Deakin University
Browse
mccaffrey-howaccuratelydo-2016.pdf (206.58 kB)

How accurately do consecutive cohort audits predict Phase III multisite clinical trial recruitment in palliative care?

Download (206.58 kB)
journal contribution
posted on 2016-04-01, 00:00 authored by Nikki McCaffreyNikki McCaffrey, B Fazekas, N Cutri, D C Currow
Context. Audits have been proposed for estimating possible recruitment rates to randomized controlled trials (RCTs), but few studies have compared audit data with subsequent recruitment rates. Objectives. To compare the accuracy of estimates of potential recruitment from a retrospective consecutive cohort audit of actual participating sites and recruitment to four Phase III multisite clinical RCTs. Methods. The proportion of potentially eligible study participants estimated from an inpatient chart review of people with life-limiting illnesses referred to six Australian specialist palliative care services was compared with recruitment data extracted from study prescreening information from three sites that participated fully in four Palliative Care Clinical Studies Collaborative RCTs. The predominant reasons for ineligibility in the audit and RCTs were analyzed. Results. The audit overestimated the proportion of people referred to the palliative care services who could participate in the RCTs (pain 17.7% vs. 1.2%, delirium 5.8% vs. 0.6%, anorexia 5.1% vs. 0.8%, and bowel obstruction 2.8% vs. 0.5%). Approximately 2% of the referral base was potentially eligible for these effectiveness studies. Ineligibility for general criteria (language, cognition, and geographic proximity) varied between studies, whereas the reasons for exclusion were similar between the audit and pain and anorexia studies but not for delirium or bowel obstruction. Conclusion. The retrospective consecutive case note audit in participating sites did not predict realistic recruitment rates, mostly underestimating the impact of study-specific inclusion criteria. These findings have implications for the applicability of the results of RCTs. Prospective pilot studies are more likely to predict actual recruitment.

History

Journal

Journal of pain and symptom management

Volume

51

Issue

4

Pagination

748 - 755

Publisher

Elsevier

Location

Amsterdam, The Netherlands

ISSN

0885-3924

eISSN

1873-6513

Language

eng

Publication classification

C Journal article; C1.1 Refereed article in a scholarly journal

Copyright notice

2016, American Academy of Hospice and Palliative Medicine