Methodology
An online questionnaire was translated from the previous PatientPartner questionnaire and interview schedule amongst patient organisations, recommendations the previous authors made were adopted in the new questionnaire. This was considered most appropriate in order to get data that could be compared with data from patient organisations. The questionnaire was furthermore validated members of the PatientPartner taskforce and a representative of Dutch pharmaceutical organisation Nefarma. Requests and reminders to participate in the questionnaire were sent to representatives of the European head-quarters of 18 leading pharmaceutical companies who were in an EFPIA (European Federation of Pharmaceutical Industries and Associations) think-tank on patient involvement, snow-ball method was used to spread the request. These representatives were repeatedly contacted by telephone with the request to participate. Request to participate in the questionnaire was furthermore sent to representatives of European pharmaceutical companies who were in the database of project partner EFGCP. Finally the letter with request to participate in the questionnaire was forwarded to Dutch pharmaceutical companies by Dutch pharmaceutical industry organisation Nefarma. The questionnaire was available to the contacted representatives between April and June 2009.
Content of questionnaire
The questionnaire consisted of two parts; the first part of the questionnaire consisted of questions on the current experience the companies had with involving patient organisations in clinical trials. The ways of involvement were arranged according to the levels of the PP ladder. On each level specific examples were given of ways in which a patient organisation could be involved. For involving patient organisations as advisors examples were given such as: “advising on a clinical-trial protocol” or “advising on informed consent forms”. The respondents could also fill out in which phase of the clinical trial they had had experience with involving patient organisations. This was done in order to make patient involvement more concrete to the pharmaceutical companies who lacked experience. In this way it was also made clear that active patient involvement and not involvement as research subject was meant.
The second part of the questionnaire included qualitative questions of the opinion of the companies on the involvement of patient organisations in clinical trials. There was asked what they perceived as benefits, barriers and recommendations towards patient involvement. These were all open style questions where room was left for explanation. This was thought to be most appropriate to be able to explore respondents’ views in more detail. The opinion part also included five statements on different aspects of patient involvement which the respondents could rate on a scale from 1-10.
Description of cooperating pharmaceutical companies
In total Twenty-two respondents from nineteen different companies filled out the questionnaire. The questionnaire was viewed 180 times, taking in account people looking at the questionnaire twice, the response rate was approximately 15%. Most respondents were medical or clinical director, clinical manager or a related function within their company.
Location of the pharmaceutical company
Of the twenty-two companies eighteen were multi-nationals, three national companies and one company that operated in two countries:
-
Sixteen respondents were from offices based in North Western Europe region; five from Holland, six from Belgium and five from the UK.
-
Two respondents were from offices based in Southern Europe region; one from Spain and one from Portugal.
-
Two respondents were from offices based in Central Eastern Europe region, both from the Czech republic.
-
Two respondents preferred to stay anonymous and did not fill out their contact information.
-
Cyprus, Malta, Latvia, Lithuania, Luxembourg and Rumania were countries where not all companies (especially smaller companies) operated.
Type of company
The three most common therapeutic areas of the companies were: oncology, cardiovascular diseases and infectious diseases. From the Twenty-two representatives, there were:
-
Fifteen representatives of pharmaceutical companies
-
One representative of a pharmaceutical/biotechnology company
-
Two representatives of biotechnology companies
-
Three representatives from a CRO
-
One representative of a local subsidizer of a pharmaceutical company
Analysis of questionnaire
Structural analysis of questionnaires was done trough pattern matching, where the questionnaires were matched. Common themes were identified and labelled. Also analytic induction was done, in which attention was paid to the context.
Interviews
Semi-structured interviews with five representatives of companies were held in order to get more in depth information on different topics. The length of each interview ranged from 30 minutes to just over an hour. Four interviews were conducted over the telephone, one interview was conducted in person. Four interviewees also had answered the questionnaire, one interviewee had not answered the questionnaire. For the interviewee who had not answered the questionnaire a validated interview list with the same themes and structure as the questionnaire was developed. For the interviewees who had answered the questionnaire, a motivation for their answers was asked and more in-depth answers were requested. One interviewee had a lot of experience with patient involvement. The questions of the interview focussed on this experience to identify best practices. Three interviewees had no experience and questions in their interviews therefore focussed on their views on this lack of involvement and possibilities for the future. One interviewee had some experience; this was first explored and then opinion and future possibilities were discussed.
Analysis of interviews
All interviews were transcribed and summarized. Interviews were send back to the interviewees for validation. Structural analysis of interviews was done trough pattern matching, where the questionnaires were matched. Also analytic induction was done, in which attention was paid to the context.