Secondary Logo

Journal Logo


HIV-1 disease progression and mortality before the introduction of highly active antiretroviral therapy in rural Uganda

Van der Paal, Lieve; Shafer, Leigh Anne; Todd, Jim; Mayanja, Billy N; Whitworth, Jimmy AG; Grosskurth, Heiner

Author Information
doi: 10.1097/01.aids.0000299407.52399.05
  • Free



Estimates of survival after HIV infection in Africa are essential for understanding the natural history of HIV-1 infection and for planning healthcare resources for those infected. Understanding the natural history of HIV is essential for patient management and counseling, and enables an assessment of the impact of interventions to care for and treat people with HIV, including the provision of antiretroviral therapy (ART).

Most of the data on survival and progression to AIDS in low and middle income countries have come from cohorts of prevalent HIV-1 infection [1] or from cohorts of specific groups, such as commercial sex workers [2–4], military conscripts [5], women [6], mine workers [7] and blood donors [8].

We have previously published data from this prospective population-based clinical cohort of HIV-infected rural Ugandan individuals [9,10]. Time to death and time to World Health Organization (WHO) stage 4 during 10 years of follow-up were 9.8 and 9.4 years, respectively.

The updated findings presented here cover the time from 1990 until the end of 2003 (after which HAART was introduced in this cohort), thus representing a substantially (3 years) longer observation period and allowing more precise estimates of times between seroconversion, symptomatic disease, the development of AIDS and death. In addition, we also compared progression from seroconversion to ART eligibility and from ART eligibility to death. With now 13 years of follow-up, this is one of the longest standing population-based cohorts to document the natural history of HIV infection in Africa, and one of the very few in which data are available from the time of infection. After the large-scale introduction of HAART, it is unlikely that data from similarly long observation times will become available again in the future. We also present data on the mortality of HIV seroconverters from the same community who were not enrolled in the clinical cohort, as mortality may differ for a number of reasons.


Selection of participants

Participants in the clinical cohort presented here were selected from a larger general population-based cohort in rural south-west Uganda. This larger cohort was established in 1989 to describe the dynamics of HIV-1 infection, and originally consisted of approximately 4500 adults in 15 villages [11]. In 1999, 10 more villages were added to the survey area, bringing the total number of adults to approximately 7000. Since the start of the study, annual house-to-house census surveys have been conducted among this population, followed within weeks by a serosurvey. Monthly birth and death reports are compiled by village-based recorders. This cohort has been described in detail elsewhere [12].

In 1990, one third of prevalent cases from the first survey round, randomly selected, were enrolled into the clinical cohort that forms the framework for the data presented in this paper. In addition, all seroconverters (13 years and older) identified during subsequent annual surveys were invited to enroll in the clinical cohort as HIV incident cases (Fig. 1). The estimated date of seroconversion was taken as the midpoint between the last negative and the first positive HIV test. Only individuals with an interval between the last negative and first positive HIV test of less than 4 years were invited into the cohort. All individuals were aged 15 years or older at enrollment, although some were between 13 and 15 years of age at the time of seroconversion. HIV-negative individuals were randomly selected from the population to match the age and sex of the HIV seroconverters and prevalent cases, in order to facilitate comparisons with the background morbidity and mortality in the study population.

Fig. 1:
Flow chart of enrollment and follow-up in Masaka cohort by HIV status.

Study procedures

Trained field workers visited the individuals identified for enrollment, explained the nature of the study and invited the participants to the study clinic for enrollment into the clinical cohort. At the clinic, a clinician explained the study and answered any questions. On enrollment, all participants gave informed consent (signed or by thumbprint).

Cohort participants were invited to attend the study clinic every 3 months for regular clinic visits, and for interim clinic visits whenever they were ill in between regular appointments. At the regular appointment, participants were seen by one of the two study clinicians, who administered a detailed medical and sexual history questionnaire and undertook a full physical examination. A blood specimen was routinely collected to monitor a variety of laboratory parameters. Since 1995, CD4 cell counts have been obtained every 6 months for HIV-positive and every 12 months for HIV-negative participants, using FACSCount (Becton Dickinson, San Jose, California, USA). Between 1992 and 1995, CD4 cell counts were sent occasionally to an external laboratory in Kampala and tested using flow cytometry. Any symptomatic disease was investigated and free treatment was provided. Referral for inpatient care was provided if indicated. Ugandan Ministry of Health guidelines were used for all treatments. Up to the end of 2003, cotrimoxazole or isoniazid prophylaxis and ART were not provided. At every routine visit, participants were staged according to the WHO staging system [13]. WHO stage 4 was used as the definition of AIDS in this analysis.

