Table 1 Summary of included studies

General Practitioners in Denmark

Treatment for gastro-oesophageal reflux disease

Drug prescription for individual patients

GPs who took part in trial significantly more likely to prescribe 'trial drug' proton pump inhibitors than other GPs.

Nurses in Canadian teaching hospital

Research working groups

Nurses attitudes to research, access to research, support of the use of research, use of research

Post intervention scores were higher for nurses involved in high or low research participation units than for nurses in the units that did not take part in these groups.

Gastrointestinal specialists

Management of Barrett's oesophagus

Adherence to international guidelines, especially in regard to tissue sampling

Adherence to guidelines increased after the trial started, particularly in trial centres.

Hospital doctors

Treatment for myocardial infarction

Medication received on discharge compared to admission

Doctors involved in trials change prescribing practice quicker than physicians in routine practice.

Surgeons in 3 adjacent counties in England

Discharge decisions following inguinal hernia and varicose veins surgery

Mean length of hospital stay

The mean length of stay decreased significantly in the trial area, with less effect in control areas.

Trial (high and low participation) and non-trial hospitals in CRUSADE (Can Rapid Stratification of Unstable Angina Patients Suppress Outcomes With Early Implementation of the American College of Cardiology/American Heart Association Guidelines), North America

Treatment for patients with high-risk non-ST-segment elevation acute coronary syndrome with unstable angina and non-ST segment elevation acute coronary syndrome

All cause mortality and adherence to guidelines

In-hospital mortality was lower in high trial participation hospitals compared to low participation and non-trial hospitals.

Greater adherence to guidelines in high trial participation hospitals compared to low participation and non-trial hospitals.

Acute care hospitals in Ontario, Canada participating in trials and those not.

Treatment for myocardial infarction

Hospital mortality and treatment information

Patients in trial hospitals fared better than patients in non-trial hospitals

Gynaecological departments in hospitals in Germany taking part in clinical trials and hospitals not taking part

Treatment for ovarian cancer

Survival from diagnosis to date of last follow up or death for up to two years after diagnosis and adherence to treatment guidelines.

Treatment in hospitals that did not participate in trials was associated with significant risk of death (adjusted for differences in stage of patients)

General practitioners in trial and non-trial practices in Denmark

Management of asthma

Physicians' adherence to international treatment recommendations and use of trial sponsors drugs

No change in adherence to guidelines in trial practices but there was an increase in prescription of sponsor drugs in trial practices

Hospitals in North America involved in the Survival and Ventricular Enlargement trial (SAVE) and in the Multicenter Diltiazem Postinfarction Trial (MDPIT) and hospitals not involved in these trials.

Treatment for myocardial infarction

Prescribing of angiotensin-converting inhibitors (ACE) at hospital discharge

No significant differences between trial and non-trial hospitals.

Cancer care facilities: community hospital, community cancer centre and teaching/research facilities in the USA

Treatment of patients with advanced stage laryngeal cancer

Chemotherapy and radiation treatment to patients

Higher percentage of patients with advanced-stage laryngeal cancer were treated with chemo-radiation in teaching/research hospitals than in community hospitals and community cancer centres

Apheresis units in major medical centres in Canada

Treatment of multiple sclerosis, thrombotic thrombocytopenic purpura, and myeloma cast nephropathy

Apheresis

Large increase in apheresis use, particularly outside trials, labelled 'jumping the gun' by the research team.

Centres taking part in ADEBAR trial of chemotherapy for women with breast cancer

Treatment for breast cancer

Changes to treatment strategies and patient care since participating in ADEBAR

Prior to ADEBAR, 63.2% of centres had not entered high-risk breast cancer patients into a clinical trial and 44.2% of patients received inadequate treatment by current standards.