Skip to main content
PMC home page
The BMJ logoLink to The BMJ
. 2000 Nov 11;321(7270):1224.

Treatment for intestinal helminth infection

Contrary to authors' comments, meta-analysis supports global helminth control initiatives

E Michael 1
PMCID: PMC1118973  PMID: 11073529

Editor—Dickson et al's meta-analysis of the effects of treatment for intestinal helminth infection on growth and cognitive performance in children has produced two important findings, only one of which has been discussed by the authors.1

The meta-analysis has helped highlight the poor quality of many of the trials carried out so far. The more important result (not remarked on) is the extraordinary finding that, despite the many systematic differences observed between the studies used in the meta-analysis, treatment unfailingly has a positive average effect on both the outcomes studied (table 3). In addition, the fact that many systematic differences were observed between the studies in the meta-analysis seriously questions the value of trying to derive global summary results for any of the comparisons on both statistical and biological grounds.

The result of the meta-analysis supports an important principle for judging causality from the results of clinical trials—that if different trials address related questions then differences are more likely to occur in the size of any effects than in their direction. When interpreted in this light it is clear that the present results provide quite firm support for the conclusions of the better designed individual trials that anthelmintic treatment may indeed significantly improve child growth and cognitive function; this is in direct contrast to the authors' own pessimistic conclusion. Thus rather than undermining the global helminth control initiatives promoted by the World Bank and World Health Organization, the present review has actually produced evidence in their support.

References

  • 1.Dickson R, Awasthi S, Williamson P, Demellweek C, Garner P. Effects of treatment for intestinal helminth infection on growth and cognitive performance in children: systematic review of randomised trials. BMJ. 2000;320:1697–1701. doi: 10.1136/bmj.320.7251.1697. . (24 June.) [DOI] [PMC free article] [PubMed] [Google Scholar]
BMJ. 2000 Nov 11;321(7270):1224.

Conclusions should have been based on broader considerations

Alok Bhargava 1

Editor—In their systematic review on the effects of treatment for intestinal helminth infections on growth and cognitive performance in children, Dickson et al note numerous shortcomings in the design of previous studies.1-1 The authors themselves, however, conclude misleadingly that developing countries should not invest in mass treatment of children against helminth infections.

Firstly, no studies have apparently been designed to disentangle the effects of helminth infections on cognitive function from the effects of other sicknesses. Using a dynamic multivariate random effects framework to explain Kenyan children's cognitive test scores, I found that an index of morbidity was significantly associated with lower scores but that hookworm egg count per gram of stool was not a significant predictor of cognitive scores.1-2 The data were not sufficiently detailed to shed light on the hypothesis that helminth infections may be negatively associated with cognitive test scores because they increase the duration and intensity of sicknesses.

Secondly, some psychologists have suggested that helminth infections may occasionally be beneficial for cognitive function.1-3 This would seem counter intuitive, and is perhaps a fallacy arising from a failure to recognise the nature of epidemiological data. It is, for example, well known that nutrient intakes exhibit day to day variation. At the other extreme, psychological tests such as the Brazelton neonatal behavioural assessment scale are subject to high within subject variation that often renders them useless unless babies have neurological defects.1-4 The magnitude of within subject variability in schoolchildren's cognitive scores is lower than that for scores on the neonatal behavioural assessment scale, but this factor should be incorporated when trials for assessing the effects of anthelmintic treatment on cognitive function are being designed. For example, the children could be tested several times before and after the treatment.

Finally, parasitic infections probably exacerbate iron deficiency anaemia in children growing up in unhygienic environments; bioavailability of iron from cereal based diets is low. Iron deficiency anaemia, in turn, is associated with poor learning.1-2,1-5 Anthelmintic treatment in most studies has been of short duration, and reinfection rates are typically high. It would therefore seem premature for Dickson et al to conclude that mass anthelmintic treatment may not be cost effective. Rather, cost benefit analysis of extended anthelmintic treatment, including the development of vaccines against certain types of infection, merit closer attention. It would be far sighted to approach anthelmintic treatment from a broad long term perspective in developing countries.

