Where is the patient’s chair? Differences in general practitioner consultation room layouts - an exploratory questionnaire

Background: Consultation room design varies from country to country. The layout of a general practitioner’s (GP’s) consulting room may influence the physician’s or patient’s experience. The aim of this study is to explore and investigate the layout of GP’s consulting rooms around the world and to describe any significant differences. Methods: Between 3rd July and 2nd August 2018, an internet-based questionnaire on Google Docs was distributed by email, social media and WhatsApp platforms to several worldwide rural medicine groups. Analysis of an internet-based questionnaire to explore possible layouts of consultation rooms within practices was performed. The questionnaire was designed with three distinct sections: first, a GP demographic profile including gender, year of graduation from medical school, country of graduation, and type of practice (private or public); second, questions relating to the office layout; third, a section for questionnaire feedback. Results: 502 responses to the questionnaire were received; 65.3% women and 34.7% men. The most common layout in Europe and America was where the physician and the patient were separated by a desk. The layout where the physician and the patient had a 90º angle facing each other was the most commonly used layout in Asia-Pacific and Africa. For GPs who graduated before 1990 and between 1990-2010, the layout where the table was between the patient and physician was preferred. However, physicians graduating after 2010 preferred a layout with the physician and the patient with a 90º angle facing each other. Conclusion: The position of the GP’s desk differs between and within countries as well as the gender of the physician and year of graduation. Next steps should focus on gathering an even greater breadth of GP input, as well as comparing and contrasting those to the preferences of our patients and communities.

The physical setting can influence our mood and how we perceive the social situation. It can also determine our likelihood of interacting with others, and also influences the form that interaction will take and how long it is likely to last 1 . General practitioner's (GP's) consulting rooms differ from country to country, not only in language and cultural influences, but also in architectural design and placement of exam tables, desks, and chairs. Communication proves integral for GPs 2 . The incorporation of computers and electronic medical records has further changed the physical environment and the way communication occurs. A GP consultation is one of the central experiences of the patient-physician interaction with both the location of the GP's desk and the furniture arrangement impacting that encounter 3 .
A clinical encounter requires a doctor, a patient, and two chairs at a minimum. Additionally, most practices will also have an office desk as well as a computer. These elements (two chairs and a desk) can be placed in various arrangements. Dannenberg & Burpee argue that the design focusing solely on safety regulations, such as following building codes and avoiding fire hazards, lose all the possibilities of improving the doctor and patient experience 4 . Though these safety aspects are extremely important, engineers should not forego consideration of the medical encounter itself. For example, the physician's desk can compartmentalize the physician and patient in opposite spaces -effectively serving as a "barrier" and interfering with communication 5 . Furthermore, the power dynamics relationship (how power affects a relationship between two or more people) can be directly affected by the GP's desk position, including the computer 6,7 and furniture arrangement that can negatively influence the physician-patient relationship. In some cases, the introduction of the computer in the consultation room has led patients and physicians to make comparisons with the past and the "two players" consultation where computers -the third party -was not present 8 . In one study that included a focus group, patients preferred to sit beside the physician with a clear view of the computer screen 9 .
The question of where the GP should sit to promote optimal patient-centered care was raised in a Google Group discussion of the World Organization of Family Doctors (WONCA) Rural Working Party. After several email exchanges, the authors decided to develop a questionnaire-based survey. The aim of this study was to explore and investigate the layout of GP's consulting rooms around the world and to describe differences encountered.

