Reply to letter by Fiona Barlow

Fiona Barlow raises interesting points that largely relate to outcome measures, i.e. things one might evaluate in clinical research (for example the feeling of calm, the sense of well-being, etc.). It is important to point out, I think, that outcome measures and study design are different matters.

Let’s assume we have a way to measure qualities like calmness or well-being (and I’m sure that there are several instruments to quantify these), and let’s assume we have a therapy that supposedly maximises these qualities. What would stop us from conducting an RCT of that therapy testing its effect on these qualities? And which other study design would generate results that are more definitive in terms of causal inferences? Surely not qualitative research! I agree with Fiona Barlow, ‘RCTs are excellent to measure that which is measurable.’ I disagree, however, that they ignore human benefits, subtle or holistic effects. True, it would be ‘arrogant to define effectiveness in terms which suit the medical profession or science’. But who is doing that? The scientists who spend their time developing and validating patient-centred outcome measures such as VAS, quality of life questionnaires, hundreds of symptom scales, measures of activities of daily life, etc., etc., etc.?