Introduction
Morbidity induced by infection with the major soil transmitted infections (STH—Ascaris
lumbricoides, Trichuris trichiura, and hookworms) results in an estimated 5.19 million
disability-adjusted life years (DALYs) [1]. The World Health Organization’s (WHO)
policy for control centres on three groups, preschool aged children (pre-SAC), school-aged
children (SAC), and women of child bearing age, on the basis that heavy infection
in these groups will have a detrimental impact on anaemia, child growth, and development.
The current WHO guidelines focus on school-aged children, both for monitoring infection
and as a target for treatment, although treatment of pre-SAC and women of childbearing
age is also recommended where sustainable delivery mechanisms exist, especially in
areas of intense transmission [2,3]. The guidelines recommend treating SAC annually
where any STH prevalence falls between 20% and 50% and twice a year where it exceeds
50% [3].
The London Declaration on Neglected Tropical Diseases in 2012 endorsed WHO goals to
scale up mass drug administration (MDA) for STH, so that by 2020, 75% of the pre-SAC
and SAC in need will be treated regularly [4]. Building on an existing roadmap, WHO
announced an intention to meet the target [2,5,6]. Progress has been good in some
areas, but less so in others. In 2012, global coverage of those in need was 37% for
SAC and 29% for pre-SAC [5]. Data for the more recent years is as yet to be published
by WHO [5], but a huge gain in coverage is not expected, despite increased drug donations
from the pharmaceutical companies who manufacture the main anthelmintics. This is
due in part to the logistical challenges in getting even donated drugs to these populations,
who are often beyond “the end of the road.” At present, many countries with endemic
STH infections are not availing themselves of the freely donated drugs to treat children.
We are still a long way from the 2020 target of 75%. Even if this target is reached,
will it be enough to eliminate transmission and the disease arising from heavy infections
with STH? If not, how should the guidelines be changed to push towards morbidity control,
and ideally, the eventual elimination of transmission?
Basing Policy on Quantitative Calculations
To answer these questions, calculations are required to assess the impact of MDA targeted
at particular age groupings, especially pre-SAC and SAC, on overall transmission in
communities with differing levels of infection exposure. In many areas of infectious
disease epidemiology and the design of interventions, the impact of control is today
assessed by simulations based on mathematical models using parameter estimates from
epidemiological studies (e.g., HIV and Plasmodium falciparum [7,8]). The neglected
tropical disease (NTD) field lags behind, in the sense of largely basing target treatment
levels on discussion and consensus, without detailed calculations. Much of the basic
framework for the study of the transmission dynamics of helminth infections was laid
down in the 1960s and 1980s [9,10]). Rather little has been achieved since that time
in model development and parameter estimation. Concomitantly, little use has been
made of the insights gained from these analyses in the design of public health policy
for the control of STH and schistosome infections. The tools now available can rectify
this shortcoming, and they can easily be adapted to include costs and benefits as
outlined in this article. They should be used to refine WHO policy on treatment to
aim for a robust framework that will eliminate transmission.
Demography and Epidemiology
A key issue concerns demography in sub-Saharan Africa and other areas with endemic
infection. Typically, there are as many adults as there are pre-SAC and SAC [11].
This implies that, depending on the distribution of worms across the age classes,
the adults themselves may be able to sustain transmission within a community, even
when a very high fraction of the children are treated effectively. This is especially
the case for hookworm where adults typically harbour the majority of worms (in some
areas, more than 80% of the total population of parasites, (Fig 1—see inset in graph
C)), but also applies for Ascaris and Trichuris, where up to 30% of worms and egg
counts are in those older than 15 years of age [11], irrespective of the intensity
of transmission in a defined area.
10.1371/journal.pntd.0003897.g001
Fig 1
Cross-sectional surveys of the mean intensity of infection in different age groupings
for A. lumbricoides (A), T. trichiura (B), and hookworm (C) based on worm expulsion
studies.
These are typical age intensity profiles for the three most important STH species,
where the colours denote the age groupings: pre-SAC (blue), SAC (red), and adults
(orange). Data for A from [25], data for B [26], and data for C [27]. The inset in
Graph C represents five worm expulsion studies of Necator americanus showing consistent
patterns in a rise in burden with age in different geographical locations [28].
The impact of this substantial reservoir of infection in adults on reinfection in
children can be better illuminated using deterministic and stochastic models of STH
transmission, based on parameter estimates derived from cross-sectional and longitudinal
worm expulsion studies and observations on demography and the mean intensity distribution
across the major age groupings (pre-SAC, SAC, and adults—see Fig 1 [12,13]). A series
of general conclusions emerge that support the observational studies based on demography
and cross-sectional epidemiological data on age specific intensities of infection.
