Acupuncture Use in the United States: Findings from the National Health Interview Survey

OBJECTIVE: Acupuncture has become an important provider-based complementary and alternative medicine (CAM) treatment. To improve understanding of its role in personal health care, an analysis of national data was conducted to examine user sociodemographics, conditions treated, and the relationship of use with conventional Western medical care. DESIGN: A nationally representative cross-sectional survey. SETTING: The 2002 National Health Interview Survey (NHIS), conducted in all 50 states and the District of Columbia. PARTICIPANTS: Thirty-one-thousand and forty-four (31,044) adults who completed the NHIS Sample Adult Core. OUTCOME MEASURES: The primary outcome measure was recent use of acupuncture, defined as use within the previous 12 months. RESULTS: In the 2002 NHIS sample, 4.1% of the respondents reported lifetime use, and 1.1% (representing 2.13 million Americans) reported recent use of acupuncture. Recent use (n = 327) was positively associated with being an Asian female, living in the West or Northeast, having poorer self-reported health status, a higher level of education, and being an ex-smoker. Among recent users, the most typical treatment regimen was two to four treatments (34.5%), with musculoskeletal complaints being the most frequently reported conditions, led by back pain (34.0%). Reports of perceived benefit were generally high. Respondents indicated that acupuncture was used both as an alternative and as a complementary therapy. A reasonable number also reported being referred to acupuncture by a conventional medical professional (25.3%). The cross-sectional nature of the data precluded analysis of transitions in health care use (between conventional and CAM treatments) over time. CONCLUSIONS: Utilization of acupuncture was somewhat lower than expected given its significant national and international recognition and its visibility in the media. This may in part be a function of provider availability and cultural factors.