Causal inference and the design of clinical trials in the community environment: a pilot study of allergen-reduction for the amelioration of childhood asthma.
Caldwell, Brent Ower
This thesis tested the hypothesis that it was possible to successfully adapt to the New Zealand community setting, the study design of an American allergen-reduction trial. The American study reported by Morgan and colleagues translated the gold-standard scientific method for testing causal associations, so that it could be taken from the ideal world of the laboratory, and applied in the ‘real life’ American community domestic environment. This thesis elucidates the key components of the scientific proof of a causal association, and outlines the issues involved in the adaptation of the American study in order to incorporate these key components, and take account of the relevant differences between America and New Zealand (such as, the kinds of allergens, nature of domestic houses, and cultural differences). The major adaptation was the development of placebo interventions because Morgan and colleagues did not take account of the placebo effect. A systematic review of allergen-reduction trials in childhood asthma is presented, and an assessment is made of the degree to which the key components of the scientific proof of causality were able to be included without compromising their integrity, in both the New Zealand adaptation, and similar studies reported in the literature. The thesis concludes that because of flaws in the designs of research performed to date (such as absence of a control group, or lack of a placebo, or inadequate randomisation protocols) there is insufficient evidence for or against the allergen-reduction hypothesis. This thesis makes a contribution by outlining the key study-design components that any future study must posses in order to scientifically test the allergen-reduction hypothesis. In the process of reviewing the literature, and critically analysing it, it became apparent that it was necessary to take a step back, to take a wider view of concepts and assumptions that lie prior to, and underpin, the American study and allergen-reduction research in general. This thesis explores logic and causality and their role in scientific studies of allergen-reduction, and points to reasons why research has been unable to provide definitive answers, and identifies the key features that any future study must possess, in order for it to conclusively accept or reject the allergen-reduction hypothesis once and for all. Research in this field to date has paid too little attention to theory, and this thesis makes a contribution by explicating the relevant theories of logic, causality, and immunology which must inform the design of a study if it is to have any chance of delivering interpretable and useful results. The outcomes of the pilot study of the New Zealand adaption of the study design of Morgan and colleagues are outlined, along with a critical discussion about the lessons learned from the pilot. The thesis concludes that extensive changes are needed to the pilot study design in order for it to have scientific validity, and to ensure it is acceptable to the study-participants and to the asthmatic children in the community to whom the results of the study should be applied.
Advisor: Howden-Chapman, Philippa; Crane, Julian
Degree Name: Master of Public Health
Degree Discipline: Public Health
Publisher: University of Otago
Keywords: Asthma; Allergen reduction; Community trials; Allergen
Research Type: Thesis