Family Physicians' Experiences of Physical Examination

Martina Ann Kelly, MA, MBBCh, FRCGP, CCFP; Lisa Kathryn Freeman, BSc (Hon), MD, CCFP, MPH, FRCPC; Tim Dornan, MA, DM, FRCP, MHPE, PhD


Ann Fam Med. 2019;17(4):304-310. 

In This Article



This study was conducted in family practices in Alberta, Canada in 2016.

Theoretical Approach

This qualitative study focused on family physicians' experience of physical examination by using a phenomenological approach. The starting point of this approach is to consider a phenomenon, in this case physical examination, and question what features make the phenomenon what it is. This study, which drew on ideas from 2 different philosophers, Merleau-Ponty and Van Manen, obtained family physicians' detailed descriptions of their everyday experiences of physical examination.

Sampling and Recruitment

Sampling and analysis were iterative. We first sent an introductory e-mail outlining the study to family physicians on family medicine faculty lists (University of Calgary) and known to us. All potential participants who e-mailed expressions of interest to M.A.K. or L.K.F. agreed to be interviewed, providing us with a convenience sample. As the study progressed, we wondered whether rural physicians having less access to diagnostic testing and more embedded relationships with patients would have different experiences. We therefore switched to snowball sampling, identifying potential interviewees from participants' recommendations. This sampling strategy was appropriate because sampling in phenomenological research is guided by participants' ability to illuminate the phenomenon. Although sampling can never be "complete,"[21] we judged recruitment to be sufficient when a wide range of participants had described a rich set of informative experiences.

Data Collection

M.A.K. or L.K.F. conducted individual interviews in person (14 participants) or by telephone (2 remote rural participants) in order to gather as much descriptive detail of individuals' experiences as possible. We conducted interviews at locations convenient to participants, such as their clinics, university offices, or coffee shops. We opened interviews by asking participants to select an actual experience of physical examination that they could recall in detail. We prompted participants to give rich details of the experience,[22] including where the examination took place, who was in the room, and how they had used the space in the examination room (see Supplemental Appendix, for interview guide). We audio recorded interviews (34 to 90 minutes) for verbatim transcription.


Our research team comprised a family physician with 20 years in practice (M.A.K.), a recent family medicine graduate and public health physician (L.K.F.), and a retired internist (T.D.). A specific feature of phenomenological research is the requirement for researchers to pay attention to their own preconceptions about the topic of interest. They examine how their assumptions, attitudes, and understandings of the topic, including the influences of gender, ethnicity, and prior experiences, influence their interpretations.[21] M.A.K. started the study because she noted that residents questioned the value of physical examination, sometimes prioritizing its predictive value over patients' requests for reassurance. This contrasted with her personal practice, in which she often performed physical examination, even if she though its diagnostic yield was low. L.K.F., a recent graduate, was interested in understanding why physicians performed exams, because she did them without much critical thought and saw them as integral to practice but was hearing more and more how others were not performing physical examinations. T.D. participated in the study because his experiences of providing secondary and tertiary care showed how technically proficient medicine improved health outcomes when it was provided within intersubjective relationships with patients.

We started the study by interviewing each other about our interest in the research question, examining whether our perceptions were influenced by our different amounts of clinical practice, training in different countries (Ireland, the United Kingdom, and Canada) with differing models of health care delivery, and having different access to diagnostics. We probed each other's interpretations as the study progressed, moving back and forth between transcripts and our analysis. This process was facilitated by individual coding of transcripts followed by group discussions (in person and via Skype), in which we questioned each other's interpretations.

We analyzed the data by using a template method.[23–25] As in thematic content analysis, we first read and reread transcripts to familiarize ourselves with the data. We then open-coded 3 interviews, grouping together relevant excerpts (tagged by interview number and line number) and developing a hierarchy of codes (see Supplemental Appendix for an early template). We then applied the newly formed template to more interviews, revising it to incorporate new codes or themes as presented by the data, reorganizing it until we agreed on a final set of codes and themes (Table 1). The resulting template provided a tool to help us reflexively question the data rather than a fixed and final interpretation.[26] To increase our confidence in the template, we reexamined the data, looking for differences between early, midcareer, and more experienced physicians, gender, and different work settings.

As advised by van Manen, we informed our interpretation by reading a range of phenomenological texts[27–33] and academic health care literature,[34,35] asking, "How has this topic been addressed elsewhere?" and, "How can these ideas inform our understanding?" We presented preliminary findings at family medicine conferences, using audience members' responses as a way of evaluating provisional interpretations.

Presentation of Findings

We have shortened phrases such as "participants experienced physical examination as intellectual and rational" to "physical examination was intellectual and rational" to make the article more readable; however, we ask readers to interpret such statements as interpretations of human experience, not as truth claims.

The Conjoint Health Research Ethics Board, University of Calgary and Health Research Ethics Board–Health Panel at the University of Alberta approved the project.