The vital status of participants who defaulted from the cohort was checked through family or neighbour reports. For reasons of confidentiality, clinic staff and field workers were unaware of participants' serological HIV status, unless the participant chose to reveal his or her status to the clinician or nurse. All participants and their partners were encouraged to use the free counseling and testing services available at the clinic or in their residential villages.

Statistical analysis

Frequencies and percentages were calculated for categorical factors describing the participants included in this analysis. For time periods, the median and interquartile range (IQR) were used. Person-years from the seroconversion date were computed, and Kaplan–Meier survival functions and life tables were used to describe the cumulative survival probabilities and 95% confidence intervals (CI) for the various endpoints (WHO stages 2, 3, AIDS, CD4 cell counts below 200 cells/μl, and death). To compare survival, a Cox regression model was used to obtain hazard ratios (HR) and 95% CI for univariate (unadjusted) analysis, and to compare groups adjusting for the effect of age at infection. Likelihood ratio tests were used to assess the significance of differences between groups. All analyses were performed using Stata 9.0 (Stata Corporation, College Station, Texas, USA).

For the analysis of time to WHO stages 2 and 3, only HIV incident cases enrolled within 2 years of the estimated date of seroconversion were included. For the analysis of time to WHO stage 4, CD4 cell counts below 200 cells/μl and death all HIV incident cases enrolled were included regardless of the time between estimated seroconversion and enrollment. All events after 31 December 2003 were excluded from the analysis. For all analyses, the endpoint was taken as the date first seen with the event at any clinic visit, or the date of censoring. For participants who did not experience the event, the date of the last clinic visit was taken as the censoring date, or 31 December 2003 if the clinic visit was after this date.

The time from seroconversion to death was studied in two ways. For the first analysis, we only included HIV incident participants enrolled in the clinical cohort. All the living incident cases were censored at the end of 2003. For the second analysis, we also included HIV incident individuals from the general population who never enrolled in the clinical cohort, in order to assess whether mortality in this group was different from those who enrolled. The date of death for individuals who never enrolled was obtained through the routine annual census and the monthly death reports in the general population as described above. The follow-up time of these individuals was also censored at the end of 2003 if they were still known to be alive.

Mortality rates for HIV incident participants of each sex were calculated. The data from the HIV negative individuals in the general population from which the clinical cohort had been recruited were directly standardized to the age and sex of the HIV-positive individuals at the time of seroconversion. The net mortality attributable to HIV infection was calculated from the lifetable approach of competing risks using the relationship:

All HIV incident cases with at least one CD4 cell count contributed to the analysis of time from seroconversion to CD4 cell count below 200 cells/μl. Participants who never reached a CD4 cell count below 200 cells/μl were censored at the last follow-up with a CD4 cell count available or at 31 December 2003 if the CD4 cell count was more than 200 cells/μl after that date.

All HIV-infected participants were included in the analysis from developing AIDS to death, except participants who presented with an AIDS-defining condition already at enrollment. Time was estimated from first being seen with AIDS to death; participants still alive were censored at the end of 2003. The same analysis was performed for the time from ART eligibility to death, using two definitions of eligibility: 1) having a WHO stage 4 event or a CD4 cell count less than 200 cells/μl (former WHO definition) and 2) a WHO stage 4 event, or a CD4 cell count less than 200 cells/μl, or a WHO stage3 event and a CD4 cell count less than 350 cells/μl (current WHO definition).

Ethical issues

Participants in the cohort were treated for opportunistic infections and referred to hospital for inpatient care if necessary. They were encouraged to know their HIV status, and free voluntary testing and counselling services were made available in the study villages by the project. ART was made available for all eligible patients from January 2004 onwards.

This study was approved by the Science and Ethics Committee of the Uganda Virus Research Institute and the Uganda National Council for Science and Technology.


