Background To allow patients to reflect about a decision to participate in a clinical trial, guidelines suggest a 24-h delay from when they are informed about the trial to when they give consent. In certain clinical settings, this is likely to hamper recruitment.
Method After oral and written information about the trial has been given in person, the patient signs the declaration of consent knowing that they will be asked again after 24 h whether they confirm or regret the decision. This procedure can be done by SMS. The investigators must document the response. The procedure was tried in a study in which the doctors were randomly assigned to receive a clinical communication skills course, and encounters with patients were videotaped before and after the course.
Results 553 patients were approached, 530 (95.8%) gave initial consent, eight of these later regretted their consent.
Discussion The low level of regrets suggests this is an acceptable procedure for patients.
Trial registration The RCT was registered before initiation – registration # ISRCTN22153332.
- clinical trial
- informed consent
- scientific research
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
The acquisition of informed consent when recruiting patients to clinical research represents a challenge; to convey much complex information, make sure that it has been understood and appreciated and that the patient makes an informed choice. This requires quite some use of time; however, there is often a time constraint because of characteristics of the condition under study or the organisation of the clinic. Therefore, 24 h of time for reflection before the patient makes the decision has been recommended.1 This severely hampers the possibility of recruitment in many clinical settings, particularly when research is about acute or subacute diseases that do not require hospitalisation. We introduce a practical solution to this dilemma, and report our experiences from a randomised controlled trial in which consultations were videotaped as part of an intervention supposed to improve doctors' communication skills.
Ethical research requires that the patient has not only received the necessary information about participation in a study, but has actually understood and appreciated it. There is evidence that as many as 40% of potential subjects do not understand essential elements of informed consent such as the risks or the right to withdraw, and this has led to the development of post-decision assessment questionnaires.2 The use of such questionnaires is still not implemented as a rule. Meanwhile, guidelines requiring that researchers give patients 24 h time to reflect increase the likelihood that patients make decisions they are comfortable with. This delay introduces the possibility to read, reflect and discuss with significant others the pros and cons of participation.
For several studies, however, this requirement imposes practical difficulties. Many studies require patients seen in outpatient clinics, and they would need a new appointment shortly after to complete the recruitment process. If they decide against participation, cancellation of this appointment is likely, leading to reduced efficiency in the clinic. The qualified person who recruits patients for the study might not have working schedules that fit. Collection of data needed for the study will often have to be postponed. All these obstacles might lead those who recruit not to approach all eligible patients.
The objective of our crossover randomised controlled trial was to study whether a communication skills programme for hospital doctors could improve their skills. To evaluate the skills, the participating doctors' encounters with patients in the hospital were videotaped before and after the intervention. The patient sample was convenience based, and patients received normal medical care, not knowing whether the doctor they met had been subject to training or not. The only extra risk patients were exposed to if they consented would be if a doctor was so distracted by the camera that it affected the quality of his/her work. However, videotaping of encounters also introduces extra stress for patients because they are aware that external individuals (researchers) will watch a potentially intimate situation. Therefore, the regional ethics committee required the recommended 24 h time for reflection for the informed consent to be valid. As we planned to film more than 70 doctors eight times in approximately 50 sites in the hospital, and wanted to have a fairly representative sample of elective and emergency encounters, the study would not be possible to conduct with this constraint.
As a solution, we suggested approaching the patients in the waiting room. If a patient, after having received oral information and read the written information, decided to agree to inclusion, data including videotaping of the encounter were collected. After data collection, the patients signed the information and consent form. The patients' cell phone number was noted. They were given a copy of the written information to take home, and informed that after more than 24 h we would send them a standardised message (SMS) in which we asked them to confirm or regret their decision by responding ‘yes’ or ‘no’. We emphasised that if they regretted the decision, all data about them would be deleted, and that regret would not have any consequences for their relationship with the doctor or the hospital. They were strongly recommended to discuss the decision with relevant individuals.
The SMS response was saved digitally as a documentation of their decision. The documentation included the date and time of the SMS, the number it was sent from and the message proper. Patients who did not have a cell phone were contacted in person (if an inpatient), by telephone or regular mail. When consent was confirmed orally, a note was made about the communication. The ethics committee accepted this procedure.
