Clinical question Should immigrants and refugees undergo screening for T. cruzi infection? Population of interest Migrants and refugees Intervention Screening for T. cruzi infection Comparison Not screening Outcomes of interest Cases of vertical transmission resulting from transfusion or transplant Mild to moderate organ involvement (heart failure or LVEF < 50%, ECG abnormalities or dysphagia/constipation) Severe organ involvement (hospitalisation due to heart failure, need for a pacemaker or diagnosis of megasyndrome) Infection Indication for trypanocidal treatment Mortality Organ involvement of any type after diagnosis Quality of life

Is the problem a priority? Yes Research evidence and remarks from the Panel High prevalence of immigrants from Latin America, especially Bolivia. Between 50,000 and 70,000 persons in Spain are thought to have the disease. Autochthonous transmission occurs in Spain, mainly by vertical transmission. Early detection in children of infected mothers indicates a very high rate of cure. Early detection improves response to treatment. Antiparasitic treatment is more effective in children and adolescents than in adults. It also makes it possible to identify women of reproductive age who have not yet had children (or who have had children and may wish to have more) and in whom treatment blocks vertical transmission. There is the possibility of reactivation in immunosuppressed patients, with severe consequences. Approximately 30% of persons with Chagas disease develop heart disease, with subsequent morbidity and mortality and added cost for the health system. Chagas disease carries a risk of stigmatisation. Better knowledge and normalisation of the disease improves the negative perception that often keeps patients from being diagnosed and treated early.

How substantial are the desirable anticipated effects? Large How substantial are the undesirable anticipated effects? Moderate Research evidence and remarks from the Panel See Table 1 for a summary of findings for the outcomes of interest. The prevalence of the disease in the study population, the rate of vertical transmission, and the rate of complications—mainly cardiovascular complications—justify screening. In the case of indication for treatment, treatment of chronic infection in children younger than 19 years is very useful, while in adults, it should be questioned owing to the lack of quality studies. The efficacy of treatment in chronic infection may be questionable, especially for patients with moderate to severe cardiopathy. The marginal effect of treatment for adults in this phase of the disease could be due to considerable differences between studies [51-55,68]. While the tolerability of parasiticidal treatment is good in children, it is important to highlight the poor tolerability of treatment in adults: up to 44% of patients treated with benznidazole experience adverse effects, which have an impact on discontinuation of treatment (11% of all treated patients) [56]. The adverse effects of screening are minimal, i.e. those associated with blood sampling and the potential psychological consequences of having the disease, which may lead to stigmatisation. However, treatment with trypanocidal agents led to a high rate of adverse effects (on occasion up to 50%), some of which were severe. There is little evidence on the real cure rates in adult patients with chronic disease, which may be around 30% [68].

What is the overall certainty of the evidence of the effects? Low Research evidence and remarks from the Panel The studies evaluated for this question are observational (generally cross-sectional) with a prospective or retrospective design. The populations evaluated are heterogeneous. The quality of the results of these studies can be affected by selection bias. Furthermore, most of the combined analyses for the outcomes of interest revealed a very notable rate of statistical heterogeneity. However, this variability decreased when studies with well-defined populations (e.g. pregnant women) were evaluated together or when variability was attributed to estimations from few studies. Therefore, in most studies it was not considered a reason for reduced quality of evidence.

Is there major uncertainty about or variability in how much people value the main outcomes? There may be major uncertainty or variability. Research evidence and remarks from the Panel A study on Bolivian women with Chagas disease in Madrid revealed relevant aspects of their knowledge of and attitudes towards the disease. Most knew that the disease was transmitted by a vector, could be asymptomatic, and could lead to severe complications such as sudden death and heart failure. While there is some confusion over treatment, the women know that it is available [69]. The study included women from a country where Chagas disease was highly prevalent. There may be major differences between people from different countries.