Impact of Current Treatment Strategies
We focus on three major issues concerning MDA, namely: who to treat, how frequently
to treat, and how long to treat. In our calculations, we use a fully aged structured
deterministic STH transmission model (described in [12,13]). Stochastic individual-based
models for the mean worm burden give identical results to the deterministic predictions.
For illustration, we compare the impact of annual and biannual treatment of pre-SAC
and SAC, with annual mass treatment, in communities with Ascaris and hookworm. As
a case study, we focus on an area of medium transmission (R0 values around 2–3, true
prevalence (not adjusted for diagnostic sensitivity) around 70% in SAC). As illustrated
in Fig 2, which records simulated MDA strategies focusing on once and twice yearly
treatment of pre-SAC and SAC combined (at a coverage level of 75%), compared with
community-wide treatment (at the same coverage level). Increasing the frequency of
MDA in children is predicted to be marginally more effective than annual mass treatment,
in terms of reducing the overall burden of Ascaris. In contrast, for hookworm, the
analyses illustrate that increasing the frequency of MDA for pre-SAC and SAC alone
has limited additional impact. Expanding the annual treatment programme to also include
adults actually reduces the intensity of hookworm infections both in children and
the community as a whole (Fig 2). This occurs because the adults have the majority
share of the infectious reservoir (i.e., are a core group), and consequently, treating
children alone does not significantly impact the level of transmission. The children
get reinfected after treatment because of the reservoir in adults. This suggests that
WHO guidelines to increase treatment to twice a year in high prevalence areas is unlikely
to have the desired impact in areas with high hookworm prevalence. The best treatment
strategy is highly dependent on the local age distribution of infection of the different
STH species.
10.1371/journal.pntd.0003897.g002
Fig 2
Impact of different treatment stratgies on the mean number of worms in different age
groupings.
The coloured lines represent different treatment strategies: green—annual community-wide
MDA (75% coverage of all age groupings (pre-SAC, SAC, and adults) and 95% drug efficacy);
red—biannual age group targeted MDA (pre-SAC and SAC with 75% coverage and 95% drug
efficacy); and blue—annual age group targeted MDA (pre-SAC and SAC with 75% coverage
and 95% drug efficacy). Graphs A–C and D–F correspond to Hookworm and Ascaris respectively.
Graphs A and D record the overall mean number of worms across all ages. Graphs B and
E record the mean number of worms in children (pre-SAC and SAC, 2–15-year-olds). Graphs
C and F record the mean number of worms in adults (>15 year olds). Calculations based
on a basic reproductive number, Ro, of 2.5 (medium to high transmission setting).
Model parameters described in [13].
It is important to note that monitoring and evaluation programmes that only measure
the impact of a school-based hookworm treatment programme in the treated age group
record a much bigger impact on the children (Fig 2b) than the true impact on the entire
population, as reflected by persisting worm burdens in adults (Fig 2c). School-age
group-based surveillance programmes can give good estimates of the reduced morbidity
in children but can lead to misleading estimates of the impact of these programmes
on overall transmission in the community. A revision in the monitoring and evaluation
(M & E) guidelines is required to address this problem. Current guidelines recommend
that monitoring is conducted in schools, as school children are the main targets of
control [14]. More recently, transmission assessment surveys (TAS) for lymphatic filariasis
have been proposed as an alternative platform for monitoring STH infection [15], and
the implementation of TAS in the wider community provides the opportunity to reliably
track STH across a range of age classes [16].
To cross the transmission “breakpoint” [10], where transmission is eliminated, requires
many years of continual treatment (Fig 2b) at moderate to high coverage (>75%), depending
on the intensity of transmission (the value of R0). Elimination requires the treatment
of adults for hookworm, and the time required to achieve this is accelerated (and
required in areas of high transmission) for Ascaris. This conclusion also applies
for Trichuris, and in most circumstances for the schistosome infections for which
praziquantel is employed in MDA. Pre-SAC treatment for the control of morbidity induced
by schistosome infections has been suggested by Stothard [17]. The practical feasibility
of providing mass treatment to adults in addition to pre-SAC and SAC is demonstrated
by the mass treatment campaigns for onchocerciasis and lymphatic filariasis (LF),
which provide albendazole plus ivermectin or diethylcarbamazine citrate to entire
communities, using community drug distributors [16]. The albendazole and ivermectin
used in these programmes are also highly effective against STH [18,19], although there
has been no systematic attempt to quantify the impact of the global onchocerciasis
and LF control programmes on the transmission of STH despite the potential insight
they afford into the impact of expanding current school-based deworming programmes.