References

  • 1-1.Dickson R, Awasthi S, Williamson P, Demellweek C, Garner P. Effects of treatment for intestinal helminth infection on growth and cognitive performance in children: systematic review of randomised trials. BMJ. 2000;320:1697–1701. doi: 10.1136/bmj.320.7251.1697. . (24 June.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 1-2.Bhargava A. A dynamic model for the cognitive development of Kenyan schoolchildren. J Ed Psychol. 1998;90:162–166. [Google Scholar]
  • 1-3.Watkins WE, Pollitt E. “Stupidity or worms?”: Do intestinal worms impair mental performance? Psychol Bull. 1997;121:171–191. doi: 10.1037/0033-2909.121.2.171. [DOI] [PubMed] [Google Scholar]
  • 1-4.Bhargava A. Modelling the effects of maternal nutritional status and socioeconomic variables on the anthropometric and psychological indicators of Kenyan infants from age 0-6 months. Am J Phys Anthropol. 2000;111:89–104. doi: 10.1002/(SICI)1096-8644(200001)111:1<89::AID-AJPA6>3.0.CO;2-X. [DOI] [PubMed] [Google Scholar]
  • 1-5.Pollitt E. Iron deficiency and cognitive function. Ann Rev Nutr. 1993;13:521–537. doi: 10.1146/annurev.nu.13.070193.002513. [DOI] [PubMed] [Google Scholar]
BMJ. 2000 Nov 11;321(7270):1224.

Studies of short term treatment cannot assess long term benefits of regular treatment

Don Bundy 1,2, Richard Peto 1,2

Editor—The Cochrane review of anthelmintic trials chiefly shows that studies of short term treatment cannot assess the long term benefits of regular treatment.2-1

When the World Health Organization, Unicef, Unesco, and the World Bank included deworming as one component of their efforts to focus resources on effective school health (the “FRESH start” partnership) they intended that infrequent but regular treatment from an early age would ensure that children avoided heavy infection throughout the vulnerable years of growth and development. A review of trials that have not evaluated such a strategy is not an appropriate basis for policy recommendations, especially since the review omitted, for example, the benefits of avoiding hookworm anaemia and Trichuris colitis.

The remarkable cost effectiveness of deworming derives not from some easily measured and immediate clinical benefit of a single intervention but from the longer term preventive value of an annual investment of less than 7p.

References

  • 2-1.Dickson R, Awasthi S, Williamson P, Demellweek C, Garner P. Effects of treatment for intestinal helminth infection on growth and cognitive performance in children: systematic review of randomised trials. BMJ. 2000;320:1697–1701. doi: 10.1136/bmj.320.7251.1697. . (24 June.) [DOI] [PMC free article] [PubMed] [Google Scholar]
BMJ. 2000 Nov 11;321(7270):1224.

Message does not follow from systematic review's findings

Ed Cooper 1

Editor—The number of parasites per host in persistent parasitic infections such as the helminthiases is characteristically dispersed in a frequency distribution that is extremely skewed compared with Gaussian or even Poisson distributions.3-1 From this follow a couple of points that are relevant to a meta-analysis such as that by Dickson et al.3-2 Firstly, heterogeneity in parasite burdens is of special importance, exceeding that of age and drug type, which the authors had hoped to take into account but could not do from the primary trials. Secondly, attempting to allow for this heterogeneity by generating a random effects model may be inappropriate since this is done by introducing an error term with an assumption of Gaussian distribution of the error.3-3

More generally—and this point is also related to the extremely skewed and overdispersed worm distributions—achieving end points of growth and cognitive development in trials of the treatment of populations, even where the prevalence of infections is high, needs thinking about. The difference between having no worms and having a few worms is probably much less clinically important than the difference between having some worms and having an enormous burden of worms (having an enormous burden is always relatively rare).

This is like the difference between carriage of Neisseria meningitidis (high percentage of the population) and invasive meningococcal disease (a few thousand cases a year in the British Isles). Currently in the United Kingdom we are vaccinating about 15 million people to prevent 1500 cases of type C meningococcal disease and 150 deaths a year. But we shall not be assessing our effectiveness by changes in the total morbidity or mortality of the 15 million population, which would be lost in the dilution, or by any measure such as numbers needed to vaccinate to prevent a case of meningococcal disease: we will count the individuals with type C meningococcal disease.

In a similar way the greatest effect of anthelmintic treatment on growth and development of children will be concentrated in those with the heaviest parasite burdens. This effect occurred in one of the trials reviewed3-4 and was considerable when only intense, severely symptomatic trichuriasis was treated,3-5 in a study where placebo control would have been unethical.

