Japanese physicians' preferences for decision making in rheumatoid arthritis treatment.

BACKGROUND: Rheumatoid arthritis (RA) is a complex chronic illness requiring continued medical care. During the past decade, the therapeutic options for RA have increased significantly; these often have a higher risk of adverse effects and are more expensive than traditional drugs. Rheumatologists may hence face difficulties when deciding on the optimal modality in initiating or changing treatment. The aim of this study was to explore the Japanese physicians' usual style of and preferences for decision making regarding RA treatment.
METHODS: This was a cross-sectional study conducted using an Internet survey. Respondents were asked about their usual style of making treatment decisions (perceived style), and their perception of the importance of physicians' actions and patients' attitudes.
RESULTS: Of the 485 physicians who were sent the questionnaire, 157 responded completely (response rate: 32.3%). Ninety-two percent of the respondents were men, and 57% were clinicians with more than 20 years of experience. Their specialties were general medicine (29%), rheumatology (27%), orthopedics (31%), and rehabilitation (12%). Sixty-one (39%) stated that they usually presented multiple treatment options to their patients and selected a decision for them, 42 (27%) shared the decision making with their patients, 34 (22%) let their patients choose the treatment, and 20 (13%) made the treatment decision for the patients. Physicians using the shared decision making (SDM) style desired for their patients to have supportive family and friends, to discuss with nurses, and to follow the doctors' directions more strongly compared with physicians using the other styles. There were no significant differences in sex, duration of clinical experience, major place of clinical work, and number of patients per month by the styles. More number of rheumatologists and physicians with specialist qualifications stated that they practiced SDM.
CONCLUSION: To enhance patient participation, physicians need to recognize the importance of discussing treatment options with patients in addition to giving them information.

Introduction

Rheumatoid arthritis (RA) is a complex chronic illness requiring continued medical care, including medication, rehabilitation, and surgical treatment in some cases. During the past decade, the therapeutic options for RA have increased significantly with the development of synthetic and biological disease-modifying antirheumatic drugs (DMARDs). Treatment strategies have fundamentally changed from controlling the symptoms to achieving remission.1 New and highly effective DMARDs, including biologics, have a higher risk of adverse effects, such as serious infections,2 heart failure, malignancy, and induction of autoimmunity. Furthermore, they are more expensive than traditional drugs. Rheumatologists may hence face difficulties when deciding on the optimal modality in initiating or changing treatment. The European League Against Rheumatism recommended the following three overarching principles for the management of RA in 2010 and 2013.3,4

Treatment of RA patients should aim to provide the best care and must be based on a shared decision between the patient and the rheumatologist.

Rheumatologists are the specialists who should primarily care for RA patients.

RA incurs high individual, societal, and medical costs, all of which should be considered in its management by the treating rheumatologist.

We previously reported that the majority of Japanese patients with RA desired to collaborate with the physicians in making treatment decisions.5,6 However, only half of the patients were actually able to share the decision making with their physicians. A survey on arthritis patients in the Netherlands showed the same results; most patients preferred shared decision making (SDM) but patients’ preferences were met in only half of the respondents.7 SDM is a collaborative effort between a patient and a physician. However, there have been very few studies to date on physicians’ preferences for treatment decision making in Japan. The aim of this study was to investigate the Japanese physicians’ usual style of decision making regarding RA treatment.

Methods

Ethical aspects

All physicians who answered the email questionnaire agreed with the purpose of this study. Based on this context, the ethics committee of Tokyo Medical University Hachioji Medical Center determined that approval was not necessary.

Survey

This was a cross-sectional study that was conducted between March 15, 2012 and March 21, 2012 using an Internet survey (PLAMED Co., Ltd., Tokyo, Japan). Emails with a link to the survey were sent to Japanese physicians who were treating RA patients. No honorarium was offered.

