The “Method of Physical Action” in Theatre Training Improves Balance and Reduces Fall Risk in Older Adults Living in Subsidized Housing: A Randomized Controlled Trial

Abstract

Background and Objectives

Residents of subsidized senior housing (SSH) need housing-based interventions to address high rates of physical inactivity and social isolation that can lead to increased risks of physical and mental health impairments. Theatre experience training (TET) programs have been shown to improve cognitive function, emotional well-being, social connection, and health-related quality of life (HR-QoL) in this population. This study evaluated the effect of an innovative TET program involving the method of physical acting on objective measures of physical functioning and emotional stress.

Research Design and Methods

This single-blind trial randomized SSH residents from 5 SSH communities to a 10-week, twice-weekly TET program or a waitlist control arm, with randomization conducted separately in each SSH. Participants’ static balance, functional mobility, lower body strength, HR-QoL, emotional stress as measured by hair cortisol levels, and fall incidents were assessed at baseline, postprogram, and 3-month follow-up.

Results

The study enrolled 81 participants; 53 were randomized to the TET arm and 28 to the waitlist arm. TET arm participants had improved static balance and reduced falls postprogram compared with control arm participants. Some carry-over effect was observed at the 3-month follow-up. No net benefit was observed for functional mobility, lower body strength, HR-QoL, or stress levels at the postprogram or 3-month follow-up assessments.

Discussion and Implications

The TET program involving the method of physical acting can improve static balance and reduce fall incidents in SSH residents. Ongoing TET programs in SSH may help residents maintain reductions in fall incidents.

Clinical Trial Registration Number: NCT04582370

Translational Significance:

The study shows that an innovative 10-week theatre experience training (TET) program can significantly improve static balance and reduce falls among residents of subsidized senior housing. Our results suggest that the TET is a feasible, engaging intervention for enhancing physical stability in the short term and may help reduce falls and fall-related healthcare costs in the long term. However, sustained benefits may require ongoing or repeated interventions. The findings support integrating theater-based programs into senior housing policies and practices to support broader fall reduction efforts and enhance resident well-being.

Studies report that, compared with older adults living in single-dwelling homes, subsidized senior housing (SSH) residents have a higher prevalence of poor health from a multitude of chronic health conditions (1,2). They also have a higher risk of falls (3,4).

A recent systematic review found that participating in performing arts can promote health and prevent or mitigate poor health among older adults (5). Theatre is a complex, multicomponent performing art. To prepare for stage performances, actors often incorporate appropriate physical movements while memorizing a script or improvising lines to enhance their expressive capacity and convey emotional depth (6,7). Therefore, acting requires as much movement as it does vocalization (8,9). Enacting stories using emotional verbal and nonverbal expression, as well as social engagement with other actors, was shown to decrease anxiety, stress, and loneliness; increase self-esteem and a sense of value and purpose; and promote personal growth and achievement (10–13). Theatre experience training (TET) typically involves improvisation, creativity, learning, emotional expression, and movement in a group setting (10,14). This multidimensional engagement can address several health care concerns common among SSH residents, including comorbidities associated with physical inactivity, mental health issues, social isolation, and cognitive decline (2).

Studies involving SSH residents who participated in TET have shown that the training enhances: (1) the physical component of their health-related quality of life (HR-QoL) (13); (2) cognitive functioning, such as attention and memory (12,15); (3) psychological and emotional well-being (12,13,16); and (4) social connectedness (17). In addition, evidence suggests that participating in theatre improves scores on the physical health component of the 36-item Short-Form Health Survey (SF-36) (13) and emotional health/well-being (12,13,16) among SSH residents. However, these outcomes have typically been measured with self-report instruments, and most of these studies used a single-group pretest–posttest design. As a result, there is still a lack of clear, reliable evidence demonstrating that theatre participation can enhance physical health and emotional well-being among SSH residents. Therefore, we designed an innovative 10-week TET program and used both subjective and objective clinical evaluations to explore the impact of the program on physical functioning and emotional stress in SSH residents. We hypothesized that SSH residents who were assigned to participate in an innovative 10-week TET program would demonstrate better physical functioning and HR-QoL and reduced emotional stress compared with waitlist controls at the conclusion of the program and at 3-month follow-up.

