Cost Sharing for Acupuncture Therapy in Commercial Insurance Plans

Abstract

Background

Cost sharing, or the extent to which patients contribute to health care spending, has been linked to various outcomes. The relationship between cost sharing and acupuncture utilization is unclear.

Objectives

To measure the association between cost sharing (eg, copays, share of spending paid out of pocket, and consumer-driven health plans marked by high deductibles) and the use of in-network acupuncturists.

Methods

Our team used commercial insurance claims. The study sample included 105 501 individuals who visited an in-network acupuncturist between 2012 and 2021.

Results

In plans with less out-of-pocket spending overall, more members used an in-network acupuncturist. Plans with $0 copays and consumer-driven health plans had less acupuncture utilization compared to plans with higher copays and non-consumer driven health plans.

Conclusion

Cost sharing had an inconsistent impact on acupuncture utilization. Our findings suggest that access to in-network acupuncturists is more important than cost sharing when it comes to increasing acupuncture utilization.

Introduction

Acupuncture is a safe and effective nonpharmacologic treatment that millions of Americans use for numerous health conditions.^1,2 While patients are increasingly using their insurance to pay for acupuncture, insurance plans use a variety of mechanisms to manage utilization, such as prior authorization. ³ For instance, Medicare Part B only approves its acupuncture coverage for patients with chronic low back pain who are using approved Medicare providers and allows patients to receive twelve sessions within a 90 day period, although 8 additional sessions are covered if patients have symptom improvement. ⁴

One of the most important mechanisms that insurers use to manage utilization is cost sharing, which refers to copays, coinsurance, and deductibles that require patients to pay for a portion of health care spending. Multiple studies have demonstrated that cost sharing can reduce the use of preventative care, the timeliness of accessing care, and adherence to life-saving medications like statins or insulin, which in turn may increase mortality.^5-7 Researchers have also found that a lower socioeconomic status can make individuals more price sensitive, thereby exacerbating the effects of cost sharing. ⁸

To our knowledge, no study has assessed the relationship between cost sharing and acupuncture utilization. Researchers have considered cost sharing for other types of pain management, however. In an observational study of more than 100 000 patients with new-onset low back pain, patients without a copay and patients with lower deductibles were more likely to select physical therapists and chiropractors as their entry-point provider rather than a primary care physician. ⁹ A second study examined cost sharing and opioid prescriptions, finding higher fill rates among postoperative Medicare patients who had less cost sharing. ¹⁰

In the current study, the relationship between cost sharing and acupuncture utilization was evaluated using a large, de-identified insurance claims database, focusing on insurance plans that include at least 1 in-network acupuncturist. Three dimensions of cost sharing were captured for acupuncture: (1) copays, which are a flat fee due at the point of service, (2) the share of total spending paid out of pocket, and (3) consumer-driven health plans, which are high deductible health plans with either a health savings account or a health reimbursement arrangement. In addition to measuring trends in cost sharing between 2012 and 2021, our team examined whether copays, deductibles, and total out-of-pocket spending were associated with acupuncture utilization. It was hypothesized that there would be less utilization in insurance plans with more cost sharing for care provided by in-network acupuncturists.

Methods

This study was approved by the University of Pennsylvania’s Institutional Review Board. Analysis was conducted between November 2024 and March 2025. Data from Optum’s de-identified Clinformatics® Data Mart Database (Optum® CDM) between 2012 and 2021 were used. The study sample included individuals with commercial insurance, which contains self-insured (ie, when an employer directly manages and funds medical claims) and fully-insured plans (ie, when an employer pays a fixed premium to an insurer, who covers medical claims). The focus was on patients who had visits with in-network acupuncturists, identified using a unique provider code in Optum® CDM. These visits included a Current Procedural Terminology code of 97810 (ie, first 15 min of acupuncture needling) or 97813 (ie, first 15 min of electroacupuncture). We examined paid claims and required that acupuncturists participate in the plan’s network (ie, were “in-network”) to ensure that cost sharing information was available.

First, the use of in-network acupuncturists over time and the characteristics of plan members who used an in-network acupuncturist were measured, including age, sex, race/ethnicity (non-Hispanic Asian, non-Hispanic Black, Hispanic, and non-Hispanic White), educational attainment (high school or less, 1-3 years of college, 4 or more years of college), annual household income (<$40,000, $40,000-$99,999, >$100,000), and region (East North Central, East South Central, Middle Atlantic, Mountain, New England, Pacific, South Atlantic, West North Central, West South Central).

Next, trends in cost sharing for in-network acupuncturists were tracked using 2 measures: (1) average copay, adjusted for inflation using the consumer price index for medical care, ¹¹ and (2) the share of total spending paid out of pocket. Out-of-pocket spending by consumers can include copays, deductibles (ie, the full amount paid for a given service, which is covered by the insured until they reach a specified limit per their plan), and/or coinsurance (ie, a partial amount paid for a given service, which generally occurs after the patient exceeds their deductible). Because in-network acupuncturists billed for multiple procedure codes per visit, including treatments like manual therapy and evaluation and management codes in addition to acupuncture, spending at the patient-provider-day level was summed.