Questionnaire

Questions were developed with reference to the decision-making examples by Shepherd et al8 and Murray et al,9 and are listed in Table S1. In Question 1, respondents were asked to select one option that most accurately reflects their usual style in making a treatment decision (perceived style). Options offered were: 1) you select a treatment for your patient on the basis of what you think is best (henceforth referred to as “paternalism”); 2) you present multiple options to your patient and make the decision for the patient; 3) you let your patient choose the treatment after presenting some treatment options; and 4) you discuss treatment options with your patient and then come to a decision together (henceforth referred to as SDM). Then, they were asked to rate the importance of physicians’ actions and patients’ attitudes, using the 5-point Likert scale (from 1= not important at all or not desirable at all to 5= very important or strongly desirable). We also asked the physicians how important they considered 12 factors when deciding on treatments. The 12 factors consisted of common indicators of disease activity in RA,10 such as swollen and tender joint count, erythrocyte sedimentation ratio or C-reactive protein (CRP) level, measures of functional status, and predictive factors,11 such as radiographic findings of the joint and presence of the anticyclic citrullinated peptide antibody. Respondents were also asked to indicate their sex, years of clinical experience, major place of clinical work, specialty, presence or absence of specialist qualifications received from the Japan College of Rheumatology or Japanese Orthopaedic Association, and the number of RA patients treated per month.

Statistical analysis

Statistical analysis was performed using SPSS statistics version 22 for Windows.

Results

Response rate

Among the 485 physicians who were sent the questionnaire, 157 responded completely (response rate: 32.3%).

Characteristics of the respondents

Characteristics of the respondents are shown in Table 1. Ninety-two percent of the respondents were men, and 57% were veteran clinicians with more than 20 years of experience. The specialties of the respondents were general medicine (29%), rheumatology (27%), orthopedics (31%), and rehabilitation (12%).

Table 1

Characteristics of the respondents

Characteristic	Total n=157 (%)
Sex
Woman	13 (8.3)
Man	144 (91.7)
Duration of clinical experience
<10 years	15 (9.6)
10–19 years	52 (33.1)
≥20 years	90 (57.3)
Specialty
General medicine	46 (29.3)
Rheumatology	43 (27.4)
Orthopedics	49 (31.2)
Rehabilitation	19 (12.1)
Major place of clinical work
University hospital	37 (23.6)
Hospital	63 (40.1)
Clinic	57 (36.3)
Specialist qualifications
Yes	78 (49.7)
No of patients treated in 1 month
<10	58 (36.9)
10–49	52 (33.1)
>50	47 (29.9)

Perceived styles of treatment decision making

Sixty-one respondents (39%) stated that they usually presented multiple treatment options to their patients and selected a decision for them, 42 (27%) shared the decision making with their patients, 34 (22%) let their patients choose the treatment, and 20 (13%) made the treatment decision on the basis of what they thought was best for the patients (Table 2).

Table 2

Perceived styles of physicians’ actions

Perceived style	Physicians n (%)	Number (%) of physicians who answered “very important”
		Provide information to patients	Present multiple options to patients	Make an effort to answer questions	Discuss options with patients
Q1:1 Select a treatment for the patient	20 (12.7)	9 (45)	0 (0)	10 (50)	5 (25)
Q1:2 Present many treatment options and select one	61 (38.9)	27 (44.3)	12 (19.7)	22 (36.1)	19 (31.1)
Q1:3 Let the patient choose	34 (21.7)	18 (52.9)	16 (47.1)	17 (50)	16 (47.1)
Q1:4 Shared decision making	42 (26.8)	27 (64.3)	19 (45.2)	24 (57.1)	25 (59.5)
Total	157	81 (51.6)	47 (29.9)	73 (46.5)	65 (41.4)
P-value		0.22	<0.01	0.18	0.01

Note: Questions are listed in Table S1.

Abbreviation: Q, question.

Perceptions of physicians’ actions by perceived styles

In all four groups of treatment decision-making styles, both giving information and making an effort to answer patients’ questions were considered important to the same degree. Presenting multiple options and discussing with patients were considered more important by the physicians of the SDM style and those who let patients choose the option, compared with the other two groups (Table 2). From a different perspective, only 25 physicians (38%) stated that they performed SDM among the 65 physicians who considered discussion with patients to be very important.

Physicians’ desire regarding patients’ attitudes

Physicians whose perceived style was SDM wanted their patients to have supportive family and friends, to tell their desires and concerns about treatments to a nurse, and to follow the doctors’ directions more strongly than physicians using the other styles (Table 3).

