Skip to 0 minutes and 21 seconds GERRY F.
Skip to 0 minutes and 21 seconds KILLEEN: The following example of how we established a locally managed larval source management programme in the Tanzanian city of Dar es Salaam is merely illustrative. It is by no means intended as a template for direct application elsewhere or even as an exemplar of best practise in Dar es Salaam itself. There are several reasons why such larval surveys for quality control purposes cannot be considered a reliable quality assurance indicator of performance or impact. First, larval service can only tell us about the larval habitats we’ve been able to find. They provide us with no information about the habitats we’ve not yet found and are biassed towards the same habitats that are most obvious to the team’s responsibility for treating them.
Skip to 1 minute and 1 second Second, they are very subjective and require a lot of work, so they’re inherently prone to biases arising from operator expectations based on environmental factors or presumed treatment status. For example, in this Dar es Salaam programme, larval surveys were far less likely to detect surviving larvae in areas that had been recently scheduled for treatment. Such biases are often exaggerated by the preferences of supervisors for reporting success rather than failure. For example, external quality assurance of CORPS larval surveys by an independent researcher revealed far poorer levels of performance than internal quality controlled surveys by municipal and city level supervisors. Third, larval surveys do not capture the effect of adult mosquito flies that allows them to disperse into larvicide treated areas from outside.
Skip to 1 minute and 50 seconds Overall larval surveys tend to paint a much rosier picture of larval source management effectiveness than is reality on the ground. In contrast, adult mosquito densities can be passively and independently surveyed by staff with no competing interest in larvicide application or active quality control surveys of larvae. Passive surveys of adult mosquitoes exploit their habit of actively self reporting when they attack people. In the case of Dar es Salaam, our preferred method for capturing human biting mosquitoes, namely the CDC light trap placed beside an occupied bed nest, did not work in this context. So for the duration of the initial pilot programme, we had to monitor adult mosquito densities with the controversial human landing catch.
Skip to 2 minutes and 32 seconds Although this tent trap caught less mosquitoes per night of sampling than human landing catch, its convenience and comfort allowed more affordable and intensive sampling over many more nights per month and locations per ward. Both methods indicated that the epidemiological impact of larval source management was achieved through modest reductions of the most important malaria vectors from the Anopheles gambiae complex plus far more impressive impacts on the most efficient malaria vector Anopheles funestus and upon secondary vectors like Anopheles ziemanni.
Monitoring & surveillance
This video case study is brought to you by Professor Gerry Killeen, AXA Research Chair in Applied Pathogen Ecology, University College Cork and the below supporting article was written by Erin Foley, Assistant Trial Manager, ARCTEC, London School of Hygiene & Tropical Medicine and Dr Thomas Ant, Research Associate, Centre for Virus Research, University of Glasgow.
Monitoring and surveillance of vector control interventions are required in order to ensure that intervention programmes are on track and to provide information for future policies/strategies and to plan future resource requirements.
With respect to vector control programmes, monitoring refers to the continuous tracking of implementation and performance1. This involves checking progress against predetermined goals and targets, and the adaptation of activities in accordance with this. Monitoring the coverage and implementation quality of vector control interventions is essential to maintaining the effectiveness of a vector control programme.
Vector surveillance involves the regular and systematic collection, analysis and interpretation of entomological or snail distribution data for health risk assessment, and for planning, implementing, monitoring, and evaluating vector control 1. Vector surveillance is used to inform vector control initiatives to ensure that appropriate interventions are used where and when they are needed and to assess the effectiveness of ongoing vector control strategies. It can also provide early warning information that an outbreak of a vector-borne disease may be likely, and can thereby trigger a pre-emptive response, such as the launch of a vaccination campaign or the initiation of appropriate vector control measures.
As well as vector control interventions, the impact of social and behaviour change strategies should also be assessed by monitoring and surveillance of the target disease and measurement of adaptation of a behaviour change strategy within the population.
Although interruption of transmission and possible elimination of the target vector-borne disease is the end goal for many programmes, it is of critical importance to ascertain whether the strategies have been implemented correctly at the initiation of a programme.
