Journal of Gerontological Nursing

Feature Article Supplemental Data

Feasibility and Efficacy of a Shared Yoga Intervention for Sleep Disturbance in Older Adults With Osteoarthritis

Diana Taibi Buchanan, PhD, RN; Michael V. Vitiello, PhD; Kimberly Bennett, PhD, PT


The purpose of the current study was to test the feasibility and efficacy of a shared yoga (SY) program for sleep disturbance in older adults with osteoarthritis (OA). Participants (ages 50 to 72) with insomnia related to OA were randomized to 12 weeks of SY (n = 9) or individual yoga (IY; n = 7). The sample was 53% male and 41% self-identified as a race other than White/Caucasian. The groups did not significantly differ in class attendance, home practice, or change scores on efficacy outcomes. Pre–post comparisons in all participants showed statistically significant improvements on the Insomnia Severity Index and Patient-Reported Outcomes Measurement System sleep disturbance scale. Findings support the overall feasibility of the program, both in the shared and individual formats. Efficacy data suggest that this yoga program may improve sleep, but given the study limitations, further research is needed to draw conclusions. [Journal of Gerontological Nursing, 43(8), 45–52.]


The purpose of the current study was to test the feasibility and efficacy of a shared yoga (SY) program for sleep disturbance in older adults with osteoarthritis (OA). Participants (ages 50 to 72) with insomnia related to OA were randomized to 12 weeks of SY (n = 9) or individual yoga (IY; n = 7). The sample was 53% male and 41% self-identified as a race other than White/Caucasian. The groups did not significantly differ in class attendance, home practice, or change scores on efficacy outcomes. Pre–post comparisons in all participants showed statistically significant improvements on the Insomnia Severity Index and Patient-Reported Outcomes Measurement System sleep disturbance scale. Findings support the overall feasibility of the program, both in the shared and individual formats. Efficacy data suggest that this yoga program may improve sleep, but given the study limitations, further research is needed to draw conclusions. [Journal of Gerontological Nursing, 43(8), 45–52.]

Osteoarthritis (OA) is among the most common and disabling conditions affecting older adults (Neogi & Zhang, 2013). In addition to pain and functional disability, more than 50% of individuals with OA experience sleep disturbances, including difficulty falling asleep, difficulty remaining asleep, awakening too early, and/or lighter sleep (National Sleep Foundation, 2003). Research shows that sleep disturbances in individuals with OA are related to but not fully explained by arthritis pain (National Sleep Foundation, 2003; Power, Perruccio, & Badley, 2005; Smith, Quartana, Okonkwo, & Nasir, 2009; Wilcox et al., 2000). Thus, pain relief alone may not be sufficient to reduce sleep disturbance. Interventions to reduce sleep disturbance and pain have potential to improve daily functioning and quality of life in older adults with OA. Yoga is a multimodal practice that may be used as a complementary therapy to concurrently address these important factors in OA. Recent surveys have shown that yoga is popular among individuals with arthritis (Cramer et al., 2016; Mielenz, Xiao, & Callahan, 2016). The appeal of yoga for management of OA rests in its holistic nature, which potentially addresses physical aspects of OA (e.g., muscle weakness, joint instability), symptoms (e.g., pain, sleep disturbance), and related psychological factors (e.g., anxiety, depressed mood) (Taibi & Vitiello, 2012).

Although the literature on yoga for OA is limited, several recent systemic reviews and meta-analyses of individuals with OA or other painful conditions have concluded that present evidence provides mild to moderate support for yoga on the outcomes of pain and function, and that yoga appears to be a generally safe intervention (Cheung, Park, & Wyman, 2016; Cramer, Lauche, Langhorst, & Dobos, 2013; Sharma, Haider, & Knowlden, 2013; Ward, Stebbings, Cherkin, & Baxter, 2013). Only two studies examined the effects of yoga on sleep in OA (Cheung et al., 2016; Taibi & Vitiello, 2011). In the current authors' pilot study, older women with OA (N = 13) showed significant improvement on the Insomnia Severity Index (ISI) after 8 weeks of yoga, but did not show improvement on actigraphy, sleep diaries, and the Pittsburgh Sleep Quality Index (PSQI) (Taibi & Vitiello, 2011). One other pilot study (N = 36) reported improvement on the PSQI after 8 weeks of Hatha yoga in women with knee OA, but the change was not significant compared to the wait-list group (Cheung, Wyman, Resnick, & Savik, 2014). The current literature provides modest evidence that yoga may reduce pain and improve sleep and function in individuals with OA, but rigorous studies are still needed.

The purpose of the current study was to examine the feasibility and initial efficacy of Healthy Arthritis Lifestyle through Yoga (HEALTHY), a structured yoga intervention designed to reduce sleep disturbance, particularly related to pain, in older adults with OA. The program was designed to be gentle and safe for individuals with OA and to address the musculoskeletal areas most affected in those with OA of the lower body (i.e., weight-bearing lower body joints), with incorporation of relaxation practices to promote sleep. In addition to testing the yoga program, the current study tested a novel shared yoga (SY) practice approach to explore the effects of incorporating social support, which is known to be important in adopting and adhering to behavioral interventions such as yoga (McMahon et al., 2017). Descriptive studies of individuals with OA and older adults have shown that higher levels of social functioning and having an active partner were associated with higher levels of physical activity (Hong et al., 2005; Rosemann, Kuehlein, Laux, & Szecsenyi, 2008; Satariano, Haight, & Tager, 2002; Steptoe, Rink, & Kerry, 2000). It was hypothesized that inclusion of a support partner may improve adherence to—and subsequently the efficacy of—a yoga program.



The current study was a randomized, parallel-group study with two groups: SY (practiced with a partner) and individual yoga (IY; practiced without a partner). All procedures were approved by the University of Washington Human Subjects Division.