Method

Research Design and Ethical Approval

This study involved a single-blind randomized controlled trial (RCT) research design with a 3-month follow-up. The study was conducted with Institutional Review Board (IRB) approval from the University of Alabama at Birmingham (protocol # IRB-300005331). The trial was registered at ClinicalTrials.gov (NCT04582370) before the enrollment of the first participant.

Participants

Participants were older adults living in subsidized housing communities in Birmingham, AL, United States. The participant inclusion criteria were: (1) aged 62 years or older; (2) able to carry on a daily conversation with or without hearing aids; (3) able to follow simple verbal instructions involving 2–3 steps (ie, the teaching artist’s verbal instructions for various activities during TET); and (4) able to walk short distances with or without aids (as outcome measures required participants to perform static balance and walking activities). Exclusion criteria were: (1) blindness or unable to see the teaching artist’s nonverbal instructions or other actors’ movements during TET; (2) severe cognitive impairment as indicated by a score of less than 6 on the Short Portable Mental Status Questionnaire (18); (3) language barriers that might impede interactions during the TET sessions, understanding of information in the study’s informed consent form or completion of the study’s psychological outcome measures in a questionnaire format; or (4) having a known assaultive or maladaptive behavior pattern.

Sample size estimation

Based on a two-group pretest–posttest design, with an alpha level of 0.05 and power set at 0.80, sample size estimates were calculated using effect sizes of 6.5 for the physical component summary and 7.9 for the mental component summary of the SF-36 (19), as reported in a prior study (13). Consequently, sample sizes of 30 and 20 participants per group were needed for the physical and mental component summaries, respectively. The effect sizes used in the calculations indicate clinically meaningful improvements (20). The larger value of 30 was used to ensure sufficient power. To account for an expected 15% attrition rate, we planned to recruit at least 70 residents.

Setting

We recruited participants from 5 of the 15 SSH communities in Birmingham, Alabama, United States. The selection of the SSH facilities was based on their proximity to the study site and the existence of an established relationship between the SSH director or service coordinator and a research team member.

Procedures

Recruitment

Upon approval from the SSH director or service coordinator, group information sessions were scheduled to introduce the study to residents. Recruitment assistance from the service coordinators included leaving flyers in each resident’s postal mailbox, posting flyers in the common areas (eg, SSH elevators) a week before the information session, and personally inviting residents to join the session. The information sessions were held in the SSH dining room or another common venue. Refreshments were served at each information session to attract residents.

During the information sessions, study staff explained the purpose of the study, including the concept of randomization and a waitlist group, to residents, who then had opportunities to ask questions. Contact information sheets were circulated at the end of the information session. Residents who were interested in being contacted or in enrolling in the study completed the contact information sheet (ie, name and telephone number). A blank consent form was given to all residents who attended the information session.

Recruitment from the SSH community was staggered, starting with the first information session, held in September 2021. We approached the director or service coordinator of each SSH in sequence to assist with recruitment. Recruitment information sessions were held consecutively, with each cohort consisting of residents from a single SSH community. Only one TET program was conducted in each community. Because only one teaching artist was involved in the study, recruitment for each subsequent cohort began after residents in the previous community completed the program.

Clinical Assessments

An evaluator (occupational therapist) blinded to study arm assignment contacted residents who expressed interest in the study to arrange times in the following week for them to complete the consent process, undergo screening to determine study eligibility, and undergo the baseline assessment. The baseline assessment consisted of a performance-based physical functioning evaluation, questionnaires addressing socio-demographic information and psychological assessment, and hair sample collection. The baseline assessment was conducted in a semi-private area in the residents’ SSH.

The same blinded evaluator assessed participants at baseline, immediately postprogram, and at 3-month follow-up. Each physical functioning evaluation was conducted in the same manner and in the same location as the baseline evaluation. The Short Portable Mental Status Questionnaire was administered at all 3 time points to evaluate participants’ cognitive function, as cognitive decline can affect physical functioning (21). The background section of the questionnaire asked participants to report any falls they experienced in the previous 3 months.