Finally, we evaluated whether cost sharing was associated with any visit and the number of visits with in-network acupuncturists using adjusted means. Specifically, the share of plan members with in-network acupuncturists who had at least 1 visit with an acupuncturist was estimated; conditional on having at least 1 visit, the number of visits in a given year was measured. Acupuncture utilization was compared across 7 categories: plans where (1) the average copay was $0, (2) the average copay was between $0 and $20, (3) the average copay was $20 or over, (4) the share of total spending paid out of pocket was under 20%, (5) the share of total spending paid out of pocket was 20% or over, (6) the plan was a consumer-driven health plan, and (7) the plan was not a consumer-driven health plan. We adjusted means for age, sex, race/ethnicity, income, and educational attainment and included year fixed effects using linear regressions and 99% confidence intervals to allow for multiple, pairwise comparisons across the groups; differences were deemed statistically significant if the confidence intervals did not overlap.

Results

The study sample included 105 501 individuals who received care from in-network acupuncturists and another 11 075 089 individuals who did not receive care from an in-network acupuncturist across 12 894 unique plans (Table 1). There was an increase in the use of in-network acupuncturists over time, from 10 411 plan members in 2012 to 18 812 plan members in 2021 (Appendix Figure 1).

Table 1.

Study Sample Stratified by whether They had One or More Visit With an In-Network Acupuncturist, 2012-2021

	Acupuncturist users		Other plan members
Age	40.0	SD = 13.0	31.9	SD = 19.0
Sex
Female	66.3%	69 946	51.7%	5 727 461
Race/ethnicity
Asian	19.9%	20 985	7.2%	799 898
Black	4.3%	4565	8.7%	963 712
Hispanic	11.5%	12 145	11.3%	1 254 162
White	56.8%	59 928	58.5%	6 480 535
Missing	7.5%	7884	14.2%	1 576 782
Household income
<$40,000	10.0%	10 517	12.7%	1 408 763
$40,000-$99,999	26.1%	27 607	27.5%	3 054 249
>$100,000	50.7%	53 515	37.5%	4 153 776
Education
High school or less	6.9%	7229	15.0%	1 655 747
1-3 years of college	41.6%	43 884	45.9%	5 083 570
4+ years of college	48.3%	49 148	26.7%	2 960 692
Region
East North Central	2.4%	2509	10.5%	1 159 536
East South Central	0.2%	254	3.3%	369 906
Middle Atlantic	6.9%	7297	10.9%	1 202 164
Mountain	8.8%	9262	10.8%	1 195 208
New England	2.3%	2470	4.0%	444 957
Pacific	54.4%	57 436	16.4%	1 812 083
South Atlantic	11.9%	12 517	19.5%	2 163 750
West North Central	3.5%	3642	7.0%	779 425
West South Central	9.5%	10 013	17.3%	1 915 153
No. of individuals	105 501		11,075 089
No. of plans	12 894

Plan members who used an in-network acupuncturist were mostly female (66.3%; n = 69 946 out of 105 501) with an average age of 40.0 years; 19.9% were non-Hispanic Asian (n = 20 985 out of 105 501), 4.3% were non-Hispanic Black (n = 4565 out of 105 501), 11.5% were Hispanic (n = 12 145 out of 105 501), and 56.8% were non-Hispanic White (n = 59 928 out of 105 501). Over half (50.7%; n = 53 515 out of 105 501) of individuals had an annual household income of over $100,000 compared to 10.0% (n = 10 517 out of 105 501) who had an annual household income below $40,000; 48.3% (n = 49 148 out of 105 501) had 4 or more years of college compared to 6.9% (n = 7229 out of 105 501) who had a high school degree or less. Plan members who did not use in-network acupuncturists were more likely to be male (48.3%; n = 5 347 628 out of 11 075 089), were younger on average (31.9 years, SD = 19.0 years), were more likely to be Black (8.7%; n = 963 712 out of 11 075 089) and less likely to be Asian (7.2%; n = 799 898 out of 11 075 089), and were less likely to be high-income (37.5%; n = 4 153 776 out of 11 075 089) and college-educated (26.7%; n = 2 960 692 out of 11,075 089).

After adjusting for inflation using the consumer price index for medical care, which accounts for increasing prices over time, the average copay declined by approximately $5, from $20.52 to $15.09, between 2012 and 2021 (Appendix Figure 2). The share of total spending paid out of pocket, which included copays, coinsurance, and/or deductibles, was relatively stable, ranging from 17.1% in 2017 to 20.8% in 2013. The distribution of copays and the share of spending paid out of pocket revealed that nearly 40% had a $0 copay, and that less than 20% had no out-of-pocket spending.

