Auricular point acupressure for older adults with chronic low back pain: a randomized controlled trial

Abstract

Objective

Efficacious modalities are limited in chronic low back pain (cLBP). We determined the efficacy of auricular point acupressure (APA) in older adults with cLBP.

Design

3-arm randomized controlled trial.

Setting

Baltimore, Maryland.

Participants

Participants, ≥60 years with cLBP, were randomized (1:1:1) to APA with ear points targeted to cLBP (T-APA, n = 92), points nontargeted to cLBP (NT-APA, n = 91), or waitlist education control (n = 89), and followed up to 6 months (6M). Participants in the APA groups received 4 weekly APA sessions; the education control group received 4 weekly educational sessions.

Intervention

APA.

Main Outcomes and Measures

Primary outcomes were pain (Numerical Rating Scale) and function (Roland-Morris Disability Questionnaire).

Results

There were 272 participants (174 women [64%]; mean [SD] age 70.0 [6.95] years; 62% non-White). Compared to control, the T-APA group had significant improvement on pain from baseline to postintervention and 1-month (1M) follow-up by 1.73 and 1.26 points (P ≤ .001) respectively. The NT-APA group achieved similar improvements in pain. The improvement in function by T-APA and NT-APA was significant at postintervention by 1.89 and 2.68 points (P = .04 and .004) respectively, minimal at 1M follow-up, but significant at 6M in both APA groups. There were no statistically significant differences in treatment responses between the APA groups. Both APA groups had higher responder rates in pain and function at postintervention and 1M follow-up compared to the control group (odds ratio ranged from 2.11 to 6.32). The APA effects were sustained at 6M follow-up.

Conclusions

APA treatments significantly improved pain and function compared to control; effects were sustained at 6M. APA should be recommended as a nonpharmacologic therapy for older adults with cLBP.

Clinical trial registration number

Trial registry: Clinicaltrials.gov; Trial ID: NCT03589703; URL: https://clinicaltrials.gov/ct2/show/record/NCT03589703

Introduction

Chronic low back pain (cLBP) is a major public health concern affecting over 45% of older adults.¹ It is the leading cause of disability and creates an enormous public health burden.² Pharmacotherapy is the predominant treatment for cLBP, but medications cause various side effects and contribute to the ongoing opioid crisis.^3,4 Both pharmacologic and nonpharmacologic therapies are recommended by the American College of Physicians’ (ACP) clinical practice guideline,⁵ yet none produce consistently large effects.⁵ In addition, implementation of these guidelines is challenging particularly for nonpharmacologic treatments.^6,7 In a recent interview of primary care physicians, the most common nonpharmacologic treatments used are physical therapy (78%), chiropractic care (21%), massage therapy (18%), and acupuncture (17%).⁶ A larger study sample surveying patients with low back pain reported that provider recommendations for nonpharmacologic treatments included self-care (79%), physical therapy (60%), and massage, acupuncture, or spinal manipulation combined (37%).⁸ Since 2020, acupuncture has been covered by Medicare for the management of cLBP.⁹ However, scale-up is challenging due to the lack of access to acupuncturists as well as time and cost constraints.^10–12

Apart from needling of body points in acupuncture therapy, auriculotherapy is another form of Traditional Chinese Medicine that is also rooted in European traditional medicine and has garnered interest particularly in pain management. Auriculotherapy is a treatment using ear point stimulation with needles, seeds, or electrical stimulation; it is based on theoretical understanding and experimentation demonstrating that the ear is a microsystem reflecting the entire body.¹³ The World Health Organization has long recognized auricular point nomenclature indicating standardized ear points to target that correspond to specific body parts.¹⁴ Functional magnetic resonance imaging supports correlations between ear points and brain pathways.^15–17

Auricular acupuncture is one method of auriculotherapy that uses specific points on the ear wherein the symptomatic body part induces skin surface changes, electrical conductivity, and tenderness in corresponding ear points.¹⁸ Acknowledging the value and simplicity of auricular acupuncture in pain management, the Department of Defense (DoD) and Veterans Health Administration partnered to offer auricular (battlefield) acupuncture training courses for providers, but widespread implementation and sustainability at the DoD has been limited owing to healthcare system and access barriers¹⁹ as well as the need to investigate other outcomes (eg, function) apart from pain severity.²⁰ Another approach to auricular acupuncture is the National Acupuncture Detoxification Association or NADA protocol, which has been shown to address substance abuse disorder and is currently being investigated for nonspecific chronic low back pain among older adults.²¹

