Check@Home: General Population Screening for Early Detection of Atrial Fibrillation and Chronic Kidney Disease

• Participation rate of the home-based screening phase will be defined as the number of persons who completed the home-based screening phase relative to the number of persons invited in the corresponding arm (intention-to-screen);
• Participation rate of the diagnostic screening phase will be defined as the number of persons who completed the diagnostic screening phase relative to the number of persons invited for the diagnostic screening phase in the corresponding arm (in participants invited based on an 'abnormal' result in the home-based screening phase, i.e. either confirmed elevated albuminuria or an irregular heart rhythm);
• Participation rate of the treatment phase will be defined as the number of persons who completed the care initiation phase (i.e., number of participants visiting their GP; or - in future phases of the screening - the diagnostic center) relative to the number of persons referred for treatment in the corresponding arm (intention-to-screen).

• Yield of home-based screening: number of participants tested positive (i.e., either confirmed positive urine test and/or abnormal heart rhythm test) relative to the number of persons participating in the corresponding arm (per-protocol analysis) and to the number of persons invited in the corresponding arm (intention-to-screen analysis);
• Yield of diagnostic screening: number of participants with (newly) diagnosed cardiovascular or chronic kidney disease, diabetes or risk factors, relative to the number of persons participating in the corresponding arm (per-protocol analysis) and to the number of persons invited in the corresponding arm (intention-to-screen analysis);
• Yield of treatment phase: number of participants completing the treatment phase relative to the number of persons participating in the corresponding arm (per-protocol analysis) and to the number of persons invited to the treatment phase in the corresponding arm (intention-to-screen analysis).

Effectiveness of the atrial fibrillation screening (the home-based screening program and follow-up strategy) will be based on the incidence of (fatal and non-fatal) ischemic stroke. This will be evaluated by comparing the event rates in the intervention group (the sample of 160,000 subjects invited for screening) and the control group (the sample of 160,000 subjects not invited for screening).

Effectiveness of the albuminuria screening (the home-based screening program and follow-up strategy), based on the incidence of the composite outcome of kidney failure events (defined as start of dialysis, receiving a kidney transplant, or death due to kidney failure without having started kidney function replacement treatment) and MACE (for albuminuria screening defined as cardiovascular mortality, stroke, and myocardial infarction, including CABG and PCI).

This will be evaluated by comparing the event rates in the population screened positive for elevated albuminuria and followed-up based on this result (i.e., participants in the standard strategy and other alternative strategies) and in the population screened positive for elevated albuminuria - at a later timepoint - and not followed-up based on this result (i.e., participants in 'alternative strategy B').

Effectiveness of the atrial fibrillation screening (the home-based screening program and follow-up strategy), based on the incidence of MACE (for atrial fibrillation screening defined as cardiovascular mortality, stroke, non-hemorrhagic stroke, and myocardial infarction, including CABG and PCI).

This will be evaluated by comparing the event rates in the intervention group (the sample of 160,000 subjects invited for screening) and the control group (the sample of 160,000 subjects not invited for screening).

Effectiveness of the albuminuria screening (the home-based screening program and follow-up strategy), based on each individual component defining MACE (for albuminuria screening defined as cardiovascular mortality, stroke, and myocardial infarction, including CABG and PCI).

This will be evaluated by comparing the event rates in the in the population screened positive for elevated albuminuria and followed-up based on this result (i.e., participants in the standard strategy and other alternative strategies) and in the population screened positive for elevated albuminuria - at a later timepoint - and not followed-up based on this result (i.e., participants in 'alternative strategy B').

Safety of the atrial fibrillation screening (the home-based screening program and follow-up strategy), based on the incidence of the composite outcome of (fatal and non-fatal) hemorrhagic stroke, other major bleeding events leading to hospitalization (including but not limited to intracranial bleeds, sight-threatening eye bleeds, and serious gastrointestinal bleeding), and all cause-mortality.

This will be evaluated by comparing the event rates in the intervention group (the sample of 160,000 subjects invited for screening) and the control group (the sample of 160,000 subjects not invited for screening).

An individual level simulation model will be developed to estimate the lifetime effects of the screening strategies for the detection of atrial fibrillation and chronic kidney disease by comparing the corresponding expected incidences of cardiovascular and chronic kidney disease, diabetes and their complications and endpoints with standard care. Incremental effectiveness will be calculated, expressed in QALYs gained for the different screening strategies. Costs will be calculated by multiplying resource use with the corresponding unit costs. The ICER between screening strategies and no screening, expressed in costs per QALY gained, will be calculated over a short-time (5-10 years after screening) and a lifetime time horizon. Finally, we will assess the chance that screening is cost-effective for a range of willingness-to-pay threshold values (range €20,000-€80,000 per QALY) and at which threshold for the screening costs screening would have a chance of ≥90% to be cost-effective.

Using registry data, we will compare event rates between the intervention group (invited for screening) and the control group (not invited for screening) 5 and 10 years after the screening. Currently foreseen is that the data on fatal and non-fatal cardiovascular disease (according to the MACE criteria; based on corresponding ICD-10 codes) and renal events (dialysis and transplantation; based on corresponding ICD-10 codes) will be obtained and compared. These data will be obtained by linkage with CBS (i.e., Dutch Hospital Data, Vektis, Zorginstituut Nederland), Zorgverzekeraars Nederland, Pharmo, National Heart Registry, and general practitioner registries. These data will be input for the health economic simulation model, replacing earlier estimates of treatment effectiveness and treatment adherence. The ICER and estimation of the likelihood of screening being cost-effective will be repeated using this new and more accurate evidence.

• False-positives: participants with an abnormal home-based test result, but normal test results at the diagnostic screening visit.
• False-negatives: participants with normal home-based test results, but abnormal test results at the diagnostic screening visit.
• Sensitivity: the ability of the home-based test to correctly identify participants with the disease.
• Specificity: the ability of the home-based test to correctly identify participants without the disease.
• Positive predictive value: the proportion of participants correctly diagnosed as positive (having the disease) out of all participants with a positive test result.
• Negative predictive value: the proportion of participants correctly diagnosed as negative (not having the disease) out of all participants with a negative test result.

Characteristics of responders and non-responders, and differences between them, will be analyzed using descriptive statistics per screening strategy. For health literacy descriptive statistical analyses will be performed on the HLS-Q12 questionnaire (per item and/or combined). Future exploratory analyses may evaluate whether health literacy is related to participation rate, health outcomes, and or characteristics of participants and non-participants (such as age and socioeconomic indicators of the neighborhood). In addition, facilitators and barriers for participation will be studied taking into account qualitative data on experiences, considerations and enabling and constraining conditions, collected both in participants and non-participants via in-depth interviews and/or observations.

Effectiveness of the atrial fibrillation screening (the home-based screening program and follow-up strategy), based on the incidence of each individual component defining MACE (for the atrial fibrillation screening defined as cardiovascular mortality, non-hemorrhagic stroke, and myocardial infarction, including CABG and PCI).

This will be evaluated by comparing the event rates in the intervention group (the sample of 160,000 subjects invited for screening) and the control group (the sample of 160,000 subjects not invited for screening).

Overall effectiveness of broader population-based screening strategies, focused on the detection (and potentially treatment) of heart failure, coronary artery disease, and type 2 diabetes in subjects aged 50-75 years. This will be evaluated based on participation rate, yield, and (cost-)effectiveness of the screening strategies, similarly to what is described in the previous sections, but then applied to screening on heart failure, coronary artery disease, and type 2 diabetes.