A qualitative study consisting of in-depth interviews was chosen to allow an intensive analysis of subjective motives, attitudes and needs. In-depth interviews are an established qualitative research method to collect information from particular groups e.g. professional target groups. An additional strength of the qualitative approach for this specific study is the chance to obtain a more realistic feel of self-perception and social self-image that cannot be experienced in the numerical data and statistical analysis used in quantitative research.

In April 2008 letters of invitation to participate in the study were sent to a sample of 160 GPs. Addresses were randomly chosen from a symposium member database including about 800 addresses of GPs in the area of Baden-Wuerttemberg. Furthermore, information leaflets were distributed to 200 GPs taking part in a symposium in May 2008 organized by the Department of General Practice and Health Service Research at the University Hospital of Heidelberg This symposium ("Tag der Allgemeinmedizin") takes place twice a year and is an independent event, i.e. not sponsored by the pharmaceutical industry offering continuous medical education for GPs [¹⁴]. Participation in the symposium is certified by the medical chamber of Baden-Wuerttemberg.

Based on these ways of recruitment 20 GPs expressed their interest in taking part in the study. Out of those 20 GPs, a mixed sample of 16 GPs was selected in terms of gender, age, number of years in practice, urban or rural setting, solo practice or group practice. Individual appointments for the interviews were arranged. The sample of 16 interviewees is summarized in table 1:

The in-depth interviews were undertaken in the practice of the GP or in the Department of General Practice and Health Service Research, University Hospital of Heidelberg, Germany. The interviews were semi-structured and conducted by a sociologist (IN), who is experienced in conducting interviews. Each interview lasted between 45-60 minutes. All interviews were recorded digitally and transcribed verbatim.

The interviews were based on the following questions:

How do you think medical specialists perceive you as a GP?

Do you agree with this perception of specialists?

How would you describe yourself in your role as a GP?

The aims of the study were explained to each interviewee. The interviewer ensured that each aspect of these questions was explained sufficiently, so that no questions or misunderstandings remained.

The ethics committee of the Heidelberg Medical School informed us that approval by an ethics committee was not necessary for a study which does not involves patient data.

The interviews were carried out between May and July 2008. Analyses were carried out using the software ATLAS.ti. Key issues were identified, summarized, labelled as codes and sorted into main and sub-categories based on the qualitative content analysis by P. Mayring [¹⁵]. The aspects of interpretation and categories are developed near to the material. Using this approach, qualitative content analysis has developed procedures of inductive category development. Each category was attributed to a quotation [¹⁶]. The interviews and the analyses were conducted simultaneously, so that the researchers could control for topic saturation. Topic saturation occurred after the 13^th interview. I.N. and S.J. independently reviewed transcripts to confirm that the codes were comprehensive and reproducible. Disagreements during this process were discussed until a consensus was achieved. The quotations cited here were translated by IN from German into English and cross-checked by SJ.

Following sub categories were defined for the main category "Impression" (see table 2):

The interviews showed a difference in GPs' views of how they are viewed by specialists who work in outpatient care, hospital specialists and medical teachers: GPs suggested that specialists who work in a hospital setting or in universities have a predominantly negative view of GPs whereas specialists who work in outpatient settings have a positive view of GPs.

Following sub categories were defined for "Reasons" (see table 3):

Following sub categories were defined for "Positive self-perception" (see table 4):

Our findings show discrepancies between GPs' self-perception of their professional role and their social self-image in relation to medical specialists. Generally, this study found that GPs have a positive self-perception regarding their professional role. They are satisfied with their multi-faceted work and its associated level of responsibility. While GPs' social self-image is perceived as being positive by specialists in outpatient care, it is negative for specialists working in hospitals and for medical teachers. The negatively social self-image originates from different factors like lower income of GPs, different skills and core competencies.

The study includes a mixed sample which is balanced regarding the most relevant demographic characteristics of German GPs (gender, age, solo or group practice, GPs from village, medium-sized town and town). However, the study was undertaken in only one region of Germany and the findings may not be transferable to other regions of Germany. A further strength of the study is the trustful ambiance during the conduction of the interviews with authentic reports about opinions and experiences of the interviewees. The different perspectives from varied researchers' professions (sociologist/physician) increase the intersubjective traceability.

A limitation of the study may be seen in the low number of 20 volunteers. However, out of the 20 volunteers it was possible to assemble a well-balanced sample in terms of gender, age, practice experience and practice site. Thus, compared to other interview studies and due to the fact that topic saturation occured after the 13^th interview, 16 interviews seem to be sufficient for an in-depth exploration of our research question.

A further limitation of the study seems to be that the interviews only present the point of view of GPs about their own profession. However, for a detailed and profound understanding it is important to analyse one perspective in a first step. For our main research question "Does the self-perception of GPs' professional role correspondent to their social self-image " the GPs' viewpoint is most relevant. In a next step it could be explored whether the social self-image of GPs equates to specialists perception about GPs' professional role. Therefore, specialists could be interviewed. This two-step approach is comparable to preceding qualitative studies from the Netherlands about specialists' and GPs' perceptions [¹⁷,¹⁸].

We know from sociological research that it is important for everyone to create a positive self-perception and to maintain it (Theory of Appreciation) [¹²]. It is a pernicious experience to learn that self-perception does not correspond with the perception of others and, moreover, that the perception of others is more negative that your own perception of yourself. Beyond the need for appreciation, everyone has a need for a correspondence between self-perception and the perception of others, which has been referred to as the Theory of Consistency [¹²].

Previous qualitative research with hospital specialists from the Netherlands shows that specialists would be interested in developing a better collaboration with GPs on a personal level [¹⁷]. However, barriers to developing such collaborations exist because specialists believe that there is not much they can learn from GPs. With respect to professional expertise, therefore, specialists do not consider GPs as equals. Another study from the Netherlands explored experiences of GPs with hospital specialists [¹⁸]. This study showed that GPs hope to gain respect from hospital specialists and that they were also interested in developing a personal relationship and direct communication for a better collaboration.

In comparison to the Netherlands, where specialists predominantly work in hospitals, Germany has almost half of all specialists work in outpatient care.

Our results show that one has to differentiate between these types of specialists. Specialists in outpatient care usually have good working relationships with GPs - indeed, the practice would suffer if they didn't.

A research study in Australia explored the frequency of "badmouthing" by specialists as medical teachers about metropolitan and rural GPs [¹⁰]. Medical "badmouthing" has been defined as "unwarranted, negative and denigratory comments made by doctors about other doctors in different branches of medicine" [¹⁰]. The underlying psychological mechanism of "badmouthing" stems from a common human need for self-aggrandisement and defining of group membership by aggressively putting down people outside the "in-group" [¹⁹].

In general, specialists working in hospitals and as medical teachers have different roles and responsibilities within their occupation. Trainee doctors and medical students learn from what their teachers say and from what they do in their clinical practice, as well teachers' comments happen by the way [²⁰]. Insufficient communication can lead to misunderstandings and create a wrong image of other professions. That's why during medical education and vocational training specialists' opinion could influence medical students' opinion.