Characteristics of enrolled participants

By 31 December 2003, 775 participants from the general population cohort had been invited to enroll in the clinical cohort and 605 (78%) of these had enrolled: 108 HIV prevalent and 240 incident (35 of whom were enrolled as HIV negative but seroconverted while in the cohort) cases, as well as 292 HIV-negative controls (including the 35 seroconverters; Fig. 1). The main reasons for not enrolling were refusal (N = 62, 36%), moving out of the study area (N = 55, 32%) and death before enrollment (N = 43, 25%). Table 1 gives the age distribution of all participants at the time of enrollment. The male incident cases were significantly older at enrollment than the female cases (P < 0.001, Mann–Whitney), reflecting the transmission pattern in this rural African community.

Table 1:
Clinical cohort population: numbers of participants in each age group, by gender and HIV status.

The characteristics of the HIV incident cases are shown in Table 2. The 240 incident cases had a median time of 12.7 months (IQR 12, 24.5) between their last negative and first positive HIV test, reflecting the fact that most incident cases were identified through consecutive annual survey rounds in the general population cohort. The date of seroconversion was estimated with similar accuracy for men (median time between last negative and first positive test 12.7 months) and women (median time 12.8 months). The median period between the estimated date of seroconversion and enrollment into the cohort was 11.6 months (IQR 8.1, 21.9). The median follow-up from seroconversion to death or censoring at 31 December 2003 was 5.1 years (IQR 3.1, 8.4).

Table 2:
Characteristics of HIV incident cases on enrollment, progression time to World Health Organization stages 2 and 3 and criteria observed that defined progression.

Follow-up in the cohort was high, with 80.5% of scheduled visits attended.

World Health Organization stages 2 and 3

Of the 240 incident cases, 201 were seen in the clinical cohort within 2 years of their estimated date of seroconversion, and this subgroup was used to estimate progression times to WHO stages 2 and 3. These cases had a similar age distribution to the whole group (median age at seroconversion 28.7 years, compared with 28.2 for the whole group). Up to 31 December 2003, 123 participants progressed to WHO stage 2 or higher, and 111 participants progressed to WHO stages 3 or 4. The median time to first symptoms (WHO stage 2 or higher) was 25.8 months (95% CI 22.8–31.3) and the median time to stage 3 was 39.0 months (95% CI 29.2–50.4). There were no sex differences for progression to stage 2 or stage 3 (HR 1.10, P = 0.6 and HR 0.9, P = 0.6, respectively). The criteria used to diagnose progression to WHO stages 2 and 3 are given in Table 2.

AIDS (World Health Organization stage 4) or death

All 239 incident cases who had an initial WHO stage of 3 or less were used to calculate the time to AIDS (WHO stage 4) or death (Fig. 2a). By the 31 December 2003, 65 had progressed to AIDS, 35 died without having a documented AIDS-defining event, and 139 were censored because they had not died or progressed to WHO stage 4. The median time from seroconversion to AIDS or death was 7.1 years (95% CI 5.9–8.5). Compared with those infected at 13–24 years of age, there was significantly (P < 0.001) faster progression to AIDS or death with increased hazard in all older age groups: 25–34 years (HR 1.78, 95% CI 1.04–3.07), 35–44 years (HR 4.27, 95% CI 2.31–7.89) and 45 years or more (HR 5.55, 95% CI 2.82–10.95). After adjusting for age at seroconversion, women had a slightly faster progression to AIDS or death than men (HR 1.48, 95% CI 0.95–2.30, P = 0.09). The most frequently diagnosed conditions for the first stage 4 event were wasting syndrome (36%), candidiasis of the oesophagus (30%), chronic herpes simplex virus infection (14%), extrapulmonary cryptococcal disease (3%), Cryptosporidium diarrhoea (6%), Kaposi sarcoma (8%), non-typhoid Salmonella septicaemia (7%), extrapulmonary tuberculosis (8%) and HIV encephalopathy (2%).

Fig. 2:
Cumulative probability of HIV events shown using Kaplan–Meier curves with 95% confidence intervals. (a) Kaplan–Meier survival curve for World Health Organization (WHO) stage 4 event or death from seroconversion in 239 subjects with initial WHO stage of 3 or less. (b) Kaplan–Meier survival curve for death in 240 seroconverters. (c) Kaplan–Meier survival curve for death in those eligible for antiretroviral therapy (ART) (WHO stage 4 event or CD4 cell count < 200 cells/μl or WHO stage 3 and CD4 cell count < 350 cells/μl).

