Scale up of treatment linked to rise in loss to follow-up, but those in care continue to do well

This article is more than 11 years old. Click here for more recent articles on this topic

Low death rates and high rates of viral load suppression have been sustained throughout the seven years of scale-up in a community-based antiretroviral treatment service in a poor area of Cape Town, South Africa.

However, the cumulative probabilities of loss to follow-up (LTFU) and virological failure increased over time suggesting a decreased capacity to support the long-term therapy needs of a growing caseload, according to Mweete D. Nglazi and colleagues in a prospective cohort study reported in the advance online edition of the Journal of Acquired Immune Deficiency Syndromes.

Over 5 million people were estimated to be getting antiretroviral treatment in low- and middle-income countries by the end of 2009, 3 million of whom are in sub-Saharan Africa with the majority in South Africa.


loss to follow up

In a research study, participants who drop out before the end of the study. In routine clinical care, patients who do not attend medical appointments and who cannot be contacted.

inter-quartile range

The spread of values, from the smallest to the largest. The inter-quartile range (IQR) only includes the middle 50% of values and measures the degree of spread of the most common values.

community setting

In the language of healthcare, something that happens in a “community setting” or in “the community” occurs outside of a hospital.


In discussions of consent for medical treatment, the ability of a person to make a decision for themselves and understand its implications. Young children, people who are unconscious and some people with mental health problems may lack capacity. In the context of health services, the staff and resources that are available for patient care.

prospective study

A type of longitudinal study in which people join the study and information is then collected on them for several weeks, months or years. 

Reports of the early successes of ART programmes in sub-Saharan Africa achieving good immunological, virological and clinical responses have been challenged by reports of high mortality rates and high losses to follow-up. This may be explained, in part, by patients diagnosed late and already seriously ill.

However, few studies have reported on the long-term outcomes of large cohorts in sub-Saharan Africa, in particular looking at how ever-increasing caseloads affect a programme’s ability to deliver and maintain quality of care over time.

Success of antiretroviral treatment depends on a team of health care workers and counsellors being able to meet often overlapping short- and long-term goals by delivering treatment effectively.

Critical short-term goals include rapid reduction of death rates by diagnosis and treatment of co-morbidities, provision of co-trimoxazole prophylaxis and getting the best clinical and virological responses to ART; long-term goals include keeping patients on treatment with good adherence and sustained suppression of viral load.

The authors undertook a prospective cohort study of treatment-naïve patients aged 15 years and over enrolled between September 2002 and September 2008 in a community-based antiretroviral treatment service in a poor area of Cape Town, South Africa. Follow-up data ended in September 2009.

By September 2008 3162 patients had started ART with a maximum annual enrolment of 783 patients in the 12 month period of 2005-2006. 67% were women and the median age was 34 years.

Median baseline CD4 cell counts changed significantly over successive enrolment periods from 87 cells (interquartile range (IQR): 45-145) in 2002-2004 to 121 cells (IQR:60-178) in 2007-2008. Median follow-up was 2.4 years (IQR: 1.2-3.8).

Ratios of patients to peer counsellors increased four-fold from 2002-2004 to 2007-2008; and doctor to patient ratios increased by close to 50% from 1:202 in 2002-2004 to 1:395 in 2007-2008.

After six years the cumulative probability of death and loss to follow-up was 37.4%, comparable to another programme in Cape Town.

Retention of over 60% of patients on treatment and in care after six years is a success, note the authors. However, with each successive year of enrolment the risk of being lost to the programme increased, raising concerns about the programme’s ability to maintain an acceptable standard of care.

The authors chose to look at the different losses separately as well as early and late virological responses to antiretroviral treatment to further understand this trend and how it affected quality of care.

The one year mortality rate of 7.9% in this cohort, with many at an advanced stage of illness, is very low for sub-Saharan Africa where rates vary between 8% and 26%. The authors stress that even as caseloads increased low mortality rates in the first year of treatment did not change over time.

Early viral suppression of <400 copies/mL at 16 weeks in 93% or more of patients also did not vary over time.  

Additionally the proportion of those with CD4 cell counts <200 cells/mm³, a subset recognised as having an ongoing high risk for early mortality, was looked at.  From 90% at baseline, this proportion decreased to approximately 20-30% after 48 weeks on antiretroviral treatment and did not vary significantly with successive years of starting ART.

The authors note that these results show that the initial care and treatment support in the first months on ART was maintained regardless of an increasing caseload.  

The authors note that one of the primary challenges after the first year of ART is to keep patients on treatment and in care.

While the cumulative probability of being lost to follow up after six years was 29.1% not dissimilar to another programme in Cape Town, the authors found that as elsewhere this proportion increased significantly with each successive year of enrolment.

This may be explained, the authors note, by an increasing caseload which brings with it increasing clinic waiting times, shorter consultations, reduced opportunities for counselling and adherence support as well as human resources further overstretched for patient tracing following missed appointments.

The implication that increasing clinic caseloads led to decreasing treatment support over time was supported by the rise of an increasing risk of virological failure along with the increasing rates of LTFU. 

At six years the probability of virological failure in the whole cohort was 23.1%, many of whom had drug resistant mutations. An increase in drug resistance leads to increased switching to protease-inhibitor second-line regimens. Regimens that are not only considerably more expensive but after which there are few if any other options.

The authors note that the South African population is highly mobile making long-term care especially difficult. They suggest more efficient referral and patient tracking systems are needed to provide quality uninterrupted care.

Strengths of the study include very complete data from a cohort of over 3000 with one of the longest follow-up periods in the context of a public sector ART programme in sub-Saharan Africa.

Limitations include, note the authors, an analysis of many end-points raising the possibility of false-positive findings.

They add that while increasing caseloads are suggestive of increased LTFU and virological failure over time they found no causal association.

The authors conclude that “successful early outcomes were sustained between sequential calendar years during seven years of scale-up. In contrast, the increasing cumulative probabilities of LTFU or virological failure may reflect decreasing capacity to adequately support patients during long-term therapy as clinic caseload increased.”


Nglazi, MD et al. Changes in programmatic outcomes during 7 years of scale-up at a community-based antiretroviral treatment service in South Africa. Advance online edition. J Acquir Immune Defic Syndr. 2010.