Getting practical: learning to work where there is no data

Post by Jamie Uhrig

A hostess café, Ethiopia. Image, (c) J.Uhrig

Implementing HIV prevention activities with sex workers in Ethiopia means learning to work where there is no data.

The HIV epidemics in Ethiopia are unique to this multi-ethnic federal state. Three Demographic and Health Surveys have been conducted since the beginning of the millennium. HIV testing was performed as part of the survey process so that a broad outline of the epidemics can be obtained. Adult seroprevalence among those aged fifteen to forty-nine in the country in 2011 was 1.5%. Prevalence is higher in women than men, higher in urban areas than rural, and highest among those who fall into the highest wealth bracket.

What about HIV among key affected populations? Here there is no data. There are no reports of people injecting drugs in the country and there is a deafening silence on the topic of men who have sex with men.

But it is common to talk about sex work. Sex work itself is not illegal in Ethiopia and the harassment of sex workers by police that occurs in former British colonies in other countries in the region is almost never reported by sex workers here. It is possible for HIV prevention programme implementers to work closely and openly with sex workers with the full support of local authorities. This makes the remarkable lack of data even more astounding.

The largest programme for sex workers in the country is currently being implemented by the local nongovernmental organisation Timret Le Hiwot. Outreach workers talk with sex workers during recruitment for regular training sessions on sexual health and over ten thousand women attend two-day trainings each year. Staff also listen to women during traditional Ethiopian ‘coffee ceremonies’ when female sex workers have a chance to talk with their peers about whatever is on their minds: violence, condom sizes, boyfriends… One thousand coffee ceremonies were held last year. That is a lot of peer support.

There has been no sentinel surveillance among sex workers performed in a long time. The Ethiopia HIV/AIDS Prevention and Control office and the World Bank noted five years ago that “routine and detailed surveillance of this high-risk population has not been carried out for fifteen years.” They also pulled no punches in writing at that time: “Clearly, new studies need to be initiated to monitor and measure the progress of the epidemic amongst this vulnerable population.”

HIV rates among female sex workers who undergo voluntary counselling and testing can be used to measure effectiveness and efficiency of HIV testing programmes but are not useful in determining prevalence.  There is a tantalizing piece of information that can be obtained by reading between the lines of the last Demographic and Health Survey. Among women who had ten or more lifetime partners, HIV seroprevalence was 23%. Included in this group are some women who practise sex work – but we don’t know what proportion.

If HIV prevalence and incidence cannot be calculated, then what about measuring behaviour trends? The funder of Timret Le Hiwot, DKT Ethiopia, regularly tracks changes in behaviour using Behaviour Change Impact Surveys among female sex workers. Consistent condom use rates with clients are regularly reported by sex workers to be over 90%. This figure has not changed in ten years. And yet, ironically this consistently high rate of reported condom use makes measuring significant changes in behaviour difficult.

What about coverage of HIV prevention programmes for sex workers? To determine coverage you need to have a population estimate. There has never been a scientific national population size estimate of sex workers in the country. UNAIDS/WHO-recommended “capture/recapture” methods have been used to make estimates in Addis Ababa and many of the largest cities in the country. Vigorous and acrimonious debate has followed. It took three years for a population estimate in the Kenyan capital Nairobi to be published in a peer-reviewed journal. Do not expect to see an estimate for Addis Ababa in public soon.

There is good news. The Ethiopian Health and Nutrition Research Institute is working with the US Centers for Disease control to conduct HIV sentinel surveillance among sex workers. Data is being collected as this piece is being written. Still, even when the report is published, we will be left thinking: “Is HIV prevalence going up or down among sex workers? And what about incidence?”

Even one HIV sentinel surveillance report can inform programming immediately. If prevalence is found to be higher in some places in the country, more programme resources can be devoted to these areas. Until prevalence rates are known, all that can be done is to assume that there is equal prevalence among all sex workers and ensure maximal reach of programmes in as many cities in the country as resources allow.



Jamie Uhrig is an independent consultant who provides technical assistance to Timret Le Hiwot and DKT Ethiopia in providing high quality HIV prevention services for female sex workers throughout Ethiopia.

He moderates the English language section of the website HIV Information for Myanmar [him] and can also be followed on twitter @himmoderator.


2 thoughts on “Getting practical: learning to work where there is no data

  1. Preliminary results of the HIV sentinel surveillance survey of female sex workers have been informally released. Aggregate HIV prevalence among over four thousand sex workers is 24.7%. This excludes Addis Ababa as results have not yet been released for the capital city.

    Jamie Uhrig

  2. Addis figure just released – 17%.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s