Survival to antiretroviral therapy eligibility

Of the 240 incident cases, 229 were known to be not yet eligible for ART at the time of enrollment (excluded were one participant withWHO stage 4 at enrollment, two had no CD4 cell counts, and eight had a first CD4 cell count < 200 cells/μl). Of these, 79 (34%) progressed to a CD4 cell count below 200 cells/μl and 60 (26%) had a WHO stage 4 defining event during follow-up to give a total of 99 (43%) eligible for ART by 31 December 2003. A total of 130 were censored before becoming eligible for ART: 100 at 31 December 2003, 10 at the date of their last CD4 cell count or WHO stage assessment and 20 died without being seen to fulfill the ART eligibility criteria. The median time to a CD4 cell count less than 200 cells/μl or WHO stage 4 was 6.2 years (95% CI 5.0, 8.8). Age at infection was significantly associated with faster progression to ART eligibility: the HR for the 25–34 year age group was 1.37 (95% CI 0.83–2.26), for the 35–44 year age group it was 2.71 (95% CI 1.47–5.02) and for the over 45 year age group it was 3.55 (95% CI 1.83–6.89) compared with the youngest age group (P < 0.001). Adding the criteria of WHO stage 3 with a CD4 cell count below 350 cells/μl, 18 more participants (8%) were eligible for ART before 31 December 2003 and 37 (16%) had an earlier date of eligibility. The median time to ART eligibility according to these criteria was 5.1 years (95% CI 4.2–6.2; Fig. 3).

Fig. 3:
Time to antiretroviral therapy (ART) eligibility comparing the old [World Health Organization (WHO) stage 4 event or CD4 cell count < 200 cells/μl] and the new (WHO stage 4 event or CD4 cell count < 200 cells/μl or WHO stage 3 and CD4 cell count < 350 cells/μl) WHO criteria.
Fig. 1
New ART definition;
Fig. 1
old ART definition.


After seroconversion, during a total of 1387 person-years at risk, of which 1199 person-years were observed from the first positive HIV test, 84 incident cases died, giving a crude mortality rate of 70.0 per 1000 person-years. Among HIV-negative participants, 24 died during 1980 person-years, giving a crude mortality rate of 12.1 per 1000 person-years. For incident HIV cases the median survival from seroconversion to death was 9.0 years (95% CI 7.5, 10.6). Removing background mortality (as measured in the HIV-negative population) from the mortality experience of the HIV-positive individuals yielded an estimate for net mortality attributable to HIV that would correspond to a median survival of 10.2 years, an increase of 5% over the observed, gross median survival. Figure 2b shows the cumulative probability of death from seroconversion. Older age at seroconversion was a risk factor for faster progression to death, as demonstrated in Figure 4. The HR for age group 25–34 years was 1.70 (95% CI 0.98–2.93), 3.77 (95% CI 2.00–7.10) for age group 35–44 years, and 5.27 (95% CI 2.66–10.43) for age 45 years and more, compared with the age group 13–24 years (P < 0.001).

Fig. 4:
Cumulative probability of death in 240 seroconverters by age in 10-year age groups.
Fig. 1
Age 13–24 years;
Fig. 1
age 25–34 years;
Fig. 1
age 35–44 years;
Fig. 1
age 44+ years.

Sixty-six individuals who seroconverted before 31 December 2003 (33 men, 33 women, median age 36 years for men, 22 years for women) were invited to join the clinical cohort, but did not enroll for various reasons: 16 refused to join, 19 joined after 1 January 2004, 17 died and 14 moved out of the study area. Seventy-three adult seroconverters were identified in the general population cohort but were not invited into the clinical cohort as the period between the last negative and the first positive HIV test exceeded 4 years, and their vital status was obtained from the annual census in the general population. Figure 5 shows that the shorter time to death among the invited but not enrolled was largely the result of deaths in the first 2 years after seroconversion. After adjusting for age, compared with the enrolled group there was a significantly increased hazard in those not enrolled (HR 2.12, 95% CI 1.21–3.69, P = 0.008) and a non-significant lower hazard in those not invited (HR 0.84, 95% CI 0.42–1.64, P = 0.6).

