Disagreement on Illnesses Between Older Patients and GPs
Disagreement on Illnesses Between Older Patients and GPs
This qualitative study is based on three focus groups with GPs and three focus groups with primary care patients, which were conducted at the end of 2013. The participants were recruited from the MultiCare Cohort Study. This is a multicenter, prospective, observational cohort study with a total of 3,189 multimorbid patients in the age group 65 to 85, recruited from general practices in Germany. Multimorbidity was defined as the coexistence of at least three chronic conditions from a list of 29 diseases published elsewhere. The methods used in this study, the patient population, and the sampling and response rates have been described in other papers.
We chose focus groups as a qualitative method because we expected these yield comprehensive explanations for disagreement due to the group dynamics and the exchange of experiences by the participants.
For this qualitative study we asked the GPs of the MultiCare Cohort Study from the study centre Hamburg (n = 52) to participate in the focus groups. The GPs were invited by letter. Fifteen GPs were able to participate. Afterwards, we also invited their patients (also study participants of the MultiCare Cohort Study) by letter (n = 106) to participate in focus groups. We received 56 responses from potential participants, which were assigned to focus groups according to planned dates and group composition. We conducted three gender-specific patient groups: a male group (n = 7), a female group (n = 7) and a mixed gender group (n = 7). Our hypothesis was that some gender related disease topics could be discussed more openly when only men or women were present. 21 patients were able to participate.
The socio-demographic characteristics of the study's participants (GPs and patients) are shown in Table 2 and Table 3.
The participating GPs were between 39 and 65 years old (mean age 53.4). Five of the 15 GPs were male. The years of practice experience ranged between 7 and 28 years (mean 14.6). Most of the GPs (42.9 %) treated between 500 and 749 patients each quarter (3 month period) and worked in single practices (35.7 %).
The patients were between 70 and 88 years old (mean age 77.0). 10 of the 21 participants were female. 12 participants were married, two divorced, six widowed and one never married. Seven patients had a low education level (CASMIN grade 1,), another seven had a medium education level (CASMIN grade 2) and six had a high education level (CASMIN grade 3).
For the qualitative study presented here, we developed an interview guide to explore the disagreement between patients' self-report about chronic conditions and doctors' diagnosis, which were identified in the preliminary study by Hansen et al.
The interview guide for GPs started with general questions about communication problems between GPs and their patients e.g. "When you think of conversations with your multimorbid, elderly patients, are there scenarios where you had the feeling that your patient did not understand everything? Or that the patient couldn't or didn't want to listen to you?" Thereafter, we presented the results from the analysis of the agreement between self-reported and general practitioner-reported chronic conditions among the participants of the MultiCare Cohort Study and asked the GPs about their impressions as well as which reasons they could imagine being responsible for any disagreements. The interview guide for patients had a similar structure. The questions, as well as the presentation of the results, were adapted for patients. To support the interaction in the focus groups we kept important keywords on a pin board. The interview guides are shown in the Additional files 1 and 2 http://www.biomedcentral.com/1471-2296/16/68/additional.
The focus groups were led by HH and IS, lasted approximately 120 min and were recorded on audiotape. HH is a health scientist and IS a sociologist; both are involved in health services research and epidemiological research on multimorbidity.
Audiotapes were transcribed verbatim by trained research assistants and, afterwards, were checked and corrected where necessary by HH.
The transcripts were analysed using the qualitative content analysis according to Mayring. This reductionistic method condenses the large amounts of data to identify the main content. Based on the literature, on the interview guide's topics and on the keywords on the pin board, we developed deductive categories for coding. For example: The category "Communication challenges between GP and patient" included the code "too little time for consultation". The coded text segments varied from short sentences to long paragraphs. The coding was performed using MaxQda11. HH and IS read and coded the transcripts independently and discussed the categories afterwards. During the coding process, we added inductive categories as they arose from the material. Afterwards HH and IS conclusively condensed the categories and summarized them into seven themes. The final set of categories and themes were determined by consensus.
The study was approved by the Ethics Committee of the Medical Association of Hamburg (approval no. 2881) on 8.11.2013. All participants gave a written, informed consent to participate in the study.
