Guidelines for protocol writing

It is important that decisions taken regarding the assessment of QL in the trial are preplanned and written into the trial protocol. Checklists to be followed when drafting a protocol include the following components:

♦ the rationale for the inclusion of QL assessment,

♦ the choice of QL instrument, and the timing of administration (including acceptable time windows),

♦ how to select patients if only a subset are required for the QL assessment,

♦ guidelines for administration, identifying a named person in each centre, method of data collection (home or clinic), when to hand out the forms, checking forms for completeness, whether help and/or proxy assessments are permitted, and what to do when the patient does not attend,

♦ adequate information about QL in the patient information sheet,

♦ statistical considerations (sample size, hypotheses, method of planned analyses, etc.),

♦ the instrument being included as an appendix.

Guidelines for QL administration

In addition it is useful if, for each trial, guidelines are prepared for local staff, to cover the importance of compliance and of checking completed questionnaires, staff responsibilities, absences and changes, and the role of the trial coordinating office. An example might include the following points:

♦ One named member of staff should be made responsible for the administration of QL questionnaires for each trial at each centre, and a named deputy should be available in their absence. The responsible staff member must explain to the patient the importance of QL assessment and how to complete the questionnaires.

♦ The first questionnaires should be completed before randomization (i.e. before the patient knows which treatment group he has been allocated). On all occasions it is preferable that the form is completed before the patient is seen by the doctor and before each cycle of treatment.

♦ Unless specified otherwise in the protocol, the patient should complete the questionnaires without conferring with a relative or member of staff.

♦ Ideally the questionnaires should be checked to ensure that all questions have been answered, and if necessary the patient should be asked to fill in any missing items. However, there may be situations where complete confidentiality is required, and patients maybe asked to place the completed form in an envelope for posting directly to the trials office.

♦ If an assessment is missed because of administrative failure, the patient should be contacted by telephone or letter and asked to complete and return (in a stamped addressed envelope provided) mailed questionnaires as soon as possible. Patients should be encouraged to ask for forms if not given to them.

♦ If a QL questionnaire is not completed as per the guidelines (for example at home, with help, or after seeing the doctor), this must be indicated on the form.

Patient information sheets

In the vast majority of cancer trials evaluating QL, patient-completed questionnaires are used, so it is important that patients are fully aware of the reasons for collecting QL data. Thus the patient information sheet should include details about the QL aspects of the trial and patients should be made aware of the fact that consent to the trial includes agreement to complete QL questionnaires, although the right of the patient to withdraw at any stage always remains.

It is important that patients feel comfortable about reporting their symptoms and side effects honestly. Patients may answer questions differently (and possibly less honestly) if they believe that the information will be seen by their clinician and/or may influence their treatment [54]. Even completing questionnaires with the help of their partner may affect the answers patients give, and the usual recommendation is that patients are found a quiet space to sit and complete the form, without conferring with staff or relatives, whilst waiting in the clinic to see the doctor.

Patient information sheets should therefore make it very clear what is to be done with the completed QL forms. Ideally, as soon as they have been checked for completeness, they should be sent straight back to the trials centre, but there may be circumstances where they are reviewed by the local nurses and clinicians, or where the data centre reviews the forms and passes information back to the local clinician.

It is usually recommended that completed QL questionnaires or at least sensitive questions are confidential and not made available to nurses or clinicians. In a trial of antihypertensive therapy [88] the responses to items relating to sexual functioning were placed in an envelope and sealed by the patient and mailed directly to the central trials office. In such situations, patients need to be reminded that it is up to them if they wish to discuss any such concerns with the clinician or the nurse. A real dilemma can occur for those collecting the data centrally, if they notice that patients are reporting serious side-effects or psychological problems, and yet this is not reflected in the clinical reports.

Decisions about what to do in these circumstances should be discussed before the trial starts, and guidelines included in the protocol.

Another ethical concern relates to the growing requirement that to be eligible for a trial, patients must be willing and able to complete QL questionnaires. Is it right to exclude a patient from a trial because they are unable to complete the questionnaires (perhaps because they are blind or do not speak English)? This of course depends on what the primary outcome measure is and whether QL is required on all patients or just a sample. One advantage of using standard questionnaires is that many are already translated into a variety of languages.

In many countries, before a trial can start, an ethics committee must approve the protocol and the patient information sheets. When QL is an integral part of a trial the patient information sheets must include specific information for the patient of the reasons for including QL and what will be expected of them.

0 0

Post a comment