The HEALTHY program was received by both groups. This program comprised 12 weekly classes, a 5-minute morning “warm up” routine, and a daily 30-minute home practice. Prior studies using 8- to 12-week yoga interventions have shown high adherence rates and significant effects on sleep (Cheung et al., 2014; Sherman, Cherkin, Erro, Miglioretti, & Deyo, 2005; Taibi & Vitiello, 2011). The current study opted for a 12-week program rather than 8 weeks to enhance the likelihood of efficacy and support the progressive sequencing of lessons. The 12 weekly classes were conducted in the evenings at 6 or 7 p.m., depending on the cohort. Each class session was 75 minutes, except for Class 1, which was 90 minutes. All classes involved a 10-minute check-in time and concluded with 5 minutes of resting pose. The remaining time was spent in instruction on the poses. Class 1 was longer because participants were taught the morning practice routine as well as the first 30-minute home practice routine.

The yoga program was designed by an expert yoga therapist and the principal investigator (D.T.B.), who is also a yoga instructor. The program used the Viniyoga style, which is a type of yoga commonly used for therapeutic purposes. Viniyoga focuses on the desired physical or mental effect of a yoga pose rather than attainment of the exact pose (Kraftsow, 1999; Taibi & Vitiello, 2012). Therefore, poses are commonly modified to address individual needs. In addition, poses are gently repeated several times rather than being held in a static position, which was considered preferable for older adults who are at risk of muscle strain or may be unable to hold poses. The program included four components: a morning yoga routine and three sequential lessons that built on each other. The morning yoga routine could be performed in 5 minutes and was aimed at reducing stiffness upon awakening. This practice could be completed in bed before arising. Each of the three sequential lessons was taught for 3 weeks and reviewed again in each of the final 3 weeks. The poses and sequencing of the lessons are described in Table A (available in the online version of this article).

HEALTHY Program Yoga Poses and Practice SequenceHEALTHY Program Yoga Poses and Practice SequenceHEALTHY Program Yoga Poses and Practice SequenceHEALTHY Program Yoga Poses and Practice SequenceHEALTHY Program Yoga Poses and Practice Sequence

Table A:

HEALTHY Program Yoga Poses and Practice Sequence

Participants and partners were asked to practice the current lesson at home on non-class days. They were provided with a mat, block, and blanket, which they were allowed to keep after completing the study. They were given an audio CD with narration that guided them through the 30-minute practice, as well as handouts with diagrams and written instructions on each pose. Classes were attended by three to seven individuals and were taught by a highly experienced yoga instructor with 500-hour yoga teaching certification in the Viniyoga style.

Sample and Recruitment

The study was advertised by flyers posted in a large health sciences complex starting in December 2012, and by advertisements posted on metro transit buses starting in February 2013. Recruitment continued until January 2014.

Inclusion criteria were: age 50 to 85; diagnosed with OA of the hip, knee, or ankle; reporting chronic pain (≥6 months) from OA; meets research diagnostic criteria for “Insomnia Due to Medical Condition” (Edinger et al., 2004); insomnia directly related to OA based on the screening interview; stable use of pain, sleep, and/or mood-regulating medications over the past month; and had a practice partner who could attend yoga classes weekly. Hip, knee, or ankle OA were required for eligibility because these sites potentially may have a major impact on mobility and sleep related to discomfort or stiffness in bed. In addition, many of the same yoga postures are used therapeutically for these areas. OA at other sites was not ground for exclusion unless it prevented performance of HEALTHY routines. Practice partners were required to be at least 18 years old and not have a medical or psychiatric condition that prevented yoga practice.

Exclusion criteria were: acute injury that causes current pain; inability to stand without assistance; self-report of untreated or poorly managed medical or psychiatric conditions; an acute episode or change in the treatment of a psychiatric problem within the past 3 months; a musculoskeletal disease other than OA, such as fibromyalgia, lupus, or rheumatoid arthritis; previous back surgery or spinal cord/nerve damage; self-report of diagnosis or symptoms of a sleep disorder other than insomnia (e.g., sleep apnea, restless legs syndrome); and cognitive impairment based on a screening tool.


Individuals who contacted the research staff were initially screened by phone. After telephone screening, potential participants and their identified partners were mailed screening questionnaires and postage-paid envelopes for return. Once returned and participants were determined to meet eligibility criteria, participants were randomly assigned to the SY or IY group. A random sequence was generated by computer for each cohort in blocks of four. Assignments were placed in sealed envelopes numbered sequentially within each cohort. Envelopes were opened when a participant and partner completed the telephone and mailed screening steps. Participants were informed of group assignment by phone prior to the baseline visit. If assigned to the IY group, partners did not participate in any study activities beyond screening.

At the baseline visit, informed consent forms were reviewed and signed with participants and the SY group's partners. Participants and partners then completed the questionnaires. Participants and partners were provided a general overview of the HEALTHY program. Participants were given a 1-week sleep diary and actigraph (i.e., a wrist-worn device that measures activity to assess sleep, given that sleep corresponds to inactive periods) and were instructed on how to use them. Both were collected after 1 week.

Participants and partners attended the yoga class once per week at a community location. Participants were called at Weeks 4 and 8 to check-in and monitor for adverse events. Participants (and SY partners) returned for a post-intervention visit within 1 to 2 weeks of Class 12. At this visit, participants and partners completed the questionnaires and post-intervention interview. Participants again completed 1 week of sleep diaries and actigraphy. Participants were paid $250 and partners were paid $140 for completing assessments.


Descriptive Measures. Demographics and health history. Participants and partners were asked to provide information on their age, race, ethnicity, education, marital status, and employment. Participants completed the Self-Administered Comorbidity Questionnaire to screen for major health issues (Sangha, Stucki, Liang, Fossel, & Katz, 2003).

Feasibility. Recruitment log. A recruitment log was used to track information on all contacts, including where the individual learned about the study, whether the individual declined to participate (and reasons if given) or was ineligible, and final screening/enrollment status.

Adverse Effects of the Yoga Program. Adverse effects of the yoga program were tracked by asking participants in their weekly yoga practice logs if they had experienced any unpleasant symptoms that they do not normally experience. Participants were also asked about the occurrence of new symptoms at weekly check-ins with the yoga instructor at the beginning of class, monthly study calls, and the post-treatment interview. Expected adverse effects were based on the current authors' prior research and general risks of gentle activities for older adults (Taibi & Vitiello, 2011, 2012), including muscle soreness, bruising, joint pain, tiredness or fatigue, dizziness, unintentional fall, blood pressure changes, and shortness of breath.