Randomization

After residents completed the baseline assessment, they were randomly assigned to 1 of 2 arms: the TET or waitlist control arm. We conducted computerized (ie, blinded) randomization of participants in each SSH using the list of residents who completed the baseline assessment. We used a 2:1 ratio of subject allocation in arm assignments. The rationale for adopting unequal allocation/randomization was that TET is a group activity that requires a minimum number of participants to make the group sessions viable and create the intended benefits (22), while the waitlist condition did not involve a planned group activity. Therefore, randomizing more participants to the TET arm allowed us to form adequate group sizes.

TET Program

The design of the TET program was based on the principles of acting as written and practiced by Constantin Stanislavski in his revolutionary text on acting: “An Actor Prepares” (23). For this TET program, we embedded the method of physical action, an acting technique developed by Stanislavski. The method of physical action is an approach emphasizing purposeful physical movement and behavior to reveal and portray a character’s emotion and psychology (24). Based on Stanislavski’s acting principles, the fundamentals addressed in the TET program included the following 3 main objectives (25):

1. Preparation for the Actor, which involves relaxation (reducing social anxiety), collaboration (becoming part of the ensemble by learning to give and receive trust, thereby breaking down social barriers), physicality (communicating need through movement and posture), and vocality.

2. Learning the Components of the Repeatable Acting Process, which involves commitment to the given circumstances (urgency), physicality, and attention and concentration (focusing on character needs).

3. Synthesizing Components into Characterization, which involves moment-to-moment creativity and creates an opportunity for true emotional expression to arise.

The TET program followed Stanislavsky’s acting method by encompassing the following elements:

• Emotional memory: Participants draw on their own past experiences and emotions to bring authenticity to their characters.

• Given circumstances: Participants understand their character’s world by considering who they are, their relationships, their location, and their time period.

• Physical action: Participants concentrate on the physical requirements of their role to become aware of a character’s reasoning.

• Sense memory: Participants use their senses (ie, sight, hearing, smell, and touch) to create a more realistic portrayal of their characters’ experiences.

The TET program was designed as a process-based rather than a performance-based course. For populations such as SSH residents, who often have little or no previous experience or interest in performance, the program’s benefits are best achieved through its emphasis on the training process. Having a product or performance focus could potentially pressure and intimidate SSH residents, which could keep them from participating. Group warmups (physical and vocal) and group ensemble exercises were used to address each of the program’s three objectives. The acting exercises were designed to stimulate participants’ physical and cognitive skills, enhance their communication and socialization skills, and encourage their emotional expression. Activities involving miming, mirroring, pantomime, pretending, endowment, and improvisation were chosen to increase participants’ awareness of their overall balance and the physical movement of different body parts and to stimulate cognitive function.

Each acting class session began with a physical and vocal warmup. This was followed by acting exercises, which included 2-person scenes, individual monologues, and group skits. The physical warmup exercise involved various body movements. The vocal warmups included breathing exercises and diction exercises using tongue twisters. Participants engaged in physical, cognitive, and emotional exercises such as those given to beginning acting students in traditional theatre schools. Participants were taught to craft and repeat behaviors. The repeatable steps through crafting the physical behaviors (ie, proximity, urgency) created an open environment that allowed honest emotion to arise. This predictable structure helped to create a sense of safety and confidence that eased the introduction of new exercises and materials. Participants were also encouraged to create characters and pursue the needs of those characters in a way that encouraged the rise of honest emotion. Because of the wide disparity in the participants’ visual functioning, the use of printed monologues and dialogue was minimal. Instead, participants made nursery rhymes a key focus of their work, suggesting appropriate costumes and props and developing characteristics of their chosen nursery rhyme character. This was a focal point of several TET sessions in which participants wore elaborate costumes and used complex props that they had created to enhance their scene (these elements were not required, however). Participants used these props and costumes to create characters, place them in modern day settings, and improvise contemporary scenes. Participants were asked to interact with each other’s characters to increase their improvisational skills and enhance awareness of the other characters’ fictional surroundings. This, in turn, increased their awareness of their own imagined environments. Improvisation and increased physical awareness were essential to this process. Participants also took part in activities that involved eye contact and exploring their personal space with an acting partner or partners to enhance communication and socialization.