Another form of auriculotherapy that is more accessible is auricular point acupressure (APA). APA involves noninvasive, acupuncture-like stimulation of ear points using pellets (nonmedicinal, Vaccaria seeds, 2 mm in diameter) taped to the ear instead of needles, providing sustained stimulation to ear points. Studies on CLBP suggested that stimulation of ear points using pressure for 3 minutes, 3 times daily, can provide pain relief.^22,23 Mechanistic studies on CLBP and other pain conditions provided support for APA,^24–28 and a pilot study demonstrated that an interventionist-administered, 4-week APA program was feasible and improved pain and function among older adults with cLBP.²² This is important because the prevalence of chronic pain, especially high-impact chronic pain, is greater among older adults.²⁹ Several other chronic pain studies and systematic reviews including meta-analysis have been conducted on auriculotherapy in general, less on APA specifically or among older adults, and most are underpowered.^30–34

We present the results of an adequately powered randomized controlled trial (RCT) to determine the effectiveness of APA in older adults with cLBP. To address ear point specificity, 2 APA protocols (ear points targeted to cLBP, T-APA, or ear points nontargeted to cLBP, NT-APA; Figure S1) were examined and compared to education control. We hypothesized that T-APA would yield significantly better cLBP outcomes (pain and function) than NT-APA or the control condition in older adults with cLBP.

Methods

This was a 3-arm (T-APA, NT-APA, and education control) crossover RCT. Participants enrolled in education control were randomly reassigned to either T-APA or NT-APA after 1 month (M); this report did not include control group data after reassignment to prevent bias. Outcomes were assessed at baseline, immediately postintervention, and at 1M, 3M, and 6M follow-ups. Recruitment occurred from March 1, 2019 to March 13, 2020, resumed from June 12, 2021 to November 18, 2021 (due to the COVID-19 pandemic), and was completed by December 1, 2022. The study protocol was approved by the Institutional Review Board of the Johns Hopkins School of Medicine; protocol details were previously published.³⁵ Approved protocol revisions after the first 20 participants were recruited included: (1) expanding age eligibility criteria from ≥65 years to ≥60 years due to many inquiries from those 60-64 years old to participate in the study and (2) re-randomizing the control group after their 1M follow-up (instead of 3M as initially planned) due to consistent requests to receive the intervention sooner having minimal improvement from education control. Guidelines from the Consolidated Standards of Reporting Trials (CONSORT) and the CONSORT extension, STandards for Reporting Interventions in Clinical Trials of Acupuncture (STRICTA, Table S1), were used to report this paper.^36,37

Participants, setting, and location

Individuals were included if they were ≥60 years of age, were able to read and write English, had cLBP that persisted at least 3M or pain on at least half of the days for the previous 6M, had worst pain intensity ≥ 4 on a 0-10 numerical rating scale (NRS) in the previous week, had no cognitive impairment (Mini-Mental State Examination >24),³⁸ and were able to apply pressure to the seeds taped on their ears. Individuals were excluded if they had malignant or autoimmune diseases (eg, rheumatoid arthritis), known acute compression fractures, or hearing aids. All participants provided written informed consent; 272 community-dwelling participants were recruited from metropolitan Baltimore, Maryland.

Randomization

After baseline assessments were obtained, participants were randomly assigned to 1 of 3 groups (1:1:1). Randomization occurred in blocks of 3 or 6 and was performed by the statistician through a random-number generator to create a list of group assignments before study recruitment began. A trained project coordinator enrolled and assigned participants to each group based on their group assignment. Participants in the education control group were re-randomized into either T-APA or NT-APA after completing their 1M follow-up. The decision to re-randomize the education control participants to one of the intervention groups was in response to emerging evidence that APA is a noninvasive and safe intervention. Offering this potentially beneficial intervention to the control group facilitated access to a treatment that may be advantageous for their pain and well-being. Without having to wait for a longer period, participants in the control group were provided with equitable access to APA.

