Epidemiological modelling for monitoring and evaluation of lymphatic filariasis control.

Monitoring and evaluation are critically important to the scientific management of any parasite control programme. If a management action is prescribed, monitoring plays a pivotal role in assessing the effectiveness of implemented actions, identifying necessary adaptations for management and determining when management objectives are achieved. Here, we focus on the control of the vector-borne parasitic disease, lymphatic filariasis, to show how mathematical models of parasite transmission can provide a scientific framework for supporting the optimal design of parasite control monitoring programmes by their ability to (1) enable the estimation of endpoint targets, (2) provide information on expected trends in infection due to interventions to allow rational evaluation of intervention effectiveness and calculation of the frequency of monitoring, (3) support the selection of indicators that permit reliable and statistically powerful measurement of the effects of interventions while minimizing costs and sampling intensity and (4) aid the interpretation of monitoring data for improving programme management and knowledge of the population dynamics of parasite control. The results also highlight how the use of a model-based monitoring framework will be vitally enhanced by adopting an adaptive management approach that acknowledges uncertainty and facilitates the use of monitoring data to learn about effective control, and which allows future decisions to be modified as we learn by doing. We conclude by emphasizing a pressing need to incorporate mathematical models coupled with changes to existing management systems in ongoing efforts to design and implement rational monitoring plans for evaluating filariasis and other parasitic disease control programmes.

Mathematical models and lymphatic filariasis control: monitoring and evaluating interventions.

Monitoring and evaluation are crucially important to the scientific management of any mass parasite control programme. Monitoring enables the effectiveness of implemented actions to be assessed and necessary adaptations to be identified; it also determines when management objectives are achieved. Parasite transmission models can provide a scientific template for informing the optimal design of such monitoring programmes. Here, we illustrate the usefulness of using a model-based approach for monitoring and evaluating anti-parasite interventions and discuss issues that need addressing. We focus on the use of such an approach for the control and/or elimination of the vector-borne parasitic disease, lymphatic filariasis.

Filarial-specific antibody response in East African bancroftian filariasis: effects of host infection, clinical disease, and filarial endemicity.

The effect of host infection, chronic clinical disease, and transmission intensity on the patterns of specific antibody responses in Bancroftian filariasis was assessed by analyzing specific IgG1, IgG2, IgG3, IgG4, and IgE profiles among adults from two communities with high and low Wuchereria bancrofti endemicity. In the high endemicity community, intensities of the measured antibodies were significantly associated with infection status. IgG1, IgG2, and IgE were negatively associated with microfilaria (MF) status, IgG3 was negatively associated with circulating filarial antigen (CFA) status, and IgG4 was positively associated with CFA status. None of the associations were significantly influenced by chronic lymphatic disease status. In contrast, IgG1, IgG2, and IgG4 responses were less vigorous in the low endemicity community and, except for IgG4, did not show any significant associations with MF or CFA status. The IgG3 responses were considerably more vigorous in the low endemicity community than in the high endemicity one. Only IgG4 responses exhibited a rather similar pattern in the two communities, being significantly positively associated with CFA status in both communities. The IgG4:IgE ratios were higher in infection-positive individuals than in infection-negative ones, and higher in the high endemicity community than in the low endemicity one. Overall, these results indicate that specific antibody responses in Bancroftian filariasis are more related to infection status than to chronic lymphatic disease status. They also suggest that community transmission intensity play a dominant but subtle role in shaping the observed response patterns.

Mathematical models and lymphatic filariasis control: endpoints and optimal interventions.

The current global initiative to eliminate lymphatic filariasis is a major renewed commitment to reduce or eliminate the burden of one of the major helminth infections from resource-poor communities of the world. Mathematical models of filariasis transmission can serve as an effective tool for guiding the scientific development and management of successful community-level intervention programmes by acting as analytical frameworks for integrating knowledge regarding parasite transmission dynamics with programmatic factors. However, the power of these tools for supporting control interventions will be realized fully only if researchers address the current uncertainties and gaps in data and knowledge of filarial population dynamics and the effectiveness of currently proposed filariasis intervention options.

The effect of eight half-yearly single-dose treatments with DEC on Wuchereria bancrofti circulating antigenaemia.

The effect of eight half-yearly treatment rounds with diethylcarbamazine (DEC; 6mg/kg bodyweight) on Wuchereria bancrofti-specific circulating filarial antigen (CFA), a marker of adult worm infection, was followed in 79 individuals who were CFA-positive before start of treatment. Half of these were also microfilariae (mf)-positive. Microfilaraemia decreased rapidly after onset of treatment and became undetectable after four treatments. Circulating antigenaemia also decreased progressively, but at a much slower rate. After two, four and eight treatment rounds, the mean CFA intensity was reduced by 81, 94 and 98%, and the prevalence of CFA positivity was 85, 66 and 57%, compared with pre-treatment, respectively. CFA clearance rates were negatively related to pre-treatment CFA intensities, and were higher among pre-treatment mf-negative individuals than among pre-treatment mf-positive individuals. Even among patients who had pre-treatment CFA intensities above the upper measuring level (32000antigen units), and who continued to have intensities above this level after treatment, a decrease in post-treatment CFA intensities was obvious from a continuous decrease in ELISA optical density values. Repeated DEC therapy thus appears to have a slow but profound and persistent macrofilaricidal effect, which in the long run may be beneficial to populations undergoing DEC-based control interventions by reducing the probability of future morbidity development.

The effect of single dose ivermectin alone or in combination with albendazole on Wuchereria bancrofti infection in primary school children in Tanzania.

