The effect of self-administered acupressure on pain severity and sleep quality of patients with knee osteoarthritis: a randomized controlled trial Authors: Elaheh Shourabi, Seyyed Abolfazl Vagharseyyedin, Hamidreza Bahrami Taghanaki, Zeinab Saremi Published: 2024 ABSTRACT AbstractBackgroundPain in the knee joint area and reduced sleep quality are prevalent symptoms of knee osteoarthritis. This study examines the impact of self-administered acupressure on pain severity and sleep quality in patients with knee osteoarthritis.MethodsIn this trial conducted in 2023, 126 patients were enrolled. They completed a demographic characteristics form, the Visual Analogue Scale (VAS), and the Pittsburgh Sleep Quality Index (PSQI). Afterward, they were randomly allocated to three groups: acupressure, sham, and control. While the control group received routine care, patients in the acupressure and sham groups were instructed to apply pressure on selected acupoints daily for one minute for a duration of one month. All participants completed the PSQI and the VAS one month after the intervention began and one month after it ended. Data analysis involved Fisher’s exact tests, chi-square, Kruskal-Wallis, and one-way ANOVA tests. Also, per-protocol analyses were conducted with linear mixed-effects models.ResultsNo significant effects associated with group allocation were observed on pain VAS scores in acupressure and sham groups compared to the control group. However, participants in the acupressure group showed a significant decrease in pain VAS scores as time progressed. PSQI scores significantly decreased over time. Interaction effects showed that PSQI scores decreased more in the acupressure than in the control group over time. Finally, PSQI scores in the acupressure group demonstrated a significant decrease from pretest to follow-up, while no significant change was observed over time in the sham group.ConclusionsSelf-administered acupressure may be useful in improving sleep quality in patients with knee osteoarthritis. This work may be informative in developing acupressure interventions with more effective outcomes in future research.Trial registrationThis trial is registered on the Iranian Registry of Clinical Trials website with the identifier IRCT20211221053475N1 (Date: 10/01/2022).Supplementary InformationThe online version contains supplementary material available at 10.1186/s12906-024-04693-x.Keywords: Acupressure, Knee osteoarthritis, Pain, Sleep quality Background Osteoarthritis is the leading cause of chronic joint pain in the elderly population. It is characterized by the progressive deterioration of articular cartilage followed by inflammation in the synovial cavity. Knee osteoarthritis (KOA) is the predominant form of osteoarthritis. It is responsible for over 80% of the total burden of osteoarthritis, and its prevalence is rising rapidly [1]. KOA is among the ten debilitating diseases in developed countries [2]. The incidence of this condition is steadily rising with increasing life expectancy and obesity. Approximately 10% of men and 13% of women aged 60 or older have characteristic KOA [3]. In Iran, KOA affects 15.3% of the urban population and 19.3% of the rural population [4]. Knee pain is the primary symptom of KOA. This pain can range from mild to severe and may be dull, sharp, intermittent, or constant. Pain and other injuries, such as reduced range of motion, can make it difficult to walk, stand, and perform household tasks. Significant disability in daily life and its negative psychological impacts lead to reduced quality of life [1, 5]. KOA treatment options include various approaches, such as pharmaceutical and non-pharmaceutical therapies, surgery, and complementary and alternative therapies. Medications commonly used for KOA treatment entail painkillers like non-steroidal anti-inflammatory drugs (NSAIDs) and narcotic painkillers. NSAIDs can potentially harm the digestive system and have adverse effects on the kidneys and cardiovascular system. On the other hand, narcotic painkillers can lead to side effects such as nausea, vomiting, dizziness, constipation, drowsiness, fatigue, and headache [6]. Sleep disturbance is another significant symptom of KOA. Previous research indicates that 28 to 80% of individuals with KOA experience sleep disturbances [7]. The connection between sleep and pain is complex and multifaceted. The pain persists throughout the night, disrupting the quality of sleep and making it challenging to start or sustain sleep. On the other hand, inadequate sleep quality contributes to pain and fatigue [8]. Thus, one can suppose that relieving knee pain may lead to improved sleep quality. Sleep disorders can lead to various