Blinding

The APA participants were blinded regarding their APA group assignment. The co-investigators, data collectors, and analysts were also blinded to group assignments.

Intervention

The APA intervention included a 15-minute session/week of interventionist-administered APA at the clinic, followed by 5 days of self-administered stimulation on the prescribed ear points at home (3 times a day for 3 minutes each time), repeated for 4 weeks. The self-administered stimulation for 5 days after each weekly interventionist session offers a sustained, ongoing treatment (extended auricular therapy).³⁹ Participants were then instructed to remove the ear seeds on day 5 of each week after seed placement to let the ears rest and desensitize prior to the next week. Ear points for seed placements in the T-APA group were specific to cLBP35 (front and back ear points corresponding to the lumbar vertebral area; unilateral or bilateral ears based on the side affected with low back pain) and 3 master points known for alleviating stress and pain (shenmen, sympathetic, and nervous subcortex; Figure S1).²² Ear points for NT-APA did not include the master points and were nonspecific to cLBP (mouth, stomach, duodenum, internal ear, and tonsil).¹⁸ Each of these points were treated per weekly session with the precise number of seeds as agreed between the participant and interventionist based on comfort and coverage of seeds on individual ears. No further instructions were provided to continue treatment after the 4 weeks of APA intervention. Education control participants received the cLBP educational booklet published by the National Center for Complementary and Integrative Health⁴⁰ and visited the research office weekly with time and attention exposure similar to those in the APA groups.

The interventionists were trained by the PI-APA expert (C.H.Y.) who had a Master of Auricular Medicine since 2011. The interventionists were a research program supervisor with a Master of Business Administration, research coordinator with a Master of Public Health, and research assistant with a Bachelor of Applied Science. Intervention fidelity was assessed using multiple approaches: APA interventionists demonstrated their proficiency via written and oral examinations, and they were observed and mentored during training. The interventionists took photos of every participant’s ear after placing the seeds for placement verification by the PI-APA expert (C.H.Y.). The ear photos for the first 20 participants placed by newly trained interventionists were reviewed to ensure at least 95% accuracy in the placement of seeds. Ear photos were then randomly selected to check for accuracy (20% of photos were reviewed monthly) during the study. The PI also conducted random visits during the intervention session to observe adherence to the protocol in real-time.

Assessment and outcome measures

Primary outcomes

The minimal dataset by the NIH Pain Consortium’s Task Force on Research Standards for Chronic Low Back Pain⁴¹ was used. The primary outcomes were changes in scores for pain and function. Pain was measured using the NRS for worst pain in the past 7 days using a 0-10 scale (0 “no pain” to 10 “worst pain imaginable”). Function was measured using the Roland-Morris Disability Questionnaire (RMDQ), consisting of 24 questions about functional limitations from cLBP.⁴²

Secondary outcomes

Based on the biopsychosocial model of pain,⁴³ secondary outcomes included pain interference (the Brief Pain Inventory subscale with 7 pain-related interference domains)⁴⁴; pain catastrophizing (Pain Catastrophizing Scale)⁴⁵; fear/avoidance beliefs about physical activity (Fear-Avoidance Beliefs Questionnaire)⁴⁶; and anxiety, depression, fatigue, and sleep disturbance (Patient-Reported Outcomes Measurement Information System [PROMIS 29], T-scores were used for analysis).⁴⁷ Other validated measures included comorbidity (Charlson Comorbidity Index),⁴⁸ treatment beliefs and expectations (Healing Encounters and Attitudes Lists [HEAL] short form),⁴⁹ and neuropathic pain (painDETECT questionnaire).⁵⁰ All measures had a Cronbach’s α  ≥  0.84 for the current study (Table S2). A paperless data entry system (REDCap) was used to collect data during 1-on-1 sessions with participants or via email (due to the COVID-19 pandemic). Ecological momentary assessment was also used to measure daily opioid use and adherence to APA defined by pressing the seeds at least ≥ 2 times/day and ≥ 2 minutes/time.

Sample size

Power analysis was based on the smallest relative effect size on the primary endpoint (Cohen’s d = 0.65) at 1M follow-up.²² With a total sample size of 270 (90 for each group), the study was powered at 0.90 to detect significant differences between APA treatment and control groups at a significance level of .05 using a 2-sided test considering an attrition rate of 25%.

