The SCORE project commenced in Niger with the goal of providing an evidence base for programmatic decisions on how best to control urogenital schistosomiasis, and ultimately eliminate the public health problem in both highly and moderately endemic areas. Depending on the prevalence of schistosomiasis infection in a given community, WHO recommends PC using either a school-based or community-wide approach [7]. The initial aim of this study was to examine the impact of SBT, CWT and treatment holidays for the control of S. haematobium in villages with both a high (⩾21%) and moderate (10–24%) prevalence. In Year 3 the study was redesigned, however, to compare annual versus biannual treatment delivered by either SBT or CWT.
The primary research question presented here is which PC strategy provided the greatest reduction in prevalence and intensity of S. haematobium infection among 9-to-12-year olds after four years of intervention in the Niger River Valley. Biannual SBT treatment was found to be significantly more effective for reducing active schistosome infection than annual SBT in high prevalence areas (Arm B1 vs. B2). There was no significant effect of treatment frequency, however, in areas with a low starting prevalence (Arm A1 vs. A2), or in higher prevalence areas receiving CWT (Arm C1 vs. C2).
Although the purpose of the eligibility survey was to ensure a high enough starting prevalence to make comparisons between treatment arms, the eligibility survey did have higher infection rates than the baseline parasitological survey whereby several villages had high enough prevalence using Hemastix® but when assessed by urine filtration the prevalence was lower than 10% (see Table 5 in Appendix). Some studies have shown that in very low prevalence settings, microhaematuria can be an unstable proxy for urogenital schistosomiasis [29–30]. On the other hand, the sensitivity of microscopic egg detection can vary according to intensity of infection, day-to-day variation in egg secretion, time the sample was collected, and number of examined samples [31–35]. Another potential reason for the difference in dipstick and microscopic examination could be the fact that the eligibility was carried out in 13-to-14-year olds and the parasitology survey among 9-to-12-year olds. Some studies have shown that 13-to-14-year olds harbour higher infection levels due to them not attending school and therefore more likely to be exposed to infection [36]. Other studies have showed that many cases of urogenital schistosomiasis stay undetected when the examination method is limited to urine filtration [37–38].
Table 5
Prevalence of S. haematobium in western Niger using Haemastix reagent dipsticks (2010–2011)
District | Number of Schools Sampled | Number of People Sampled | Number of Schools Eligible | Number of People Sampled in Eligible Schools | Method: Dipstick (Eligibility survey among 13-14-year olds) | Method: Urine filtration (Baseline data collection among 9-12-year olds) |
Prevalence (%) | 95% CI | Prevalence (%) | 95% CI |
CUN5 | 3 | 150 | 1 | 50 | 58.00 | NA | 4.00 | NA |
DS1 | 1 | 50 | 1 | 50 | 54.00 | NA | 32.00 | NA |
DS3 | 1 | 50 | 1 | 50 | 30.00 | NA | 8.00 | NA |
Filingue | 93 | 4,528 | 44 | 2,141 | 20.10 | 16.71, 23.50 | 5.72 | 2.93, 8.52 |
Kollo | 87 | 4,235 | 52 | 2,558 | 42.43 | 36.35, 48.51 | 26.68 | 19.72, 33.64 |
Loga | 54 | 2,641 | 26 | 1,281 | 13.75 | 12.03, 15.47 | 1.70 | 0.78, 2.62 |
Say | 76 | 3,568 | 31 | 1,501 | 37.19 | 31.79, 42.60 | 13.07 | 7.90, 18.24 |
Tera | 45 | 2,242 | 41 | 2,042 | 39.92 | 33.87, 45.96 | 18.79 | 12.51, 25.07 |
Tillaberi | 32 | 1,632 | 28 | 1,399 | 38.12 | 27.22, 49.03 | 24.29 | 13.10, 35.49 |
Total | 392 | 19,096 | 225 | 11,072 | | | | |
Both age and gender influenced infection with these findings supported by that of other studies, which have shown that the highest prevalence and intensities of infection occurred in males and among older adolescents, with infection decreasing in adulthood [36, 39–40]. There was no significant difference, however, in response to different treatment approach by age and sex.