The systematic review is useful, but the message in the “What this study adds” panel that “There is little evidence to support the use of routine anthelmintic treatment to improve growth and cognitive performance in children in developing countries” does not follow from its findings.

Footnotes

Dr Cooper is an author of one of the trials reviewed.

References

  • 3-1.Anderson RM, May RM. Helminth infections of humans: mathematical models, population dynamics and control. Adv Parasitol. 1985;24:1–101. doi: 10.1016/s0065-308x(08)60561-8. [DOI] [PubMed] [Google Scholar]
  • 3-2.Dickson R, Awasthi S, Williamson P, Demellweek C, Garner P. Effects of treatment for intestinal helminth infection on growth and cognitive performance in children: systematic review of randomised trials. BMJ. 2000;320:1697–1701. doi: 10.1136/bmj.320.7251.1697. . (24 June.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3-3.Greenland S. Meta-analysis. In: Rothman K, Greenland S, editors. Modern epidemiology. 2nd ed. Philadelphia: Lippincott-Raven; 1998. pp. 643–673. [Google Scholar]
  • 3-4.Simeon D, Callender J, Wong M, Grantham-McGregor S, Ramdath DD. School performance, nutritional status and trichuriasis in Jamaican schoolchildren. Acta Paediatr. 1994;83:1188–1193. doi: 10.1111/j.1651-2227.1994.tb18279.x. [DOI] [PubMed] [Google Scholar]
  • 3-5.Callender JEM, Grantham-McGregor SM, Walker SP, Cooper ES. Treatment effects in trichuris dysentery syndrome. Acta Paediatr. 1994;83:1182–1187. doi: 10.1111/j.1651-2227.1994.tb18278.x. [DOI] [PubMed] [Google Scholar]
BMJ. 2000 Nov 11;321(7270):1224.

Review needed to take account of all relevant evidence, not only effects on growth and cognitive performance

Lorenzo Savioli 1,2,3,4,5,6,7,8,9,10,11,12,13, Maria Neira 1,2,3,4,5,6,7,8,9,10,11,12,13, Marco Albonico 1,2,3,4,5,6,7,8,9,10,11,12,13, Michael J Beach 1,2,3,4,5,6,7,8,9,10,11,12,13, Hababu Mohammed Chwaya 1,2,3,4,5,6,7,8,9,10,11,12,13, David W T Crompton 1,2,3,4,5,6,7,8,9,10,11,12,13, John Dunne 1,2,3,4,5,6,7,8,9,10,11,12,13, John P Ehrenberg 1,2,3,4,5,6,7,8,9,10,11,12,13, Theresa Gyorkos 1,2,3,4,5,6,7,8,9,10,11,12,13, Jane Kvalsvig 1,2,3,4,5,6,7,8,9,10,11,12,13, Martin G Taylor 1,2,3,4,5,6,7,8,9,10,11,12,13, Carlo Urbani 1,2,3,4,5,6,7,8,9,10,11,12,13, Feng Zheng 1,2,3,4,5,6,7,8,9,10,11,12,13

Editor—Dickson et al's paper reflects the public health importance of helminth infections, particularly in children, adolescent girls, and women of childbearing age.4-1 We are concerned, however, that on the basis of limited evidence the reviewers “would be unwilling to recommend that countries or regions invest in programmes that routinely treat children with anthelminthic drugs.” As the authors state in their introduction, the World Health Organization, Unicef, the World Bank, and the World Food Programme together with partners and collaborators have strongly recommended such interventions, having regard to a substantial body of supportive evidence, for the past 25 years.

The impact of population based chemotherapy depends on many factors. Local patterns of mixed nematode infections transmitted in soil, and their clinical consequences, show important variations. Whereas hookworm may be associated primarily with iron deficiency anaemia, Ascaris lumbricoides may be associated mainly with stunting of growth. Intensity of transmission, nutritional intake, and retreatment schedules are among other variables of fundamental importance. Assessing the impact of regular anthelmintic chemotherapy must be related to these multiple effects. Only then can proper policy implications and recommendations be given to countries or regions.