Acceptability. Attendance and adherence. Attendance was documented each week by the yoga instructor. Adherence to home practice was reported by participants on weekly practice logs.

Post-Intervention Interview. At the post-intervention visit, participants were asked whether the program affected their pain or sleep and if it was useful to include the support partner.

Efficacy Outcomes. Sleep diary. The sleep diary was completed in the morning upon awakening. The diary assessed bedtime, rise time, estimated time to fall asleep, number of awakenings, time spent awake during the night, perceived sleep quality, and how rested the participant felt upon awakening.

Actigraphy. An actigraph is a wrist-worn device that detects the presence or lack of movement as indicators of wake (activity) and sleep (inactivity) behavior. Wrist actigraphy corresponds well with polysomnography for summary measures of sleep and wake, including total sleep time, wake time, and number of awakenings (Lichstein et al., 2006; Taibi, Landis, & Vitiello, 2013). Actiwatch-2™ actigraphs were used, and the associated software (Actiware 5.57) was used to analyze actigraphic sleep outcomes. The devices were configured to record activity in 1-minute epochs. Diary bedtimes and rise times were manually entered in the analysis software by the principal investigator (D.T.B.) and a trained research assistant. These times were empirically validated against other available data using the University of Washington Center for Research on Management of Sleep Disturbance standardized analysis protocol. These data included bedtime and rise time event markers (i.e., time-stamped markers placed in the digital actigraphy data by participants pressing a button on the actigraph at bedtime or rise time), activity levels viewed on graphical representation of the actigraphy data (activity should sharply decline at bedtime and increase at rise time), and light levels viewed on graphical representation of the actigraphy data (the timing of lights being turned off or on can be used as evidence of bedtime and rise time, respectively). Once these times were entered, the software applied a well-validated mathematical algorithm to score activity data as sleep or wake.

Patient-Reported Outcomes Measurement System (PROMIS) Sleep and Fatigue Assessments. Standardized and validated PROMIS short-forms (access were used to assess sleep disturbance (short-form 8b), sleep-related impairment (short-form 8a), and fatigue (short-form 7a) (Buysse et al., 2010; Garcia et al., 2007; Yu et al., 2011). PROMIS is a system of standardized health status measures developed by a consortium with support from the National Institutes of Health. The sleep disturbance short-form includes eight items on which respondents rate their perceived sleep quality, depth, and restoration. The sleep-related impairment short-form includes eight items on which respondents rate perceived alertness, sleepiness, tired-ness, and sleep-related functional impairments during the day. The sleep disturbance and sleep-related impairment forms have concurrent validity with established measures of sleep disturbance (PSQI) and sleepiness (Epworth Sleepiness Scale), respectively (Buysse et al., 2010). The fatigue short-form includes seven items on which respondents rate the frequency, intensity, and duration of fatigue as well as its impact on functioning. Concurrent validity was demonstrated with the multidimensional fatigue symptom inventory short-form and the brief fatigue inventory (Ameringer et al., 2016). All three PROMIS scales assess experiences over the past 7 days, with respondents rating statements on a 5-point scale.

Insomnia Severity Index (ISI). The ISI is a seven-item questionnaire on which respondents rated insomnia symptoms experienced over the past week on a scale of 0 to 4 (Bastien, Valliéres, & Morin, 2001). The ISI is well-validated and shown to be sensitive to treatment response in individuals with insomnia (Morin, Belleville, Bélanger, & Ivers, 2011). The items are summed for scores of 0 to 28, with higher scores indicating more severe insomnia symptoms.

Western Ontario McMaster (WOMAC) Osteoarthritis Index. The WOMAC is a 24-item scale that measures physical function (17 items), activity-related pain (five items), and stiffness (two items). Items are scored on 5-point Likert scales. The WOMAC has been extensively validated and is widely used in clinical trials of OA treatments (Bellamy, Buchanan, Goldsmith, Campbell, & Stitt, 1988; Gandek, 2014). Scores on the total WOMAC and subscales are computed by summing item scores. Higher scores indicate greater functional difficulty.

Patient Health Questionnaire-8 (PHQ-8). The PHQ-8 is a reliable and valid scale that provides an assessment of the presence and severity of depression (Kroenke, Spitzer, & Williams, 2001; Kroenke, Spitzer, Williams, & Löwe, 2008). Eight indicators of depression are rated on their frequency of occurrence, from 0 = not at all to 4 = every day. Item ratings are summed to generate an overall score of 0 to 32. Higher scores indicate more frequent depressive symptoms.

Statistical Analyses

Data were double-entered and checked using Microsoft Excel® databases. The data were imported into SPSS 17.0 for analysis. Recruitment, enrollment, and retention data were tallied and summarized in a CONSORT format study flowchart (Moher, Schulz, & Altman, 2001) (Figure). Demographics, retention rates, adherence, and adverse events in each study group were summarized using descriptive statistics. Descriptive statistics are reported on pre- and post-intervention outcomes, and pre–post comparisons in the sample were calculated using Wilcoxon signed-rank tests. Group comparisons of change scores pre- to post-intervention were examined using Mann–Whitney U tests.

Study recruitment, enrollment, and retention. Among 107 callers who did not meet screening criteria, the most common reasons were lack of a partner (n = 35), no osteoarthritis diagnosis (n = 18), no hip/knee/ankle osteoarthritis (n = 15), or age younger than 50 (n = 9).


Study recruitment, enrollment, and retention. Among 107 callers who did not meet screening criteria, the most common reasons were lack of a partner (n = 35), no osteoarthritis diagnosis (n = 18), no hip/knee/ankle osteoarthritis (n = 15), or age younger than 50 (n = 9).



Recruitment and Screening. A total of 187 individuals were called about the study, of which 37 qualified based on phone screening (Figure). The most common source from which individuals learned of the study was bus advertisements (n = 95), followed by word-of-mouth (n = 10), flyers (n = 5), and the medical center's research trials website (n = 2). Eighteen individuals completed the telephone and mail steps of the screening process and were randomized, but one who was randomized to the IY group disclosed a disqualifying condition after randomization. Thus, the final sample included 17 individuals.

