Date of Award


Degree Type


Degree Name

Doctor of Philosophy


Human Ecology

Major Professor

Robert J. Pursley

Committee Members

Michael Betz, Robert E. Levey, Thomas Hood, James Neutens


The purpose of the study was to determine the factors associated with the use of acupuncture for pain and to compare these factors to the use of other complementary and alternative medicine (CAM) and conventional medical services. The treatment of pain is expensive. In addition, use of surgery and medication increases risk of medical errors, complications, side effects and addiction. Clinical trials of acupuncture for pain have demonstrated results at least as effective as conventional treatments with negligible side effects and complications. Only 4% of the United States population has used acupuncture. The healthcare field is studying means to predict health-related behaviors by identifying characteristics of patients. This study identifies those characteristics that increase the likelihood of adoption of acupuncture for pain.

A review of the literature provided factors that were associated with use of acupuncture, CAM, and people with pain conditions. The Behavioral Model of Health Services Use is a predictive framework that classifies factors as predisposing, enabling and need variables. Data from the national probability sample 2002 National Health Information Survey (NHIS) were used in the study. The NHIS 2002 consisted of 31,044 adults aged 18 and older residing in the United States. The survey included a special section on CAM modalities including acupuncture. The response rate for the survey was 74.3%.

Survey respondents were classified as having pain conditions if they reported having at least one of eight conditions within the past year. The pain sample size was 18,103. Respondents were then categorized as acupuncture (n=257), other CAM (n=302) or conventional medicine (n=6,812) users based on use of each service within the past year.

SPSS was used to obtain descriptive and inferential statistics. Differences in type of service were analyzed using Chi-square for categorical data and ANOVA for continuous data. A multiple logistic regression model was used to determine those factors predictive of acupuncture use as compared to conventional medicine.

The study found that predisposing factors that predicted use of acupuncture were being younger, Asian, and having a higher level of education. Enabling factors were residing in the Eastern or Western regions of the country, and exercising at a moderate level. Need factors were higher numbers of pain conditions, and having functional limitations. The logistics regression model correctly identified 26.5% of acupuncture users and 93.8% of conventional medicine users. To improve the predictive ability of the Behavioral Model additional research is needed on beliefs and attitudes.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."