An Exploration of General Practitioners' Reassuring Strategies
An Exploration of General Practitioners' Reassuring Strategies
We used a qualitative observational design combining two sets of stimulated recall interviews with GPs (dataset A and dataset B). To identify reassuring strategies that GPs consider as useful we explored the strategies they used during videotaped consultations with their own patients. Dataset A is an existing set of interviews that explored how GPs selected communicative actions during their patient encounters. Since GPs try to reassure patients in 70% of consultations, we expected a secondary analysis of this dataset to yield first insights with relevance to our research question. Dataset B consists of interviews conducted for the study in which GPs were prompted to reflect on reassurance in order to deepen and broaden the findings of dataset A and obtain a rich dataset of reassuring strategies. We performed a thematic network analysis using principles of grounded theory, with an iterative process of data collection and analysis and a constant comparison method.
The Medical Ethics Committee of Maastricht University Medical Centre granted approval for our study. The participating GPs and patients gave written informed consent. Verbatim transcripts of the recorded interviews were anonymized with codes.
Dataset A. The GPs in the existing dataset were purposively sampled in order to obtain a variety in age, gender, number of working years and practice settings. Each GP was interviewed about two videotaped consultations that were selected by the researcher to obtain a maximum variation sample with respect to the patients' age, gender, complaint, type of consultation and GPs' communication techniques assessed by the instrument MAAS-Global.
Dataset B. We contacted eighty GPs in the Southern part of the Netherlands, with at least five years of experience in general practice and who were not recently approached for other studies by our university. GPs were sent an invitational letter containing information about the study, followed by a telephone call. If the GP worked in a group practice, his/her colleagues were approached as well. We aimed to include 10–15 participants as this number was considered to be sufficient for stimulated recall interviews.
A researcher watched and videotaped all consultations conducted on one morning by each of the GPs. GPs were asked to rate on a scale ranging from 0 to 10 the importance of reassurance in each of the consultations. Before their consultation, patients rated their level of concern on a scale ranging from 0 to 10. For the stimulated recall interviews, the researcher used the rating on the GPs' reassurance scale and the rating of patients' concern to select two consultations of each GP in which the GP aimed to reassure and the patient was concerned. In case we could not find two of such consultations, we prioritized the GPs' rating. Reassurance needed to be a goal of the GP, otherwise they could not be interviewed about how they reassured their patients.
For both datasets, trained interviewers conducted the interviews shortly after the consultations. The interviewers discussed the interview procedure in depth before and during the period of data collection. GPs were asked to watch two videotaped consultations and stop the tape whenever they wished to reflect upon their thoughts, intentions and actions regarding their communication (dataset A) or reassurance (dataset B). Once the video was stopped, they were prompted further to clarify these, e.g. what did you want to achieve here, why/how do you think that works? In case they did not stop the tape at a moment the researcher considered important, the researcher could do so and invited them to reflect upon their behaviour. At the end of the interviews of dataset B, GPs were asked whether, in retrospect, they would have reassured differently and if yes, they were asked why and how. Additionally we asked them whether they use reassuring strategies in their daily practice other than those used during the two discussed consultations.
All the interviews were audiotaped and transcribed verbatim. We firstly analysed dataset A. Text fragments about reassurance were selected and coded using Atlas-ti software for qualitative data analysis. Actions, mechanisms and goals were extracted from the GPs' reflections during thematic coding. In line with thematic network theory, we constructed networks to structure and visualize relationships between global themes, consisting of the goals pursued by GPs, organizing themes, consisting of mechanisms that explained how GPs' actions promoted the goals, and basic themes, that is the actions performed by GPs. The networks helped us to obtain first insights into GPs' reassuring practices and aspects that needed to be explored further in subsequent interviews in order to acquire a better understanding of GPs' strategies. During the thematic analysis of the interviews of dataset B, the networks were revised and expanded. The networks were eventually used to develop a schematic table presenting a framework of goals, mechanisms, and actions related to reassuring patients.
All the transcripts were analysed independently by at least two researchers with different backgrounds: health sciences (EG) or medicine (WV/CL/DW). The researchers reached consensus on the coding through discussion. The thematic networks and the schematic table were validated through in-depth discussions between the researchers.