Statistical analysis

Intention-to-treat was used with 2-sided tests of α = 0.05. Baseline characteristics were compared using analysis of variance (continuous variables) and $χ2$ tests or Fisher exact tests (categorical variables). Mixed models for repeated measures (MMRM) for continuous outcomes and generalized linear mixed models (GLMM)^51,52 for dichotomous outcomes were used to examine study outcomes by appropriately specifying the contrast parameters. The dependent variables for MMRM included the change score of the study’s primary outcomes from baseline, and the independent variables included group, time, and group-by-time. An unstructured covariance was used to account for the correlation of repeated measurements within participants. The MMRM provided an unbiased estimate for missing data (missing at random), so we did not impute missing values. Data were analyzed for up to 6M follow-up. To compare the proportion of participants with a clinically meaningful response (ie, responder analysis), the primary outcomes were dichotomized and tested using GLMM. Responders were defined as achieving more than 1.5 points of pain reduction from baseline^41,53–56 or more than 2.5 points improvement from baseline in function.^53,54,56,57 Sensitivity analyses were used to assess the robustness of the study conclusions at 1M follow-up for the primary outcomes (see the Supplementary Material for details). Analyses for the primary outcomes were not adjusted for multiple tests because the 2 primary outcomes, pain and function, quantify different aspects of treatment effects. All data analyses were performed using SAS 9.4 and R-4.2.0. No interim analyses were conducted.

Results

Recruitment

A total of 700 individuals were screened, 178 were not eligible, 122 could not be reached, and 128 declined to participate (Figure 1). Of the 272 enrolled, 92 were randomly assigned to T-APA, 91 to NT-APA, and 89 to education control. All enrolled participants provided signed informed consent. Of those enrolled, 220 (81%) completed the intervention; 60 (65%) in the T-APA group and 49 (54%) in the NT-APA group completed the 6M assessment (40% attrition rate occurred primarily during the pandemic at 6M).

Participant characteristics

The trial groups were similar with respect to the demographic and clinical characteristics at baseline (Table 1). The mean (standard deviation) age of the participants was 70 (6.95) years; 174 were women (64%), and 170 (62%) were African American.

Primary outcomes

Compared with education control, T-APA improved pain by 1.26 points (95% CI, −2.01 to −0.51; P = .001) and function by 1.46 points (95% CI, −3.42. to 0.51; P = 0.15) at 1M follow-up. NT-APA had improved pain by 1.12 points (95% CI, −1.90 to −0.33; P = .006) and function by 2.10 points (95% CI, −4.13 to 0.06; P = .04) at 1M (Table 2 and Table S3). For pain intensity, the effect sizes (Cohen’s d) were 0.91 and 0.59 at postintervention and 1M follow-up, respectively, while for the NT-APA group, these were 0.76 postintervention and 0.44 at 1M. The effect sizes were smaller for function in both groups at both timepoints. The improvement in function was not significant or borderline significant at 1M follow-up for both T-APA and NT-APA groups; interestingly, the improvement in function was significantly greater at 6M follow-up in both groups (Figure S2, Tables 3 and 4). The responder analysis shows that the responder rate, ie, the proportion of patients achieving clinically significant improvements in pain (≥ 1.5-point)^41,53–56 and function (≥ 2.5-point)^53,54,56,57 in both T-APA and NT-APA was much higher at postintervention and 1M follow-up compared to the education control group (the odds ratio ranging from 2.11 to 6.32; Table 3). Similar effects for pain and function in both T-APA and NT-APA treatments were sustained up to 6M follow-up (Tables 3 and 4). Sensitivity analyses did not yield different conclusions (Table S4). No statistically significant difference was found between T-APA and NT-APA (Tables 2-4).

