Abstract



Introduction

In Japan, although primary-care physicians are widely accessible, the health-care system also gives most people easy access to specialists.¹ Unlike in some other developed countries, the barriers against self-referral are few and easily overcome. Although this system gives access to medical care to much of the population, it also facilitates ‘doctor shopping’ and other inefficient uses of medical resources.^2–4 One possible solution to these problems is to ensure that, as much as possible, primary-care physicians meet their patients’ (and potential patients’) needs and expectations.^5,6 Primary care, in the sense of care given by expert generalist practitioners, is still in a very early stage in Japan. Hence we need to make our own distinctive program for education of primary-care physicians by both knowledge from other countries and the needs of our culture.
We therefore wanted to find out what Japanese people expect of primary-care physicians. (In Japanese, a doctor one visits for most of one’s medical needs can be called a kakaritsuke-i. That is the term we use in Japanese, and here we translate it as ‘primary-care physician’.) According to the results of a survey done in 1993, Japanese people valued easy access to care, technical competence, and communication skills.⁷ Those three categories were not suggested by the participants in that study, nor were they derived by the investigators from the participants’ responses. Rather, they were the items chosen most frequently from a list that had been prepared by the investigators. Details regarding content validity were not reported in depth, and we suspected that those three categories were insufficient. In the US, primary care has been defined by the Institute of Medicine⁸ and tools to assess the quality of primary care have been developed in the US and in Europe.^9–12 However, whether such definitions and measurement tools can be reasonably applied in Japan, and in other non-Western cultures, is not known. Believing that patients’ perceptions and expectations are important, we used qualitative methods to develop a profile of the ideal primary-care physician from the viewpoint of Japanese people.


Method

Instutitional approval

The board of directors of the Tokyo Health Co-operative reviewed and approved the plans for the study.

Timing

We first conducted three individual interviews in Tokyo during October and November of 2000. We used the results of those individual interviews to plan focus-group discussions, which were held in Tokyo between November 2000 and April 2001.

Recruitment

All participants were more than 20 years old. For the individual interviews, we recruited one participant from among the acquaintances of a friend, one patient chosen by a staff member of a clinic, and one person recommended by one of the directors of a hospital (who was not a physician). Participants in the first two group interviews were recruited by convenience sampling. For the first group, we asked one of the directors of the hospital where one of us works to recruit participants. For the second group, we asked a staff member of the clinic where one of us works to recruit participants. Because the participants in the first two groups were mainly elderly, for the third group we sought to include younger people: mothers of children at a kindergarten (‘theoretical sampling’, as described by Glaser and Strauss).¹³ For this group, we asked a friend who uses the kindergarten to recruit participants. For the fourth group, we again used theoretical sampling: we asked our acquaintances to help us recruit and choose self-employed and company-employed workers, because after the third session we found that housewives and non-working elderly people would otherwise have been over-represented. After working with the texts from these four groups we conducted a session including employed men, who were chosen by one of the directors of the clinic where one of us works, to make the overall sample more representative of the population.

Setting

All interviews were held in Tokyo, Japan. They were held in an office at a university, and in meeting rooms in a clinic, a hospital, a civic hall, a kindergarten, and a newspaper office.

Individual interviews

The three people interviewed individually were women aged 66, 45, and 57 years; one of them had a primary-care physician. Each individual interview lasted 30–60 minutes; two were conducted by MO and one by MS. After a self-introduction, the interviewer explained the purpose of the study, and asked whether the interviewee had a primary-care physician. (The one who did was then asked about that doctor’s age and specialty, and how long she had been seeing him.) The interviewer then asked the main question: ‘What do you want in a primary-care physician?’ Once the individual interviews were finished, we developed the plan for the focus-group interviews.

Group interviews

The focus groups had a total of 32 participants, 12 men and 20 women; their mean age was 58 years (range, 28–89), and 19 (59%) had a primary-care physician (Tables 1,2). Each session lasted approximately 2 hours. To help the interviewees respond to the main question (above), we added before it the question ‘What kind of doctor do you think of when you hear the term “primary-care physician”?’ For the final question in the fifth session, we showed the participants the list of characteristics of an ideal primary-care physician that resulted from our review of the comments in the previous groups, and asked them for ideas to add to that list. Very little information was added, so we held no further sessions.

