Woody Caan1, Michael Cole. 1. Department of Public Health, Anglia Ruskin University, Cambridge, UK. wcaan@rsph.org.uk
Abstract
OBJECTIVE: To determine how often clinical research from doctoral degree programmes is unpublished and to see if one can characterise differences between those researchers who do or do not publish their work or between the institutions where they studied. DESIGN: Secondary research with data in the public domain: combining a national dataset of healthcare doctoral theses with related publications traceable in two widely accessible databases. Data sources The Index to Theses was searched for UK doctoral theses and the databases Web of Knowledge and SCOPUS searched by author name for literature derived from each thesis. Confirmatory details about student names or their doctoral programmes were obtained from university websites, if required. Eligibility criteria Search terms were chosen to identify only projects directly related to human health. Thesis records included 'clinical trials' or 'clinical research'. The awards were all doctorate-level degrees between 2000 and 2010. 'Related' publication in the Web of Knowledge or SCOPUS was interpreted very broadly to include publications with any research commonality with the degree project as recorded in the Index to Theses. RESULTS: A sample of 82 theses from 39 universities provided the initial material for a literature search. The Web of Knowledge contained more publications from more students than SCOPUS held. Of the doctoral research samples, 39 out of 82 theses (47.6%) produced no discernable publications. The 43 students whose research was published typically began publishing before completing their degree (mean time 1.3 years preceding their award). Not all types of doctorates were associated equally with publication: all 11 MD theses in our sample had related publications (100%). CONCLUSIONS: Research evidence associated with doctoral degrees is often left unpublished, across many settings. Sharing such evidence is a scientific and ethical duty. Supporting students in publishing preliminary work while they are still in doctoral programmes may be the most productive response. The large number of doctorates awarded in 'medical' disciplines every year represents a substantial investment in training and a resource for evidence-based medicine that should not be squandered.
OBJECTIVE: To determine how often clinical research from doctoral degree programmes is unpublished and to see if one can characterise differences between those researchers who do or do not publish their work or between the institutions where they studied. DESIGN: Secondary research with data in the public domain: combining a national dataset of healthcare doctoral theses with related publications traceable in two widely accessible databases. Data sources The Index to Theses was searched for UK doctoral theses and the databases Web of Knowledge and SCOPUS searched by author name for literature derived from each thesis. Confirmatory details about student names or their doctoral programmes were obtained from university websites, if required. Eligibility criteria Search terms were chosen to identify only projects directly related to human health. Thesis records included 'clinical trials' or 'clinical research'. The awards were all doctorate-level degrees between 2000 and 2010. 'Related' publication in the Web of Knowledge or SCOPUS was interpreted very broadly to include publications with any research commonality with the degree project as recorded in the Index to Theses. RESULTS: A sample of 82 theses from 39 universities provided the initial material for a literature search. The Web of Knowledge contained more publications from more students than SCOPUS held. Of the doctoral research samples, 39 out of 82 theses (47.6%) produced no discernable publications. The 43 students whose research was published typically began publishing before completing their degree (mean time 1.3 years preceding their award). Not all types of doctorates were associated equally with publication: all 11 MD theses in our sample had related publications (100%). CONCLUSIONS: Research evidence associated with doctoral degrees is often left unpublished, across many settings. Sharing such evidence is a scientific and ethical duty. Supporting students in publishing preliminary work while they are still in doctoral programmes may be the most productive response. The large number of doctorates awarded in 'medical' disciplines every year represents a substantial investment in training and a resource for evidence-based medicine that should not be squandered.