complications, including depression, anxiety, reduced family welfare, decreased performance in waking-life activities, and low quality of life. Thus, the quantity and quality of sleep are crucial factors in maintaining both mental and physical health, making them vital considerations in any treatment plan [9]. Pharmaceutical therapies are often used to treat sleep disorders. Pharmaceutical therapy options include histamine type 1 receptor blockers, benzodiazepines, melatonin receptor agonists, and orexin receptor antagonists (ORAs) [10]. Typically, pharmaceutical therapy can lead to certain complications, including drowsiness, difficulty concentrating, memory problems, and dependency [11]. Given the side effects of medications used to treat pain and sleep disorders in patients with osteoarthritis, it may be beneficial to utilize non-pharmaceutical therapy options. Non-pharmaceutical approaches, including complementary and alternative medicine like acupressure, can offer a practical alternative for addressing these problems. Acupressure involves using the fingers to apply pressure to specific points along the body’s energy meridians. As a non-invasive technique, acupressure is simple to learn and apply in practically any situation [12]. Prior research has indicated that acupressure can potentially enhance the production of natural sedatives by activating the autonomic nervous system, resulting in decreased pain [13, 14]. However, a limited number of studies have used acupressure to alleviate KOA-associated pain. The study by Akbarnejad et al. (2019) demonstrated the potential effectiveness of acupressure in reducing pain and stiffness and improving knee function in individuals with osteoarthritis [15]. Likewise, Sorour et al. (2014) showed that acupressure exerts a more pronounced effect on pain in osteoarthritis patients than isometric exercise [16]. However, in other studies, acupressure has exhibited limited effectiveness in reducing KOA-associated pain. For instance, the effectiveness of acupressure in lowering pain severity among osteoarthritis patients was not found to be significant in the studies conducted by Zhang et al. (2012) and Cheung et al. (2020) [14, 17]. Researchers suggest that while there is promise in using acupressure to address the effects of KOA, further studies are needed to fully understand its potential [18]. Importantly, the researchers themselves have implemented acupressure in the majority of studies investigating acupressure’s impact on KOA pain. As a result, further research is needed to determine the feasibility of using acupressure by KOA patients. Methods The aim, design, and setting of the study The study was a mono-centric, randomized controlled trial with a single-blinded, three-arm parallel design. The study aimed to examine the impact of self-administered acupressure on pain severity and sleep quality in individuals with KOA. The study recruited patients with KOA who visited the rheumatology clinic affiliated with Birjand University of Medical Sciences, Birjand, Iran, in 2023. The characteristics of participants and description of materials The sample size in the present study was determined based on a previous study [15], taking into account α = 0.05, β = 0.2, and an attrition rate of 15%. As such, the sample size was determined to be 42 individuals per group. Samples were selected using convenience sampling. The inclusion criteria were as follows: patient consent to participate in the study; KOA grades 2 and 3 based on the Kellgren-Lawrence (K-L) scale; absence of autoimmune diseases; no history of substance abuse; absence of psycho-cognitive disorders; and absence of inflammation, infection, or sensory disorder, including neuropathy at the site of acupressure. Other inclusion criteria were absence of acupressure in the past six months, no history of knee or hip replacement surgery, no intra-articular injections (steroid, platelet-rich plasma, prolotherapy, etc.), body mass index (BMI) < 30, no pregnancy and breastfeeding, no use of anticoagulants, no history of coagulation disorders, absence of chronic diseases (such as diabetes, neuropathic pain, and kidney disorders), and a score of 5 or more on the Petersburg Sleep Quality Index (PSQI). Exclusion criteria included dependence on joint surgery or joint physiotherapy during the intervention. Moreover, refusal to continue participating in the study and acupressure practiced for fewer than three sessions per week were considered as post-randomization exclusion criteria. Sampling was performed between January and March 2023. The samples were assigned to the study groups by an independent research assistant who was not involved in the study. The study groups included the