The participants were guided through the 10-week TET program by a professional teaching artist who had more than 20 years of experience in developing and implementing theatre-arts programming for older adults. The teaching artist met with participants for 1-hour sessions twice a week (Tuesday and Thursday, either late morning or early afternoon), for a total of 20 hours across 10 weeks. The training sessions were held in a common meeting room within each SSH. Twenty sessions over a 10-week period provided a sufficient trajectory for participants to acclimate themselves to the collaborative process as well as to the fundamentals of theatre. We began implementing the TET program in SSH in October 2021 and concluded in April 2023.

Study Outcome Measures

The primary outcome measures used for evaluating the impact of the TET program were physical functioning, HR-QoL, and emotional stress. The secondary outcome measure was the number of falls in the 3 months experienced by participants after they completed the baseline evaluation.

Physical functioning included static balance, functional mobility/dynamic balance, and lower body strength. Static balance was assessed using the standing balance test component of the Short Physical Performance Battery (26). Functional mobility/dynamic balance was measured using the 8-foot timed up-and-go test (27). Lower body strength was assessed using the 30-second sit-to-stand test (28).

For the standing balance test, participants are asked to remain free standing for 10 seconds with their feet in three different positions: side-by-side, semi-tandem, and tandem (26). Participants were timed until they moved their feet or 10 seconds had elapsed. Using previously defined and validated cut-points based on timed quartiles of each stance position, scores of each position were summed to form an overall balance score ranging from 0 (null balance) to 4 (no balance impairment). Participants received a maximum score of 4 if they could maintain their feet in the side-by-side (1 point), semi-tandem (1 point), and tandem (2 points) in standing positions for 10 seconds each (26).

For the timed up-and-go test, participants were asked to perform the following sequence: sit in a stable armchair (positioned against a wall), stand up, walk 8 ft, turn around, walk back, and sit down (27). Participants were instructed to sit toward the front of the chair with their back straight and to keep their feet flat on the floor and hands on the armrests of the chair before the start of the timed up-and-go test. A cone marker was placed 8 ft from the front of the chair. On the signal “go,” the participant rose from the chair, walked as quickly and safely as possible around both sides of the cone, and sat back down in the chair. Time to complete the task was recorded.

For the 30-second sit-to-stand test, participants were asked to perform as many sit-to-stand repetitions as possible in 30 seconds (28). Participants were instructed to sit toward the front of a stable chair (positioned against a wall, with a seat height of about 18 inches) with their back straight, feet flat on the floor, and arms crossed at the wrists and held against the chest. On the signal “go,” the participant rose to a full stand, returned to a fully seated position, and repeated this procedure as fast as possible for 30 seconds. The total number of sit to stands was recorded.

HR-QoL was measured using the Medical Outcomes Study SF-36 (29). The SF-36 is a self-administered questionnaire that consists of 36 questions, 35 of which are condensed into 8 multi-item scales: physical functioning, role physical, bodily pain, general health, vitality, social functioning, role emotional, and mental health. Raw scale scores were transformed into scores from 0 to 100, with higher scores indicating better functional health status. Scores of the eight scales were summarized into 2 dimensions: the physical component summary and the mental component summary (19). The SF-36 has been well-validated worldwide in various populations (30).