Methods
General Design
We used a qualitative observational design combining two sets of stimulated recall interviews with GPs (dataset A and dataset B). To identify reassuring strategies that GPs consider as useful we explored the strategies they used during videotaped consultations with their own patients. Dataset A is an existing set of interviews that explored how GPs selected communicative actions during their patient encounters. Since GPs try to reassure patients in 70% of consultations, we expected a secondary analysis of this dataset to yield first insights with relevance to our research question. Dataset B consists of interviews conducted for the study in which GPs were prompted to reflect on reassurance in order to deepen and broaden the findings of dataset A and obtain a rich dataset of reassuring strategies. We performed a thematic network analysis using principles of grounded theory, with an iterative process of data collection and analysis and a constant comparison method.
Ethical Approval and Informed Consent
The Medical Ethics Committee of Maastricht University Medical Centre granted approval for our study. The participating GPs and patients gave written informed consent. Verbatim transcripts of the recorded interviews were anonymized with codes.
Selection Procedures
Dataset A. The GPs in the existing dataset were purposively sampled in order to obtain a variety in age, gender, number of working years and practice settings. Each GP was interviewed about two videotaped consultations that were selected by the researcher to obtain a maximum variation sample with respect to the patients' age, gender, complaint, type of consultation and GPs' communication techniques assessed by the instrument MAAS-Global.
Dataset B. We contacted eighty GPs in the Southern part of the Netherlands, with at least five years of experience in general practice and who were not recently approached for other studies by our university. GPs were sent an invitational letter containing information about the study, followed by a telephone call. If the GP worked in a group practice, his/her colleagues were approached as well. We aimed to include 10–15 participants as this number was considered to be sufficient for stimulated recall interviews.
A researcher watched and videotaped all consultations conducted on one morning by each of the GPs. GPs were asked to rate on a scale ranging from 0 to 10 the importance of reassurance in each of the consultations. Before their consultation, patients rated their level of concern on a scale ranging from 0 to 10. For the stimulated recall interviews, the researcher used the rating on the GPs' reassurance scale and the rating of patients' concern to select two consultations of each GP in which the GP aimed to reassure and the patient was concerned. In case we could not find two of such consultations, we prioritized the GPs' rating. Reassurance needed to be a goal of the GP, otherwise they could not be interviewed about how they reassured their patients.
Interviews
For both datasets, trained interviewers conducted the interviews shortly after the consultations. The interviewers discussed the interview procedure in depth before and during the period of data collection. GPs were asked to watch two videotaped consultations and stop the tape whenever they wished to reflect upon their thoughts, intentions and actions regarding their communication (dataset A) or reassurance (dataset B). Once the video was stopped, they were prompted further to clarify these, e.g. what did you want to achieve here, why/how do you think that works? In case they did not stop the tape at a moment the researcher considered important, the researcher could do so and invited them to reflect upon their behaviour. At the end of the interviews of dataset B, GPs were asked whether, in retrospect, they would have reassured differently and if yes, they were asked why and how. Additionally we asked them whether they use reassuring strategies in their daily practice other than those used during the two discussed consultations.
Data Analysis
All the interviews were audiotaped and transcribed verbatim. We firstly analysed dataset A. Text fragments about reassurance were selected and coded using Atlas-ti software for qualitative data analysis. Actions, mechanisms and goals were extracted from the GPs' reflections during thematic coding. In line with thematic network theory, we constructed networks to structure and visualize relationships between global themes, consisting of the goals pursued by GPs, organizing themes, consisting of mechanisms that explained how GPs' actions promoted the goals, and basic themes, that is the actions performed by GPs. The networks helped us to obtain first insights into GPs' reassuring practices and aspects that needed to be explored further in subsequent interviews in order to acquire a better understanding of GPs' strategies. During the thematic analysis of the interviews of dataset B, the networks were revised and expanded. The networks were eventually used to develop a schematic table presenting a framework of goals, mechanisms, and actions related to reassuring patients.
All the transcripts were analysed independently by at least two researchers with different backgrounds: health sciences (EG) or medicine (WV/CL/DW). The researchers reached consensus on the coding through discussion. The thematic networks and the schematic table were validated through in-depth discussions between the researchers.