Fig. 5:
Cumulative probability from seroconversion to death, in 240 seroconverters enrolled into the rural clinical cohort, 66 seroconverters invited but never enrolled into the rural clinical cohort, and 73 seroconverters identified from the general population, but not invited into the cohort because the period between the last HIV-negative test and the first HIV-positive test exceeded 4 years. (a) Unadjusted for age at infection; (b) adjusted to age 25–34 years at infection.
Fig. 1
Enrolled into rural clinical cohort;
Fig. 1
invited but not enrolled;
Fig. 1
not invited into rural clinical cohort.

Survival from AIDS or antiretroviral therapy eligibility

For the 117 (53 prevalent, 64 incident) cases who developed AIDS (WHO stage 4), the median survival from AIDS to death was 8.5 months (95% CI 4.7–12.0). Using the current WHO definition, 174 (117 incident and 57 prevalent cases) HIV-infected participants became eligible for ART (CD4 cell count < 200 cells/μl or WHO stage 4 event or WHO stage 3 and CD4 cell count < 350 cells/μl) before 31 December 2003. Of the 174 eligible for ART, 107 died and 67 were censored at the end of 2003 with the median time to death of 34.7 months (95% CI 27.8–46.0). Using the previous WHO definition (CD4 cell count < 200 cells/μl or WHO stage 4 event) 147 participants (48 prevalent, 99 incident) were eligible for ART before 31 December, and the median survival to death was 25.7 months (95% CI 21.4–30.5). Figure 2c shows the survival after ART eligibility using the current WHO recommendations.


We have provided an update of our results on the natural history of HIV infection in a rural African context. These current estimates take into account a longer follow-up (until 2003) since the last publication, which included data until 2000 [9]. HAART has been made available to participants of both the clinical and the general population cohort since January 2004, and these estimates are probably the best available evidence from the African continent. It is unlikely that similar studies will be conducted in the future, as it would be unethical to deny HAART to study participants.

Survival from seroconversion in our cohort before the introduction of HAART was similar to what has been described in high income countries before the widespread use of HAART (8.3 to approximately 13 years) [14–17], and we also found that older age at seroconversion was a risk factor for faster progression, as has been reported from a large multicentre study from industrialized countries [18].

Symptom-free survival in our cohort was substantially shorter than described in a cohort of blood donors in Abidjan [8]: 79.3% of the participants in that west African study had no symptoms after 3 years, whereas in our cohort only 42.7% had no symptoms after 3 years. The Abidjan cohort used the Centers for Disease Control classification, whereas we use the WHO classification, and 95% of the subjects in Abidjan were prescribed primary prophylaxis with trimethoprim–sulfamethoxazole (TMS). Although cohort participants were encouraged to know their HIV results through the village-based counseling offices, only approximately 25% of our cohort participants were aware of their serostatus by 2003, and TMS prophylaxis was only offered for secondary prophylaxis during the observation period as the official guidelines in Uganda did not recommend the use of TMS until recently. The median age at seroconversion in both cohorts was similar (29 years).

The median time to a CD4 cell count below 200 cells/μl or WHO stage 4 in our cohort was slightly lower than observed among Thai sex workers (6.2 years versus 6.9 years) [3]. Data from developed countries showed significant differences between geographical areas: the Tricontinental Seroconverter Study showed median times from seroconversion to a CD4 cell count less than 200 cells/μl varying between 7.1 and 10.2 years between different sites [17].

According to the new WHO criteria, our study participants were found to be eligible for ART more than one year earlier compared with the old WHO eligibility criteria. If this is representative of the situation in other low and middle income countries, more individuals could benefit from ART than reflected in the current estimates of ART need, and more resources will be required to address this need.

The survival time from ART eligibility (CD4 cell count below 200 cells/μl or WHO stage 4) in our cohort was over 2 years. This is more than double the time found in another Ugandan cohort of mostly symptomatic HIV patients, who had a median survival of 9 months with a CD4 cell count below 200 cells/μl [19]. The median age at enrollment in that predominantly female seroprevalent cohort was approximately 30 years. All patients were aware of their HIV status and most presented with symptoms suggestive of HIV infection. The survival time in our cohort is comparable with what was found in a cohort in Baltimore, USA, who reported survival times between 600 and 820 days with CD4 cell counts less than 200 cells/μl [20].

Cohorts from low and middle income countries have reported comparable survival times after seroconversion [3,5,7]. Those cohorts consisted of specific groups, e.g. sex workers, army recruits and miners, who tend to be younger than the general population. Five-year survival among Thai female sex workers was 85.9%, 82.3% among Thai army conscripts, and 89% among South African miners, compared with 80.5% in our general population-based cohort.

The survival time for HIV incident cases who were invited but never enrolled in the cohort was shorter than for those who did enroll. Almost 20% of those who never enrolled died within 2 years of seroconversion, indicating a possibly very rapid progression among these individuals: it is easy to miss these individuals in similar cohort-based studies, and this might introduce a bias, overestimating median survival and progression times. We therefore thought it important to present the survival analysis including individuals who never enrolled, although we lacked detailed clinical information about those individuals. From interviews with relatives postmortem, however, it seemed that most of those individuals were older and died from what could have been an AIDS-related illness. Only one of the deaths was caused by trauma. Individuals who were never invited into the cohort because they had an interval of more than 4 years between the last negative and the first positive HIV test had a non-significantly longer survival than those who were invited. This indicates that there is a risk of overestimating survival in population-based cohorts with longer periods between consecutive HIV tests.

As we only assessed CD4 cell counts every 6 months, participants could have crossed the thresholds of 200 and 350 cells/μl in the time interval between two measurements. We preferred not to make assumptions on when this occurred, as CD4 cell levels tend to be influenced by physiological and pathological events occurring in the same interval. Consequently, we might have overestimated the time from seroconversion to a defined CD4 cell threshold, and underestimated the time from a defined CD4 cell level to death. As we took 6-monthly measurements, however, we do not think this error is large. The same applies for WHO staging, which was only recorded at the regular visits every 3 months. Here again, we think the error is probably small, because of the frequency of the measurements.

The risk of becoming HIV infected may not be constant over the interval between the last negative and the first positive HIV test. Taking the midpoint of this interval as the estimated date of seroconversion could possibly cause a bias in the estimates of disease progression, especially if the interval is longer than 2 years. Approximately 75% of all incident cases had an interval of up to 2 years, and survival analysis restricted to these incident cases did not change the results (data not shown).

This updated analysis of disease progression and survival in a rural African cohort will provide useful information for modelling the HIV epidemic and public health planning. It will also serve as a background to assess the impact of treatment and care programmes in sub-Saharan Africa.


The authors would like to thank all the participants, the clinic and laboratory staff as well as staff involved in data entry and management. They are grateful for comments on an earlier draft by Shabbar Jaffar. The authors would also like to thank the anonymous reviewers for useful comments on the paper.

Sponsorship: This work was supported by a grant from the Medical Research Council, UK.

Conflicts of interest: None.


