a community HIV/AIDS treatment and care programme: the experience of the AIDS support organization (TASO), Kampala, Uganda. BMC Health Services Research 8:241.  Ferradini L, Jeannin A, Pinoges L, et al. (2006) scaling up of Public Health Research 2013, 3(6): 169-177 177 highly active antiretroviral therapy in a rural district of Malawi: an effectiveness assessment. Lancet 367: 1335–42.  Moha R, Danela C, Messoua E, et al. (2007) Incidence and determinants of mortality and morbidity following early antiretroviral therapy initiation in HIV-infected adults in West Africa. AIDS 21: 2483–2491.  Lawn SD, Badri M, Wood R. (2005) Tuberculosis among HIV-infected patients receiving HAART: long term incidence and risk factors in a South African cohort. AIDS 19:2109–2116.  Xueyan J, Hongzhou L, Yuexin Z, et al. (2008) A Cross-Sectional Study of HIV and Tuberculosis Co infection Cases in Mainland China. Southern Medical Association 101(9):914-917.  Lo´pez-Gatell H, Cole SR, Margolick JB, Witt MD, Martinson J, Phair JP, Jacobson LP. (2008) Effect of tuberculosis on the survival of HIV-infected men in a country with low tuberculosis incidence. AIDS 22:1869–1873.  Galai N, Vlahov D, Bareta JC, Wang C, Cohn S, Sterling TR. (2005) Prognostic Factors for Survival Differ According to CD4+ Cell Count Among HIV-Infected Injection Drug Users: Pre-HAART and HAART Eras. J Acquir Immune Defic Syndr 38:74-81.  Liu C, Johnson L, Ostrow D, Silvestre A, Visscher B, Jacobson LP. (2006) Predictors for Lower Quality of Life in the HAART Era among HIV-Infected Men. J Acquir Immune Defic Syndr 42:470-477.
... pages left unread,continue reading
Free reading is over, click to pay to read the rest ... pages
0 dollars，0 people have bought.
Reading is over. You can download the document and read it offline
0people have downloaded it