Does the balance between desired and undesired effects favour the intervention or the comparison? The balance between desired and undesired effects favours the intervention. Research evidence and remarks from the Panel The expected benefit is centred on three basic aspects: Intervention prevents vertical transmission, with treatment offered after serology-based diagnosis in women of reproductive age from endemic areas. Furthermore, detection of the disease in a pregnant woman enables early diagnosis and treatment of the newborn, thus ensuring cure rates close to 100% during the first year, with very few adverse effects. Early diagnosis and treatment of the disease could prevent mainly cardiovascular complications, although evidence is scarce. The risks associated with treatment can be controlled to a large extent with appropriate follow-up at specialised clinics during the administration period. Screening of donors to prevent transmission through blood products and organs is addressed in current Spanish legislation (Royal Decree 1088/2005, dated 16 September 2005 (BOE-A-2005–15514))

Is the intervention acceptable to key stakeholders (population, professionals)? Probably Research evidence and remarks from the Panel The participants in the study on Bolivian women living in Madrid showed a certain degree of indifference and a lack of understanding of the risk of contracting the disease [69]. Participants who presented symptoms or had relatives with symptomatic disease expressed their concern for the more severe complications of the disease and the possibility of vertical transmission. In addition, fear of rejection by their social circle because of the disease was detected. In a similar study in the US, participants also expressed their concern about the lack of knowledge of the disease by health professionals and about more logistic aspects, such as difficulties reaching health centres or having sufficient time to attend an appointment. Of note, the participants in this study were at risk of social exclusion [70]. Stigmatisation because of the disease in specific groups should be taken into consideration when screening. Education, knowledge of the disease and easy access to diagnosis and treatment are necessary in order to overcome this problem.

Does the cost-effectiveness of the intervention favour the intervention or the comparison? They probably favour the intervention. Research evidence and remarks from the Panel An economic evaluation [11] evaluated the most efficient strategy for controlling Chagas disease among Latin American immigrants living in Spain. The study showed that not screening was more expensive and less effective and was dominated by other strategies from the societal perspective and from the perspective of the Spanish National Health System (SNHS). One of the most efficient strategies from both perspectives was screening of pregnant women, their newborns and first- and second-degree relatives of disease-positive mothers. From the SNHS perspective, moving from the strategy of screening ‘mother and newborn’ to the strategy of including ‘relatives of a disease-positive mother’ would involve a mean increase of EUR 301 per patient and QALY gained. In addition, moving from a strategy of ‘relatives of a disease-positive mother’ to ‘relatives of a disease-negative mother’ would involve a mean increase of EUR: 30,844 per patient and QALY gained. The parameters whose modification had the greatest effect on the results were efficacy of treatment of chronic disease and disease prevalence. We observed that it could prove efficient to extend a programme of this type to relatives of disease-negative mothers if the efficacy of treatment increased or the programme targeted the population at greatest risk. Results were recently published for an economic evaluation of systematic screening for Chagas disease in the Latin American population receiving primary care in Europe [71]. The evaluation consisted of a decision-making approach based on a Markov model in which the impact of systematic screening (all asymptomatic Latin Americans seen at primary care setting, treatment and follow-up of positive cases) was compared with screening and treatment of symptomatic persons. The model was based on a simulation in a cohort of 100,000 Latin American citizens, with a T. cruzi infection rate of 4.2% (95% CI: 2.2–6.8) and a maximum of 5 years of treatment in positive cases. The results of the analysis according to the probabilistic models revealed a total cost of systematic screening of EUR 32,163,649 (95% CI: 31,263,705–33,063,593) compared with EUR 6,904,764 (95%: CI: 6,703,258–7,106,270) for screening of symptomatic persons. The difference in QALY gained was 4,758.62 (95% CI: 4,618.42–4,898.82) for systematic screening compared with testing only symptomatic patients, with an incremental cost (ICER) of EUR 6,840.75 (95% CI: 6,255.75–7,425.75) for each QALY gained, treatment efficacy of 20% and EUR 4,243 for each QALY gained when efficacy was estimated to be 50%. Therefore, systematic screening in primary care would prove cost-effective, even in scenarios with a very low disease prevalence (0.05%). Another economic evaluation compared a decision-making model for screening the newborn or the mother against the alternative of no screening [72]. All of the models revealed screening to be the dominant (clinically superior and cost saving) situation compared with not screening. The ICER for the screening in the newborn was EUR 22 per QALY gained compared with EUR 125 per QALY gained in the case of not screening. In the case of mothers, the ICER was EUR 96 per QALY gained compared with EUR 1,675 per QALY gained for not screening. Screening strategies were cost-effective, even in settings with a higher prevalence of Chagas disease (0.9%) and lower risk of vertical transmission (2.2%), where the ICER increased by only EUR 37.50 per QALY gained for both strategies. It would be necessary to perform studies on the costs for the health system of a patient with symptomatic Chagas disease, especially when this involved heart disease.