Enrollment. Participant demographics are shown in Table 1. Median age of participants was 55 (range = 50 to 72 years) and 53% were male. Participants identified as Latino (29%); Caucasian/White (59%); African American (35%); and Asian, American Indian/Alaskan Native, or Pacific Islander (24%). More than one half of the sample was unmarried or not living with a partner (59%), and 77% had completed at least some college or graduate school.

Participant Demographics and Health Information (N = 17)Participant Demographics and Health Information (N = 17)

Table 1:

Participant Demographics and Health Information (N = 17)

Demographics were examined for all partners who completed screening, regardless of the accompanying participant's group assignment. Partners' median age was 48 (range = 25 to 80 years). Eighty percent of partners were female, and only female participants had a male partner (no male participant had a male partner). Ethnic/racial identification was less diverse than among primary participants; partners identified as Latino (7%); Caucasian/White (73%); African American (7%); and Asian, American Indian/Alaskan Native, or Pacific Islander (20%). Sixty percent of partners were unmarried or not living with a partner, and 73.3% had completed at least some college or graduate school.

Retention. Two IY participants and one SY participant dropped out. None of these individuals responded to calls, so no reason could be obtained. No demographics differed between those who did and did not drop out.

Adverse Effects of the Yoga Program. Eight of 14 participants who completed the study reported adverse effects of the yoga program, all of which were minor and expected. In all cases, the symptom reported was pain/soreness, including overall soreness (n = 4), knee pain (n = 4), hand/wrist pain (n = 2), hip pain (n = 1), and side/rib pain (n = 1). In the cases of overall soreness, participants improved by either continuing practice or reducing repetitions of each pose. In the other cases, reported pain resolved with either pose/practice modification or elimination of a pose. Seven specific poses were reported to contribute to pain/soreness (Table 2).

Poses Reported to Cause Pain/Soreness

Table 2:

Poses Reported to Cause Pain/Soreness


Attendance and Adherence. There were no significant differences in class attendance or frequency of home practice by study group. Of 12 weekly classes, the IY group attended a median of nine classes (range = 7 to 10 classes), and the SY group attended 10 classes (range = 2 to 12 classes). Partners attended a median of nine classes (range = 2 to 12 classes). The median number of days of practice per week was 6.2 (range = 4.8 to 7 days/week) in the IY group and 6.3 (range = 2.5 to 7 days/week) in the SY group. The SY group practiced a median of 2.5 days/week (range = 0.8 to 6.1 days/week) with their partners.

Post-Treatment Interview

When asked whether the program helped their pain or sleep, 10 participants reported that it was helpful for their pain and nine reported improvements in their sleep. When asked whether it was helpful to have a partner, eight SY participants stated that it was helpful, but one stated that it was not helpful. Benefits of having a partner, as noted by SY participants, included providing motivation to go to class (n = 3), helping adhere to or continue the program (n = 2), having assistance/coaching during home practice (n = 2), being nice to have someone at class with them (n = 2), and making it easier to ask questions in class (n = 1).


The IY and SY groups did not significantly differ on any study outcomes at baseline, except for PROMIS sleep-related impairment, which was slightly worse in the IY group (p = 0.02). The SY and IY groups were compared on change scores from pre- to post-treatment; no significant between-group differences were found (Mann–Whitney U, all p > 0.05). To explore the overall effects of the yoga program, changes in outcomes from pre- to post-treatment were analyzed using Wilcoxon signed-rank tests (Table 3). ISI and PROMIS sleep disturbance significantly improved from pre- to post-intervention, but no other outcomes were significantly changed. Although data were analyzed using non-parametric statistics due to the small sample, the effect size of the mean difference was calculated (pre- to post-intervention, accounting for paired data) on the significant outcomes. Effect sizes were large (ISI = 1.41, PROMIS sleep disturbance = 0.98).

Efficacy Outcomes Pre- and Post-Intervention

Table 3:

Efficacy Outcomes Pre- and Post-Intervention


Findings from the current study support aspects of the HEALTHY program and SY approach, and provide information on aspects that could be improved. Overall, the program was viewed positively by participants and was well-attended. Although most SY participants favored inclusion of a partner, it was not reported as a deciding factor in enrolling in or adhering to the program. Attendance and home practice were similar in the IY and SY groups. Therefore, a SY program may be appealing for middle-aged to older adults with OA, but evidence does not suggest that it is a strong factor in promoting adoption of or adherence to a yoga program. It was apparent that male and female participants tended to favor female practice partners—a trend warranting further exploration.

Community recruitment resulted in an unexpectedly diverse sample but was labor-intensive with a high screening-to-enrollment ratio. The diversity of recruited participants was surprising in light of the fact that yoga is most commonly performed by highly educated, middle-class, White/Caucasian women (Cramer et al., 2016). The current sample was 53% male and 41% self-identified as a race other than White/Caucasian. Several factors may have promoted this diversity. First, the study was advertised on metro buses that served a range of residential areas in which individuals of varied backgrounds lived. The second factor may have been that the potential of bringing a partner to the shared program made the program more appealing. Overall, the diverse sample demonstrated interest in yoga by individuals beyond the demographic that typically performs yoga, which argues for further research on the efficacy of yoga in diverse groups as well as further examination of factors that serve as barriers or facilitators to yoga participation among these groups.

Data from the current study showed a higher rate of adverse effects than those in other studies of yoga for individuals with OA (Cheung et al., 2014; Kolasinski et al., 2005; Moonaz, Bingham, Wissow, & Bartlett, 2015; Park, McCaffrey, Dunn, & Goodman, 2011), but was comparable to the current authors' pilot study of a different gentle yoga program (Taibi & Vitiello, 2011). Among the few yoga studies that reported adverse effects, three reported no yoga-related adverse effects (Cheung et al., 2014; Kolasinski et al., 2005; Moonaz et al., 2015), and one reported one adverse event but did not describe the event or state whether it was related to yoga (Park et al., 2011). The higher rate of adverse effects of yoga in the current study was likely related to a combination of the approach to monitoring adverse effects and certain characteristics of the program. It is plausible that the higher rate may have been due to closer tracking as well as having participants report any unusual symptoms versus reporting only injuries, as was done in one other study (Cheung et al., 2014). The other studies reporting adverse effects did not state how these were defined or measured.

