Method: Profile of interviewees
The Principal Investigators will identify 10-20 voluntary interviewees to participate in Sub-study 1 from a range of relevant medical services of the teaching hospital chosen. Each interviewee will be interviewed approx. three times, allowing for individual increase or decrease of number of interviews, as appropriate. The individual interviewee should be interviewed on different weekdays, covering possible important variations in tasks over the week. However, interviews should only be carried out on days where the interviewee has had patient contact.
Use of tape recorders
Tape recorders will be used for the interviewing to ensure storing the data and preserving the narrative detail for subsequent analysis. The Principal Investigators should inform the interviewees on the use of tape recorders during the identification of participants. Use of tape recorder is considered the best way to ensure full correspondence between the information given during the interview and the data analyzed.
The tape recorder should be of good quality since the accuracy of the transcription depends on the quality of recording. Separate rather than in-built microphone should be used, the interviewer should pretest the sound of each interview using headphones, and the tape recorder should have a connection for a pedal operated remote control if a separate transcribing machine is not available at the centre.
Interview format
Semi-structured interviewing will be used for the interviews in Sub-study 1. A semi-structured interview is understood as a specific kind of conversation where the interview-process is partially guided by the interviewer's intention of covering a specific area of interest, while at the same time the informant or interviewee is encouraged to talk freely about the subject. The interviewer must stress that (s)he is interested in the interviewee's thoughts, ideas and experiences with - in this case - ethical decision making in the clinic and the interviewee must be allowed to take the lead in the interview.
The interviewees will be asked to describe briefly their work during the day of the interview and subsequently, whether they saw any ethical problems in relation to patient care during that day. An ethics problem will be defined as "a question of what one ought to do, rather than what is usually done or can be done, and which requires a resolution of value choices, as opposed to resolving merely factual or scientific matters".
The interviewer will use various probing tools to investigate in some depth the ethical dilemmas identified by the interviewee.
In addition, the physician will be asked about age, sex, specialization, working experience in yrs., formal training in medical ethics and position in hospital. Also, the age, sex, principal diagnosis, length of stay and discharge disposition of patients described in relation with ethical dilemmas will be recorded.
Transcription
The tapes will be transcribed verbatim and computerized parallel with the interviewing process. If interviews are not conducted in English the interviews must be translated verbatim to enable subsequent centralized analysis at SEARO. It is recommended that the tapes are translated during the transcription by a competent translator to reduce amount of time and manpower involved. The translation must subsequently be checked against the tape by a different person than the translator, and possible errors must be discussed to ensure that the intended meaning is conveyed in the translation, and corrected as appropriate.
For the transcription any PC wordprocessor may be used. To facilitate subsequent data analysis the following format should be used:
Confidentiality
Measures should be taken to ascertain the anonymity and confidentiality of the interviewee. This includes establishing a coding system where the tapes are numbered and the key linking number and person is known only to the interviewer. Immediately following the last interview with an interviewee the key for this person is destroyed by the interviewer. While the Principal Investigator should not have a copy of the key, it is the responsibility of the Principal Investigator that the key is destroyed.
While the name of the interviewer is protected in this way, the base data that form an important part of the information, may serve to identify the person. This risk should be minimized by including 4-5 persons with similar specializations, thus limiting the number of participating units/specializations in order to protect the anonymity of each interviewee.
Profile of interviewer
The interviewer chosen by the Principal Investigator should, if possible, have the following profile:
After the interviews have been conducted, and the tapes have been transcribed and translated into English, the narratives will be analyzed. If the centres are able to identify a suitable researcher for analysis, this may be carried out at the each centre. If this is not the case, the centre(s) will send the computerized data to SEARO for centralized analysis. If some centres carry out the analysis locally while others send data to SEARO for centralized analysis, close coordination of the indexing procedure is necessary. The exact procedures in this regard will depend on the ratio between local and centralized analysis. However, the Principal Investigator should inform SEARO as soon as possible and before 1 October 1997 whether local or centralized analysis is preferred to enable SEARO to prepare the necessary.
The interview data will be indexed, using indexing categories according to contents. While the categories therefore cannot be predetermined, certain main categories, including "base data", "agency", "topic" and "contextual features" may be foreseen. The "agency category" classifies the narrative by the agent viewed as central to the problem (professional, patient, or external agent). The "topic" category classifies the narrative by topics such as refusal of treatment, informed consent, and disclosure of information, which will be further broken down in subcategories. Thus, each main category will consist of a number of sub-categories, the exact classification of which depends on the contents of the narratives.
This method will be used in order to extract maximum information from the data. A computer package for qualitative research (Q.S.R.Nud.Ist) will be procured for the purpose.
Profile of researcher for analysis
The profile of the research analyst, if the data analysis is to conducted at the centre, should be as follows: