Cotrimoxazole prophylaxis is cost-effective for people starting ART in sub-Saharan Africa

Carole Leach-Lemens
Published: 10 February 2012

Achieving full coverage of cotrimoxazole prophylaxis during the first six months of antiretroviral therapy would be a highly cost-effective way of reducing early death among those with advanced HIV infection in sub-Saharan Africa, researchers report in the advance online edition of the Journal of Acquired Immune Deficiency Syndromes.

The researchers developed a decision-analytic model from a health care perspective to compare costs and outcomes. Full cotrimoxazole prophylaxis coverage at an estimated additional cost of $3.29 for each person on ART prevented an additional 22 deaths compared to the base-case scenario (from 94 to 72 deaths per 1000 patients) at a cost of $146.91 for each death prevented over the first six months.

Potential cost savings for specific opportunistic infections (OIs) prevented by cotrimoxazole prophylaxis were also calculated.

Prevention of 45 new malaria episodes per 1000 persons treated would save between $69.95 and $203.32 per case averted, while prevention of 22 severe bacterial infections per 1000 persons would save between $68.62 and $126.71 per case averted. Prevention of four new cases of pneumocystis pneumonia would save between $75.69 and $88.41per case averted.

An intervention is considered very cost-effective by the World Health Organization if the incremental cost per life-year saved is no greater than the GDP per capita; in the case of the poorest countries in Africa this was calculated at $1695 in 2005. This analysis is not strictly comparable because it calculates cost savings in deaths averted.

Over the past decade the increasing availability and access to ART in resource-poor settings has resulted in reductions in AIDS-related deaths.

Yet, in sub-Saharan Africa people continue to present for care at an advanced stage of illness resulting in high rates (8-20%) of early death after starting ART compared to North America and Europe. Common causes of death include tuberculosis, pneumonia and diarrhoeal illnesses.

In North America and Europe it is common practice to give cotrimoxazole prophylaxis to those who present for care with advanced HIV, primarily to prevent PCP. Its use in African settings, however, appears to protect against a wider range of infections and is not restricted to those with advanced HIV.

Recent studies in sub-Saharan Africa, while not randomised, have shown a consistent reduction in death where people on ART got cotrimoxazole compared to no cotrimoxazole, note the authors. In particular cotrimoxazole has been shown to reduce the risk of tuberculosis and of malaria in people taking antiretroviral therapy. A meta-analysis of seven studies shows that cotrimoxazole prophylaxis reduced the death rate in people taking antiretroviral therapy by almost 60%.

CTX, a relatively cheap antibiotic, has been shown to be cost-effective and cost-saving among those not on ART, and cost-effective when provided to all HIV-infected adults in a treatment setting where ART was available for those with advanced illness.

The authors noted there have been no cost-effectiveness analyses of cotrimoxazole provision and starting ART. So they chose to look specifically at the effect of cotrimoxazole on reducing death rates among those with advanced HIV illness in resource-poor settings during the first six months on ART.

Six months was chosen because studies have shown the risk of death to be the greatest during this time. Additionally the DART trial showed the effect of cotrimoxazole on reducing death might lessen over time with no further benefit after 72 weeks of ART.

Because cotrimoxazole is already provided in a number of settings the authors chose to estimate the current average rates of cotrimoxazole administration and compare it with close to full coverage (65% and 97%, respectively) to look at the potential benefits of increasing cotrimoxazole coverage in reducing deaths and OIs. Those allergic to or intolerant of cotrimoxazole were excluded.

So the decision-analytic model was developed to look at the additional cost, deaths and cases of OI prevented and the incremental cost-effectiveness ratio (ICER). 

The ICER was estimated as the difference in health care utilisation costs at base-case and full coverage scenarios divided by the difference in outcomes.

Health care costs (for people getting ART) included routine outpatient visits and drugs (ARVs and cotrimoxazole). Total cost was estimated as the total drug costs and six months outpatient care assuming one routine visit.

The authors note the difficulty in determining cost-effectiveness when using death as an outcome measure. So their findings are not applicable to a willingness to pay (WTP) context. That is, the maximum amount of monies a person is willing to pay to achieve the desired result.

However, these findings are comparable with studies looking at life-years saved/gained as the outcome measure to show the effectiveness of cotrimoxazole for people with HIV. Studies showing cotrimoxazole cost-effective among adults with HIV had ICERs ranging from $150 to $1180 for each year of life gained compared to no cotrimoxazole.

Generic cost information for ARVs and OI drugs were used since median costs for such goods are traded on the international market so costs would be consistent across settings, note the authors.

Outpatient and inpatient costs were more difficult to generalise so the authors chose region-specific costs to reflect a variety of low-income settings in sub-Saharan Africa; so limiting the findings to comparable settings.

Other limitations include the absence of randomisation among the studies that informed the assumption about death rates and use of cotrimoxazole, the authors add. Ethically a randomised trial is highly unlikely. However, they add the reductions in death were fairly consistent but may not be generalizable to other settings where rates of OIs may differ.

The authors did not look specifically at a common issue in many HIV treatment settings of drugs meant for people with HIV being used for other conditions. This results in frequent stock-outs of cotrimoxazole so increasing the costs and cost-effectiveness.

However, in further analysis the authors doubled the base-case drug costs and the resulting favourable ICER meant that a considerable supply of cotrimoxazole could be used for other conditions but would still be cost-effective for ART patients.

They conclude “if [high cotrimoxazole coverage] is achieved, expansion of access to cotrimoxazole during ART initiation could have an important impact on the effectiveness of on-going HIV initiatives.”


Abimbola TO and Marston BJ. The cost-effectiveness of cotrimoxazole in people with advanced HIV infection initiating antiretroviral therapy in sub-Saharan Africa.  JAIDS advance online edition, doi: 10.1097/QAI.0b013e31282478dc0, 2012.

Community Consensus Statement on Access to HIV Treatment and its Use for Prevention

Together, we can make it happen

We can end HIV soon if people have equal access to HIV drugs as treatment and as PrEP, and have free choice over whether to take them.

Launched today, the Community Consensus Statement is a basic set of principles aimed at making sure that happens.

The Community Consensus Statement is a joint initiative of AVAC, EATG, MSMGF, GNP+, HIV i-Base, the International HIV/AIDS Alliance, ITPC and NAM/aidsmap