Methods
Study Design
This qualitative study is based on three focus groups with GPs and three focus groups with primary care patients, which were conducted at the end of 2013. The participants were recruited from the MultiCare Cohort Study. This is a multicenter, prospective, observational cohort study with a total of 3,189 multimorbid patients in the age group 65 to 85, recruited from general practices in Germany. Multimorbidity was defined as the coexistence of at least three chronic conditions from a list of 29 diseases published elsewhere. The methods used in this study, the patient population, and the sampling and response rates have been described in other papers.
We chose focus groups as a qualitative method because we expected these yield comprehensive explanations for disagreement due to the group dynamics and the exchange of experiences by the participants.
For this qualitative study we asked the GPs of the MultiCare Cohort Study from the study centre Hamburg (n = 52) to participate in the focus groups. The GPs were invited by letter. Fifteen GPs were able to participate. Afterwards, we also invited their patients (also study participants of the MultiCare Cohort Study) by letter (n = 106) to participate in focus groups. We received 56 responses from potential participants, which were assigned to focus groups according to planned dates and group composition. We conducted three gender-specific patient groups: a male group (n = 7), a female group (n = 7) and a mixed gender group (n = 7). Our hypothesis was that some gender related disease topics could be discussed more openly when only men or women were present. 21 patients were able to participate.
Characteristics of the Study Participants: Patients and GPs
The socio-demographic characteristics of the study's participants (GPs and patients) are shown in Table 2 and Table 3.
The participating GPs were between 39 and 65 years old (mean age 53.4). Five of the 15 GPs were male. The years of practice experience ranged between 7 and 28 years (mean 14.6). Most of the GPs (42.9 %) treated between 500 and 749 patients each quarter (3 month period) and worked in single practices (35.7 %).
The patients were between 70 and 88 years old (mean age 77.0). 10 of the 21 participants were female. 12 participants were married, two divorced, six widowed and one never married. Seven patients had a low education level (CASMIN grade 1,), another seven had a medium education level (CASMIN grade 2) and six had a high education level (CASMIN grade 3).
Interview Guide and Data Collection
For the qualitative study presented here, we developed an interview guide to explore the disagreement between patients' self-report about chronic conditions and doctors' diagnosis, which were identified in the preliminary study by Hansen et al.
The interview guide for GPs started with general questions about communication problems between GPs and their patients e.g. "When you think of conversations with your multimorbid, elderly patients, are there scenarios where you had the feeling that your patient did not understand everything? Or that the patient couldn't or didn't want to listen to you?" Thereafter, we presented the results from the analysis of the agreement between self-reported and general practitioner-reported chronic conditions among the participants of the MultiCare Cohort Study and asked the GPs about their impressions as well as which reasons they could imagine being responsible for any disagreements. The interview guide for patients had a similar structure. The questions, as well as the presentation of the results, were adapted for patients. To support the interaction in the focus groups we kept important keywords on a pin board. The interview guides are shown in the Additional files 1 and 2 http://www.biomedcentral.com/1471-2296/16/68/additional.
The focus groups were led by HH and IS, lasted approximately 120 min and were recorded on audiotape. HH is a health scientist and IS a sociologist; both are involved in health services research and epidemiological research on multimorbidity.
Coding and Analysis
Audiotapes were transcribed verbatim by trained research assistants and, afterwards, were checked and corrected where necessary by HH.
The transcripts were analysed using the qualitative content analysis according to Mayring. This reductionistic method condenses the large amounts of data to identify the main content. Based on the literature, on the interview guide's topics and on the keywords on the pin board, we developed deductive categories for coding. For example: The category "Communication challenges between GP and patient" included the code "too little time for consultation". The coded text segments varied from short sentences to long paragraphs. The coding was performed using MaxQda11. HH and IS read and coded the transcripts independently and discussed the categories afterwards. During the coding process, we added inductive categories as they arose from the material. Afterwards HH and IS conclusively condensed the categories and summarized them into seven themes. The final set of categories and themes were determined by consensus.
Ethical Considerations
The study was approved by the Ethics Committee of the Medical Association of Hamburg (approval no. 2881) on 8.11.2013. All participants gave a written, informed consent to participate in the study.