Methods
After a literature review, a specific questionnaire exploring this topic was not found and the authors decided to develop one. This was an exploratory questionnaire intended to examine the possible layouts of consultation rooms within practices.
Between July 3rd and August 2nd 2018, an internet-based questionnaire on Google Docs (Extended data) was distributed by email and WhatsApp to several worldwide rural medicine social media and WhatsApp groups as this allowed reaching a varied number of countries. The communities form a part of the WONCA Rural Working Party network and contain approximately 1200 GPs and trainees.
The questionnaire was designed with three distinct sections. First, the GP demographic profile was collected including gender, year of graduation from medical school, country of graduation, type of practice (private or public), if they were in a rural/remote or urban area, and if there was a computer in their practice. The second part included questions relating to the office layout ( Figure 1). We proposed four scenarios as suggested in the Google group discussions. If the practice design was not represented in the scenarios proposed, the participants had the option to describe their own set up. Finally, there was a section for questionnaire feedback. This included a free text section and a question about satisfaction of the questionnaire (ranked from extremely dissatisfied to extremely satisfied).
The four scenarios proposed with a computer were ( Figure 1 We analyzed the results of the study questionnaire using Chi-squared tests. The results were considered significant with p<0.05. The statistical programs Epi Info v7.2.2.1 and SPSS v23 were used for the statistical analyses. We did not analyze the qualitative responses in this study as they were intended to improve the design of a future questionnaire.

Results
In total, 502 responses were collected during the 30-day period the questionnaire was active; 328 of the respondents (65.3%) were women and 174 (34.7%) were men. The country of practice with the most respondents was Romania with 199 responses (39.6%), followed by Spain with 140 (27.9%), Brazil with 51 (10.2%), Mexico with 32 (6.4%) and India with 16

Amendments from Version 1
Minor clarifications have been made to the main body text, and one new reference added.
Any further responses from the reviewers can be found at the end of the article

REVISED
(3.2%). We asked respondents to describe their practice area as rural/remote or urban; 114 respondents described their area as rural/remote (22.7%) and 379 as urban (75.5%). A total of 296 respondents described their practice as public (59%) and 184 as private (36.7%).
Regarding computers at the practice, 482 out of 502 (96%) respondents had a computer at their practice and 20 (4%) did not. All except one of the respondents who didn't have a computer were from developing countries -mainly Nigeria (40%), India (15%), and Brazil (15%).
We analyzed the countries of practice in four main geographical regions: Europe, Asia-Pacific, the Americas (North and South America), and Africa.
In relation to practice layout, the most common setup in Europe was scenario 3, with the doctor and the patient separated by a desk (66%). This is also the most widely used layout in the Americas (46%), closely followed by scenario 4, with the GP and the patient with a 90° angle facing each other (41%). Scenario 4 was also the most common layout in Asia-Pacific (80%) and Africa (56%).
When comparing geographical regions, scenario 3 was used significantly more often in Europe than in the Americas (p=0.0003), whereas scenario 4 is used significantly more often (p<0.0005) in the Americas than in Europe. We did not analyze differences in the layouts between Asia-Pacific and Africa as we had so few responses. Layouts of practice by geographical are reported in Table 1.  There were no statistically significant differences in the layout of the practice depending upon whether the practice was rural or urban (scenario 1, p=0.411; scenario 2, p=0.627; scenario 3, p=0.953; scenario 4, p=0.360).
We analyzed responses according to the respondents' year of medical school graduation and established three cohorts: before 1990, between 1990 and 2010 and after 2010. Amongst those who graduated before 1990, the most common scenario was number 3 (54%), followed by scenario 2 (27%). Amongst those who graduated between 1990 and 2010, the most common scenario was number 3 (71%). Amongst those who graduated after 2010, the most common layout was scenario 4 (55%), followed by scenario 3 (31%). Scenario 2 was used significantly more often by respondents who graduated before 1990 than by respondents who graduated between 1990 and 2010 (p=0.003) and those who graduated after 2010 (p=0.003). Scenario 3 was used significantly more often by respondents who graduated between 1990 and 2010 than respondents who graduated before 1990 (p=0.0002) and those who graduated after 2010 (p<0.0005). Finally, Scenario 4 was used significantly more often by respondents who graduated after 2010 than by respondents who graduated before 1990 (p=0.0002) and respondents who graduated between 1990 and 2010 (p=0.0002).
Layouts of practice by year of graduation are reported in Table 2.
Most female GPs reported using scenario 3 (66%), followed by scenario 2 (16%). Male GPs also used scenario 3 most often (44%), but scenario 4 (28%) was the second most popular layout. Scenario 1 was used statistically significantly more often by male rather than female respondents (p=0.01), while scenario 3 was used statistically significantly more often by female than male respondents (p<0.0005). In contrast, scenario 4 was used more often by male than female respondents (p=0.0002). Layouts of practice by gender are reported in Table 3.
Finally, the participants rated the questionnaire with 104 (20.9%) extremely satisfied, 187 (38.6%) moderately satisfied, 185 (37.1%) neither satisfied or nor dissatisfied, 16 (3.2%) moderately dissatisfied and 6 (1.2%) extremely dissatisfied. In addition, the participants' main feedback of how to improve the questionnaire was to allow participants to design their own consultation room and/or to have picture illustrations.

