Development of a framework for transferring tests evaluated in secondary care to primary care settings: a Delphi study

Context: It takes an average of nine years to develop new diagnostic tests. These tests are primarily evaluated in secondary care (SC) before they are introduced into primary care (PC). Test evaluation in PC is costly and time-intensive. Transferring diagnostic tests from SC to PC however, is hindered by factors such as variation in sensitivity and specificity between settings.

Objective: To achieve consensus on prioritizing criteria for selecting tests that are used in SC and not yet in PC, but that have clinical potential in PC.

Study Design and Analysis: This Delph study comprises two rounds of feedback and input from a panel of diverse experts: clinicians, methodologists and test developers. A preliminary list with prioritizing criteria was developed by the steering group, based on a non-systematic literature search and expert opinion. Panelists rated the proposed criteria for inclusion on a five-point Likert Scale (strongly agree to strongly disagree), followed by general questions for feedback and input, reflecting a comprehensive approach. Consensus was determined at 70% agreement. Criteria were categorized as inclusion, exclusion, or reassessment for the subsequent round based on feedback and agreement levels.

Setting or Dataset: Primary care.

Population Studied: General practitioners (GP’s), test developers, specialists in SC, and methodologists from in- and outside the Netherlands.

Intervention/Instrument: Questionnaires in RedCap.

Outcome Measures: Consensus on prioritizing criteria.

Results: The preliminary list consists of several parts and themes. The first part asks future users to describe the diagnostic situation, e.g. the target patient population, differential diagnoses and potential for improvement in the diagnostic pathway. This is followed by the themes with the criteria to be rated. The proposed themes are “Burden” (e.g. epidemiology and costs), “Clinical Pathway,” (e.g. test role and potential) “Clinical Performance” (e.g. sensitivity and added value), “Impact” (e.g. improving management), and “Feasibility” (e.g. safety and support). The final list will be available this summer.

Conclusions: Based on the results of this Delphi study, we will develop a framework that researchers can use to prioritize tests used in SC that have clinical potential in PC. GPs and their patients will benefit from earlier availability of suitable tests, positively impacting the diagnostic pathway and thereby improving clinical management in PC