Table 3

Physicians’ desires regarding patients’ attitudes

Perceived style	Patients, n	Have correct information	Actively research information	Number (%) of physicians who answered “strongly desired”
				Have supportive family and friends	State desires to the physician	State desires to the nurse	Ask questions	Follow the doctor’s directions
Q1:1 Select a treatment for the patient	20	7 (35)	3 (15)	4 (20)	4 (20)	1 (5)	5 (25)	9 (45)
Q1:2 Present many treatment options and select one	61	26 (42.6)	7 (11.5)	8 (13.1)	14 (23)	6 (9.8)	21 (34.4)	28 (45.9)
Q1:3 Let the patient choose	34	15 (44.1)	5 (14.7)	8 (23.5)	8 (23.5)	4 (11.8)	8 (23.5)	12 (35.3)
Q1:4 Shared decision making	42	23 (54.8)	7 (16.7)	17 (40.5)	13 (31)	12 (28.6)	20 (47.6)	28 (66.7)
Total	157	71 (45.2)	22 (14)	37 (23.6)	39 (24.8)	23 (14.6)	54 (34.4)	77 (49)
P-value		0.46	0.9	0.02	0.74	0.03	0.12	0.04

Note: Questions are listed in Table S1.

Abbreviation: Q, question.

Univariate analysis of perceived style by physicians’ characteristics

The initial four styles of decision making were collapsed into two: SDM and non-SDM. We analyzed the data using crosstabs and chi-square tests to identify significant predictors of decision-making style. Results of the univariate analysis are presented in Table 4. There were no significant differences in sex, duration of clinical experience, major place of clinical work, and number of patients per month by the styles. Fewer physicians whose specialty was general medicine and who were working at clinics stated that they practiced SDM. More number of rheumatologists and physicians with specialist qualifications stated that they practiced SDM (Table 4).

Table 4

Univariate analysis of perceived style by physician characteristics

Characteristic	SDM 42 n (%)	Non-SDM 115 n (%)	χ2 (df)
Sex
Woman	1 (7.7)	12 (92.3)	=3.27 (1)
Man	41 (28.5)	103 (71.5)
Duration of clinical experience
<10 years	5 (33.3)	10 (66.7)	=0.68 (2)
10–19 years	15 (28.8)	37 (71.2)
>20 years	22 (24.4)	68 (75.6)
Specialty
General medicine	7 (15.2)*	39 (84.8)*	=7.23 (3)
Rheumatology	17 (39.5)*	26 (60.5)*
Orthopedics	14 (28.6)	35 (71.4)
Rehabilitation	4 (21.1)	15 (78.9)
Major place of clinical work
University hospital	14 (37.8)	23 (62.2)	=4.96 (2)
Hospital	18 (28.6)	45 (71.4)
Clinic	10 (17.5)*	47 (82.5)*
Specialist qualifications
Yes	27 (34.6)	51 (65.4)	=4.94 (1)**
No	15 (19.0)	64 (81.0)
No of patients treated in 1 month
<10	14 (24.1)	44 (75.9)	=3.08 (2)
10–49	11 (21.2)	41 (78.8)
>50	17 (36.2)	30 (63.8)

Notes:

Absolute value of adjusted residual >1.96;

P<0.05.

Abbreviations: SDM, shared decision making; df, degrees of freedom.

Physicians’ views of the importance of various factors for treatment decision making

Adverse effects were considered to be equally important as the effects of a treatment in deciding on a treatment by the physicians. Factors indicating disease activity were also considered important. The financial burden of the treatment on the patients was considered important to the same degree as disease activity. The rate of physicians who considered that the patients’ preference for treatment was very important was 14% (Table 5).

Table 5

Physicians’ perception of the importance of various factors in deciding on a treatment

Physicians’ perception	Number of physicians who answered each option
	1	2	3	4	5	Rate of no 5 (%)	Rate of no 4+5 (%)
	Not important				Very important
Q4:11 Information about adverse effects of the treatment	0	3	6	87	61	38.9	94.3
Q4:10 Information about treatment effects	0	2	9	89	57	36.3	93.0
Q4:6 Acute-phase reactions (abnormal serum CRP or ESR)	0	2	8	91	56	35.7	93.6
Q4:3 Radiographic changes of typical erosion in the affected joints	0	2	16	83	56	35.7	88.5
Q4:8 Functional status (disability), usually assessed by HAQ score	0	2	12	94	49	31.2	91.1
Q4:2 Number of swollen or tender joints	0	2	17	92	46	29.3	87.9
Q4:5 Serum anti-CCP antibody	0	2	22	88	44	28.0	84.1
Q4:7 Activity of daily living at home, school, workplace, etc	0	2	11	103	41	26.1	91.7
Q4:12 Patients’ financial burden of treatment	0	3	16	101	37	23.6	87.9
Q4:1 Patients’ age	0	6	20	94	35	22.3	82.2
Q4:9 Patients’ preferences for treatment	0	5	24	106	22	14.0	81.5
Q4:4 Serum rheumatoid factor	0	10	41	83	18	11.5	64.3

Note: Questions are listed in Table S1.

