Page 45 - Epidemiological studies on tuberculosis control and respiratory viruses

some time and efforts before such biomarkers can be used in a diagnostic strategy on a large scale. Short-term efforts should therefore include the collection of new evidence before scaling up LTBI screening and preventive treatment among contacts of PTB patients.
The introduction of the IGRA is unlikely to have major influence on the predictive value of
contact investigations for identifying recent transmission of M. tuberculosis. Therefore,
at the moment, the clinical benefit of a diagnostic test in contact investigations is
not so much dependent on the diagnostic accuracy of the test itself, but has to rely
on the ability of contact tracing strategies to identify individuals infected through a
recent transmission event, and thus most likely to benefit from preventive treatment.
Limited resources and the urgency of a contact investigation require prioritization of
contact tracing among infectious TB patients by weighting factors associated with
increased risk for recent TB infection or disease. However, previous studies have shown
that lack of compliance to contact investigation guidelines is not uncommon, and
that nonadherence could be the result of ambiguous recommendations, competence
gaps of public health nurses, and index case-related or contact-related obstacles (24)
(25) (26). Existing screening strategies might be improved if prediction models for evidence-based decision making would be used during contact investigations. Such
models, using known risk factors of recent transmission, can provide estimates of
the probability of a recent transmission event and inform decisions about preventive
treatment (27). The utilization of such models to predict disease transmission and
to aid prioritization of contact investigation is already used for selective screening of
case-finding activities for sexual transmitted infections (28). Furthermore, Mamiya
and colleagues have shown that it is feasible to use prediction models to estimate
the probability of a newly diagnosed TB case being involved in a recent transmission
chain, which can be a valuable tool in public health practice (27). These type of models
can also be used to predict disease transmission among contacts of TB patients, 7 provided that sufficient clinical outcomes of contacts are routinely captured in
electronic surveillance records. For example, Chan and colleagues have shown that routinely collected data in contact investigations by public health nurses can be integrated into a predictive risk score and can help to prioritize active case finding or preventive treatment among children exposed to TB (29). Such efforts can make a significant contribution to current screening strategies, by identifying an increased proportion of contacts that will progress to TB in the absence of preventive treatment, which will improve the impact of preventive treatment in contact investigations.
General Discussion
125