1-3-7 strategy in China and its adaptations
China launched its malaria elimination programme in July 2010 with a plan to achieve elimination by 2020 [
21]. China’s 1-3-7 RACD strategy had time-bound targets for case reporting, investigation and foci response activities. The “1-3-7” refers, respectively, to reporting of malaria cases within 1 day using the web-based China Information System for Disease Control and Prevention (CISDCP); their confirmation by double reading of slides by expert microscopists, and, where possible, quality assured PCR at a provincial laboratory, and investigation and classification is completed within 3 days; and the appropriate public health response to prevent further transmission to be done within 7 days [
22].
Focus investigation and action are carried out within 7 days irrespective of whether a case is classified as local or imported. The area around a case (the “focus”) is investigated to evaluate the risk of local transmission. Different actions, triggered based on the results of the investigation, are completed within 7 days. RACD is done based on the classification of a focus; in “inactive” (areas that do not support transmission due to the absence of vectors during the non-transmission season or without capable vectors) and “pseudo” (imported cases reported in a malaria-free area) foci, RACD screening is carried out in contacts of the case (‘‘hot populations’’), such as co-workers who travelled to the same area. Where viable vectors are identified and ecological conditions are suitable for malaria transmission, the focus is classified as an ‘‘active focus’’, and more intensive RACD (up to 200 neighbours) and vector control are initiated. RACD is conducted using RDTs for immediate results, with filter paper blood spots collected from all persons screened for later molecular (PCR) testing to detect low-density infections that may be missed by RDTs [
9,
16,
20]. If local transmission is possible or confirmed, targeted action to seek out other infections and reduce the chance of onward transmission is completed within seven days [
22]. The performance of the strategy has improved from the control to the elimination phase in terms of higher percentages of compliance to the “1-3-7” time lines [
21,
22].
China’s 1-3-7 strategy or its variations have been adopted and adapted in many other countries [
8,
23]. Along the China-Myanmar border, the approach was well executed except for the “3” indicator, which was 96% accomplished on average in the 18 border counties. While acknowledging the need for a well-planned and executed surveillance system for malaria elimination, given the few malaria cases RACD detected, the authors state that there is no evidence to suggest that it was effective, and if effective, the extent of its effectiveness [
24].
In the Asia Pacific region, the practice of case investigation varies widely, the trigger typically being a single case report or a defined threshold of multiple cases [
4]. It was reported that case investigation is part of surveillance activities where a broad array of demographic data is collected using different definitions for imported cases. The spatial range of screening varies from a specific number of households to an entire administrative unit (e.g., village) but the optimal radius is unclear [
4]. The strategy is labour intensive and expensive; in addition, the common detection methods used, microscopy or a rapid diagnostic test, may miss low-density infections that are still capable of transmitting malaria infections [
4].
Screening tests used in RACD
RDTs were the commonest test used, the others being microscopy, DNA based polymerase chain reaction (PCR) and loop-mediated isothermal amplification (LAMP). In this review, 10 studies used RDTs, 7 used PCR and 18 used microscopy. LAMP was not conducted in any of the countries as part of routine investigation; the three instances in which LAMP was used were only for research purposes (Table
2). The most widely used diagnostic technique in most of the countries is microscopy.
Table 2
Methods used for screening and case confirmation by different countries
China | | X | X | | |
Cambodia | X | X | | | |
Myanmar | X | X | X | | |
Viet Nam | X | X | | | |
Eswatini | X | X | | X | |
Senegal | X | | | | |
Colombia | | X | X | | X |
Indonesia | | X | | X | |
Ethiopia | | X | X | | |
Kenya | | X | X | | |
Zambia | X | X | | | |
Thailand | X | X | X | | |
Bhutan | | X | | | |
Malaysia | | X | | | |
Nepal | X | X | | | |
Philippines | | X | | | |
Korea | X | X | X | | |
Solomon Islands | | X | | | |
Sri Lanka | X | X | | | |
The detection limit of microscopy or RDTs is typically 100 parasites/microlitre and in low endemic settings, a high proportion of asymptomatic infections fall below this threshold [
40,
41]. Outside of research settings, more sensitive detection methods, such as PCR are impractical due to high cost, sophisticated training and resources required, and log turnaround time (several hours). The loop-mediated isothermal amplification (LAMP) provides the sensitivity of PCR with fewer requirements, but its costs and cost effectiveness in RACD is unclear. However, testing is not at point-of-care and requires a laboratory, but the assay is simple, does not require sophisticated equipment and can be performed within a day [
10,
42].
In Eswatini, compared to RDT, LAMP showed a threefold and 2.3-fold higher yield to detect infections (1.7% vs 0.6%) and hotspots (29.7% vs 12.7%), respectively. Hotspot detection improved with ≥ 80% target population coverage and response times within 7 days [
10].