The key entomological parameters frequently collected include an assessment of the vector species occurrence and density. Vector occurrence is the presence or absence of a species known to support the development of a given pathogen. Vector occurrence is a particularly useful measure in the case of invasive species which, by their very presence, may be sufficient to warrant the triggering of a vector control response. For more established vectors, the population density in a given area and its abundance relative to other species in a region may determine the level of risk and which species should be priority targets for vector control.
Collections of vectors may be through a variety of different methods depending on the species of interest. In a comprehensive mosquito surveillance program both the immature stages and adult stages are sampled. Sampling of mosquito larvae is often performed by systematic and routine examination of a variety of potential larval habitats. For Aedes species, several standardised indicators based on larval density have been developed for the purposes of arbovirus surveillance, and include the household index (the proportion of houses positive for Aedes larvae), the container index (the proportion of positive containers) and the Breteau index (the number of positive containers per 100 houses inspected). Such indices are useful measures for inter-location and temporal comparisons and allow for quantitative thresholds for action to be defined.
Adult collections are routinely carried out through trapping of host-seeking vectors at defined sentinel sites, and this allows for an estimation of the density of biting adults. The type of trap used will vary from species to species and can incorporate different baits. The CDC light trap, for example, uses a light source as the primary attractant, and is particularly useful in capturing night-biters, such as the malaria transmitting Anopheles mosquitoes. The BG sentinel trap has been designed to allow the incorporation of chemical lures, and is particularly useful at capturing day-biting Aedes species.
The density of a vector may be coupled to other parameters, such as the presence/absence or prevalence of a pathogen in a vector population. This can be a useful indicator as some pathogens can be detected in a vector population before its circulation in humans is noted (as is the case for some zoonotic infections, such as West Nile virus).
Once entomological samples have been collected, the specimens are usually identified using either specific morphological characteristics – such as wing patterns - or by using molecular markers, often based on differences in ribosomal DNA sequences. Pathogen detection methods vary depending on the type of pathogenic organism and how the sample has been stored; for example, malaria parasites are frequently detected using a Plasmodium antigen detection method that is able to identify the presence of malarial proteins in a mosquito sample, whereas arbovirus detection is usually based on sensitive PCR methods that are able to identify the presence of viral nucleic acid sequences.
Additional parameters often surveyed include an assessment of vector receptivity. The potential for transmission of a pathogen can vary significantly between separate populations due to local differences in behaviour. These may include the host preference of a target species (how often it bites humans), its peak biting time (linked to whether the species encounters humans when they are not protected by bed nets), and its preferred biting and resting locations (whether it likes to bite or rest indoors or outdoors). Such behavioural differences are important to assessing the level of risk and guiding the selection of appropriate vector control interventions.
A further critical parameter frequently monitored is the susceptibility of a target vector to specific insecticides. There are molecular tests that can be used to monitor the abundance of resistance-associated mutations, for example the kdr allele that is widespread among Anopheles populations in Africa and provides resistance to DDT and the pyrethroids used in treated bed nets. However, as resistance may arise through a variety of potential mechanisms, surveillance performed by laboratory exposure of captured wild insects to defined doses of an insecticide can give a more functional assessment of the susceptibility of a population. Because of the importance of insecticides in controlling malaria, surveillance of Anopheles populations for resistance forms a key component of entomological surveillance of Anopheles species in Africa and other malaria-endemic regions. Several standardised functional tests, also known as bioassays, have been developed for use in the field by both the WHO (the WHO-tube test) and CDC (the CDC-bottle bioassay).
Data and evidence-based decision making
At a national level, evidence-based decision making requires not only entomological data but also epidemiological and intervention data. These data should be linked in order to stratify the risk of transmission. The WHO GVCR document states that such linkages can be supported through the use of a single, flexible data storage system to collate, validate, analyse and present aggregate statistical data required for vector control planning and implementation 1.. An example of one such data source is the Global Vector Hub, which we discussed earlier this week.