Discussion
This study found that the position of the GP's desk differs internationally, within countries, and according to year of graduation   as well as the gender of the GP. Each approach has a social and cultural context (Figure 2).
The position of the chairs, computer and desk can directly affect the doctor-patient relationship 3,6 . Jacobs has previously described that arranging chairs perpendicularly (e.g. Scenario 1) implies teamwork and support, chairs placed side-by-side augments education and instruction, while chairs arranged face-to-face (e.g. Scenario 2) show expertise and can be used to gain compliance 10 . Scenario 3 was the setup most widely used in Europe. Because of the computer interference and the table in between, this layout does not necessarily support 'expertise' of the physician while simultaneously distancing the patient 10 .
In this study, women preferred scenario 3 (66%), followed by scenario 2 (16%). Generally, female GPs reassure and encourage patients more compared to their male counterparts 2 who most often used scenario 3 (44%) -noting that scenario 4 (28%) was the second most popular. The choice of a more "protective" layout with the desk separating physician and patient could be a gender-related as GPs are increasingly facing violence, harassment, and threatening behavior 11 .
Computers can increase the accuracy of the patient record and enhance physician efficiency but can also negatively interfere with the physician-patient interaction. The majority of GP's (96%) in this study reported having a computer in the room. Frankel et al. listed the possible aspects in which a computer may interfere with a consultation: organization of the visit, verbal and nonverbal behavior, computer navigation and spatial navigation of the consulting room. Frankel also stated that "computering" could affect communication, which is the most powerful instrument available to the physician 12 . Sinsky and Beasley compared the risk of texting while driving to the risk of "texting while doctoring"; physicians should not practice "computer-centered medicine" 13 . Therefore, consulting rooms must be designed so that the computer is not the center of care but so that it is convenient and suited to the ergonomics of the physician. The multi-tasking involved in using a computer and the flow of the patient-physician-computer interaction can be complex and stressful 12 . A study conducted in Belfast examining the content of consultations with a significant psychological component showed that when the patient had a psychological complaint, physicians focused more on the patient rather than the computer and balanced their use of the computer and communication 14 .
Moreover, it has been shown that changing the layout of a consultation room has the potential to improve communications between patients and doctors 15 . Clinicians who have more barriers integrating the computer in the consulting room could have more difficulties dealing with multi-tasking and this could have a negative effect on the physician-patient relationship 12,16 .
The patient experience is constructed consciously and unconsciously, and design affects patient perception 5,10 . Why is design layout in Europe different from the Americas? Why is scenario 3 preferred in Europe but scenario 4 preferred in the Americas? Although there is no definitive answer for these questions, it is interesting to reflect on the learning matrix from different countries and experiences.
Also note the dynamic reflected by the year of graduation. For those who graduated before 1990, the table between the patient and general practitioner is preferred (54%), and this increases to 70% for graduates between 1990-2010, and then decreases to 55% for younger GPs (graduating after 2010) who prefer scenario 4. This may reflect the emphasis on a more patientcentered approach the authors have noticed in recent years.
It has been suggested that consultation spaces need to be more adaptable so that additional family members can be included when necessary 10 and -when using the computer -preferably the patient should be able to see the computer screen 17 . However, the need to understand each community and cultural competency are linked to the consulting room design. Modifying current consulting room designs should open-up new opportunities of interaction between the GP and patients.
This study has several limitations. First, the number of scenarios was limited and it was difficult to determine patient and GP locations from participants' descriptions. This cohort of respondents represents those who have access to the Internet or knowledge of this survey -noting limited numbers of respondents from Africa, North America, and Northern Europe. The results were concentrated in some countries and continents more than others where a broader representation would likely offer better data. Furthermore, other factors such as different designs to those in the proposed scenarios, satisfaction of doctors and patients in the chosen design for the consulting room and cultural differences were not covered by this research could influence consulting room design. The studies that exist have focused mainly on the developed world experience and we were unable to expand the scope much into developing countries. Lastly, the patient's viewpoint and preferences were not taken into consideration. Nevertheless, this study serves as an interesting starting point for further investigations to understand consulting room design worldwide and reflect upon it.