The study was approved by the Regional Committee for Medical Research Ethics of South-East Norway (1.2007.356), and privacy measures were accepted by the privacy ombudsman for research in Norwegian universities (NSD approval 16423/2007).
A total of 553 patients was approached by one of the two researchers (PG or BFJ, who are both doctors), 530 (95.8%) decided to participate and gave informed consent (figure 1). As a result of a procedural failure invalidating the videotape, three of these were not asked for confirmation of their consent. Of the rest, 519 (98.5%) confirmed their consent; 365 (69.3% of 527) had a cell phone. When responding to the standardised message, four (1.1% of 365) regretted their decision and 30 (8.2%) did not respond. These were contacted by telephone (23) or in person (seven), and one of those contacted by telephone regretted consent. Three (1.9%) of those contacted by other means regretted their decision. Figure 1 gives the details.
The low percentage of regrets suggests that this way of including patients in a study is acceptable, but the fact that there were regrets also highlights the importance of giving patients the opportunity to think their decisions over. Current practice in clinical research is always to allow patients, at any point, to change their mind and withdraw from studies and deny the use of data collected about them, so our method does not introduce a major change.
The advantage in our particular study was that after the encounter, the patient would know what was exposed on the videotape and relate the decision to participate to that. Such a withdrawal would not affect our results due to a robust design, which included that the patients did not know whether the doctor had been given the intervention or not. In trials in which the patient withdraws after treatment allocation, study results may be contaminated, and this has led to an array of randomisation procedures.3 Our approach is no substitute for a carefully planned randomisation procedure. However, when the design implies the acquisition of informed consent before randomisation, our method might facilitate the inclusion process.
A possible argument against the procedure is that recruiting researchers might adapt a more superficial style of oral information, referring to the possibility of post-recruitment regret. Also, it is not unlikely that patients in doubt would be more willing to agree to inclusion in the light of the possibility of regretting their consent. Persuading recruiters is a well-known problem in clinical research, because the need for a sufficient sample size, or even a conviction that the intervention under study is an improvement, may lead to an imbalanced oral presentation of the pros and cons of participating in the trial. One could argue that once consent has been given, patients might find it harder to withdraw. We believe that it will be easier for patients to withdraw when they do not have to face the person who recruited them when they send this message, and that our solution to some extent actually counteracts the problem of persuasion. However, the frequency of regret in our study was far too small to support or reject this assumption.
One possible weakness of the solution is that we cannot be certain that the person who answers the SMS or the phone is the same person who was actually recruited for the trial. However unlikely to occur, this is a relevant objection in our study in which there is no patient follow-up. On the other hand the patients in our study were already well aware of which exposure they had agreed to when they gave their written consent. If they regretted and did not receive the standardised SMS from us, they would have an opportunity to make contact because our phone numbers were written on the patient information and consent form. In clinical trials in which patients are subjected to the intervention this identity check does not represent a problem. The SMS confirmation of consent can then be checked with the patient at the first follow-up. If patients are already lost to the first follow-up we think the confirmation of consent should be deemed invalid.
This study was conducted in Norway, which adheres to European Union law concerning privacy legislation. All Norwegian studies that include personal information must obtain approval from the relevant authority in advance. Therefore, the data files that include the SMS messages and notes about personal communications with the patients must be stored at the same security level as the videotapes. We do not think our solution represents a violation, nor an extra challenge in this respect.
To secure patients' understanding of the pros and cons of consenting to participation in randomised controlled trials, a 24-h delay between information giving and consent is required.
This important procedure may hamper recruitment to clinical studies.
Mandatory confirmation by SMS of consent given at the time of information is a possible simplification of the procedure.
Funding The study was funded by the Regional Health Enterprise for Specialist Health Care in south-east Norway. There were no other study sponsors. The researchers are employees of Akershus University Hospital, which is an enterprise within the Regional Health Enterprise. Research papers are not subject to control by the administration.
Competing interests None.
Ethics approval The study was approved by the Regional Committee for Medical Research Ethics of south-east Norway (1.2007.356), and privacy measures were accepted by the privacy ombudsman for research in Norwegian universities (NSD approval 16423/2007).
Provenance and peer review Not commissioned; externally peer reviewed.