To support their conclusion the reviewers needed to take account of all relevant evidence, not only the effects on growth and cognitive performance. They seem to have failed in this crucial requirement. For instance, whereas they refer to the work of Stoltzfus et al,4-2 they exclude reference to the main finding that, in an area where hookworm infection predominated, “this deworming program prevented 1260 cases of moderate-to-severe anemia and 276 cases of severe anemia in a population of 30 000 schoolchildren in 1 year.”4-3

Finally, two of the authors of this letter (LA and MA) are cited in the paper as being members of the advisory panel to the authors. Including their names may give the impression that they agreed with the content of the paper; in fact, they were not consulted before publication.

References

  • 4-1.Dickson R, Awasthi S, Williamson P, Demellweek C, Garner P. Effects of treatment for intestinal helminth infections on growth and cognitive performance in children: systematic review of randomised trials. BMJ. 2000;320:1697–1701. doi: 10.1136/bmj.320.7251.1697. . (24 June.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4-2.Stoltzfus RJ, Albonico M, Tielsch JM, Chwaya HM, Savioli L. School-based deworming program yields small improvement in growth of Zanzibar school children after one year. J Nutr. 1997;127:2187–2193. doi: 10.1093/jn/127.11.2187. [DOI] [PubMed] [Google Scholar]
  • 4-3.Stoltzfus RJ, Albonico M, Chwaya HM, Tielsch JM, Schulze KJ, Savioli L. Effects of the Zanzibar school-based deworming program on iron status of children. Am J Clin Nutr. 1998;68:179–186. doi: 10.1093/ajcn/68.1.179. [DOI] [PubMed] [Google Scholar]
BMJ. 2000 Nov 11;321(7270):1224.

Authors' reply

Paul Garner 1,2,3,4, Rumona Dickson 1,2,3,4, Colin Demellweek 1,2,3,4, Paula Williamson 1,2,3,4, Shally Awasthi 1,2,3,4

Editor—Michael points out that the weight changes favour the intervention. But the differences in weight gain between the groups are often small. Michael is incorrect in saying that the evidence presented provides firm support that anthelmintic treatment significantly improves cognitive function. It is certainly not the case, as he implies, that treated children invariably do better in cognitive and academic tests than control children. Control children taking placebo have shown greater gains in some tests than treated children, and in some cases this difference has been significant. On this topic, Bhargava seems to imply that the failure to find an effect may be due to the unreliability of the tests. But most tests used have had adequate to good reliability.

Bundy and Peto comment that our review does not evaluate infrequent but regular treatment from an early age. We sought trials that repeated treatments, and this was defined in our protocol. But there were few such trials, and the data were limited. The current large cluster randomised trial in Lucknow will help provide some answers to the effectiveness of these strategies.

Cooper notes that the random effects estimate may be inaccurate because of the skewed dispersion of worms in a population. His argument suggests that the uncertainty around effect estimates is increased. In our protocol we sought to conduct subgroup analyses by intensity of worm burden, but no trials provided the data necessary for us to do this.

The letter from the World Health Organization with 13 authors states that they disagree with how we interpreted the data, but again they do not provide substantive evidence to support their past and current recommendations. Savioli and Albonico provided helpful input to the protocol development for this review. We did not intend to imply that they had agreed with the results of the review, only to acknowledge their valuable input in the review process; we will make this explicit in the Cochrane review.

Several authors comment on the fact that the review was not able to draw conclusions about the effects of long term treatments. We were unable to find any randomised controlled trials that evaluated long term benefit, and the evidence of short term benefit was not, for us, convincing. We therefore stand by our conclusion that it was premature to recommend this widely, and for countries to borrow money from the World Bank to routinely implement national population based policies of routine repeated treatment. We believe that the introductory statement to a World Health Organization publication—that “regular chemotherapy of infected populations reduces mortality and morbidity in pre-school children, improves nutritional status and school performance of school children”—is not based on current available evidence.5-1

Routine treatment with anthelminths could well be an exciting and important intervention. But we need the results of larger, well designed trials, such as the current trial in India, before lending money to already poor countries to invest in an intervention where there are doubts about its wholesale benefit.

References

  • 5-1.World Health Organization. Report of the WHO informal consultation on the use of chemotherapy for the control of morbidity due to soil transmitted nematodes in humans. Geneva: WHO; 1996. [Google Scholar]

Articles from BMJ : British Medical Journal are provided here courtesy of BMJ Publishing Group

RESOURCES