Page 59 - Htain Manual

T
here is an increasing interest in including patient-reported outcome measures

(PROs) in clinical studies specifically those in tandem with economic evaluations

or as part of an HTA. PROs include measures like patient satisfaction and their

health-related quality of life (HRQoL) and these capture certain treatment effects that are not
captured by the main clinical outcomes. Moreover, since quality-of-life (QoL) measures focus

on treatment effects that primarily impact the patient’s well-being, their relevance becomes

all the more important for HTA studies. QoL measures are thus divided into two categories.

• Disease-specific, or condition-specific measures: These concentrate on the main QoL
impacts of a particular disease. E.g., the EORTC instrument for cancer, CatQuest for

Cataract

• Generic measures: These do not focus on the impacts of a particular disease. Rather,
they consider a broad range of dimensions of quality of life that, in principle, could be

impacted by any disease, including physical function, mental well-being, social

function, and pain. The most widely used measure of this type is the EQ5D (-3L or 5L)

What is a good measure for health outcomes?

An important characteristic required for a measure of health is that it should be able

to compare changes across diseases and interventions. Now, as a cost-effectiveness analysis
looks to compare the costs and effects of alternative interventions for one disease it does beg

the question; why is comparison important? Consider that there is one intervention with a

higher cost but with better overall health outcomes as compared to other interventions. The
dilemma that arises now is whether to invest in such an intervention or not? To decide one

would need to know if the additional health benefits are large enough to merit the additional

expense that the costlier intervention provides.. Thus, even though the evaluation might

focus on a single disease, the resource allocation and service outcomes needs to be looked at

with a comparative lens.

Another characteristic required is that the health measure have a scale with interval

properties. The reason for this is that firstly, we need to know by how much the health change

has occurred, be it for better or for worse. Secondly, these health changes need to be related

to the monetary changes to be scaled either up or down. In this, the scale may be binary,
ordinal or cardinal depending on the conditions and requirements of the study.

47