In a Namibian study, the sensitivities of RDTs and LAMP compared to nPCR were 9.30% and 95.50%, respectively, and specificities were 99.27 and 99.92%, respectively; LAMP carried out on collected RDTs had a sensitivity and specificity of 95.4% and 99.9% compared to nPCR carried out on drop blood spots (DBS). LAMP performed equally to nPCR for the identification of
P. falciparum infections [
43].
Among 2802 persons in Cambodia who were screened by PACD, 33 cases were detected by PCR (6 by RDT) (1.07%). Subsequent RACD activities among 273 persons yielded 3
P. falciparum cases (1.1%) by PCR (0 by RDT) [
44].
In Zambia, testing by PCR revealed a
P.falciparum gametocyte rate of 2.4% (2/87) among index case household members and 0% (0/141) among other contacts (p = 0.145) [
9]. PCR identified four cases which were missed by routine microscopy in a study in Thailand [
34]. Four rounds of microscopy-based RACD in the Amazon-basin of Brazil identified 49.5% of the infections diagnosed by qPCR, comprising 76.8% of the total parasite biomass circulating in the proximity of index households. Control households accounted for 27.6% of qPCR-positive samples; 92.6% of them were from asymptomatic carriers beyond the reach of RACD [
33].
Challenges and limitations of RACD
Even though countries have successfully implemented various RACD techniques, there have been many challenges. A major challenge has been adhering to timelines. Achieving the three-day target in China was difficult due to logistic reasons such as sample collection, transportation and expedition by hospital staff to local CDC staff. Since PCR confirmation takes place in centralized laboratories, obtaining results and household investigations takes more than 3 days. Although household investigations should be done by local CDC’s staff, very often it is the hospital staff that does the investigations [
22].
Reduced capacity due to limited diagnostic skills, shortage of primary healthcare staff and decreasing vigilance of malaria cases have been reported by field staff in China as a challenge for case reporting within 1 day. Lack of knowledge on malaria and its early presentation also hampers early detection. Even though RDTs are easier to perform, they are sometimes not available at primary healthcare centres. Despite a mobile-phone based short messaging alert system being in place to notify a case to the local Chinese Centre for Disease Control and Prevention [
8], China’s web-based reporting system begins at township level and village clinics cannot fully participate in the surveillance system [
45].
Challenges have been posed at case investigation level as well. Complexity of the procedures, difficulties in transportation, limited working time and other individual aspects may lead to delays in case confirmation. Old equipment coupled with limited experience of health workers hindered correct diagnosis and species identification. Case classification is done based on travel history alone [
8] and is sometimes problematic due to incomplete travel histories. Even though, genotyping is helpful in distinguishing the geographical origin of the infection, standardized genotyping methods are not yet defined and unlikely to be completed within the three-day window [
8]. Quality control of case investigations is difficult and sometimes accuracy and classifications are doubtful [
45]. When PCR is used, confirmation of all networks to detect sub-microscopic infections within 7 days is difficult to achieve [
8].
Foci investigation in 7 seven days is especially difficult during the transmission season when locally transmitted case numbers are highest and determination of foci is the most difficult [
8], lack of human resources to make a professional judgement especially when a course of action needs to be determined due to differences of opinion and due to transportation issues in remote areas. Lack of a SOP often led to uncertainty in decision-making (for example, the radius to carry out RACD) [
45]; sometimes, personnel who carried out RACD activities did not have a clear idea about the minimum geographic screening radius [
31]. Along the China-Myanmar border, 65% of healthcare workers correctly stated that case investigation and RACD should occur in 3 and 7 days, respectively, 76% stated that all household members should be tested during RACD and 42% knew that it has to be conducted by visiting each household; knowledge of the minimum geographic radius to test around an index case household varied greatly [
46].
Declining motivation for detection of cases and conducting investigations, increasing number of returning migrant workers from malaria-endemic countries, and the complexity of establishing functioning multi-sectoral collaboration teams have also been identified as challenges [
45].
A pilot tool to evaluate RACD activities in Jiangsu province in China revealed that proper SOPs, organizational structures and documentation protocols for index cases and RACD were available in all healthcare facilities in the province [
31]; 100% of RACD activities were completed during the specified time frame [
31]. Implementation of the 1-3-7 strategy had varying success rates based on the area it is being implemented in.
A major limitation of the “1-3-7” strategy is when foci investigation are carried out by hospital doctors which may influence quality of data [
22]. Another challenge was the difficulty in consolidating data on focus investigations and response for imported cases due to lack of clear implementation guidelines for county staff [
22]. In China, IRS and RACD in potential active/active foci in foci investigation have been identified as challenges [
45]. For RACD, standard operating procedures were lacking, the radius within which RACD should be conducted was not specified (ranging from 50–200 m) and community acceptance was poor. Active screening of migrant workers and their peers upon return to China was identified as a lesson learned. Furthermore, screening the workers’ social networks, potentially through information provided by export labour companies, facilitated the detection of potential cases [
45].