Abstraction of categories from the texts

Each interview was tape-recorded with the participants’ consent, and all individual and focus-group interviews were transcribed verbatim. We used the grounded-theory approach (i.e., the ‘constant comparative’ method of arriving at categories13). Specifically, two investigators (MO and MS) read the texts and independently abstracted the characteristics of an ideal primary-care physician. Then they discussed the characteristics mentioned in the first individual interview, and classified them into nine provisional categories. These provisional categories were used to organize information from the next interview. New categories were added and old categories were changed as necessary. After the fifth group interview only a few new categories emerged, so we stopped holding focus groups. By that time we had identified 49 categories. Then we (four of the authors) discussed the categories and the relations among them; we combined some into larger categories until we reached a consensus. We also sent to each of the 32 participants a copy of the list of categories and some illustrative quotations, and a sheet on which we asked them for their opinions of the results. Two responded, and one of the others was interviewed by MO. They indicated that the list of categories was valid.


Results

By the end of our reviews of the participants’ comments, we had organized the characteristics of an ideal primary-care physician into four large categories: physical accessibility, medical competence, psychological accessibility, and familiarity with the patient (Table 3).

Physical accessibility

People wanted doctors who were physically accessible; easy access to care is important. Ideally, a primary-care physician should be near the patient’s home, office hours should fit in well with patients’ schedules, and waiting times should be short.

‘We go to doctor A’s office because it is near. We also go to doctor B’s office, because it is not busy.’ (57-year-old woman)

Some participants, mothers of young children in particular, said that they value physicians who make house calls, and who are available for consultation by telephone and outside of their regular office hours.

‘He came to our home whenever we called…we were lucky. When my children were very young, I could have our doctor examine them at any time, even when we went into the office through the back door.’ (49-year-old woman)

Medical competence

Participants told us that they wanted primary-care physicians to be competent medical professionals. The physician should be able to diagnose and treat their condition skillfully; keep current; be able to manage most medical problems; have a well-equipped office with a capable staff; do emergency treatment if necessary; and consult and advise patients on minor concerns, family health, health promotion, and lifestyle. Some participants wanted evidence of competence, such as a good reputation or some official form of certification.

‘I want a doctor who has a wide range of knowledge and clinical experience with almost every illness, and can give us proper advice; unlike many specialists in university hospitals, who tend to have deep but narrow knowledge and experience only in their specialties.’ (75-year-old man)

However, the participants did not expect all of their medical problems to be solved completely by their primary-care physician alone. They wanted to be referred to others as necessary, but they wanted continuity of medical information between facilities.

‘I want my community doctor to refer me to another hospital if my illness is serious. After I get back from the hospital, I hope the doctor can obtain my record from the hospital and continue taking care of me.’ (38-year-old man)

Psychological accessibility

Patients are often tense, and many hesitate to talk openly with physicians. The participants in this study wanted not to feel separated from their doctor by this psychological barrier.
A psychologically accessible physician is one who communicates with patients well and treats them kindly and empathetically. The participants wanted their primary-care physician to listen to their explanations, to understand their expectations, and to respect their opinions. They also wanted their primary-care physician to explain medical conditions and treatments.

‘My ideal doctor would be kind and easy to talk with about all sorts of things. For example, when I talk about a minor skin problem on my hand, the doctor might say “Don’t worry. It’s all right.”’ (38-year-old woman)

‘I had a case of sudden hearing loss, so I went to a university hospital, but when I asked what caused it they said only “we don’t know”, and I just rested for a month. Probably a primary-care physician working in his or her own private clinic would have taken me seriously.’ (60-year-old woman)

‘I want a doctor who will give me a satisfying explanation, and will do it without condescending or assuming that my ideas are uninformed.’ (38-year-old man)

Some participants, particularly younger women, mentioned a physician’s gender as an important factor.

‘When I was a teenager, I became embarrassed about being examined by a male doctor. So I looked in a telephone book for a female doctor.’ (40-year-old woman)

Included in this category were participants’ desires for doctors with a strong sense of responsibility for their care. They wanted a primary-care physician in whom they could place their trust.

