Plans to harmonise data collection and reporting of TB/HIV collaborative activities

WHO, UNAIDS, PEPFAR and the Global Fund have agreed to simplify and harmonise the data they ask of programmes regarding the implementation of collaborative HIV/TB activities, according to an announcement at the HIV Implementers’ Meeting in Kampala this June.

While this may not sound like as a much of a breakthrough as a new drug or treatment strategy, data collection and reporting is actually one of the biggest technical challenges on the ground in the delivery of services to reduce the burden of TB/HIV coinfection.

Consistently programmes have described data recording and reporting as a problem. It takes training and time to fill in multiple and sometimes redundant registers — which increases the workload of the healthcare delivery staff. As a result, programmes are reporting incomplete or inaccurate data, international partners are reporting conflicting figures, and national leadership doesn’t know what is or isn’t being done in their countries, or where they should be investing more effort and funding.

But the technical agencies have realised that they were adding to the problem by using different monitoring indicators for the same activities, and collecting the data from different partners on the ground rather than from the local ministries of health.

“We acknowledged that deficiency and have now harmonised monitoring and evaluation indicators for TB/HIV — which is already feeding the monitoring and evaluation revision of the Global Fund. But we really have to make sure national authorities, national administrators of funds, take leadership in owning this information and this data, because although there was anecdotal evidence that many HIV/TB activities are going on, the data was not being captured by the national authorities, not coming to WHO or helping to monitor the global, national and the local progress in implementation,” said Dr Haileyesus Getahun of WHO.

The challenge at the local level

But it is important that the word of these changes gets out to the country and implementing partner level, because at present there are a number of different recording practices and procedures going on which will have to be revised. And of course training will have to be revised and staff retrained.

It is important to keep in mind that “TB/HIV” variables were only added to programme monitoring in the last few years, and there are major problems in their implementation.

For example, Dr Sydney Kololo, working as technical officer in the national TB control programme of Botswana told the conference that although a few HIV variables have been included in the national electronic TB register in 2005, there is still a lot of missing data.

One crucial variable is the number of people with TB who have been tested for HIV. In Botswana, all TB patients are supposed to be tested; and the country reports good uptake of provider initiated testing and counselling overall. But in the TB register, although reporting went up between 2005 and 2007, data about whether the test was administered is still missing in over 40% of the patients.

“The ETR system is flexible enough to allow inclusion of new variables. Integration is slow, but improving; However, TB/HIV surveillance requires significant investment in time, resources and political commitment; and the introduction of new variables to a system needs more training and support to be implemented,” he said.

And these are just a few variables reported by the national TB programme in one country.

“But what we are having now, is TB/HIV data being produced by all these different organisations,” said Dr Christian Günneberg of the TB, HIV and Drug Resistance Unit at the Stop TB Department of WHO. The data now is coming in from different sources, and reported separately for different time periods.

“It is easy when you get one source of data that is your source of data and you do not have to worry about anything. When other people start producing data, it gets interesting.”

An example of how interesting? In 2005, PEPFAR reported that over 35,000 people with HIV were receiving TB treatment through its partner programmes. But WHO had less than 10,000 people reported as being coinfected in those countries. The next year, WHO’s figures went up, but there was still a gap of over 10,000 people.

Even within WHO, the HIV/AIDS department and Stop TB Department sometimes have different numbers for the number of TB/HIV coinfected, with the STOP TB Department getting its figures from the National TB Programme and the HIV/AIDS department getting those figures from the ART registers. And what is being reported by UNAIDS is about twice as much what is being reported to WHO

“What should be happening, theoretically, is for accurate harmonised TB/HIV data to come from the National Ministry of Health source to all partners, collected by all partners through the Ministry of Health and then sent on to WHO or whoever is collecting the data internationally. But this is not happening,” said Dr Günneberg.

Recommendations

To move forward, the key technical partners and funding agencies have now agreed that TB/HIV data should be comparable, consistent, accurate and comprehensive, based on countries’ Ministry of Health data sources and reporting cycles.

To achieve this, there will need to be an international-level regular review process, including WHO’s TB and HIV departments, UNAIDS, PEPFAR and the Global Fund. At the national level, there will have to be regular national TB/HIV review meetings, including the TB and HIV departments and implementing partners, and supported by technical donor partnership (PEPFAR etc).

The proposed ‘harmonised’ TB/HIV indicators

To decrease the burden of HIV in people with TB (in the TB register)

• Number of people with TB with documented HIV-serostatus (out of the total number of registered TB patients during the reporting period).
• Number of HIV-positive TB patients starting or continuing CPT during reporting period (out of total number of TB/HIV patients)
• Number of HIV-positive TB patients starting or continuing ART during reporting period (out of total number of TB/HIV patients)

To decrease the burden of TB in people with HIV patients (in the ART/pre-ART register)

• Number who had TB status assessed at last visit (out of the total number in HIV care and treatment)
• Number starting IPT within recording period (out of the total number of newly enrolled patients in whom active TB has been excluded and who are not on TB treatment)
• Number who started TB treatment (out of total number of people with HIV in HIV care and treatment during the reporting period)

Harmonising these indicators “should make things simpler within the country” said Dr Günneberg. “If you are only reporting for one monitoring and evaluation system, it means that you streamline recording and reporting, so you have less work for the health workers. And when you are comparing recording and reporting nationally and internationally, you have one language for comparison — and the indicator harmonisation will lead to the use of routine data to improve quality of services at clinic district and national level.”

But as the example from Botswana shows, agreement at the technical partner level may be the easy part. Revising existing national and partner systems, and changing current health worker practice on the ground will require additional training and support.