The capturing of routine quantitative data generated by a vector control programme is required to carry out the monitoring of these strategies effectively, and involves the input of programme records. To ensure the quality of these data, it is important that they are subject to entry by multiple operators, and cleaned (i.e. checking for discrepancies) before uploading to any data repository. Data must also be provided in a timely fashion in order to flag any potential issues that may arise.
Data records and the maintenance of these are important tools in defining and monitoring indicators which measure the programmes progress.
Other data outside of the public health sector should also be collected as part of the monitoring and surveillance process. This may include, but is not limited to, meteorological data, urban planning and sanitation data. This information can be used to identify or predict changes in the vector population or the risk of disease transmission, allowing for adaptations and prioritization of vector control interventions and can provide planners with a lead-time to knowing when interventions should be implemented or scaled up. The WHO GVCR document states that monitoring of human demographic and socioeconomic changes is also imperative given the association of vector-borne diseases with societal factors such as unplanned urbanization and migration.
Monitoring and surveillance in vector control
Vector control programmes across the globe must also be aware of the entomological and vector-borne disease situation in neighbouring countries and localities and more broadly in their regions (and indeed globally). This knowledge of public health trends allows for vigilance against the potential threat of pathogen importation, the emergence of insecticide resistance, or invasion of a new vector species of re-emergence of a previously controlled infection. A case study for the malaria control programme in Dar es Salaam is available in the ‘see also’ section of this step. Communication is also a key component of monitoring programmes, and is essential in order to promote collaboration and sharing of experiences. WHO has established a global database on insecticide resistance in malaria vectors based on reports from national programmes through WHO country and regional offices 2.. These data are managed at regional and headquarters levels and are used to track this biological threat in order to inform policy updates, with frequent reporting in technical fora and regional or global reports such as the World Malaria Report, and collation in the WHO Global Health Observatory 1..
Monitoring, surveillance and evaluation are the core responsibilities associated with vector control programmes. Adequate infrastructure, capacity and human resources are required at national and subnational levels to support these activities 3. A fundamental part of this is that the operational structure involved is both well-defined and robust. The aim of this is to support systemic surveillance of vector species, and is integral to ensure a proactive response to any arising issues or new concerns with both the programme itself, and the wider environment. This may be facilitated by the provision of ongoing training, for example in the monitoring and management of insecticide resistance, as we covered in Week 5 of this course.
Evaluation of vector control programmes, their processes and outcomes is necessary in order to document whether the activities and interventions implemented are leading to or have led to the desired impact on public health and whether programmes have been run to time and within the allowable resources. Evaluation is required both periodically and at the end of the programme. National surveillance information/WHO reports may be sufficient to establish whether a change has occurred in the target setting, however further research and data collection may also be necessary.
Basic analysis such as plotting case numbers over time may be sufficient to measure outcomes of a vector control programme. However, if improvements are not notable, or there are too many confounding factors, further statistical testing may be carried out in order to infer causality. Evaluation data may include household surveys (in direct comparison with the surveys conducted prior to programme introduction) and also health facility surveys (reporting on several factors including proportion of cases of suspected malaria or other disease pre-programme when compared with post-programme cases). The measurement of impact may occur at the conclusion of a programme with a defined end-date, or in the case of a continuous programme, this may be assessed at regular intervals (i.e. monthly, yearly etc.)
Monitoring and surveillance must be ongoing throughout the control programme, and it is recommended for assessments/analysis to be carried out at multiple timepoints, for example; one ‘midterm’ assessment roughly 2/3 into the timeline, and one final evaluation when the programme is completed, (provided that the programme has a definitive end). In longer-term programmes, plans for fixed-point follow-ups should be made. Following final evaluation, it is important to continue monitoring in order to prevent resurgence and identify the potential for outbreaks/epidemics. This must be rigorously maintained. Where programmes are successful, this monitoring must be continued for a fixed period of time (in malaria, this is a period of at least 3 years) in order for the country in which this programme is implemented within to qualify to receive WHO elimination certification.
The evaluation of vector control programmes will be covered in greater detail in the next step.
© London School of Hygiene and Tropical Medicine 2020