Secondary outcomes

Compared to education control, T-APA and NT-APA groups had significantly improved anxiety at 1M follow-up (2.93 points with P = .02 in T-APA; 4.02 points with P = .003 in NT-APA; Table 2). Additionally, the NT-APA group had significantly improved pain interference by 0.90 points (P = .007), PCS by 3.38 points (P = .05), depression by 3.43 points (P = .002), and fatigue by 3.18 points (P = .01) at 1M follow-up (Table 2); the improvements in pain interference, PCS, and fatigue were sustained at 6M follow-up (Table 4 and Table S5). Opioid use (Table 4) decreased from 48% at baseline to 27% at 1M and 23% at 6M in the T-APA group and from 49% at baseline to 24% at 1M and 18% at 6M in the NT-APA group, but did not reach statistical significance, compared to the education control group (36% at baseline and 19% at 1M).

Group assignment, adherence to APA, and adverse events

No significant differences were noted between the groups regarding their responses to treatment beliefs and expectations. Blinding of group assignment was successful; 83% (n = 129) of participants in the 2 APA groups believed they had received T-APA at the postintervention assessment. For APA use at week 4, 77% in the T-APA group and ≥ 70% in the NT-APA group used APA ≥ 3 times/day to stimulate their ear points, while 68% in the T-APA group and ≥60% in NT-APA group self-stimulated ≥ 3 minutes/time (Table S6). APA adherence was slightly higher in the T-APA group.

No serious adverse events were reported, and there were no complaints of bruising or infection on the ear point sites of seed placement. Mild side effects were reported, including itching or discomfort on the ear where the seed was placed in both APA groups (47% in T-APA and 31% in NT-APA, P < .05 in weeks 1-3, but not week 4) (Table S7). Participants indicated that these side effects were temporary and resolved spontaneously without lasting effects. Five participants (3%) were withdrawn from the study owing to reports of ear sensitivity from the seeds and allergic reaction to the adhesive tape (Figure 1).

Discussion

Notably, we recruited 62% non-White and 64% female participants, consistent with the national data on back pain¹; this diversity is important because many pain studies lack representation of females and non-White participants⁵⁸ thereby contributing to pain care disparities. Furthermore, the willingness and enthusiasm of older adults to engage in research for their cLBP despite the COVID-19 pandemic was highly encouraging, with target recruitment being completed within 18 months instead of 36 months as planned.

We found that 4 weeks of APA led to significant improvement in pain and function in the 2 groups receiving different APA protocols (ear points targeted to cLBP and ear points nontargeted to cLBP) compared to education control, and these were sustained at the 6M follow-up. The findings were less compelling in the secondary outcomes except for anxiety and fatigue, which significantly improved at 1M in both APA groups compared to control, and improvement in pain interference and fatigue was significantly sustained at 6M in the NT-APA group.

Our study findings are clinically important and durable. First, this is the first fully powered RCT to our knowledge that investigated the efficacy of interventionist-delivered APA in older adults with cLBP to improve pain and function. Second, the moderate and sustained effects at 6M achieved by APA in pain and function exceeded the effects of nonpharmacologic therapies recommended by the ACP (a mean difference of 1-2 points in pain and 2-5 points in function [RMDQ]). Third, APA is a scalable intervention, administered by trained interventionists with no prior professional medical background, conducted in four 15-minute weekly sessions and combined with instructions for self-stimulation the rest of the week. Furthermore, the lack of acupuncture or acupressure-related prior expertise of the interventionists underscores the potential accessibility of APA. This intervention regimen produced significant and sustained improvements from 1M to 6M without further sessions after the 1-month intervention period. Fourth, the simplicity made APA an attractive treatment, providing patients with a greater sense of control over their cLBP and allowing them to resume their daily tasks, especially with immediate pain relief after ear stimulation.⁵⁹

Of note, both APA treatments yielded similar improvements in cLBP outcomes, commonly seen in acupuncture studies^54,60,61 and even in other pain treatment modalities.^54,62 However, there was no significant difference noted between both APA treatments. Key points to be considered include nonspecific and APA treatment effects (stimulation, selected ear points, and/or placebo effect). To control nonspecific effects, questionnaires were used to evaluate participants’ beliefs and expectations toward APA; there were no significant differences among the 3 groups, so it may be unlikely that APA’s impact on cLBP was from nonspecific or placebo effects.