Emotional stress was measured using hair cortisol levels. Hair samples were taken from the posterior vertex of the participant’s head and cut with a pair of scissors as close to the scalp as possible. The amount of hair collected was about the diameter of a drinking straw (31). For cortisol extraction, segmented hair samples (trimmed to about 3 cm in length) were processed as previously reported (32), with a few minor modifications. In brief, each 3-cm hair segment was cut into small pieces, washed twice with isopropanol, and allowed to dry. Next, 25–75 mg of each hair sample was finely ground using stainless steel grinding balls and extracted overnight in methanol. After centrifugation, the supernatant was collected and dried at room temperature, reconstituted with 400 mL of phosphate buffered saline per 60 mg of hair samples, and vortexed for 1 minute. Cortisol extracts were then analyzed using a commercial high-sensitivity enzyme immunoassay kit (Salimetrics, LLC, Philadelphia, PA) following the manufacturer’s instructions. All samples were analyzed in duplicate, and the mean values were used for statistical analysis to meet established quality control standards. The assay readout was converted to picograms per milligram (pg/mg). The interassay coefficients of variation (ie, the average coefficient of variation from plate control means) and the intra-assay coefficients of variation (ie, the average coefficient of variation between duplicates) of the present sample assays were <10%.

Fall incidents were the number of falls in the past 3 months that participants reported in the background information section of the questionnaire.

Data Analysis

All quantitative data were assessed for normality using the Shapiro–Wilk test. Age was found to follow a normal distribution. The independent-samples t-test was conducted to compare the baseline age between the TET and waitlist arms, while Mann–Whitney U-tests were used to compare other quantitative variables (eg, years of living in SSH, numbers of visitors in the past 2 weeks). Chi-squared tests or Fisher’s exact tests were applied to compare qualitative variables between the TET and waitlist arms. Multiple imputation by chained equations was used to impute missing values using the R “mice” package, which is used with the random forest method for the multiple imputation (33). The random forest method is a nonparametric method that does not assume normality distributions or predictive equations. The multiple imputations were used to generate 10 imputed datasets and the mean values from the 10 imputations were used for analysis. Since the outcome measures were either ordinal measurement scales or their scores did not meet the normality assumptions for a standard analysis of covariance (ANCOVA), a Quade nonparametric ANCOVA was conducted to examine the variances in 2 arms (34). The p-value is based on an F-distribution with (k − 1) and (N − k) degrees of freedom with k = number of arms and N = sample size. The Wilcoxon signed-rank tests were used to evaluate significant differences between pre- and postprogram scores and postprogram and 3-month follow-up scores within-arm per change measure. Statistical significance was set at p < .05, 2-sided. All data analyses were conducted using the IBM SPSS Version 29 (IBM Corp, Armonk, NY).

Results

Recruitment

The total number of residents in the 5 SSH was 490, with individual community sizes ranging from 50 to 145. Attendance at the information sessions ranged from 5 to 26 residents per community (10%–26%), with a total of 89 (18%) from the 5 SSH. Of these attendees, 87 signed up to be contacted after the information sessions. An additional 16 residents who did not attend the information session were referred to by their SSH service coordinators and provided with contact information for a research team member. Thus, a total of 105 residents expressed interest in study participation, which represents a recruitment rate of 21% (105/490) among all residents from the 5 SSH.

Eighty-four residents completed the screening, and 3 did not meet the eligibility criteria (two were younger than 62 years and a third did not achieve a score above 5 on the Short Portable Mental Status Questionnaire). A total of 81 residents enrolled (ie, signed an informed consent form and completed a baseline assessment), which represented 77% (81/105; ranging from 61% to 100% in each of the 5 SSH) of those who provided contact information. The enrollment rate was 16.5% (81/490), ranging from 12% to 18% in each of the 5 SSH. Table 1 shows the characteristics of the 81 participants and their breakdown in the 2 study arms. No significant differences in any of the included background characteristics at baseline between the 2 study arms were observed.

Table 1.

Participant Characteristics at Baseline (N = 81)