Although the HEALTHY program was designed to be safe and feasible for older adults, certain characteristics of the program may have contributed to adverse effects. Several participants reported experiencing general soreness at the beginning of the program, suggesting that this discomfort was due to deconditioning. Thus, it seems advisable to instruct participants to recognize the difference between soreness from strength building versus injury. Other participants reported that they repeated some poses beyond the number of times instructed while the teacher was giving explanations. The HEALTHY program used the Viniyoga style, in which poses are gently repeated several times rather than holding for a short period. The latter approach is characteristic of the yoga styles used in the studies reporting fewer side effects, with those styles including Hatha (Cheung et al., 2014; Moonaz et al., 2015), Iyengar (Kolasinski et al., 2005), or chair yoga (Park et al., 2011). Thus, it is important when using the Viniyoga approach to remind participants to stop once they have finished the recommended number of repetitions. It may also be helpful to instruct participants on an activity, such as practicing a breathing exercise, if they finish practicing a pose before others. Finally, to provide adequate monitoring during instruction, small class sizes and/or a second instructor or assistant could be helpful.

Efficacy outcomes showed improvements on sleep questionnaires (PROMIS sleep disturbance and ISI), but no effect on the sleep diary, actigraphy, WOMAC, or PHQ-8. These findings suggest that the program may have improved perceptions related to sleep quality but may not have affected actual sleep. However, firm conclusions about efficacy cannot be reached, as the current study was aimed primarily at exploring feasibility. To broadly explore interest in and adherence to the program, exclusion criteria were not stringent. Therefore, extraneous factors could have obscured potential findings. Furthermore, several participants experienced life events and injuries unrelated to the yoga program that could have affected the current findings, including a minor car accident, ground-level fall, and dehydration requiring treatment. Future research with a larger sample and possibly more stringent eligibility criteria is needed to determine the efficacy of this yoga program.

Another limitation is the lack of a control group. Furthermore, the full “dose” of yoga was not known because participants reported whether they performed home practice, but not the duration. It also would have been informative to have obtained an objective assessment of functional ability, such as the Timed Up-and-Go test (Podsiadlo & Richardson, 1991) or more focused self-report scales, such as the Lower Extremity Functional Scale (Binkley, Stratford, Lott, & Riddle, 1999).


The current study supports the overall feasibility of the program, both in the shared and individual formats. The study also demonstrated interest in yoga of a surprisingly diverse sample. Inclusion of a partner in yoga did not seem to improve attendance, adherence, or efficacy, but was also not a barrier. Efficacy data suggest that the HEALTHY program may improve certain self-reported sleep outcomes, but given the study limitations, further research is needed to draw conclusions on whether the HEALTHY program improves sleep and OA symptoms.