Conclusions
The position of the patient's chair differs between and within countries in addition to distinctions based on gender and year of graduation. Modifying current consulting room designs according to patient's and doctor's preferences could open up new opportunities for interaction between GPs and their patients, while continuing to remain conscious of both community and cultural differences during planning. Next steps should focus on gathering an even greater breadth of GP input as well as comparing and contrasting those to the preferences of our patients and communities.

Ethical approval
The study protocol has been approved by the Catalan To ensure confidentiality, all data has been anonymized. number of questions, the fact that among them 296 respondents described their practice as public and 184 as private and that means there were some respondents missing. The number of countries that participated in the study, why are Asia and Australia are included in the same group? Apart from the graduation year, the age of participants is not included, and it would have been of interest to have a more complete respondents' profile.

Data availability
In the discussion section the authors mention that "The position of the chairs and desk can directly affect the doctor-patient relationship". This statement should be supported by several references.
In the conclusion section we read that "Modifying current consulting room designs opens up new opportunities for interaction between GPs and their patients…" What kind of modifications are the authors referring to and in which of the scenarios described? What kind of opportunities? It seems that it is of great interest to describe these points even briefly.

Ferdinando Petrazzuoli
Center for Primary Health Care Research, Clinical Research Center, Lund University, Lund, Sweden This manuscript addresses an interesting topic: how orientation and seating arrangements may influence the doctor patient's relationship. The manuscript is interesting but a bit poor in terms of references. The topic has a lot to do with personal communication and I strongly advice to introduce the concepts and the references illustrated below.
Orientation refers to body angles adopted when people talk, such as directly facing or shoulder to shoulder. It concerns the position of the trunk, rather the head, and it marks the degree of intimacy in the conversation and the levels of friendship. It is useful to look at proximity and orientation together since it has been found that there can be an inverse relationship between them that is, direct face to face alignment is linked to greater interpersonal distance and sideways angling to closer distance. This would be expected in situations where orientation was being used to compensate for excessive closeness. Orientation can also be used to include or exclude others from the group during discussion. Early studies of seating behaviour by Sommer in North America, replicated by Cook in the UK, point to some interesting differences in seating arrangements if individuals are given a choice of where to sit when involved in different sorts of activities. Environmental factors are also important. The physical setting can influence our mood and how we perceive the social situation. It can also determine our likelihood of interacting with others, and also influences the form that interaction will take and how long it is likely to last. Hall distinguished between fixed-feature and semi-fixed feature elements of the environment. The former includes everything that is relatively permanent or not easily modifiable, such as the architectural layout of a house, size and shape of rooms, and the material used in their construction while semifixed features are much easier to move around or modify, and include furniture, lighting, temperature. According to Knapp and Hall based on such characteristics, we form impressions of our surroundings organized around 6 dimensions. These are:

Formality: concerns cues leading to decisions about how casual in what is said and done or if a more ritualized or stylized performance is demanded.
Warmth: here one feels comfortable, secure, and at ease in what are regarded as convivial surroundings.

Constraint: concerns perceptions of how easy it is to enter and leave the situation.
Distance: address how close, either physically or psychologically, we feel to those with whom we share the space.
These perceptions shape the types of interactions we engage in and how we experience them. The way in which work space is arranged and utilised sends strong signals about the status and authority of occupants, the kinds of tasks and activities being implicitly proposed, and last but not least the desirability and appropriateness of focused communication in that situation.