acupressure group (n = 42), Sham group (n = 42), and control group (n = 42). The block randomization method with a block size of 6 and an allocation ratio of 1:1:1 was used to assign participants to study groups. For this, a researcher assistant used sequentially numbered opaque envelopes. Each of these opaque envelopes contained a card marked with the group assignment (acupressure, sham, or control group). Once a participant was found eligible, the researcher assistant opened the next sealed envelope in the sequence, and the participant was assigned to the study group indicated on the card (Fig. 1). All participants in the acupressure and sham groups were blinded to the study groups. The respondents completed a demographics questionnaire that inquired about the following: age, gender, occupation, marital status, BMI, level of education, duration of illness, smoking history, average frequency of taking sleeping pills in the past month, and average frequency of painkiller use during the past week. The Visual Analogue Scale (VAS) was used to assess the severity of knee pain as the primary outcome. This tool is widely used for pain assessment. The scale features a horizontal line measuring 100 mm, with “no pain” on the left end and “the worst possible pain” on the right end. Participants are requested to indicate the severity of their pain by marking a point on the scale. The researcher measures the distance between the “no pain” point and the point specified by the patient in millimeters, which is then considered as the pain severity score [19]. Research has indicated that the VAS is a more sensitive measure of pain severity in osteoarthritis than the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) pain subscale [20]. To assess the reliability of this scale in our study, we had fifteen patients complete the scale on two occasions, with a 24-hour gap between each completion. The intraclass correlation coefficient was calculated to be 0.82, indicating that the scale has acceptable test-retest stability. The PSQI was used to evaluate the participants’ sleep quality as a secondary outcome. It is a self-report questionnaire that assesses the quality of sleep in patients over the previous month. Patients answer 19 questions, the results of which are used to assess their overall sleep quality. The questionnaire has seven components: subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleeping medication, and daytime dysfunction. These components are scored in a range from 0 to 3. The scores are added to yield a total score ranging from 0 to 21, with a higher score indicating lower sleep quality. Patients with a total score of 5 or higher are classified as poor sleepers [21]. This tool has exhibited good reliability when assessing the sleep quality of osteoarthritis patients in Iran [22]. Cronbach’s alpha coefficient was computed in this study to assess questionnaire reliability, which was found to be 0.8. Intervention Initially, one of the researchers (first author) received a certificate of acupressure from a traditional Chinese medicine specialist with 14 years of experience in teaching and practicing clinical traditional Chinese medicine, including acupuncture (third author). Afterward, she visited the rheumatology clinic affiliated with Medical Sciences, Birjand, Iran. Using convenience sampling, the researcher selected 126 patients for the study. Afterward, the patients completed the demographics questionnaire, VAS, and PSQI (pretest). The participants were assigned to one of three groups—acupressure, sham, and control—by an independent research assistant who was not involved in the study. The block randomization method was used for this assignment. In the acupressure group, the researcher initially instructed the participants on the points employed for pain relief. Subsequently, patients received practical one-on-one instruction on how to apply acupressure to the selected, namely ST34, ST36, SP9, SP10, and GB34 [23–25]. Following previous research and expert opinion (third author), patients were instructed to apply pressure to the designated acupoints on the painful knee(s) for one minute [17] while sitting, using the pad of their thumb, until they experienced a sensation of heaviness and numbness. The identification of these acupoints and mastering the technique of applying pressure on them persisted until it was confirmed that the whole process was executed accurately. Finally, patients were given a pamphlet that included the locations of the designated acupoints. Participants completed the intervention on a daily basis before going to bed for a duration of one month. The sham group consisted of individuals who were instructed to apply pressure to