1. Crampin AC, Floyd S, Glynn JR, Sibande F, Mulawa D, Nyondo A, et al. Long-term follow-up of HIV-positive and HIV-negative individuals in rural Malawi. AIDS 2002; 16:1545–1550.
2. Anzala OA, Nagelkerke NJ, Bwayo JJ, Holton D, Moses S, Ngugi EN, et al. Rapid progression to disease in African sex workers with human immunodeficiency virus type 1 infection. J Infect Dis 1995; 171:686–689.
3. Kilmarx PH, Limpakarnjanarat K, Saisorn S, Mock PA, Mastro TD. High mortality among women with HIV-1 infection in Thailand. Lancet 2000; 356:770–771.
4. Lavreys L, Baeten JM, Chohan V, McClelland RS, Hassan WM, Richardson BA, et al. Higher set point plasma viral load and more-severe acute HIV type 1 (HIV-1) illness predict mortality among high-risk HIV-1-infected African women. Clin Infect Dis 2006; 42:1333–1339.
5. Rangsin R, Chiu J, Khamboonruang C, Sirisopana N, Eiumtrakul S, Brown AE, et al. The natural history of HIV-1 infection in young Thai men after seroconversion. J Acquir Immune Defic Syndr 2004; 36:622–629.
6. Leroy V, Msellati P, Lepage P, Batungwanayo J, Hitimana DG, Taelman H, et al. Four years of natural history of HIV-1 infection in african women: a prospective cohort study in Kigali (Rwanda), 1988–1993. J Acquir Immune Defic Syndr Hum Retrovirol 1995; 9:415–421.
7. Glynn JR, Sonnenberg P, Nelson G, Bester A, Shearer S, Murray J. Survival from HIV-1 seroconversion in Southern Africa: a retrospective cohort study in nearly 2000 gold-miners over 10 years of follow-up. AIDS 2007; 21:625–632.
8. Salamon R, Marimoutou C, Ekra D, Minga A, Nerrienet E, Huet C, et al. Clinical and biological evolution of HIV-1 seroconverters in Abidjan, Cote d'Ivoire, 1997–2000. J Acquir Immune Defic Syndr 2002; 29:149–157.
9. Morgan D, Mahe C, Mayanja B, Okongo JM, Lubega R, Whitworth JA. HIV-1 infection in rural Africa: is there a difference in median time to AIDS and survival compared with that in industrialized countries? AIDS 2002; 16:597–603.
10. Morgan D, Mahe C, Mayanja B, Whitworth JA. Progression to symptomatic disease in people infected with HIV-1 in rural Uganda: prospective cohort study. BMJ 2002; 324:193–196.
11. Mbulaiteye SM, Mahe C, Whitworth JA, Ruberantwari A, Nakiyingi JS, Ojwiya A, Kamali A. Declining HIV-1 incidence and associated prevalence over 10 years in a rural population in south-west Uganda: a cohort study. Lancet 2002; 360:41–46.
12. Mulder DW, Nunn AJ, Wagner HU, Kamali A, Kengeya-Kayondo JF. HIV-1 incidence and HIV-1-associated mortality in a rural Ugandan population cohort. AIDS 1994; 8:87–92.
13. World Health Organization. Acquired immune deficiency syndrome (AIDS): interim proposal for WHO staging for HIV-1 infection and disease. Wkly Epidemiol Rec 1990; 65:221–228.
14. Multicohort Analysis Project Workshop. Part I. Immunologic markers of AIDS progression: consistency across five HIV-infected cohorts. AIDS 1994; 8:911–921.
15. UK Register of HIV Seroconverters Steering Committee. The AIDS incubation period in the UK estimated from a national register of HIV seroconverters. AIDS 1998; 12:659–667.
16. Koblin BA, van Benthem BH, Buchbinder SP, Ren L, Vittinghoff E, Stevens CE, et al. Long-term survival after infection with human immunodeficiency virus type 1 (HIV-1) among homosexual men in hepatitis B vaccine trial cohorts in Amsterdam, New York City, and San Francisco, 1978–1995. Am J Epidemiol 1999; 150:1026–1030.
17. Veugelers PJ, Page KA, Tindall B, Schechter MT, Moss AR, Winkelstein WW Jr, et al. Determinants of HIV disease progression among homosexual men registered in the Tricontinental Seroconverter Study. Am J Epidemiol 1994; 140:747–758.
18. Collaborative Group on AIDS Incubation and HIV Survival including the CASCADE EU Concerted Action. Time from HIV-1 seroconversion to AIDS and death before widespread use of highly–active antiretroviral therapy: a collaborative re-analysis. Concerted Action on SeroConversion to AIDS and Death in Europe. Lancet 2000; 355:1131–1137.
19. French N, Mujugira A, Nakiyingi J, Mulder D, Janoff EN, Gilks CF. Immunologic and clinical stages in HIV-1-infected Ugandan adults are comparable and provide no evidence of rapid progression but poor survival with advanced disease. J Acquir Immune Defic Syndr 1999; 22:509–516.
20. Veugelers PJ, Schechter MT, Tindall B, Moss AR, Page KA, Craib KJ, et al. Differences in time from HIV seroconversion to CD4+ lymphocyte end-points and AIDS in cohorts of homosexual men. AIDS 1993; 7:1325–1329.

AIDS; CD4 cell count; HIV-1; rural Africa; survival

© 2007 Lippincott Williams & Wilkins, Inc.