More recently, some information that could affect TB
programmes and create new roles for nurses and lay providers was produced by the
ZAMSTAR study, a very large trial that looked at whether two different case
finding approaches in communities (or a combination of the two), could increase
early TB diagnoses — and whether either would have a greater impact on a
community’s TB burden than passive case finding. The preliminary results were reported
at the 42nd Union World Conference on Lung Health held last October in Lille, France.
Over the better part of a decade it took to plan, prepare
for and perform the ZAMSTAR study, it seemed like it was much more than a
clinical study— it was more like a cultural phenomenon, particularly in Zambia,
where the study mostly took place. It was huge, had an innovative design, and
one of the intervention approaches was very novel — requiring a process of
community selection, preparation, mass community mobilisation, training and
participation, on a scale never seen before in sub-Saharan Africa.
Many of the
most enduring lessons from ZAMSTAR may come out of this process, and the
lessons learned about bringing together so many diverse people with a stake in
improving TB case management, from elementary school students to expert
patients, to nurses and newspaper reporters, to government ministers and
brilliant researchers to the multi-billionaire visionary who paid for it all.
Somehow, they pulled off a rigorous study which reached the clear conclusion
that implementing one of the active TB case finding interventions in a
community could, after several years, reduce its overall burden of TB.
That’s right— the
community as a whole did better. Unlike most other studies, ZAMSTAR did not
randomly assign individuals to different care or treatment arms. Instead, it
randomised entire communities to one of four arms, each with a different TB
One of the arms was supposed to serve as a control — where
the community received TB case finding in just the same way as they always had:
through passive case finding at the clinic. Only, the researchers pointed out
that this wasn’t entirely true — as part of the preparation for the study, all
the communities had significant upgrades to their local health system, with
additional staff and training to recognise and manage TB appropriately. This
included rapid updates to current policy when new recommendations emerged, like
the recommendation to perform intensified case finding in all people living with HIV). And, very
importantly, they also got better laboratories to provide faster and more
reliable diagnoses than before. Every community got these upgrades, but each
community knew whether it was implementing one or both of the new interventions
or in the case of the control arm, neither.
So after strengthening the capacity of communities to
diagnose and manage TB, ZAMSTAR randomised communities in Zambia and the
Western Cape province of South Africa into four arms, no intervention versus one or both
of two different interventions to increase the early detection of TB cases. The
investigators then evaluated whether there was less transmission of TB in the
household or a lower prevalence of TB in the community.
One of the active case finding arms included an intensive
form of ‘household contact tracing’ which involved sending counsellors to the
homes of TB patients to perform contact tracing, screening household members
for TB, and offering home-based HIV counselling and testing (HCT), with repeat
visits to deliver TB results, educate the household about TB, and to make
certain that anyone in the household with TB or HIV was effectively linked to
treatment, care and support.
In communities where household contact tracing took place, TB prevalence was reduced by 22%.
Household contact tracing is nothing new — it
has been an established and recommended TB control practice for some time — but
one that has been rarely implemented in resource-limited settings. It has just
been put on the stack of things that clinics and staff never got around to
organising. The intervention included HIV testing, and involved repeated visits
with more counselling and support than usual, but nothing really revolutionary
— simply a slight update to, and consistent performance of, good TB control practice
that programmes ought to have been doing all along.
The other intervention was called ‘enhanced case finding’ (ECF) which involved
a number of activities to make a TB diagnosis easier to access and encourage
people with symptoms to seek out diagnosis. In this case, sputum collection
centres were set up within the community where people could collect or deliver
their sputum specimens without having to go to the clinic.
The community was inundated with advocacy, communication and
social mobilisation activities (community drama, radio) to educate children and
adults about TB and the need to get a diagnosis. The improved labs were supposed to turn their results around very
quickly — within 48 hours — but had trouble keeping up with the number of the
specimens at some sites.
The ECF intervention with all the various community
activities was of course the one that everyone got excited about, and that
communities wanted to continue after the study was over. But it didn’t seem to
do any better than the control arm — it even did a little bit worse.
On the other hand communities assigned to implementing the household
contact tracing intervention saw a 22% reduction in the prevalence of
culture-positive TB. This was statistically significant — and one of the first
times a study has found a TB intervention to have an effect across the entire population of a community in the era of HIV.
The household contact tracing intervention arm also appeared
to reduce TB transmission (as measured by tuberculin skin test conversions in
children) by 55%, but this was not quite statistically significant (the limited
number of communities made it hard to prove this but it would seem to be
consistent with the strategy's effect on TB burden).
Enhanced case finding identified up to a quarter of the TB
cases in the communities randomised to the intervention, but as implemented in
this study at least, did not lead to a reduced TB prevalence when compared to
the communities that did not receive the intervention. Maybe it would have done
better with newer, faster and more sensitive lab technology such as the Xpert MTB/RIF test now being introduced in some countries. Even improved
fluorescent microscopy can be insensitive for TB especially in people living with
HIV. It is possible that people who handed in their sputum at a collection
centre could have interpreted a negative result as meaning that they didn’t
have TB — when maybe they did. If this resulted in a delay going to the clinic,
the approach might have backfired. If so, this underscores how very important
it is for clinical teams and health educators to better communicate and support
people with symptoms of TB in the quest for a diagnosis for smear-negative TB,
until the cause of their symptoms is either determined or their symptoms