Nermina Ferizovic | Health Economics Expert

During the process of cost-effectiveness analysis, the most relevant and appropriate health outcomes need to be identified and modelled to determine the impact of the proposed health intervention on these relevant and appropriate outcomes. These outcomes are largely identified from the published scientific literature or known natural history of the specific disease area. In cases where the published literature is uncertain or there is disagreement, it is common to seek validation of the proposed outcomes. This validation is in the form of clinical validation through either clinician interviews or advisory boards whereby clinicians are asked to confirm whether or not the proposed outcomes, reflected in the model health states, are reflective of the presenting disease. Although it is commonplace to seek clinical validation for cost-effectiveness analyses regardless of available evidence as this is deemed to add weight to both the structure and inputs of the cost-effectiveness analysis. However, is basing the health outcomes assessed in a cost-effectiveness analysis on published or clinically validated outcomes the most relevant and appropriate method by which to determine cost-effectiveness of a health intervention? Let’s consider the alternatives.

An experienced and specialist clinician would have first-hand knowledge of the disease area and would have treated a reasonably large number of patients with this disease. Thus, she would be aware of relevant health outcomes and the impact of current treatment on these outcomes as well as possessing vast knowledge of the disease natural history. However, the clinician does not experience the disease first-hand and may only have spent minimal time with patients. It is also likely that their validated outcomes may be based only on clinically relevant outcomes, in effect ‘classic textbook’ outcomes. On the other hand, the patient has first-hand experience of the condition through living with it on a daily basis and could, therefore, be best placed to highlight or identify the key health outcomes at specific health states of the disease. ‘Key’, in this instance, is defined as the outcomes that are of most importance to a patient with the disease. 

Let’s use neuromuscular diseases as an example to highlight the potential differences between clinicians and patients in terms of determining relevant outcomes. A patient with a neuromuscular disorder may deem that one of the most relevant and appropriate outcomes may be their ability to stand on their own two feet, essentially, weight bear. A clinician may dis-regard that as being irrelevant and insignificant, thus weight bearing would not be ‘validated’ for inclusion in a cost-effectiveness analysis either through published literature or advisory boards. A drawback of this is that the potential impact of a proposed health intervention aimed at these patients would not be assessed against this particular outcome.

Thus, should the outcomes modelled in cost-effectiveness analyses be based on clinical and published scientific evidence or should we be explicitly modelling outcomes that the patients deem the most relevant and assess the potential impact of a health intervention against these patient-derived health outcomes? In terms of welfare economics, this may perhaps provide us with the most accurate estimate of cost-effectiveness.