		Theatre (n = 53)	Control (n = 28)
Age, year, Mean ± SD (range)		75 ± 7 (63–92)	74 ± 8 (63–90)
Sex	Female	45 (85%)	21 (75%)
	Male	8 (15%)	7 (25%)
Race	Black	28 (53%)	14 (50%)
	White	23 (43%)	13 (46%)
	Other	2 (4%)	1 (4%)
Education	≥Bachelor’s degree	14 (26%)	4 (14%)
	<Bachelor’s degree	37 (70%)	23 (82%)
	<High school	2 (4%)	1 (4%)
Marital status	Married	1 (2%)	0 (0%)
	Widowed	17 (32%)	8 (29%)
	Never married	13 (25%)	5 (18%)
	Divorced	22 (42%)	15 (54%)
Living situation	Alone	50 (94%)	28 (100%)
	With someone	3 (6%)	0 (0%)
Employment	Full Time/Part Time	2 (4%)	3 (11%)
	Retired/homemaker	45 (85%)	18 (64%)
	Disability	6 (11%)	7 (25%)
Walk outside	Cannot walk outside	6 (11%)	4 (14%)
	≤1 city block	19 (36%)	8 (29%)
	Several city blocks	18 (34%)	11 (39%)
	≥1 mile	10 (19%)	5 (18%)
Self-rated general health	Fair or Poor	13 (25%)	11 (39%)
	Excellent to Good	40 (76%)	17 (61%)
Chronic conditions†	0–1 condition	14 (26%)	8 (29%)
	2 conditions	17 (32%)	6 (21%)
	3+ conditions	22 (42%)	14 (50%)
Hearing impairment		10 (19%)	6 (21%)
Depression		12 (23%)	4 (14%)
Number of falls	None	35 (66%)	19 (68%)
	Once	9 (17%)	4 (14%)
	More than once	9 (17%)	5 (18%)
SPMSQ	Normal mental functioning	49 (93%)	25 (88%)
	Mild cognitive impairment	4 (8%)	3 (11%)
Significant life events in the past 3 months	No	44 (83%)	26 (93%)
	Yes	9 (17%)	2 (7%)
Years of living in SSH, Mean ± SD (range)		6 ± 6 (>0–33)	4 ± 4 (>0–15)
Number of times having visitors in the past 2 weeks, Mean ± SD (range)		2.7 ± 3 (0–14)	2.3 ± 2 (0–12)

Note: SPMSQ = Short Portable Mental Status Questionnaire. SSH = subsidized senior housing. Reported as number (percent), unless otherwise noted.

^†Sum of eight chronic conditions: heart disease/heart attack, stroke, diabetes, hypertension, arthritis, osteoporosis, chronic lung disease, and cancer.

Of the 81 participants, 53 were assigned to the TET arm and 28 assigned to the waitlist arm. The size of the theatre groups in the TET arm ranged from 4 to 18, with a mean and median group size of 11. The mean attendance rate for the TET program was 69%, with 35 (66%) participants completing at least 15 of 20 sessions. Eleven participants attended fewer than 3 sessions, and 4 attended no sessions.

The overall dropout rate at the postprogram evaluation was 15%, with rates of 21% in the waitlist arm and 11% in the TET arm. Among the 6 participants lost to follow-up in the TET arm, 2 did not attend any sessions, 3 attended fewer than 75% of the sessions (5–12), and 1 attended 18 sessions. At the 3-month follow-up evaluation, the overall dropout rate increased to 26%, with rates of 21% in the waitlist arm and 28% in the TET arm (see Figure 1). There was a significant relationship between the proportion of TET participants who attended fewer than 75% of the sessions and the proportion of those who dropped out at the postprogram evaluation (X² (1,53) = 7.35, p = .007) and at the 3-month follow-up evaluation (X² (1,53) = 9.98, p = .002).

Figure 1.

CONSORT flow diagram showing screening, randomization, and 3-month follow-up.

Main Findings

Descriptive statistics of the outcome measures at baseline, postprogram, and 3-momth follow-up are shown in Table 2. The Quade nonparametric ANCOVA revealed a significant difference in static balance scores (p = .005) and fall incidents (p = .046) between the TET and control arms after adjusting for the baseline scores as covariates (see Table 3). In the TET arm, balance scores increased from 2.51 at baseline to 2.75 at postprogram (z = 1.79, p = .07), whereas in the waitlist arm, these scores decreased from 2.93 to 2.54 (z = 1.91, p = .06). Fall incidents in the TET arm significantly decreased from 0.57 at baseline to 0.28 at postprogram (z = 2.41, p = .02), while in the waitlist arm, they increased slightly from 0.71 to 0.75 (z = 0.26, p = .79). Comparisons between baseline and postprogram in the SF-36 (mental or physical component), timed up-and-go test, 30-second sit-to-stand test, and hair cortisol levels showed no significant changes after completion of the TET program. Comparisons between the TET and waitlist arms at the 3-month follow-up evaluation revealed no significant differences in any outcome measures (p > .05).

