Chair: John Varigos, PPD Developments, Melbourne, Australia

Paul Newman

CM Reid*, M McMurchie, F DeLooze, D Gleave, P Beckinsale, J Newberry, M Nelson and L Wing, on behalf of the ANBP2 investigators

W Hague*, S Magrie and RJ Simes, for the LIPID Group

Anthony Rodgers

Gemma Ritchie

Linda L Reaby

3.1

Paul Newman

Large trials are necessary for the detection of modest but clinically important treatment effects. The scale of such trials, for example from 10 000 to 40 000 subjects, poses several challenges for effective recruitment, but also provides opportunities to apply recruitment strategies that would not be appropriate in smaller studies.

The use of occupational and targeted screening, registers of trials, clinicians and patients, and mass media or direct mailing to access patients has been shown to effectively provide a large patient reserve to support eligibility screening.

Key success factors that are necessary for recruiting to large trials include meticulous planning and pilot testing of recruitment methods, monitoring of patient accrual with respect to a priori targets, regular feedback of recruitment progress to recruiting centres, and the development of a risk management plan to respond to failure to achieve targets.

Detailed understanding of the stages in the recruitment process allows the identification of steps where potential patients are lost to the study, and provides a framework for the development of alternative tactics to enhance recruitment.

The implications if a large trial fails to achieve recruitment targets include cost escalation and delays, inability to detect subtle treatment effects and the failure to meet ethical responsibilities to patients and clinicians.

Large trials that have recruited patients successfully have employed an enthusiastic, committed and talented staff, who can apply initiative and adaptability to recruitment problems.

3.2

CM Reid*, M McMurchie, F DeLooze, D Gleave, P Beckinsale, J Newberry, M Nelson and L Wing, on behalf of the ANBP2 investigators

ANBP2 is a cardiovascular outcome trial of the treatment of hypertension in the elderly, which is being conducted entirely in general medical practices across Australia.¹ No study of this size and nature has been undertaken in Australian general practice and the response of general practitioners to becoming involved in long-term cardiovascular research was unknown.

Academic departments and divisions of general practice were approached to support the project. General practitioners were approached by letter of invitation, and once an expression of interest was made, the GP was contacted by a regional medical coordinator, either at a face-to-face meeting or by telephone.

For each subject randomised to ANBP2, payment of $100 was made to the GP. This payment was based on the covering the difference between the Medicare rebate and the schedule fee for subjects attending for a minimum of 12 visits over the recruitment and the five-year monitoring phase of the study.

The visit costs for ANBP2 subjects were covered by the Health Insurance Commission Medicare scheme under an agreement established for this project.²

At the close of recruitment to ANBP2 at 30 June 1998, 1938 GPs from 950 practices had registered as investigators in the project.

A total of 62 divisions of general practice were approached to support the study in five mainland states, with 39 (63%) participating.

States varied in the extent of divisional involvement (range: 185–100), with Victorian divisions being the most involved and New South Wales divisions being more reluctant to participate.

Thirty divisional or promotional dinner meetings were held, with 56% (368 of 658) of those attending registering as investigators. Of the 8098 GPs who were sent a letter of invitation to participate, 17% (1578) expressed interest in the study and eventually enrolled as investigators; the proportion ranged from 8% in Queensland to 28% in New South Wales.

Of the GPs who had a personal face-to-face contact with the regional medical coordinator, 96% (696 of 724) registered in the study.

1. Management committee, Australian competitive outcome trial of ACE inhibitor and diuretic based treatment of hypertension in the elderly (ANBP2). Objectives and protocol. Clin Exper Physiol Pharmacol 1997; 24; 188–192.

2. Ried CM, Graham D, Wing LMH, on behalf of the ANBP2 Management committee. A new paradigm for funding for cardiovascular outcome research in general practitioners in the 2nd Australian National Blood Pressure Study. Med J Aust 1998; 169: 345–350.

3.3

W Hague*, S Magrie and RJ Simes for the LIPID Group

The Long-Term Intervention with Pravastatin in Ischaemic Disease (LIPID) trial was a multicentre, randomised, double-blind, placebo-controlled study to determine whether long-term cholesterol reduction could reduce coronary mortality in patients with coronary artery disease. LIPID was the largest randomised trial to be conducted in Australia and New Zealand, and recruitment to this study was very successful.

Recruitment commenced in April 1990, and over 9000 patients were randomised from 86 centres by close of recruitment in December 1992.

The recruitment strategies adopted by the group included:

• invitation of keen and successful investigators
• development of a comprehensive investigator's manual
• organising investigator recruitment meetings
• provision of numerous study aids to the centres
• maintaining contact with the centres via the telephone
• regular monitoring of recruitment progress and strategies (with subsequent feedback to the centres via the study newsletter and meetings for both the investigators and study nurses).

Early in the study (January 1991) a survey of 61 centres was conducted to identify successful recruitment strategies. Factors identified as important at each centre by univariate analysis were:

1. the study nurse being employed for longer hours (P<0.001)

2. the use of a coronary care register to identify eligible patients (P=0.001)

3. a systematic recruitment plan with targets and timetable (P=0.02)

4. the invitation of patients by both a personal letter and follow-up phone call (P=0.09).

The study nurse being employed for longer hours remained significant in the multivariate model. Other factors considered important were regular contact with the patients' usual doctors and adequate funding of centres.

3.4

Anthony Rodgers

Clinical trials typically require several thousand participants to reliably assess prevention of serious but relatively rare events, such as fatal or life-threatening complications after hip fracture. Trials of this size require collaboration between large numbers of orthopaedic centres if they are to be completed in a reasonable time.

The Pulmonary Embolism Prevention (PEP) trial of low-dose aspirin randomised 17 500 patients undergoing major hip or knee surgery from 150 centres in five countries between March 1992 and July 1998.

In all stages of the trial, success was in large part due to the efforts of the principal collaborating nurse and surgeon, and other orthopaedic ward staff. Particular challenges in the setting up of the trial included extensive and varying regulatory approval, poorly established use of national mortality registers in some countries, lack of previous research experience and busy clinical workloads.

However, collaboration was aided by the fact that there was widespread interest in the research question, participation did not require any change to usual management practices, and toll-free telephone randomisation and a simple one-page outcome form each took only a few minutes to complete.

Recruitment was improved by factors including well-organised and committed principal nurses, ongoing staff training, annual collaborators’ meetings, regular communication and target setting.

Only minimal financial recompense for staff time was available and many potentially eligible patients could not be recruited because of the lack of staff time to dedicate to research.

Clinical demands on already overstretched ward staff increased during the course of the trial. Nonetheless, there was considerable enthusiasm for further collaboration.

3.5

Gemma Ritchie

3.6

The Australian/New Zealand Breast Cancer Trials Group Consumer Advisory Panel

Mortality rates for breast cancer have fallen consistently—about 1–2% per year—since 1990. This is in a large part because of treatment advances that have been shown to be beneficial by clinical trials.

However, in Australia, only about 3% of available women are participating in breast cancer clinical trials.

The Consumer Advisory Panel is working to develop new strategies to improve breast cancer clinical trials accrual, both in individual clinics and nationally, and to participate in the review of new research protocols. The intent is not to turn panel members into scientists, but to give them a knowledge base to take part in scientific discussion from the consumer’s point of view.

The presentation will focus on the immediate and long-term goals of the panel and strategies that have been identified to achieve the stated goals.

It is envisaged that consumers and scientists can create a powerful synergy that will hasten achievement of a mutual goal: ending the breast cancer epidemic. This will happen by thinking, talking, and planning together in an atmosphere of understanding, respect and shared commitment.

11 November 1999