Costs of Different Strategies
A key issue in scaling up treatment to the whole community is cost, and this in turn
depends on demography (the proportions of the population in each age grouping), who
is treated, at what coverage level, how often, and for how long. If transmission is
not interrupted by only focusing on pre-SAC and SAC, treatment must continue forever
if no other conditions change [13]. Water, sanitation, and hygiene (WASH) programmes
have the potential to radically change the picture, if they can be designed to permanently
supress transmission, but progress has been limited in many areas of endemic infection
[20]. Hopefully, this situation may change in the coming decades as economic growth
improves in parts of Africa and Asia, leading to better sanitation and hygiene.
Under the pessimistic assumption that treating adults is twice as expensive as treating
children (many studies reporting a much lower difference [21]), we calculate that
the total costs of community-wide treatment over a 20 year period (discounting at
3% per year) are much less (approximately 60%) than that for repeated annual treatment
of pre-SAC and SAC, given that the former has only to continue for three years (with
the assumptions in Fig 2), while the latter must continue beyond the 20-year time
horizon in hookworm transmission areas of intermediate intensity (Turner et al. manuscript
in preparation).
New Treatment Guidelines and a New Strategy
Our calculations suggest that the current guidelines need modification, particularly
regarding the recommendation to increase treatment frequency if the prevalence of
any STH is greater than 50% in SAC. Calculations suggest that in most areas where
hookworm is the dominate infection, it is better to broaden treatment across all age
classes instead of treating children twice a year. This leads to the more efficient
use of limited resources in the longer term. For high transmission Ascaris areas,
the best option is both increasing frequency and broadening coverage across all age
classes.
Drug efficacy for Trichuris is known to be low with monotherapy using either Albendazole
or Mebendazole (the standard treatments for STH) [22], and hence in settings where
Trichuris is the dominant STH, alternative approaches will be required. One such approach
that has been shown to notably increase the treatment efficacy against Trichuris is
coadministering Ivermectin with the benzimidazoles or using other drug combinations
[23]. With dual therapy, broadening the coverage of age groupings to include adults
may also be beneficial (depending on the local age intensity profile [Fig 1]).
What are the barriers to changing the quidelines for treatment, monitoring and evaluation?
The school has many advantages for treatment delivery since high treatment coverage
can be achieved for regular attenders and for surveillance as schools provide a ready
sampling frame. In contrast, in some settings it has proven difficult to achieve high
coverage (and good surveillance) in adults for STH (although the experience with MDA
for lymphatic filariasis argues that it can be achieved [24]), and the cost of treating
one adult may at times be higher when compared with treating children in a school
setting. Community-wide coverage also requires an increase in drug donations. Despite
this, the longer term cost calculations are compelling, given that one strategy has
to be continued forever while the other offers the hope of interrupting transmission
permanently. On this basis, it would seem highly desirable to change the WHO guidelines,
with a concomitant emphasis on education, sustainability of current WASH programmes
(to reduce transmission intensity and thereby enhance the impact of MDA), communication
to encourage high treatment uptake amongst adults and better integration of STH control
with that of LF where community-wide coverage has been a target for some time. The
coverage levels predicted to eliminate LF transmission are much less stringent than
those required for hookworm or Ascaris, so integrated STH and LF control is desirable,
but reported coverage and frequency of treatment with Albendazole for LF must be increased
to stop STH transmission. A revision of the guidelines is especially desirable when
hookworm is the dominant infection, since most worms are typically harboured by adults.
The cross-sectional age intensity profiles for Ascaris and Trichuris (Fig 1), and
for the schistosomes, suggest that in high transmission areas, infections across the
community may be maintained by adults even when children are effectively treated twice
a year at high coverage levels. Of course, treating the whole community will also
lead to more rapid reductions in transmission. But the effect will not be as extreme
as for hookworm. To achieve the 2020 goals, treatment coverage in children must be
increased significantly, but in many areas reductions in morbidity, and the highly
desirable goal of stopping transmission, would both be more likely and much more rapid,
if coverage is broadened to encompass adults. The debate on what is the best strategy
to manage STH infection should shift from morbidity control to transmission interruption.
Concomitantly, there is a need to broaden the scope of research to investigate the
cost-effectiveness and feasibility of alternative treatment strategies in achieving
the interruption of transmission across a range of settings. Linked to any shift from
age group targeting to community-wide control is the risk of enhancing selection for
drug resistant strains of the parasites, where the refugium of untreated adults no
longer dilutes the gene pool of those parasites exposed to selection. But the experience
with community-wide control in LF programmes, where the Albendazole drug also impacts
STH, suggests this concern may not materialize in practice. However, this needs careful
monitoring.