the ST40, SP6, and LV3 points [23] on their painful knee(s) using the same procedure as the acupressure group. Furthermore, individuals in this group exerted pressure on specific areas corresponding to the dimensions of the patient’s four attached fingers—the index, middle, ring, and little fingers—above the ST34 and SP10 points. Notably, all acupoints and non-acupoints were selected based on previous research [24, 25] and traditional Chinese medicine specialist opinions (third author). A full description of the location of the selected acupoints is provided in Table S1. In order to ensure proper application of pressure on the specified acupoints, patients were given a logbook and instructed to record their daily practice. Furthermore, the participants received regular follow-up calls at the end of each week for a month. These calls served as reminders for self-practice and for recording any adverse events. Participants in both groups were instructed to continue their regular medical care for KOA, which included taking prescribed medications and attending medical appointments. The control group received routine care delivered at the rheumatology clinic. Both the acupressure and sham groups completed the PSQI and VAS one month after the start of the intervention (posttest) and one month after the end of the intervention (posttest). In the control group, one month after the pretest, the patients completed these questionnaires (posttest). In addition, this group completed these instruments again one month after the pretest for follow-up. At post-test and follow-up, a research assistant not involved in the study collected and coded the data for the three groups. Thus, the data analyzer was blinded to group allocation. Follow-ups were conducted from May through June 2023. Statistical analysis The collected data were analyzed using SPSS software version 24.0 and the lmer package in R software version 4.3.2. Descriptive statistics (e.g., mean, median, standard deviation, and frequency distribution) and inferential statistics were used to analyze the data. Considering participants’ age, BMI, and duration of disease, the three groups were compared using one-way ANOVA. Also, the study groups were compared using a chi-square test to analyze the differences in gender, occupation, and smoking history. Fisher’s exact test was used to compare the three groups regarding the levels of education, the average frequency of using sleeping pills in the past month, and the mean frequency of using painkillers for knee pain in the past week. The analysis was conducted on a per-protocol basis. Mixed-effects models were developed to examine the effect of acupressure on VAS and PSQI. The independent variables were assignment (categorical variable: acupressure group, control group, and sham group), time (categorical variable: pretest, posttest, and follow-up), and the time by assignment interaction. Individual patients’ intercepts for acupressure on VAS and PSQI were allowed to vary as random effects. Another mixed-effects model was developed solely to assess changes in VAS and PSQI in three groups. The independent variable was time (categorical variable: pretest, posttest, and follow-up). The random effects of the intercept were also entered. All linear mixed-effects models were calculated using the lmer package in R software version 4.3.2. The significance level for all statistical analyses was set at P < 0.05. Results A total of 114 patients were evaluated in this research, with 38 individuals assigned to the acupressure group, 38 to the sham group, and 38 to the control group. In this study, no adverse events were reported by any participant. Table 1 details the demographic characteristics of the three groups. Table 1Demographic and clinical characteristics of the participantsVariable GroupAcupressure (n = 38)Sham (n = 38)Control (n = 38)P-valueGender, n (%)Male4 (10.5)6 (15.8)6 (15.8)0.72 ^Female34 (89.5)32 (84.2)32 (84.2)Marital status, n (%)Single/ spouse deceased4 (10.5)3 (7.9)3 (7.9)> 0.9 ^^Married34 (89.5)35 (92.1)35 (92.1)Education level, n (%)Middle school or lower15 (39.5)16 (42.1)17 (44.7)0.37^High school diploma7 (18.4)5 (13.2)12 (31.6)Associate degree5 (13.2)4 (10.5)3 (7.9)Bachelor’s degree or higher11 (28.9)13 (34.2)6 (15.8)Occupation, n (%)Employee/self-employed11 (28.9)11 (28.9)7 (18.4)0.75 ^Housewife22 (57.9)21 (55.3)26 (68.4)Retiree5 (13.2)6 (15.8)5 (13.2)Smoking history, n (%)Yes0 (0)0 (0)3 (7.9)0.11^^No38 (100)38 (100)35 (92.1)Average frequency of taking sleeping pills in the past month, n (%)027 (71.1)25 (65.8)32 (84.2)0.31 ^14 (10.5)7 (18.4)4 (10.5)2 or more7 (18.4)6 (15.8)2 (5.3)Average frequency of painkiller use during the past week, n (%)010 (26.3)13 (34.2)12 (31.6)0.41 ^113 (34.2)12 (31.6)18 (47.4)215 (39.5)13 (34.2)8 (21.1)Age in years (Mean ± SD)48.8 ± 68.8751.10 ± 13.0647.11 ± 61.480.31†BMI (kg/m2)24.2 ± 90.1126.2 ± 04.4625.1 ± 2.700.51†Duration of illness-years (Mean ± SD)3.98 ± 5.364.63 ± 4.622.47 ± 2.240.51††Baseline VAS score (mm)60.47 ± 18.2260.54 ± 21.4855.24 ± 15.040.36†Baseline PSQI score9.21 ± 2.209.81 ± 2.369.34 ± 1.690.42††Analyzed using the one-way ANOVA test^Analyzed using the Chi-squared test^^Analyzed using the Fisher’s exact test††Analyzed using the Kruskal-Wallis test Demographic and clinical characteristics of the participants †Analyzed using the one-way ANOVA test ^Analyzed using the Chi-squared test ^^Analyzed using the Fisher’s exact test ††Analyzed using the Kruskal-Wallis test We applied linear mixed-effects models to investigate the impact of time and group (acupressure, sham, and control) on pain VAS scores and PSQI scores. As displayed in Table 2., group assignment had no significant main effects on pain VAS scores for the acupressure group (β = 7.01, P = 0.12) and sham group (β = 6.49, P = 0.15) compared to the control group. Participants in the acupressure group showed a significant decrease in pain VAS scores over time (β=-6.19, P < 0.001). The sham group also exhibited a decrease in VAS scores over time, but this decrease was not statistically significant (β=-1.50, P = 0.16). These results suggest that while time alone did not influence pain VAS scores, the interaction between time and group assignment significantly impacted pain VAS score reduction in the acupressure group. Table 2.Effect of acupressure on pain VAS scores and PSQI scores: mixed-effects modelsVariablePain VAS scoresPSQI scoresPretestPosttestFollow-upPretestPosttestFollow-upMean± SDMean± SDMean± SDMean± SDMean± SDMean± SDAcupressure group60.47±18.2242.53±16.2650.55±19.879.21±2.207.29±2.976.05±3.19Sham group60.54±21.4859.64±22.2660.02±22.309.81±2.369.71±2.619.45±2.95Control group55.24±15.0456.76±14.7457.71±14.269.34±1.6911.00±2.3110.29±2.38Linear mixed-effects modelsEstimate (95% CI)P valueEstimate (95% CI)P valueTime1.23 (-0.26 to 2.74)0.11-0.47 (-0.87 to .0.08)0.02Acupressure vs control group7.01(-1.87 to 15.89)0.12-1.41(-2.59 to -0.23)0.01Sham vs control group6.48 (-2.39 to 15.37)0.15-0.76 (-1.94 to 0.41)0.20Time× acupressure group-6.19 (-8.31 to -4.07)<0.001-2.05 (-2.61 to -1.49)<0.001Time× sham group-1.50 (-3.61 to 0.62)0.16-0.66 (-1.21 to -0.10)0.02 Effect of acupressure on pain VAS scores and PSQI scores: mixed-effects models Statistical analysis indicated that group assignment had significant main effects on PSQI scores for the acupressure group (β=-1.41, P = 0.01) compared to the control group. However, group assignment has no significant main effects on PSQI scores for the sham group (β=-0.76, P = 0.20) compared with the control group. Interaction effects showed that PSQI scores significantly decreased in the acupressure group (β=-2.05, P < 0.001) and the sham group (β=-0.66, P = 0.02) over time (Table 2.). Table 3. presents the linear mixed-effects model to examine changes in pain VAS scores and PSQI scores in the acupressure group. The pain VAS scores in this group showed a significant decrease in the posttest and follow-up phases compared to the pretest, while PSQI scores significantly decreased from the pretest to the follow-up. Table 3.Change in pain VAS scores and PSQI scores in the acupressure group: mixed-effects modelsVariablePain VAS scoresPSQI scoresPretestPosttestFollow-upPretestPosttestFollow-upMean± SDMean± SDMean± SDMean± SDMean± SDMean± SDAcupressure group60.47±18.2242.53±16.2650.55±19.879.21±2.207.29±2.976.05±3.19Linear mixed-effect modelEstimate (95%CI)P valueEstimate (95%CI) P valueTime: Follow-up vs pretest-9.92 (-18.09 to -1.74)0.02-3.15 (-4.42 to -1.89) <0.001Time: Follow-up vs posttest8.03 (-0.24 to 16.29)0.06-1.24 (-2.51 to 0.04) 0.06Time: Posttest vs pretest-17.94 (-26.12 to -1.74)<0.001-1.92 (-3.18 to -0.65) 0.003 Change in pain VAS scores and PSQI scores in the acupressure group: mixed-effects models As presented in Table 4., there was not a significant difference in VAS scores and PSQI scores across the three measurement time points in the sham group. Table 4.Change in pain VAS scores and PSQI scores in the sham group: mixed-effects modelsVariablePain VAS scoresPSQI scoresPretestPosttestFollow-upPretestPosttestFollow-upMean± SDMean± SDMean± SDMean± SDMean± SDMean± SDShams group60.54±21.4859.64±22.2660.02±22.309.81±2.369.71±2.619.45±2.95Linear mixed-effect modelEstimate (95%CI)P value Estimate (95%CI)P valueTime: Follow-up vs pretest-0.52 (-10.42 to 9.38)0.92 -0.37 (-1.56 to 0.82)0.54Time: Follow-up vs posttest0.37 (-9.64 to10.38)0.94 -0.26 (-1.47 to 0.94)0.66Time: Posttest vs pretest-0.89 (-10.79 to 9.01)0.86 -0.10 (-1.30 to 1.09)0.86 Change in pain VAS scores and PSQI scores in the sham group: mixed-effects models Finally, the statistical analysis revealed a significant increase in PSQI scores at the posttest compared to the pretest in the control group (Table 5.). Table 5. Change in pain VAS scores and PSQI scores in the control group: mixed-effects modelsVariablePain VAS scoresPSQI scoresPretestPosttestFollow-up PretestPosttest Follow-upMean± SDMean± SDMean± SDMean± SD Mean± SD Mean± SD Control group55.24±15.0456.76±14.7457.71±14.269.34±1.6911.00±2.3110.29±2.38Linear mixed-effect modelEstimate (95%CI)P value Estimate (95%CI)P valueTime: Follow-up vs pretest2.47 (-4.14 to 9.08)0.460.94 (-0.02 to 1.91)0.06Time: Follow-up vs posttest0.95 (-5.74 to 7.63)0.77-0.71 (-1.68 to 0.27)0.15Time: Posttest vs pretest1.53 (-5.08 to 8.14)0.651.66 (0.69 to 2.62)0.001 Change in pain VAS scores and PSQI scores in the control group: mixed-effects models Discussion Pain in the knee joint is the most common symptom of osteoarthritis in the knee. Sleep disturbance is another common symptom of KOA [1, 5, 7]. This study sought to examine how self-administered acupressure affects pain severity and sleep quality in patients with KOA. In this study, no significant group differences were observed in the pain VAS score change. Nevertheless, the knee pain severity significantly decreased in the posttest and follow-up compared to the pretest in the acupressure group. Consistent with our research, a study conducted by Cheung et al. (2020) examined the impact of self-administered acupressure and knee health education on knee pain in middle-aged and older adults with KOA. The study found that while both groups experienced significant reductions in knee pain intensity, there were no statistically significant differences between the two groups [14]. Moreover, in Li et al.‘s study, older adults in the self-acupressure and sham groups experienced significant improvements in knee pain. However, the study did not find any significant differences between the two groups [26]. Zhang et al. (2012) investigated the feasibility of 12-week self-administered acupressure practices in postmenopausal women with KOA. The study conducted by Zhang et al. revealed that these practices can be easily learned and safely performed at home. However, when it came to clinical outcomes such as knee pain, no statistically significant difference was found between the control group and the acupressure [17], which is inconsistent with the findings in this study. In our study, the knee pain severity significantly improved at posttest, and this was sustained at follow-up in the acupressure group. However, there were no significant differences in pain VAS scores over time in the sham and control groups. In line with this finding, Sorour et al. (2014) found that self-administered acupressure was more effective than isometric exercise in reducing pain in women with KOA. On the other hand, the isometric group had significantly lower scores for stiffness and physical dysfunction than the acupressure group. The authors proposed that their study findings be validated through a randomized clinical trial [16]. In a study conducted by Akbarnejad et al. (2019), acupressure exhibited a significant impact on reducing pain and physical dysfunction in older adults with KOA. In the above-mentioned study, the acupressure group experienced greater improvements in the general WOMAC index than the placebo group and routine care [15]. Notably, in the study of Akbarnejad et al., acupressure was performed by a research colleague rather than the patients. A recent study noted that auricular acupressure led to a significant reduction in pain in elderly people who have osteoarthritis, as compared to those in the placebo group [27]. It is believed that acupressure stimulates certain triggers in the immune system, whereby β-endorphin is released. This release then causes vasodilation and ultimately helps reduce pain [28]. Furthermore, studies have demonstrated changes in functional connectivity among individuals experiencing chronic pain in the medial prefrontal default-mode network. Research using functional magnetic resonance imaging has demonstrated that acupuncture has the potential to enhance functional connectivity between different brain networks, including the frontoparietal network, in patients with KOA [29, 30]. As is evident from the studies mentioned above, there have been conflicting reports about the effect of acupressure on KOA-associated pain. The findings of these studies provide opportunities to improve future study designs. In this study, sleep quality significantly improved in the acupressure