Book traversal links for 2.7 Monitoring, evaluating and modifying the programme
A monitoring and evaluation plan should be part of any screening programme. General conditions and risk group-specific conditions for discontinuing screening should be established from the outset – for example, in relation to yield, contribution to overall case detection and improvement in treatment enrolment and outcomes, cost per case detected or some combination of these. Indicators should be chosen, and digital forms created for collecting data or adapted to the specific objectives and local conditions. To monitor yield and the NNS in each targeted risk group, an appropriate information system should be developed to generate data about the number of people diagnosed with TB in relation to the number of people approached, screened and tested. This information should be assessed periodically and the mix of approaches adjusted appropriately.
The general epidemiology of TB, the importance of different risk groups and the epidemiology of TB in each group may change over time, and prioritization for screening will have to be adapted accordingly. As some members of a particular risk population will eventually find their way to diagnosis through the patient-initiated pathway if they are not screened, it is of interest to evaluate the impact of screening a particular group on overall additional notifications in a larger basic management unit or group of basic management units. This will require analysis of notification trends, preferably with comparisons to control areas.
It is also important to measure whether screening is simply concentrating case-finding in a few facilities, which may be the case if a specific intervention is seen as beneficial and information about it spreads through the community. This can result in increased notification in one area and decreased notification in another.
As one objective of screening is early detection, it may be useful to measure delays in diagnosis and treatment, which will require special surveys. Treatment outcomes and mortality rates among detected TB patients can, however, be captured and evaluated more easily.