Cost-effectiveness of Acupuncture in an Employee Population: A Retrospective Analysis

This retrospective study conducted at the Mayo Clinic investigated the cost-effectiveness of acupuncture as adjunct treatment for employees and their dependents presenting with pain symptoms. The research used data from January 2009 through June 2013, analyzing 466 patients who received acupuncture compared with 466 propensity-score-matched controls who did not receive the treatment. The primary objective was to determine whether acupuncture could generate savings in total healthcare costs by reducing the need for other medical care. The methodology employed a difference-in-differences analysis, comparing healthcare costs from 1 year before treatment with up to 14 months after acupuncture initiation.

Investigators used health insurance data and electronic medical records, including all medical and pharmaceutical costs. To make the groups comparable, an index date was established based on medical visits with diagnoses that would indicate acupuncture, such as headaches, back pain, joint pain, and anxiety-related conditions. The results demonstrated no cost savings associated with acupuncture treatment. In fact, monthly medical costs were consistently higher in the acupuncture group both before and after treatment.

The acupuncture group had medical costs roughly $500 higher per month in the year before treatment and more than $1,000 higher in the year afterward. When comparing pre- and post-treatment periods, monthly medical costs increased by $44 for acupuncture patients, while they decreased by $582 for controls. Subgroup analyses focused on patients who completed at least 6 treatments within 2 months, considered "better responders." Even this group showed no significant cost savings compared with controls. Approximately one quarter of patients were treated for back pain and another quarter for joint or limb pain.

Fewer than 8% received treatment for conditions not related to pain. The clinical implications of this study are complex. Although no cost savings were demonstrated, this does not necessarily invalidate the clinical benefits of acupuncture. The study has several important limitations that may have prevented detection of economic benefits.

Because it was retrospective, it was not possible to fully control for the factors that influenced the indication of acupuncture for some patients and not others. The heterogeneity of conditions treated may also have diluted potential positive effects that would be more apparent in studies focused on specific conditions. Another relevant factor is that not all patients completed the full course of treatments, which may have reduced overall effectiveness. The study reflects real-world conditions, in which factors such as work schedules and limited provider availability can affect treatment adherence.

In addition, acupuncture was delivered as an isolated treatment, unlike traditional Chinese practice, which typically includes nutritional counseling, exercise recommendations, and herbal treatments. Methodological limitations include the relatively small sample size given the high variability in healthcare costs, the presence of outliers in the data, and the fact that propensity scores can adjust only for measurable differences. The study population was predominantly White and female at a Midwestern academic medical center, which may limit generalization of the findings to other populations.