Examination of 1829 children from 6 primary schools in coastal Tanzania revealed overall Wuchereria bancrofti microfilaria (mf) and circulating filarial antigen (CFA) prevalences of 17.3% and 43.7%, respectively. A randomized double-blind field trial with a single dose of ivermectin (150-200 microg/kg body weight) alone or in combination with albendazole (400 mg) was subsequently carried out among these children. Both treatment regimens resulted in a considerable decrease in mean mf intensities, with overall reductions being slightly but statistically significantly higher for the combination than for ivermectin alone. The difference in effect between the two treatment regimens was most pronounced at 6 months, whereas it was minor at 12 months after treatment. The relative effect of treatment on mean CFA units was less pronounced than on mf. For both treatment regimens, reductions in CFA intensity appeared to be higher in children who were both CFA and mf positive before treatment, which may suggest that treatment mainly affected the survival and/or production of mf, rather than the survival of adult worms. New cases of infection appeared after treatment with both regimens among the pre-treatment mf and CFA negative children. Adverse reactions were few and mild in both groups, and mainly reported from pre-treatment mf and CFA positive children. The alarmingly high prevalence of W. bancrofti infection in primary school children highlights the importance of also determining the reversibility of already acquired early lesions, and the development of new measures and strategies to specifically protect children from later developing clinical disease.

Mathematical modelling and the control of lymphatic filariasis.

The current global initiative to eliminate lymphatic filariasis represents one of the largest mass drug administration programmes ever conceived for the control of a parasitic disease. Yet, it is still not known whether the WHO-recommended primary strategy of applying annual single-dose mass chemotherapy with a combination of two drugs for 4-6 years will effectively break parasite transmission from all endemic communities. Here we review recent work on the development and application of a deterministic mathematical model of filariasis transmission, to show how models of parasite transmission will help resolve the key currently debated questions regarding the ultimate effectiveness of the global strategy to control filariasis. These critical questions include the required duration of mass treatment in different endemic areas, the optimal drug coverage required to meet control targets within prescribed timeframes, the impact and importance of adding vector control to mass chemotherapy regimens, and the likelihood of the development of drug resistance by treated worm populations. The results demonstrate the vital role that integrating these models into control programming can have in providing effective decision-support frameworks for undertaking the optimal design and monitoring of regional and global filariasis-control programmes. Operationally, the models show that the effectiveness of the strategy to achieve filariasis control will be determined by successfully addressing two key factors: the need to maintain high community treatment coverages, and the need to include vector control measures especially in areas of high endemicity.

The effect of repeated half-yearly diethylcarbamazine mass treatment on Wuchereria bancrofti infection and transmission in two East African communities with different levels of endemicity.

The effect of repeated half-yearly mass treatment with diethylcarbamazine (DEC, 6 mg/kg body weight) on infection and transmission of Wuchereria bancrofti was assessed and compared in communities with high and low endemicity in eastern Africa, with pretreatment microfilaria (mf) and circulating filarial antigen (CFA) prevalences of 29.4% and 53.2% in the high endemicity community and 3.1% and 18.7% in the low endemicity community, respectively. Human infection was monitored by repeated cross-sectional surveys, and transmission by weekly light trap collection of vector mosquitoes in selected houses in each community. Treatments resulted in a progressive decrease in microfilaremia and circulating antigenemia in both communities, with relative reductions being considerably higher for mf than for CFA. Among pretreatment mf-positive individuals, more than 60% were diagnosed as mf negative and mean mf intensities were reduced by 99% in both communities after two treatment rounds. In contrast, only moderate reductions were seen in circulating antigenemia among pretreatment CFA-positive individuals, with mean intensities still being 24-39% of pretreatment values after two treatment rounds. Among the pretreatment mf/CFA-positive individuals, clearance to a CFA-negative status was negligible. Complete CFA clearance was only observed among pretreatment CFA-positive but mf negative individuals who also had much lower initial mean CFA levels than the mf-positive individuals. After treatment, the intensity of transmission decreased in the high-endemicity community, but this appeared mainly to be a consequence of a drought-induced reduction in vector density rather than to reduced mf load in the human population, since the proportion of mosquitoes carrying infective larvae was not reduced. No change in transmission or mosquito infectivity was observed after treatment in the low-endemicity community. Implications of these observations for the control of Bancroftian filariasis are discussed.

Filaria dance sign and subclinical hydrocoele in two east African communities with bancroftian filariasis.

During population-wide cross-sectional surveys for Wuchereria bancrofti microfilaremia, circulating antigenaemia, and clinical disease in a high and a low endemicity community in East Africa in 1998, a portable ultrasound scanner was used simultaneously to examine the scrotal tissue of the male populations (n = 422 and 328, respectively) for signs of adult worms. The overall microfilaria (mf) and circulating filarial antigen (CFA) prevalences in the scanned males were 30.8% and 53.6% in the high and 4.3% and 19.8% in the low endemicity community, respectively. During ultrasound examination, the filaria dance sign (FDS)--indicating the presence of live adult W. bancrofti worms--was observed in 16.1% and 6.7% of the males in these communities, respectively. This examination also revealed that subclinical hydrocoele (fluid accumulation in the scrotal sac, not detected during physical examination for clinical hydrocoele) was very common, affecting 25.3% and 15.5% of the examined males in the high and low endemicity community, respectively. Both of these ultrasonographic signs started to appear around the age of puberty and were most common in adults. In the high endemicity community, the prevalence and mean intensity of mf and CFA were considerably higher in FDS-positive than in FDS-negative adult males, whereas no obvious difference in these parameters was noted between adult males with and without subclinical or the combination of clinical and subclinical hydrocoele. Associations were less clear in the low endemicity community, probably because of the low number of infected individuals. The application of ultrasonography as a tool in bancroftian filariasis epidemiological field studies thus indicated that scrotal pathology may be much more common in endemic areas than hitherto reported.