Abbreviations: Q, question; ESR, erythrocyte sedimentation rate; HAQ, health assessment questionnaire; CRP, C-reactive protein; CCP, cyclic citrullinated peptide.

Discussion

Although 41% of the physicians answered that discussion between physicians and patients was very important, only 27% stated that they were actually practicing SDM in the RA outpatient setting. A survey of doctors practicing mainly oncology in Australia showed that 62% practiced SDM,8 and a study of physicians in the USA reported that 73% practiced SDM.9 A recent systematic review reported that physicians’ attitudes toward SDM tended to vary by specialty.12 German researchers reported that cultural differences had a significant influence on patients’ behavior regarding medical encounters.13 However, there are no studies to our knowledge regarding the role of cultural differences in determining the physicians’ attitudes toward decision making for treatment.

Most physicians considered providing information and answering patients’ questions to be very important regardless of whether or not they practiced SDM. A survey in the UK found that RA patients had a high desire for information.14 We believe that Japanese patients also want to be informed similarly, and therefore physicians should respond to this need.

We did not analyze the reasons of discordance between perceived style and preferred style of treatment. A systematic review12 divided the barriers against implementing SDM into three types: physician-related, patient-related, and condition/treatment-related. Nota et al15 investigated the barriers against arthritis patient participation by in-depth semi-structured interview, and reported three types: 1) doctor-related barriers, mostly concerning communication skills and a paternalistic attitude; 2) patient-related barriers, such as lack of knowledge, lack of awareness of having a choice, and lack of assertiveness; and 3) contextual barriers, such as too little time to decide.

Limitations

There are some limitations to this study that need to be considered. First, the data of this study were based on self-reports from physicians rather than observation. Second, the small sample size, particularly of female physicians and physicians with less than 10 years of clinical experience, limited our ability to detect differences among groups. Third, the numbers of RA patients treated in a month by the physicians were relatively small. Despite these limitations, our findings provided important information about the overall attitude of Japanese physicians.

Conclusion

To enhance patient participation, physicians need to recognize the importance of discussing treatment options with patients in addition to providing them with information. In the future, we would like to survey the factors preventing the widespread practice of SDM for RA treatment in Japan.

Table S1

Questions of the Internet survey

Q1	Which of the following is your usual method of deciding on a treatment for RA in the outpatient setting? Please select 1 of the 4 options.
	1	You select a treatment for your patient on the basis of what you think is best.
	2	You present multiple options to your patient but make the decision for the patient.
	3	You let your patient choose the treatment after presenting some treatment options.
	4	You discuss treatment options with your patient and then come to a decision together.
Q2	How important are the following actions for physicians when they decide on treatments? Please rate the importance on a 5-point scale (5: very important – 1: not important at all).
	1	Physicians give information about treatment to patients.
	2	Physicians present multiple treatment options to patients.
	3	Physicians make an effort to answer patients’ questions.
	4	Physicians discuss treatment options with patients.
Q3	How much do you want your patients to do the following for effective RA treatment? Please rate the importance on a 5-point scale (5: strongly – 1: not at all).
	1	Have correct information about treatments.
	2	Actively research information about treatments.
	3	Have supportive family and friends.
	4	State their desires and concerns about treatments to the physician.
	5	State their desires and concerns about treatments to a nurse.
	6	Ask medical staff questions about the disease and treatments.
	7	Follow the doctor’s directions.
Q4	How important is the following information for you to decide on a treatment? Please rate the importance on a 5-point scale (5: very important – 1: not important at all).
	1	Patient’s age
	2	Number of swollen or tender joints
	3	Radiographic changes of typical erosion in the affected joints
	4	Serum rheumatoid factor
	5	Serum anti-CCP antibody
	6	Acute-phase reactants (abnormal serum CRP or ESR)
	7	Activity of daily living at home, school, workplace, etc
	8	Functional status (disability), usually assessed by HAQ score
	9	Patient’s preferences for treatment
	10	Information about the effects of the treatment
	11	Information about adverse effects of the treatment
	12	Financial burden of the treatment on the patient

Abbreviations: RA, rheumatoid arthritis; CCP, cyclic citrullinated peptide; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; HAQ, health assessment questionnaire.