When Does Health Technology Assessment Need Deliberation? Evidence, Algorithms and the Limits of Consultation.

Health technology assessment (HTA) increasingly depends on formal methods such as cost-effectiveness analysis, QALYs, thresholds, structured evidence reviews and decision criteria. These tools are indispensable, but they do not remove the need for judgement. Decisions about health technologies often involve heterogeneous evidence, uncertainty, contested values, institutional constraints, professional responsibilities and consequences for both identifiable patients and anonymous others whose care may be displaced. In this article, we ask when deliberation is likely to be needed in HTA, and what distinguishes it from algorithmic reasoning, consultation and commenting. We argue that deliberation should not be defended simply because it is democratic, inclusive or procedurally attractive. Its value is conditional. Deliberation becomes useful where evidence does not combine itself, where values shape what counts as relevant evidence, and where recommendations must be justified in ways that can withstand disagreement. We distinguish context-free scientific evidence, context-sensitive scientific evidence, and colloquial evidence, and show why their interpretation cannot be reduced to evidence accumulation or modelled calculation. We then identify the circumstances in which deliberation is most likely to add value: (1) when evidence is heterogeneous, incomplete, contested or context dependent, (2) when values, outcomes, equity or fairness are disputed, (3) when implementation is uncertain, and (4) when professional ownership, public credibility or political trust matter. We argue that HTA does not need deliberation everywhere, but where judgement must be exercised, exposed and justified.