  • Ameringer, S., Elswick, R.K. Jr. , Menzies, V., Robins, J.L., Starkweather, A., Walter, J. & Jallo, N. (2016). Psychometric evaluation of the Patient-Reported Outcomes Measurement Information System fatigue-short form across diverse populations. Nursing Research, 65, 279–289. doi:10.1097/NNR.0000000000000162 [CrossRef]
  • Bastien, C.H., Valliéres, A. & Morin, C.M. (2001). Validation of the Insomnia Severity Index as an outcome measure for insomnia research. Sleep Medicine, 2, 297–307. doi:10.1016/S1389-9457(00)00065-4 [CrossRef]
  • Bellamy, N., Buchanan, W.W., Goldsmith, C.H., Campbell, J. & Stitt, L.W. (1988). Validation study of WOMAC: A health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of the hip or knee. Journal of Rheumatology, 15, 1833–1840.
  • Binkley, J.M., Stratford, P.W., Lott, S.A. & Riddle, D.L. (1999). The Lower Extremity Functional Scale (LEFS): Scale development, measurement properties, and clinical application. North American Orthopaedic Rehabilitation Research Network. Physical Therapy, 79, 371–383.
  • Buysse, D.J., Yu, L., Moul, D.E., Germain, A., Stover, A., Dodds, N.E. & Pilkonis, P.A. (2010). Development and validation of patient-reported outcome measures for sleep disturbance and sleep-related impairments. Sleep, 33, 781–792.
  • Cheung, C., Park, J. & Wyman, J.F. (2016). Effects of yoga on symptoms, physical function, and psychosocial outcomes in adults with osteoarthritis: A focused review. American Journal of Physical Medicine & Rehabilitation, 95, 139–151. doi:10.1097/PHM.0000000000000408 [CrossRef]
  • Cheung, C., Wyman, J.F., Resnick, B. & Savik, K. (2014). Yoga for managing knee osteoarthritis in older women: A pilot randomized controlled trial. BMC Complementary and Alternative Medicine, 14, 160. doi:10.1186/1472-6882-14-160 [CrossRef]
  • Cramer, H., Lauche, R., Langhorst, J. & Dobos, G. (2013). Yoga for rheumatic diseases: A systematic review. Rheumatology, 52, 2025–2030. doi:10.1093/rheumatology/ket264 [CrossRef]
  • Cramer, H., Ward, L., Steel, I., Lauche, R., Dobos, G. & Zhang, Y. (2016). Prevalence, patterns, and predictors of yoga use: Results of a U.S. nationally representative survey. American Journal of Preventive Medicine, 50, 230–235. doi:10.1016/j.amepre.2015.07.037 [CrossRef]
  • Edinger, J.D., Bonnet, M.H., Bootzin, R.R., Doghramji, K., Dorsey, C.M., Espie, C.A. & Stepanski, E.J. (2004). Derivation of research diagnostic criteria for insomnia: Report of an American Academy of Sleep Medicine Work Group. Sleep, 27, 1567–1596.
  • Gandek, B. (2014). Measurement properties of the Western Ontario and McMaster Universities Osteoarthritis Index: A systematic review. Arthritis Care & Research, 67, 216–229. doi:10.1002/acr.22415 [CrossRef]
  • Garcia, S.F., Cella, D., Clauser, S.B., Flynn, K.E., Lad, T., Lai, J.S. & Weinfurt, K. (2007). Standardizing patient-reported outcomes assessment in cancer clinical trials: A patient-reported outcomes measurement information system initiative. Journal of Clinical Oncology, 25, 5106–5112. doi:10.1200/JCO.2007.12.2341 [CrossRef]
  • Hong, T.B., Franks, M.M., Gonzalez, R., Keteyian, S.J., Franklin, B.A. & Artinian, N.T. (2005). A dyadic investigation of exercise support between cardiac patients and their spouses. Health Psychology, 24, 430–434. doi:10.1037/0278-6133.24.4.430 [CrossRef]
  • Kolasinski, S.L., Garfinkel, M., Tsai, A.G., Matz, W., Van Dyke, A. & Schumacher, H.R. Jr.. (2005). Iyengar yoga for treating symptoms of osteoarthritis of the knees: A pilot study. Journal of Alternative and Complementary Medicine, 11, 689–693. doi:10.1089/acm.2005.11.689 [CrossRef]
  • Kraftsow, G. (1999). Yoga for wellness. New York, NY: Penguin.
  • Kroenke, K., Spitzer, R.L. & Williams, J.B. (2001). The PHQ-9: Validity of a brief depression severity measure. Journal of General Internal Medicine, 16, 606–613. doi:10.1046/j.1525-1497.2001.016009606.x [CrossRef]
  • Kroenke, K., Spitzer, R.L., Williams, J.B. & Löwe, B. (2008). The Patient Health Questionnaire somatic, anxiety, and depressive symptom scales: A systematic review. General Hospital Psychiatry, 32, 345–359. doi:10.1016/j.genhosppsych.2010.03.006 [CrossRef]
  • Lichstein, K.L., Stone, K.C., Donaldson, J., Nau, S.D., Soeffing, J.P., Murray, D. & Aguillard, R.N. (2006). Actigraphy validation with insomnia. Sleep, 29, 232–239.
  • McMahon, S.K., Lewis, B., Oakes, J.M., Wyman, J.F., Guan, W. & Rothman, A.J. (2017). Assessing the effects of interpersonal and intrapersonal behavior change strategies on physical activity in older adults: A factorial experiment. Annals of Behavioral Medicine. Advance online publication. doi:10.1007/s12160-016-9863-z [CrossRef]
  • Mielenz, T.J., Xiao, C. & Callahan, L.F. (2016). Self-management of arthritis symptoms by complementary and alternative medicine movement therapies. Journal of Alternative and Complementary Medicine, 22, 404–407. doi:10.1089/acm.2015.0222 [CrossRef]
  • Moher, D., Schulz, K.F. & Altman, D. (2001). The CONSORT statement: Revised recommendations for improving the quality of reports of parallel-group randomized trials. Journal of the American Medical Association, 285, 1987–1991. doi:10.1001/jama.285.15.1987 [CrossRef]
  • Moonaz, S.H., Bingham, C.O., Wissow, L. & Bartlett, S.J. (2015). Yoga in sedentary adults with arthritis: Effects of a randomized controlled pragmatic trial. Journal of Rheumatology, 42, 1194–1202. doi:10.3899/jrheum.141129 [CrossRef]
  • Morin, C.M., Belleville, G., Bélanger, L. & Ivers, H. (2011). The Insomnia Severity Index: Psychometric indicators to detect insomnia cases and evaluate treatment response. Sleep, 34, 601–608.
  • National Sleep Foundation. (2003). 2003 Sleep in America poll. Retrieved from
  • Neogi, T. & Zhang, Y. (2013). Epidemiology of osteoarthritis. Rheumatic Diseases Clinics of North America, 39, 1–19. doi:10.1016/j.rdc.2012.10.004 [CrossRef]
  • Park, J., McCaffrey, R., Dunn, D. & Goodman, R. (2011). Managing osteoarthritis: Comparisons of chair yoga, Reiki, and education (pilot study). Holistic Nursing Practice, 25, 316–326. doi:10.1097/HNP.0b013e318232c5f9 [CrossRef]
  • Podsiadlo, D. & Richardson, S. (1991). The timed “Up & Go”: A test of basic functional mobility for frail elderly persons. Journal of the American Geriatrics Society, 39, 142–148. doi:10.1111/j.1532-5415.1991.tb01616.x [CrossRef]
  • Power, J.D., Perruccio, A.V. & Badley, E.M. (2005). Pain as a mediator of sleep problems in arthritis and other chronic conditions. Arthritis and Rheumatism, 53, 911–919. doi:10.1002/art.21584 [CrossRef]
  • Rosemann, T., Kuehlein, T., Laux, G. & Szecsenyi, J. (2008). Factors associated with physical activity of patients with osteoarthritis of the lower limb. Journal of Evaluation in Clinical Practice, 14, 288–293. doi:10.1111/j.1365-2753.2007.00852.x [CrossRef]
  • Sangha, O., Stucki, G., Liang, M.H., Fossel, A.H. & Katz, J.N. (2003). The self-administered comorbidity questionnaire: A new method to assess comorbidity for clinical and health services research. Arthritis and Rheumatism, 49, 156–163. doi:10.1002/art.10993 [CrossRef]
  • Satariano, W.A., Haight, T.J. & Tager, I.B. (2002). Living arrangements and participation in leisure-time physical activities in an older population. Journal of Aging and Health, 14, 427–451. doi:10.1177/089826402237177 [CrossRef]
  • Sharma, M., Haider, T. & Knowlden, A.P. (2013). Yoga as an alternative and complementary treatment for cancer: A systematic review. Journal of Alternative and Complementary Medicine, 19, 870–875. doi:10.1089/acm.2012.0632 [CrossRef]
  • Sherman, K.J., Cherkin, D.C., Erro, J., Miglioretti, D.L. & Deyo, R.A. (2005). Comparing yoga, exercise, and a self-care book for chronic low back pain: A randomized, controlled trial. Annals of Internal Medicine, 143, 849–856. doi:10.7326/0003-4819-143-12-200512200-00003 [CrossRef]
  • Smith, M.T., Quartana, P.J., Okonkwo, R.M. & Nasir, A. (2009). Mechanisms by which sleep disturbance contributes to osteoarthritis pain: A conceptual model. Current Pain and Headache Reports, 13, 447–454. doi:10.1007/s11916-009-0073-2 [CrossRef]
  • Steptoe, A., Rink, E. & Kerry, S. (2000). Psychosocial predictors of changes in physical activity in overweight sedentary adults following counseling in primary care. Preventive Medicine, 31, 183–194. doi:10.1006/pmed.2000.0688 [CrossRef]
  • Taibi, D.M., Landis, C.A. & Vitiello, M.V. (2013). Concordance of polysomnographic and actigraphic measurement of sleep and wake in older women with insomnia. Journal of Clinical Sleep Medicine, 9, 217–225. doi:10.5664/jcsm.2482 [CrossRef]
  • Taibi, D.M. & Vitiello, M.V. (2011). A pilot study of gentle yoga for sleep disturbance in women with osteoarthritis. Sleep Medicine, 12, 512–517. doi:10.1016/j.sleep.2010.09.016 [CrossRef]
  • Taibi, D.M. & Vitiello, M.V. (2012). Yoga for osteoarthritis: Nursing and research considerations. Journal of Gerontological Nursing, 38(7), 26–35. doi:10.3928/00989134-20120608-04 [CrossRef]
  • Ward, L., Stebbings, S., Cherkin, D. & Baxter, G.D. (2013). Yoga for functional ability, pain and psychosocial outcomes in musculoskeletal conditions: A systematic review and meta-analysis. Musculoskeletal Care, 11, 203–217. doi:10.1002/msc.1042 [CrossRef]
  • Wilcox, S., Brenes, G.A., Levine, D., Sevick, M.A., Shumaker, S.A. & Craven, T. (2000). Factors related to sleep disturbance in older adults experiencing knee pain or knee pain with radiographic evidence of knee osteoarthritis. Journal of the American Geriatrics Society, 48, 1241–1251. doi:10.1111/j.1532-5415.2000.tb02597.x [CrossRef]
  • Yu, L., Buysse, D.J., Germain, A., Moul, D.E., Stover, A., Dodds, N.E. & Pilkonis, P.A. (2011). Development of short forms from the PROMIS sleep disturbance and sleep-related impairment item banks. Behavioral Sleep Medicine, 10, 6–24. doi:10.1080/15402002.2012.636266 [CrossRef]

