We are aware that the methodology, interpretation and evaluation of research on spirituality and mental health are complex matters which are not without their controversies and difficulties. The measurement of spirituality (which is to be contrasted with religiosity in this regard) is necessarily subjective, and easily prone to confounding with psychological variables. Alongside healthy forms of spirituality there are pathological forms of spirituality, and harmful forms of religion and religious practice, which are clearly detrimental to mental well-being. ^{Reference Crowley, Jenkinson, Cook, Powell and Sims1} Not surprisingly, therefore, there are negative as well as positive associations in the research literature charting the relationships between spirituality, religiosity and mental health. For this reason we did not suggest in our article ^{Reference Dein, Cook, Powell and Eagger2} that ‘research unequivocally shows an association between religiosity and well-being’ ^{Reference Hansen and Maguire3} but rather stated that ‘research demonstrates largely positive associations between religiosity and well-being’. Our cited reference in support of this contention was the Handbook of Religion and Health, a volume written by Harold Koenig and his colleagues, which reviewed 1200 studies that were critiqued according to methodology and outcome. ^{Reference Koenig4} We might also have quoted Koenig's more recent review in the Canadian Journal of Psychiatry, ^{Reference Koenig5} which reaches a similar evidence-based conclusion that, although there are undoubtedly unhealthy forms of religious and spiritual involvement, the usual associations are with better coping and healthier functioning.

The editorial by Dan Blazer in the same issue of the Canadian Journal of Psychiatry, ^{Reference Blazer6} cited by Hansen & Maguire, ^{Reference Hansen and Maguire3} does not suggest that there is an unequivocal association between religiosity and well-being. Nor does it claim that ‘The research findings are wildly contradictory and it would be unreasonable to draw any firm conclusion on the basis of current knowledge’. ^{Reference Hansen and Maguire3} Rather, Blazer summarises Koenig's review as showing that ‘studies to date generally support a positive association between (religion and (or) spirituality) and mental health’. However, Blazer does importantly acknowledge that this association is ‘a tough topic to research’. One of the reasons he gives for this is that it is difficult to be objective on matters about which we hold very deep beliefs. Blazer goes on to warn that ‘Neglect of the religious dimension, not to mention refusal to discuss religious matters with our patients, may seriously cut off meaningful communication and significantly undermine the therapeutic relationship’. He concludes that ‘even though the task is tough, neglect is even more difficult to justify’.

Clinicians, researchers and patients do hold very deep beliefs on these matters. It is for this reason that it is important not to rely only on impressions derived from clinical experience but also to refer to evidence-based research and reviews. If we cannot eliminate bias in our interpretation of these findings, we can at least minimise it.