‘When I was a child and had a high fever, our doctor quickly came to our house outside of his office hours. If a doctor looks after me with such kindness, the doctor’s specialty won’t matter to me. I would trust him because being warm and caring is more important than medical specialty.’ (36-year-old man)

Psychological accessibility is also reflected in people feeling that they are close to a physician, almost as they would be to a family member.

‘When I just chat with a doctor or talk about family matters, not medical problems, I feel that we are close.’ (63-year-old man) ‘When the doctor passed away, I felt as if I had lost someone as close to me as a member of my family.’ (78-year-old woman)

Familiarity with the patient

The participants said that it was important for physicians to know their patients well.

‘My daughter’s doctor understands her allergic rhinitis very well. He advises me to give her medicine twice a day, morning and evening usually, and to give it to her three times a day when it gets worse. I want to take her to a doctor who understands all the details of her situation.’ (38-year-old-woman)

The participants wanted their primary-care physician to understand not only their medical condition but also their familial and social context. They wanted to have a continuing relationship with their doctor. They saw value in the doctor and the clinic staff working in the same clinic for a long time, and in continuity of information among doctors and staff members in the clinic.

‘Through a long relationship, I came to trust my doctor. I want my doctor to know about my life and my relationship with my family members and my neighbors.’ (75-year-old-man)

One woman illustrated this with a metaphor:

‘When a new doctor replaced my previous doctor, I felt as if the stairs we had built up just collapsed, and I had to start building them all over again with the new doctor.’ (70-year-old-woman)

The relative importance of the four categories above could be expected to vary with each patient’s medical condition and personal preference. Perhaps because the Japanese health-care system makes it possible for people to have many primary-care physicians, the participants seemed to choose their doctor carefully. Their decisions seemed to depend on their experiences with each doctor and on their perception of each doctor’s strengths and weaknesses. One participant, for example, seemed to weigh physical accessibility against her own judgements of a physician’s competence and of her medical need:

‘I have a pediatrician of a hospital as a primary-care physician for my child because the hospital has an emergency department that is open through the night. Also, I have another primary-care physician whose specialty is surgery, whom I go to because his clinic is never busy. I visit him when my child gets a slight illness like a cold in winter… When my child did not get well after treatment by the doctor at the hospital, we finally had to go, without a referral, to the National Children’s Hospital a few times.’ (37-year-old woman)