Participant Demographics and Health Information (N = 17)

VariableShared Yoga (n = 10)Individual Yoga (n = 7)
Age (years), median (range)54 (50 to 68)56 (50 to 72)
n (%)
  Male6 (60)3 (43)
  Female4 (40)4 (57)
  Caucasian/White7 (70)3 (42.9)
  African American4 (40)2 (28.6)
  Latino3 (30)2 (28.6)
  Asian, American Indian/Alaskan Native, or Pacific Islander2 (20)2 (28.6)
  Primary through high school4 (40)
  Associate degree or some college4 (40)2 (28.6)
  College graduate2 (20)5 (71.5)
Relationship status
  Single7 (70)4 (57.1)
  Married/partnered3 (30)3 (42.9)
Joints affected by OA
  Knee6 (60)6 (85.7)
  Hip5 (50)4 (57.1)
  Hand4 (40)1 (14.3)
  Back3 (30)
  Ankle3 (30
  Shoulders2 (20)
  Neck1 (10)1 (14.3)
  Elbow1 (10)1 (14.3)
  Wrist1 (10)1 (14.3)
n (%)
Other health conditions
  Depression5 (50)1 (14.3)
  Hypertension4 (40)2 (28.6)
  Hypothyroidism2 (20)
  Posttraumatic stress disorder2 (20)1 (14.3)
  Seizure disorder2 (20)
  Otherb2 (20)2 (28.6)
  Diabetes1 (10)1 (14.3)
  Gastroesophageal reflux1 (10)1 (14.3)
  Scoliosis or spondylosis1 (10)1 (10)
  Back pain2 (28.6)

Poses Reported to Cause Pain/Soreness

PoseSpecific Problem (n)
BridgePain in back (1)
Cobra (on floor)Pain in hands (1), pain in back (1)
Clam shellPain in weight-bearing (dependent) hip (1)
Knees to chestSoreness in wrists and hands (1)
Leg cradleSoreness in knee (1)
Side angle posePulled muscle in side/ribs (1)
Warrior 1Pain in knees (2)

Efficacy Outcomes Pre- and Post-Intervention

VariableMedian (Range)p Value
Insomnia Severity Index (0 to 28)15.5 (11 to 23)12.5 (3 to 23)0.006
Patient Health Questionnaire-8 (0 to 24)7.5 (3 to 14)7 (0 to 19)0.387
Western Ontario McMaster Osteoarthritis Index
  Total (0 to 96)37 (20 to 70)35.5 (12 to 74)0.396
  Pain (0 to 20)9 (5 to 15)8 (2 to 17)0.313
  Stiffness (0 to 8)4 (2 to 6)4 (2 to 6)0.952
  Function (0 to 68)25 (9 to 53)25.5 (8 to 51)0.413
Patient-Reported Outcomes Measurement System
  Sleep disturbance (8 to 40)26 (20 to 34)22.5 (14 to 33)0.033
  Sleep-related impairment (8 to 40)22.5 (14 to 35)22.5 (13 to 31)0.379
  Fatigue (7 to 35)19.5 (12 to 24)18.5 (11 to 24)0.253
Sleep diarya
  Time in bed (hours)8.3 (6.2 to 11.4)8.3 (6.2 to 13.9)0.65
  Total sleep time (hours)6.7 (3.5 to 8.6)6.4 (4.1 to 12.2)0.463
  Sleep latency (minutes)19.3 (6.4 to 145.7)22.5 (7.1 to 96)0.861
  Wake after sleep onset (minutes)44.2 (8.3 to 100.8)79.7 (62 to 90.1)0.753
  Sleep efficiency (%)80.4 (42.2 to 98.3)79.7 (62 to 90.1)0.753
  Sleep quality (1 to 9)b5.6 (2.6 to 7.8)6 (3.3 to 7.4)0.576
  Time in bed (hours)8.3 (7.8 to 9.8)8.1 (6 to 10.4)0.241
  Total sleep time (hours)6.9 (4.3 to 8.5)7.1 (2.7 to 8.6)0.075
  Sleep latency (minutes)12.1 (2.4 to 48.1)17.3 (1.7 to 45.2)0.594
  Wake after sleep onset (minutes)56.7 (30.1 to 182.4)54.6 (22.9 to 141.4)0.534
  Sleep efficiency (%)85.4 (53.8 to 93)82.9 (42.8 to 93.3)0.213