To evaluate the potential longer-term impact of the program, we also examined the infections in first-year students, some of whom had not received treatment previously in SBT communities as they had only recently started attending school. It is important to note that S. haematobium infection in this group of children decreased significantly over four rounds of treatment, implying that the level of environmental transmission in these areas could have been reduced because of overall reductions in excreted eggs from infected children.
Currently the justification for targeting adults in a schistosomiasis control programme is based on the prevalence of infection in SAC [7]. A study in Nigeria, however, has shown that monitoring and evaluation of activities based on SAC was not a successful indicator of the burden of infection in adults [41]. The burden in adults will likely be driven by many local behavioural and cultural factors, and is, therefore, context specific. This makes it difficult to make a universal SAC infection prevalence threshold for switching to CWT. For this reason, adults were also sampled in this study. The findings here demonstrated that biannual CWT was not significantly more effective at reducing prevalence and intensity of infection. This lack of statistical significance, however, may be attributed to the difference in starting prevalence of the two groups whereby Group C1 was 14.3% and C2 was 23.4%. An important reason why studies find contrasting results regarding the benefit of CWT is the variation in the relative worm burden harboured by adults across different geographical settings; indeed, it has been modelled by Turner et al that the higher the pre-control burden in adults, the larger the benefit of switching to CWT [13]. Interestingly, there was a significant reduction in prevalence of S. haematobium infection seen among adults even in Group B implying the rate of transmission in the community had been decreased, even where only school children have been treated. These findings of reduction in infection among adults even in SBT communities was also seen in Mozambique [42].
Research in Niger has shown that the full economic delivery cost of SBT in 2005/06 was $0.76 and CWT was $0.46. Including only the programme costs the figures were $0.47 and $0.41 respectively [43]. Differences at sub-district were more marked. This is partly explained by the fact that a CDD treats 5.8 people for every one treated in school [43]. Although CWT did show a significant impact on the prevalence and intensity of S. haematobium infection among adults, given that prevalence also decreased among adults in SBT areas and twice-yearly CWT did not have a significantly greater impact on prevalence reduction with respect to annual CWT, these findings both have significant logistical and cost-saving implications for a national control programme and the justification for CWT.
The success of PC programmes may be influenced by a wide range of factors such as initial levels of endemicity and transmission intensities in the local environment; treatment frequency, coverage and compliance, among others [44]. Coverage data among different age groups is an important key determinant for achieving programme targets. Although average treatment coverage in the Niger SCORE programme has been shown to be well above the recommended WHO 75% threshold, in some areas it was above 100% and therefore implies an unreliable denominator. There may have been other external issues that might have affected the success of the biannual treatment. For example, there may have been a considerable gap between coverage and compliance [45]. Even when the proportion of eligible people who received tablets reach a significant fraction of the target population (coverage), those ingesting all the tablets at the same time (compliance) may be a better indicator of how well PC is being implemented. In this study, SBT was directly observed as the survey teams were present in the schools during the treatment. For CWT, however, there is less of a guarantee that the CDD supervised actual consumption of the treatment. In addition, frequent migration of people particularly among fishing communities or the nomadic Peule communities, some of whom move between neighbouring countries, remain significant hurdles to the successful implementation of the control program, particularly in terms of ensuring repeated treatment. Finally, there was an increase in prevalence seen across all Groups in Year 3. This is likely the consequence of severe flooding across the country in 2012, which would have increased water contact as well as affecting the snail population, in addition to subsequent migration.
There were limitations to the study. One of which was the treatment coverage, as discussed above, particularly where SAC numerators were higher than SAC denominators where adults who presented for treatment at the school were registered in the MDA. The other limitation was the differential in time periods between MDAs and parasitological surveys, which were not conducted precisely at once and six-monthly time points due to the logistical challenge of conducting such large surveys and treatment in such a short time frame. It is not clear whether this contributed to the findings as it was difficult to account for this in the analysis.
Given that there is a shifting emphasis towards transmission elimination by WHO, the findings here show that with good coverage biannual SBT may be the best strategy for schistosomiasis control in high prevalence areas. In low transmission settings, however, the lack of statistical impact of two rounds of treatment indicate that it may not be possible to break transmission using through PC alone and using other strategies such as health education, WASH, and snail control, should also be considered [46, 47].