International Journal of Technology Assessment in Health Care

General Essays

Perspectives on the National Institute for Health and Clinical Excellence's recommendations to use health technologies only in research

Irfan A. Dhallaa1, Sarah Garnera2, Kalipso Chalkidoua2 and Peter Littlejohnsa2

a1 University of Toronto and St. Michael's Hospital

a2 National Institute for Health and Clinical Excellence


Background: The concept of using public funds to pay for healthcare interventions only when provided in the context of ongoing research is receiving increasing attention worldwide. Nevertheless, these decisions are often controversial and implementation can be problematic.

Objectives: The aim of this study was to investigate the views of United Kingdom stakeholders on the current arrangements for implementing “only in research” (OIR) decisions and to investigate how improvements might be made.

Methods: After an internal review of previous OIR decisions issued by the National Institute for Health and Clinical Excellence (NICE), deliberations by NICE's Citizens Council, and an international workshop convened by NICE and the United States Agency for Healthcare Research and Quality, thirteen key stakeholders and experts from academia, industry, government, and the National Health Service (NHS) were interviewed using a semistructured interview guide. Interview transcripts were subjected to a framework-based analysis using computer-assisted qualitative data analysis software.

Results: All interviewees endorsed the use of the OIR option. There was a high degree of consensus for several suggestions regarding how the use of the OIR option might be improved. For example, there was universal agreement that a formal process should be established to prioritize research needs arising from OIR decisions and that funds for publicly funded research projects should be channeled in a manner that would better motivate healthcare providers to participate in OIR-related research.

Conclusions: The findings of this study suggest several potential modifications of the OIR pathway in the United Kingdom and may also be helpful to health technology assessment agencies in other countries that already use or are considering using an OIR-like option to reduce the uncertainty inherent in health technology assessment.


We are grateful to the interviewees and to Amy Kelsey, Andrew Hoy, and Clifford Middleton at NICE for assistance facilitating this work. Irfan Dhalla received financial support from a Commonwealth Scholarship and the Canadian Institutes of Health Research.