HEALTHY Program Yoga Poses and Practice Sequence

Lesson (class session)Sequence and Poses
Lesson 1: focus on core strength and stability (Classes 1–3, 10)Start in a seated positionSeated mountain pose – seated upright with alignment of shoulders over hips and knees over ankles. Breath awareness – conscious awareness of one's breathing, noting the pattern, flow, and any tension. Seated wheel pose – gentle flexion and extension of the spine. Transition onto the floor, prone positionCobra pose – from a prone position, the head and shoulders are lifted slightly off the floor. Transition to supine positionKnees to chest pose – the knees are drawn toward and away from the chest, with the knee bent. Extended leg pose & ankle circles – one leg is drawn with the knee bent toward the chest and then extended toward the ceiling. On the last extension, the ankle is circled several times clockwise and counter-clockwise. Butterfly pose – with the knees bent and feet on the floor, the knees open away from each other then are drawn back together. Clam Shell – in a side-lying position, the upper knee is lifted and lowered to strengthen the hip muscles. Leg Cradle – in a supine position with knees bent, one ankle is placed on the opposite knee, which is gently drawn toward the chest for a hip stretch. Bridge Pose – with knees bent and feet placed on the floor, the buttocks and back are lifted into a gentle back bend. Hip Circles – with knees bent and feet lifted off the floor, the knees are moved in circular motions to rotate the hip. Resting Pose – resting supine with support under the neck and knees, the individual relaxes and focuses awareness on the body. Transition back onto the chairSeated breathing exercises and breath awareness – individuals practice slow, paced breathing. After a few breath cycles, the exhalation is partitioned with a very brief (<1 second) pause mid-exhale to promote breath awareness and control.
Lesson 2: focused on healthy joint alignment (Classes 4–6, 11)Start in a seated positionSeated mountain pose1Breath awareness1Thunderbolt Pose – seated in the chair, the individual moves into a forward bend and place the backs of the hands on the small of the back. The individual then sweeps the arms forward and extends the arms overhead while sitting upright. Cobra Pose (in chair) – from a seated forward bend, the individual lifts the shoulders and head parallel to the floor. Seated leg extensions – one leg is extended and the heel placed on a yoga block. The ankle is flexed and extended, then rotated clockwise and counter-clockwise. Hip flexes and circles – one leg at a time is lifted using engagement of core muscles (using hands to assist as needed). The knee is moved in clockwise and counter-clockwise rotations to move the hip joint in its socket. Seated wheel pose1Seated side bend – sitting upright, one arm is swept out to the side and overhead. The arm and torso are then gently extended toward the opposite side. Wide-legged forward bend – the knees are moved slightly wider than the seat of the chair, then the individual moves into a forward bend. Seated breathing exercises and breath awareness1Transition onto the floor and resting pose1Transition back onto the chair
Lesson 3: focused on muscle engagement for stability (Classes 7–9, 12)Start in a seated positionSeated mountain pose1Breath awareness1Thunderbolt Pose2Transition to standingStanding Mountain Pose – standing with alignment of hips over ankles and shoulders over hips. Arms are relaxed at the sides. Warrior1, Part 1 – the individual stands with one leg forward from the other. The front knee bends slightly, and the arms sweep out to the sides with hands pointed upward and elbows bent. Warrior 1, Part 2 – the individual remains with feet in Warrior 1 position and bends the front knee. Instead of sweeping the arms, the individual forward-bends over the front leg, placing the hands on the back of a chair for support. Chair Pose – standing with the feet under the hips, the individual bends both knees as if sitting. The hands are placed on the back of a chair for support. Progressive standing forward bend – standing with the feet under the hips, the individual forward-bends from the hips, first to a half-way point, then to the furthest comfortable point. The hands are placed on the back of a chair for support. Seated breathing exercises and breath awareness1Transition onto the floor and resting pose1Transition back onto the chair
Morning practice (Class 1)Start supine (in bed, before arising)Knees to chest pose1Extended leg pose & ankle circles1Individual hip circles – performed like hip flexes and circles, but done one leg at a time. Transition to sitting on side of bed or in a chairSeated wheel pose1Full arm circles – both arms are simultaneously rotated front to back in small, then in increasingly larger, circles. Arm circles are then repeated circling back-to-front. Seated mountain pose1

Dr. Buchanan is Associate Professor, Department of Biobehavioral Nursing & Health Informatics, School of Nursing, Dr. Vitiello is Professor, Department of Psychiatry & Behavioral Sciences, and Dr. Bennett is Lecturer, Physical Therapy Curriculum, Department of Rehabilitation Medicine, School of Medicine, University of Washington, Seattle, Washington.

The authors have disclosed no potential conflicts of interest, financial or otherwise. The study was funded by the National Institute of Nursing Research (P30 NR04001; Center for Research on the Management of Sleep Disturbance). This was not an industry-supported study.

The authors thank Robin Rothenberg for her collaboration and contributions to the design of the HEALTHY program, Dorothy Hiestand for teaching the HEALTHY program, Katina Velloth for her assistance with the data collection and actigraphy analyses, and the participants who gave their time.

Address correspondence to Diana Taibi Buchanan, PhD, RN, Associate Professor, Department of Biobehavioral Nursing & Health Informatics, School of Nursing, University of Washington, Box 357266, Seattle, WA 98195-7266; e-mail:

Received: July 25, 2016
Accepted: February 16, 2017
Posted Online: April 11, 2017


Sign up to receive

Journal E-contents