than in the control group. Statistical analysis revealed that the sleep quality significantly improved at the posttest and sustained at follow-up in the acupressure group. Also, sleep quality diminished at the posttest compared to the pretest in the control group. Individuals experiencing chronic pain tend to experience a higher frequency of sleep disturbances. Sleep disturbances affect 40.7–86% of patients suffering from chronic pain. Chronic pain can disrupt a patient’s sleep as they struggle to fall asleep or stay asleep. This pain often hinders their ability to achieve a restful night’s sleep. Chronic pain and sleep have neurophysiological mechanisms in common [31]. Consequently, it can be supposed that the patients’ sleep quality improved in the acupressure group concurrently with the reduction of knee pain. According to a study conducted by Vellucci et al. (2021), reduced pain severity is linked with better sleep quality and overall quality of life [32]. Researchers suggest that there is a reciprocal relationship between chronic pain and sleep disturbances. While pain can contribute to sleep disorders, it is also common for individuals with chronic insomnia to experience chronic pain [33]. Lee and Park (2022) examined the impact of auricular acupressure on sleep and pain among older adults with osteoarthritis residing in nursing homes. The results revealed a significant improvement in pain levels among the participants in the acupressure group compared to those in the placebo control group. In addition, the group experienced a significant increase in melatonin levels after the intervention, and their sleep quality scores exhibited a marked rise compared to the placebo control group [27]. In the present study, a non-significant improvement in sleep quality was found in the sham group. Similar to this finding, Tsay and Chen (2003) examined the effectiveness of acupressure on the sleep quality of patients with end-stage renal disease, noting that sleep quality improved in the acupressure group as well as in the sham group [34]. In the current study, no participants in the acupressure or sham groups reported experiencing any adverse effects from the intervention. On the other hand, the attrition rate in the acupressure group was approximately 10%. Based on the findings of this study, it can be concluded that self-administered acupressure is a feasible option for managing symptoms in community-living middle-aged adults. Future studies may consider exploring the potential of applying this intervention to individuals of varying age groups with KAO. In this study, all data collection instruments were self-report instruments. Thus, the collected data might be susceptible to reporting bias such as the recall bias. Using objective measures such as polysomnography to monitor the objective sleep quality can be considered in future research. Moreover, the present study involved patients who were referred to one rheumatology clinic. Future studies can enhance the generalizability of findings by including patients from multiple rheumatology clinics. Conclusions Using self-administered acupressure as a non-pharmacological approach may be a helpful tool in enhancing sleep quality for individuals with KOA. Supplementary Information Supplementary Material 1. Acknowledgements Our special thanks goes to the Deputy for Research and Technology Birjand University of Medical Sciences, Birjand, Iran, for its support during the current study. We would also like to thank to all participants who voluntarily participated in the study. Abbreviations Body mass index Kellgren-Lawrence scale Knee osteoarthritis Non-steroidal anti-inflammatory drugs Orexin receptor antagonists Petersburg Sleep Quality Index Visual Analogue Scale Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Authors’ contributions Study design: SAV, ESh, and HBT, and ZS; data collection: ESh; analysis and statistics: SAV, manuscript preparation: SAV and ESh. All authors approved the final version for submission. Funding This study was nfunded. Data availability The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate The present study’s proposal received approval from the ethics committee of Birjand University of Medical Sciences, Birjand, Iran [ethics code: IR.BUMS.REC.1400.224 (Approval date was 08/11/2021)]. The participants signed written informed consent forms to take part in the research. Participants were informed that their participation in the study was voluntary. Furthermore, they were assured that any information provided by them would be kept confidential. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1. Data Availability Statement The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.