In Kenya, a number of logistic challenges in conducting RACD have been highlighted as challenges [
25]. The household location of each clinical case had to be recorded and communicated to teams ready to conduct follow-up activities. The identification of the size of the focus, and thus the number of households to target, required a detailed understanding of transmission epidemiology. The authors concluded that following-up index cases helps to identify asymptomatic cases but is unlikely to have a major impact on transmission in a hyper-endemic environment. They also highlighted the fact that given the logistic challenges to achieve high coverage of RACD, control programmes need to weigh the increased chance to detect secondary cases vs. activities targeting the whole community, which might be more cost effective [
25].
Within the first year of implementation of a reactive screen-and-treat programme in Zambia, community health workers followed up 32% of eligible index cases; 66% of residents were at home in the index case households and 58% in neighbouring households [
13]. Forty-one neighbourhood households of 26 index case households were screened, but only 13 (32%) were within the 140-m screening radius as specified in the country’s guidelines [
13]. The authors conclude that with limited resources, coverage and diagnostic tools, reactive screen-and-treat will likely not be sufficient to achieve malaria elimination in this setting. They also surmise that high coverage with reactive focal drug administration could be efficient in decreasing the reservoir of infection and should be considered as an alternative strategy [
13].
Although RDT is a quick and easy method to screen, often it misses detecting infections; PCR has shown to be better at detecting cases. Microscopy is also used in a majority of countries. However, lack of training and unavailability of equipment can be challenging when using this method. In Zambia, microscopy is limited to urban areas and referral hospitals. In rural health centers and at community level, the majority of the cases are confirmed by RDT [
47]. In Myanmar, basic health staff and village health volunteers primarily use RDTs rather than microscopy for diagnosis of cases [
48]. The primary and sole use of RDT for diagnosis may miss low-density malaria infections and elimination targets [
22,
32,
45].
PCR is superior to microscopy in detecting malaria infections. PCR detected 2.2-fold higher
Plasmodium infections as compared to microscopy [
32]. In Kenya, PCR detected a higher number of cases than microscopy among members of the index case household, neighbours and persons living 500 m away from the index case [
25].
Hustedt et al. found 0.5% positive cases by RDT and 1.1% using PCR during RACD. In the households assessed as a comparison group, 0 cases were identified by RDTs and 25 were identified by PCR [
29].
Cambodia uses RDTs to identify infected persons. When RDTs were used to screen household members, only 1 case was found to be positive, whereas the number increased to 20 with PCR [
44]. When the target population was expanded to include co-exposed individuals, RDT detected 6 cases whereas PCR detected 11. When screening was done in a wider target population using PCR, the overall positivity rate increased to 3.9% (31 out of 785) with the highest positivity rate reported among co-exposed. There was a significant association between the test used and the detection rate, with PCR having the higher detection rate (6.8% vs 3.2%, p = 0.03) [
44]. Moreover, 75% of the samples from co-exposed individuals showed the same genotype as the sample from the index case.
In Colombia, the number of cases detected by PCR (93 cases) was significantly higher (5.6 times) than that detected by microscopy (16 cases) [
27] strengthening the need to re-evaluate the diagnostic methods used in different types of epidemiological settings. Unavailability of resources is a major problem in accommodating these techniques in RACD approaches. In Ethiopia, detection by PCR is largely confined to PCR facilities and not currently used in the national malaria control programme [
32].
Studies have shown that loop-mediated isothermal amplification (LAMP) is a better option that can be used to detect malaria infections. In an Indonesian study, one out of three who were detected positive for malaria by microscopy was false positive by LAMP. Of 1492 negative by microscopy, 5 were false negatives by LAMP. LAMP was more costly when compared to microscopy but more cost-effective for the detection of infections in scenarios with higher prevalence of infection using more sensitive diagnostics [
49].
The confirmed presence of sub microscopic infections represents an important public health problem, as the unidentified positive cases will not receive treatment, and will maintain transmission [
50]. RACD combined with parasite genotyping allows a better assessment of the transmission patterns [
50].
RACD, though widely implemented [
51], is operationally challenging requiring significant human resources, commodities, and time of an “on-call” team to conduct screenings in villages, often travelling long distances to reach remote locations [
49]. There are also limitations with the standard diagnostics used, microscopy or RDTs, to detect low-density infections. Highly sensitive diagnostics are available but the costs and cost-effectiveness of using them is unclear [
49].
There is little evidence available to support countries in deciding which methods to maintain, change or adopt for improved effectiveness and efficiency [
4]. The development and use of common evaluation metrics for these activities will allow malaria programmes to assess performance and results of resource-intensive surveillance measures and, may benefit other countries that are considering implementing these activities [
4]. There is very little information on how RACD programmes work in practice, if they achieve their goal, and if they are cost-effective, with little evidence to guide practice.