Discussion

We realize that some of the participants might have been influenced by group and peer pressure. They might have hesitated to reveal or discuss thoughts about primary care medicine that were based on unpleasant, private, or sensitive experiences. They also might have been affected by knowing that the researchers were physicians. For example, this knowledge might have made them hesitant to express negative opinions of primary-care physicians. It is also conceivable that the abstraction of categories from the texts could have been biased because it was done by physicians. More individual interviews, and inclusion of non-physicians in the collection and handling of the texts might be useful.
We also cannot be completely sure about the representativeness of the participants, because some of them were recruited by convenience (as in many other qualitative studies). For example, the participants might have been unusually concerned about their health, and thus they may have had particularly strong opinions about primary-care physicians. We also note that all of the participants lived in the Tokyo metropolitan area. Studies of people in other areas of Japan might reveal different concerns. For example, people living in rural areas might have concerns arising directly from the rural depopulation of recent years. Nonetheless, we believe that these results can be generalized to much of the population of Japan.
The participants in this study wanted their primary-care physicians to be physically accessible, medically competent, psychologically accessible, and familiar with them. Although our reflections on the content of the transcripts led us clearly to four distinct categories, it is equally clear that these categories are interdependent, and that this interdependence requires further study. For example, psychologically accessibility might easily lead to greater familiarity; and doctors who are perceived to be medically very competent might thereby attract many patients and thus become busier and less accessible.
We compared these results with those of previous studies done in Japan. The findings of the 1993 survey are encompassed by only three of the categories we identified; the earlier survey did not identify people’s desire for their doctor to know them well (Table 3). Physicians generally recognize the benefits of continuity of care; here we found that continuity is important also from the point of view of potential patients. In accord with this finding, Sebata et al. pointed out that elderly people in Japan expected continued medical treatment by the same doctor.^14,15 Arborelius’ qualitative study done in Sweden showed that patients want doctors to treat them as whole persons, not only as patients, and to ask questions about things other than the immediate medical topic, for example about the patient’s family or work.¹⁶
We are not aware of comparable studies done in any other east-Asian countries. We did compare our categories to those described in the reports of two studies: the Primary Care Assessment Survey (PCAS), which was published in the US in 1994⁸ and was used as a model for the General Practice Assessment Survey in the UK;¹⁰ and the list of aspects of health care by Wensing et al.¹⁷ The contents of that list were reflected in a study done to evaluate general practice in 10 European countries, which was carried out by the European Task Force on Patient Evaluation of General Practice Care (EUROPEP).¹² Each scale in the PCAS and the aspects of care in Wensing’s list correspond to one of our four categories (Table 3), which suggests that people of different cultural backgrounds nonetheless have generally similar desires of their primary-care physicians, and that these categories can be useful in cross-cultural studies. The first systematic review of the literature on patient priorities for general practice care included studies done in the US, the UK, Australia, the Netherlands, Canada, and Scandinavian countries.¹⁸ Our results indicate that EUROPEP instruments may also be useful in Japan.
We also note that the organization and financing of medical care entails trade-offs among the four categories. For example, in Japan, because of easy access to care at all levels of specialization, a person with a headache may choose not to go to the nearby internist who knows her well, and may go instead directly to a neurologist or a neurosurgeon at the outpatient clinic of a famous university hospital: Japan’s system of health insurance facilitates the sacrifice of physical accessibility and continuity of care for perceived medical competence. This system may be applauded because it gives patients a wide range of options, and also criticized as an inefficient use of medical resources. By contrast, if primary-care physicians function as gatekeepers, continuity of care might be maintained at the expense of perceived medical competence or psychological accessibility. No single system for organizing medical care is appropriate in all contexts, and patients’ concerns can help to guide policies and goals including education for primary care physicians. With that in mind, future studies might be directed at finding out whether (and, if so, how) Japanese people prioritize their desires among the categories described here.
The categories documented here can be a vantage point from which different systems may be viewed, to ensure that decisions in health-care policy and in medical practice incorporate patients’ perspectives. As one example, information about patients’ priorities might be used together with measures of quality of care to illuminate areas combining low quality with high priority, which would then naturally be seen as the most appropriate areas for quality improvement.


Summary of implications for practice

Primary care will likely become more important in Japan, but little is known of Japanese people’s views of an ideal primary care physician.
This study revealed Japanese people want primary care physicians who are physically accessible, medically competent, psychologically accessible, and familiar with them. Compared to results in Western countries, these results suggest that people of different cultural backgrounds nonetheless have generally similar desires of their primary-care physicians, and that these categories can be useful in cross-cultural studies and to ensure that decisions in health-care policy and in medical practice incorporate patients’ perspectives.


Acknowledgments

We thank Thomas S. Inui, Jon Adams, George K. Freeman and Alison P. Hill for their valuable advice and assistance with this study and Byotai seiri kenkyujo for funding.