The relationship between cost sharing arrangements and acupuncture utilization was inconsistent (Figure 1). In plans with a $0 copay, 0.33% of plan members used an in-network acupuncturist at least once; in plans with a copay between $0 and $20, 0.57% of plan members used an in-network acupuncturist at least once; and in plans with a copay of $20 or more, 0.64% of plan members used an in-network acupuncturist at least once. Among in-network acupuncturist users, the relationship between copays and the number of visits was U-shaped, with the highest number of visits occurring among individuals with a copay between $0 and $20. All differences were statistically significant (P < 0.001).

Figure 1.

Share of Individuals With Any Acupuncturist Visit and Mean Number of Acupuncturist Visits Based on Cost Sharing Arrangement, 2012-2021. Note. Means are adjusted for age, sex, race/ethnicity, income, education, and year. 99% confidence intervals are included. Consumer-driven health plans are marked by high deductibles with a savings option.

The relationship between out-of-pocket spending and acupuncture utilization was as expected. In plans with a share of out-of-pocket spending between 0% and 20%, 0.77% of plan members used an in-network acupuncturist at least once and had an average of 7.6 visits. In plans with a share of out-of-pocket spending of 20% or above, 0.39% of plan members used acupuncture at least once and had an average of 6.9 visits. The relationship between consumer-driven health plans and acupuncture utilization was also as expected: plan members with a high deductible were less likely to use acupuncture than plan members with a lower deductible. All differences were statistically significant (P < 0.001).

Discussion

More Americans are using their insurance to pay for acupuncture therapy, and these trends are likely to continue given Medicare Part B’s expansion of insurance coverage for chronic low back pain in 2020. ⁴ Yet information on coverage decisions is often unclear. For instance, Bleck et al ¹² (2021) performed a scoping review of insurance coverage for acupuncture, concluding that most states did not publish information regarding reimbursement.

Patients’ out-of-pocket spending affects a wide array of health care utilization, ranging from cardiac rehabilitation to mental health consultations.^13,14 Here, it was revealed that cost sharing correlated with the use of in-network acupuncturists in unexpected ways. Insurance plans with $0 copays had a smaller share of members who used acupuncture compared to plans with higher copays. Accounting for overall cost sharing, which included copays, coinsurance, and/or deductibles, and consumer-driven health plans, which are marked by a high deductible, a more expected relationship was seen: a smaller share of spending paid out-of-pocket and lower deductibles were associated with more acupuncture utilization. Combined, our findings suggest that acupuncture coverage by insurance plans is more important than cost sharing when it comes to individuals increasing their use of in-network acupuncturists.

The inconsistent relationship between cost sharing and acupuncture utilization may reflect other differences in plan design. For instance, plans with $0 copays may be more likely to have other types of “utilization management,” such as prior authorization or a cap on the number of visits, or have fewer acupuncturists who are proximate and in-network. It could also be that the socioeconomic status of acupuncture users, who were predominately high-income in our study sample, makes them less price sensitive to copays. Due to our reliance on insurance claims, which lack information on the plan’s benefits and individuals’ price sensitivity, assessing these possibilities was not possible.

Of note, only 12 894 insurance plans included an in-network acupuncturist during the study period, which amounted to less than 2% of the total number of insurance plans in Optum® CDM. These plans accounted for roughly 20% of the total number of covered lives, however. This information adds important context to the previous finding that half of acupuncturist visits are covered by insurance plans—in fact, the vast majority of plans in our study did not have an in-network acupuncturist.

There were important limitations. Our analysis relies on insurance claims, which have not been verified with patient reports of their cost sharing arrangement, are not subject to chart review, and may be subject to administrative error. Moreover, we restricted our sample to commercially-insured patients who have insurance plans with in-network acupuncturists in the U.S., which comprised a minority of insurance plans and which limits the generalizability of findings. More research on other commercial and public insurance plans could unveil different approaches to acupuncture coverage. We were not able to capture services that were not received due to high levels of cost sharing. Finally, we were unable to discern other aspects of insurance plan design, such as the use of prior authorization, which could have an important impact on the utilization of acupuncture, nor could we account for patient selection into insurance plans.

Conclusion

Using a large, de-identified commercial insurance claims database, this study found that patients with insurance plans with a larger share of out-of-pocket spending and with higher deductibles had less utilization of in-network acupuncturists. Unexpectedly, it was discovered that insurance plans with higher copays had more utilization of in-network acupuncturists. This study contributes to a large literature on cost sharing by showing that it can have an inconsistent relationship with acupuncture utilization.

Supplemental Material

Supplemental Material - Cost Sharing for Acupuncture Therapy in Commercial Insurance Plans

Supplemental Material for Cost Sharing for Acupuncture Therapy in Commercial Insurance Plans by Molly Candon, PhD, Jeffery A. Dusek, PhD, Arya Nielsen, PhD, Martin Cheatle, PhD, Rachel M. Werner, MD, PhD and David Mandell, ScD in Global Advances in Integrative Medicine and Health