Comparison of Opioid Prescribing Patterns in the United States and Japan: Primary Care Physicians' Attitudes and Perceptions

Article Figures & Data

Table 1.

Demographic Characteristics of Survey Respondents and Self-Reported Frequency of Opioid^* Prescribing

	Japanese Respondents (n = 435)^{^†}	US Respondents (n = 198)	P Value^{^‡}
Sex			<.001
Female	70 (16.4)	96 (49.0)
Male	358 (83.6)	100 (51.0)
Age (years)			<.001
<40	184 (43.0)	57 (28.9)
41–50	132 (30.8)	64 (32.5)
51–60	95 (22.2)	54 (27.4)
>60	17 (1.0)	22 (11.2)
Use opioids to control acute pain^{^§} or chronic pain^{^‖}			<.001^{^¶}
Never/seldom	105 (24.1)	2 (1.0)
Occasionally to all the time	330 (75.9)	196 (99.0)
Use opioids to control acute pain			<0.001
Never/seldom	220 (50.6)	6 (3.0)
Occasionally to all the time	215 (49.4)	192 (97.0)
Use opioids to control chronic pain			<.001
Never/seldom	158 (36.3)	18 (9.1)
Occasionally to all the time	277 (63.7)	180 (90.9)
Frequency of opioid use in own country^**			<.001
Seldom/just right	396 (93.4)	9 (4.6)
Too often	28 (6.6)	187 (95.4)
Opioids are indicated for chronic pain^{^††}	317 (73.9)	115 (58.7)	<.001^{^‡‡}

Data are n (%). Totals may vary because some respondents did not complete all parts of the questionnaires.
↵* Opioids include tramadol, codeine, buprenorphine, pentazocine, morphine, fentanyl, methadone, and hydromorphone.
↵† A total of 461 responses were received, but 26 respondents (5.4%) had no opioid prescribing license, leaving 435 prescribers to analyze.
↵‡ P values from a χ² test unless otherwise specified.
↵§ Acute pain includes acute nonobstructing kidney stones and sprain or strain of any joint.
↵‖ Chronic pain includes chronic nonspecific back pain, osteoarthritis of any joint, fibromyalgia, or other, similar conditions.
↵¶ P value from Fisher exact test because of the small cell size.
↵** Personal belief that opioids are used too seldom/used just right versus opioids are used too often.
↵†† Very strongly/strongly/mildly agree versus neutral/disagree/strongly disagree.
↵‡‡ P values for adjusted odds ratios.

Table 2.

Physician Self-Report of Factors That Influence Opioid Prescribing for Acute Pain in the United States and Japan

	Agree/Strongly Agree, n (%)	OR^* (95% CI)	P Value^{^†}
Medically indicated
Japan	130 (52.6)	1.00 (ref)	—
United States	168 (86.6)	5.33 (3.15–9.05)	<.001
Standard of care		1.00 (ref)
Japan	66 (27.3)	4.57 (2.91–7.17)	—
United States	128 (66.0)		<.001
Patient expectation
Japan	121 (49.2)	1.00 (ref)	—
United States	89 (45.9)	0.94 (0.62–1.44)	.790
Maintaining patient satisfaction
Japan	119 (48.4)	1.00 (ref)	—
United States	98 (50.8)	1.22 (0.80–1.87)	.360
Legal expectation
Japan	16 (6.6)	1.00 (ref)	—
United States	29 (15.1)	2.47 (1.22–5.04)	.012

↵* Odds ratios (ORs) were estimated from logistic regression models that adjusted for sex (female, male) and age category (<40, 41–50, 51–60, >60 years).
↵† P values for adjusted odds ratios.
CI, confidence interval; ref, reference category.

Table 3.

Self-Report of Factors That Influence Opioid Prescribing for Chronic Pain in the United States and Japan

	Agree/Strongly Agree, n (%)	OR^* (95% CI)	P Value^{^†}
Medically indicated
Japan	211 (69.2)	1.00 (ref)	—
United States	146 (75.6)	1.29 (0.83–2.03)	.260
Standard of care
Japan	130 (42.9)	1.00 (ref)	—
United States	107 (56.0)	1.60 (1.07–2.39)	.023
Patient expectation
Japan	193 (63.1)	1.00 (ref)	—
United States	109 (56.5)	0.83 (0.55–1.25)	.380
Maintaining patient satisfaction
Japan	201 (65.3)	1.00 (ref)	—
United States	103 (53.6)	0.68 (0.45–1.02)	.061
Legal expectation
Japan	25 (8.6)	1.00 (ref)	—
United States	33 (17.6)	2.03 (1.09–3.77)	.026

↵* Odds ratios (ORs) estimated from logistic regression models that adjusted for sex (female, male) and age category (<40, 41–50, 51–60, >60 years).
↵† P values for adjusted odds ratios.
CI, confidence interval; ref, reference category.