APA effects rely on 2 key variables: Stimulation and selected ear points. Both APA groups were instructed to stimulate their ear points at identical dosages, indicating the stimulation itself may have produced a powerful effect to induce improvement regardless of the points stimulated. Patients with chronic pain who take an active role in their treatment have been shown to have superior outcomes compared to those who take a passive approach.⁶³ The initial seed application among participants was passive (ie, the interventionist places the seeds with no effort required from the participant). However, participants’ subsequent active study participation, effort, and having to stimulate the seeds to achieve sustained analgesia encourage an active approach to self-manage their pain. This active approach may have influenced the outcome for both groups.

Different ear points were used to address point specificity; however, our study findings did not support our hypothesis that T-APA would render better cLBP outcomes than NT-APA. Evidence suggests that stimulating the ear points related to the gastrointestinal tract (ie, mouth, stomach, and duodenum; points used in NT-APA that are located in the auricular concha, [where the auricular branch of the vagus nerve is distributed]) promotes gastrointestinal function and improves nutritional status.⁶⁴ We speculate these improvements led to improved fatigue resulting in better function as demonstrated in our study findings. Existing evidence also suggests that stimulation of the auricular vagus nerve resulted in improved fatigue⁶⁵ and antinociception,^65–67 indicating a high potential for clinical application. The stimulation of NT-APA points could modulate vagal activity and interfere with pain response.^21,68–70 Consequently, NT-APA was unlikely an appropriate comparator or a true inert placebo; the true impact of the intervention may have been underestimated,⁶⁰ or stimulation of the vagus nerve may have been sufficient to achieve significant results that point specificity may not be relevant. While we were not able to provide support for ear point specificity between the 2 APA groups, our study findings showed that APA can work either by way of active stimulation of ear points that are specific to cLBP or via vagal nerve stimulation using ear points along the auricular vagus nerve as demonstrated in the NT-APA group. Additional experimental studies are warranted to differentiate the effects of influential variables (ie, ear point specificity or stimulation) on cLBP outcomes.

Apart from the impact of NT-APA on pain and function, another novel and unexpected finding from this study was the significant results from NT-APA over T-APA on some secondary outcomes (eg, interference and fatigue at 1M, catastrophizing and depression at postintervention and 1M). Historically, based on Paul Nogier’s (father of auriculotherapy, French engineer and physician) ear mapping, the points for the NT-APA group align with zones that are not only associated with gastrointestinal and other internal organs but also associated with emotional regulation,⁷¹ which may have modulated pain perception and led to the improvement of other pain-related outcomes noted in this study. As noted, one of the proposed mechanisms by which auriculotherapy works is by vagal nerve stimulation⁷⁰; the selected NT-APA ear points are innervated by the vagus nerve. Research within the past decade has demonstrated the impact of auricular vagus nerve stimulation beyond pain management to other outcomes such as depression, anxiety, fatigue, and general well-being^21,65,68 by influencing various physiological processes suggesting mechanisms related to neurotransmitter and endorphin release as well as brain plasticity.⁶⁸

Although NT-APA showed significance compared to control in some outcomes over the T-APA group, there was no statistically significant difference between the T-APA and NT-APA groups. This could be related to variability in the outcomes (overlapping of confidence intervals as seen in Figure S2), small difference between the 2 APA groups (also shown in Figure S2 where the difference in the mean changes from baseline comparing both APA groups were small), or both instances maybe occurring at the same time. Future studies and clinical application may continue preference for the T-APA intervention approach due to well-documented, standardized ear point locations for low back pain,¹⁴ our prior studies on chronic low back pain,^{22,23,26,72–74} as well as the impact and sustainability of the intervention on pain relief and function at 6M noted in this RCT. In addition, future studies are warranted to further investigate and replicate the NT-APA ear points as to this approach’s impact on low back pain, function, and secondary outcomes given the magnitude of some effect sizes demonstrated in this RCT. Moreover, while APA has shown efficacy in various studies including this current study, evidence remains scarce as to the anatomical basis and physiological support for mapping the body onto the pinna (somatotopic theory) in auriculotherapy.^75,76 More research could be valuable in validating the somatotopic theory to enhance the scientific foundation of auriculotherapy.