References

1 Ikegami N, Campbell JC. Health care reform in Japan: the virtues of muddling through. Health Aff (Millwood) 1999; 18: 56–75.
2 Starfield B. Primary Care: Balancing Health Needs, Services, and Technology. New York: Oxford University Press, 1998: 335–56.
3 Guo Y, Kuroki T, Yamashiro S, Koizumi S. Illness behaviour and patient satisfaction as correlates of self-referral in Japan. Fam Pract, 2002; 19: 326–32.
4 Arai Y, Ikegami N. Health care systems in transition. II. Japan, Part I. An overview of the Japanese health care systems. JPublic Health Med, 1998; 20: 29–33.
5 Ministry of Health and Welfare. Kokumin Eisei no Doko [Health State of the Nation, Annual Report]. Tokyo: Kosei Tokei Kyokai, 1999: 178–81 [In Japanese].
6 Otaki J. Considering primary care in Japan. Academy of Med, 1998; 73: 662–8.
7 Kakaritsukei kenkyukai (JP). Kakaritsukei kino no hyoka ni kansuru kenkyu in 1993–4. Tokyo: Kakaritsukei kenkyukai, 1994. [In Japanese].
8 Institute of Medicine (US). Defining primary medical care. An interim report. Washington, DC: National Academy Press, 1994.
9 Safran DG, Kosinski M, Tarlov AR. et al. The primary care assessment survey: test of data quality and measurement performance. Medical Care, 1998; 36: 728–39.
10 Shi L, Starfield B, Xu J. Validating the adult primary care assessment tool. J Fam Pract, 2001; 50: 161–75.
11 Ramsay J, Campbell JL, Schroter S, Green J, Roland M. The General Practice Assessment Survey (GPAS): tests of data quality and measurement properties. Fam Pract, 2000; 17: 372–9.
12 Grol R, Wensing M, Mainz J. et al. Patient in Europe evaluate general practice care: an international comparison. Br J Gen Pract, 2000; 50: 882–7.
13 Glaser BG, Strauss AL. The Discovery of Grounded Theory. New York: Walter De Gruyter Inc, 1967.
14 Sebata K, Sugisawa Y, Fetters MD, Maezawa M. [Individual interviews with elderly in a rural area: expectations for medical care services] Nippon Koshu Eisei Zasshi. 2002; 49: 739–48 [In Japanese].
15 Sebata K, Sugisawa Y, Fetters MD, Hiraga A, Ohtaki J, Maezawa M. [Focus group investigation on expectations of elderly people from health care facilities and physicians] Nippon Koshu Eisei Zasshi. 2002; 49: 114–25 [In Japanese].
16 Arborelius E, Bremberg S. What does a Human Relationship with the Doctor mean? Scandiavian Journal of Primary Health Care, 1992; 10: 163–9.
17 Wensing M, Grol R, van Montfort P, Smith A. Indicators of the quality of general practice care of patients with chronic illness: a step towards the real involvement of patients in the assessment of the quality of care. Quality Health Care, 1996; 5: 73–80.
18 Wensing M, Jung HP, Mainz J, Olesen F, Grol R. A systematic review of the literature on patient priorities for general practice care. Part 1: Description of the research domain. Society of Science Medicine, 1998; 47: 1573–88.


Table 1 Composition of focus-groups

Group number	No. participants	Mean age	No. participants having a primary-care physician
	(Male, female)	(years)
1	5 (2,3)	68	4
2	9 (3,6)	74	7
3	8 (0, 8)	47	6
4	5 (2,3)	34	0
5	5 (2,3)	60	2
All groups	32 (12 20)	58	19

Table 2   Characteristics of participants in the focus-group discussions

	Number of participants
Age (years)	20–39
7
40–59	9
60 or over	16
Sex
Male	12
Female	20
Employment status
Employed	15
Unemployed	15
Not known	2
Having a primary-care physician
Yes	19
No	13

Table 3 Characteristics of an ideal primary-care physician

Our results	Items in Japan survey (1993)	Scales in PCAS (1996)	Wensing et al. (1996)
Physical accessibility	Doing consultation out of office	Financial access	Waiting times
	hours and on holiday	Organizational access	Flexibility
	House call		Telephone consultations
	Near the house		Physical accessibility
	Available to contact and get		Financial accessibility
	advice whenever patient wants		Premises
Medical competence	Managing all problems	Preventive counseling	Cooperation
	Well-equipped surgery	Integration	Efficiency
	Referring to an appropriate	Physical examinations	Special services available
	physician as necessary		Effectiveness
	Being skillful		Burden
	Doing health consultation		Competent/accuracy
			Stimulating self-help
			Supporting patients’ relatives
Psychological	Explaining condition of the	Communication	Exploring patients’ needs
accessibility	illness and treatment well	Interpersonal treatment	Patients’ involvement in
	Listening to the patient well	Trust	decisions
	Trustworthy		Patients’ privacy
			Time for patient care
			Informativeness
Familiarity with the	(Nothing)	Longitudinal continuity	Continuity
patients		Visit-based continuity	Humaneness
		Contextual knowledge of patient	Counseling

^top