Our findings have key implications for clinical practice. Adherence is key to achieving long-term effects especially in interventions having a behavioral component as patients need to maintain behavior change; in effect, adherence remains challenging. For example, in a large meta-analysis of meditation interventions, long-term follow-ups were problematic, and only 64% of participants were able to adhere to the assigned amount of home practice (30 minutes per day, 6 days per week).⁷⁷ Since APA requires stimulating ear points for only 9 minutes (3 times/day for 3 minutes/time), it may be more feasible than other pain interventions that incorporate behavioral change and adherence in the long term.

Overall, we provide support that APA has a clinically relevant, persistent effect on pain and function among older adults with cLBP. Incorporating APA into clinical practice for older adults with cLBP offers an accessible, viable, efficient option and a sustainable approach to pain management. Improving pain and physical function, with decreased reliance on pain medications, potential reduction of pharmacotherapy with accompanying side effects, and reduced necessity for other costly pain interventions, can be transformative for the quality of life of older adults. APA is also considered to be the safest form of extended auricular therapy as it does not involve embedded needles, inadvertent needlesticks, or potential infection from contaminated roaming sharps.³⁹

Training and education for healthcare professionals and others in APA may be warranted to ensure its effective integration into clinical practice, offering real-world evidence. For example, we successfully demonstrated the feasibility of training nurses to integrate APA and offer this option to manage cancer-related pain after theoretical training and APA practice.⁷⁸ Lastly, we envision that patients can do APA by themselves to self-manage their pain as we develop a smartphone app that is a self-guided tool for seed placement and ear point stimulation. We demonstrated the feasibility of supporting self-administration of APA to self-manage pain.^23,79,80 This app presents an exciting opportunity to facilitate greater access to APA for pain relief, especially considering existing disparities in pain access and care. We propose that APA be considered and recommended as a nonpharmacologic therapy for older adults with cLBP.

Limitations

Our study was conducted in an academic medical center with participants recruited from the Baltimore metropolitan area, primarily identifying as African American or White; thus, the findings may have limited generalizability to individuals from other regions of the United States or other racial/ethnic groups. The interventionists were not blinded to the APA groups but majority of the participants believed they received T-APA when queried at the postintervention assessment. In-person study visits were stopped due to the COVID-19 pandemic, causing higher attrition than initially estimated particularly at the 6M follow-up. High attrition rate at the 6-month follow-up could decrease efficiency leading to a loss of power to detect a significant difference in results. Intention-to-treat analyses and mixed-effects modeling were applied to account for missing data. Sensitivity analysis was conducted to test robustness of the results to the assumption that data are missing at random.

The primary study endpoint was at 1M since the education control group was re-randomized, limiting our ability to compare the APA groups to control at the 3 and 6M follow-up assessments. Furthermore, it was determined from this study that the NT-APA group was not a true placebo comparator, potentially influencing effect estimates. Regardless, we found evidence that APA is beneficial in cLBP and study findings showed sustainability in the improvement of the primary outcomes up to the 6M follow-up.

Conclusions

In this RCT, our hypothesis was partially supported; 4 weeks of APA effectively improved pain and function compared to control, with sustained effects to 6M follow-up in older adults with cLBP. While noninferiority was not shown between the 2 APA groups, the impact of APA is established for cLBP and should be recommended as part of a comprehensive pain care for older adults. Future studies are warranted to investigate APA in pragmatic implementation trials and other pain populations. Study findings are crucial for pain care practitioners and policymakers to incorporate APA toward improving efficacious and accessible pain management options.

Acknowledgments

It is with profound sadness that we acknowledge the passing of our esteemed Principal Investigator, Professor Dr Chao Hsing Yeh, whose unwavering dedication and visionary leadership were instrumental in the completion of this study. Though she is no longer with us, her work lives on in every facet of this manuscript. We are forever grateful for the opportunity to have worked alongside her and to have benefited from her wealth of expertise and wisdom. This manuscript is a testament to her enduring legacy and we dedicate it to her memory as a token of our profound respect, admiration, and commitment to honoring her by continuing the pursuit of knowledge and scientific discovery.

Supplementary material

Supplementary material is available at Pain Medicine online.

Funding

The research reported in this publication was supported by the National Institute on Aging of the National Institutes of Health under award number R01AG056587. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Conflicts of interest: One of the authors, C.C., is the Deputy Editor-in-Chief of Pain Medicine. All other authors have no conflicts of interest to declare.

References