Table 2.

Descriptive Statistics at Baseline, Postprogram, and 3-Month Follow-Up

Outcome measure	Theatre Experience Training Arm (N = 53)									Waitlist Control Arm (N = 28)
	Baseline			Postprogram			3-Month follow-up			Baseline			Postprogram			3-Month follow-up
	Mean	SD	Median	Mean	SD	Median	Mean	SD	Median	Mean	SD	Median	Mean	SD	Median	Mean	SD	Median
Static balance	2.51	1.30	2.00	2.75	1.21	3.00	2.98	1.29	4.00	2.93	1.30	3.00	2.54	1.26	2.00	3.07	1.18	4.00
Fall incidents	0.57	0.93	0.00	0.28	0.66	0.00	0.36	0.76	0.00	0.71	1.27	0.00	0.75	1.24	0.00	0.36	0.68	0.00
TUG (sec)	14.98	12.64	13.00	14.18	8.77	13.00	13.57	5.68	13.00	11.91	4.23	12.00	12.36	4.13	12.00	12.29	3.40	11.50
30CST	8.38	5.21	10.00	8.23	5.73	9.00	7.25	5.45	8.00	9.11	5.01	9.50	9.57	4.95	10.00	8.93	4.22	8.00
SF-36 MCS	69.57	20.92	71.75	68.08	19.92	74.79	67.54	20.72	70.38	69.30	21.87	73.11	66.17	22.50	70.84	67.10	19.03	67.57
SF-36 PCS	55.59	23.87	54.38	54.65	21.95	55.63	57.87	24.18	58.75	51.39	26.15	46.88	55.45	25.59	63.44	56.97	19.30	56.57
Hair cortisol level (pg/mg)	47.09	78.75	3.30	32.12	61.31	2.55	21.45	52.25	1.08	59.39	83.11	9.25	22.00	43.80	3.04	24.62	52.15	1.86

Note: TUG = 8-foot timed up-and-go test; 30CST = 30-second chair stand test or 30-second sit-to-stand test; SF-36 = 36-item Short-Form Health Survey; MCS = mental component summary; PCS = physical component summary.

Table 3.

Results From Quade Nonparametric Analysis of Covariance (Quade’s ANCOVA) (N = 81)

Outcome measures	Theater (n = 53)	Control (n = 28)	F (1, 79)	p-value
	Mean difference§ (95% CI)	Mean difference§ (95% CI)
Static balance	4.01 (−0.47, 8.49)	−7.59 (−14.90, −0.27)	8.337	.005
Fall incidents‡	−2.67 (−7.07, 1.72)	5.06 (−1.55, 11.67)	4.113	.046
TUG (sec)	−0.79 (−5.41, 3.83)	1.49 (−3.66, 6.64)	0.388	.536
30CST	−0.29 (−3.31, 2.73)	0.55 (−3.05, 4.14)	0.118	.733
SF-36 PCS	−2.18 (−5.94, 1.58)	4.12 (−2.80, 11.05)	3.142	.080
SF-36 MCS	−0.02 (−4.97, 4.93)	0.04 (−5.87, 5.95)	0.000	.988
Hair cortisol level (pg/mg)	12.73 (−393.29, 418.75)	−24.79 (−499.15, 450.24)	0.013	.908

Notes: TUG = 8-foot timed up-and-go test; 30CST = 30-second chair stand test or 30-second sit-to-stand test; SF-36 = 36-item Short-Form Health Survey; MCS = mental component summary; PCS = physical component summary. The F test evaluates the effect of the original randomized group assignment. This test is based on the linearly independent pairwise comparisons among the estimated marginal means.

^‡Controlled for baseline fall score and balance score.

^§Mean difference = mean differences of the unstandardized residuals from ranking analysis.

Post-hoc Analysis

Within the TET arm, a slight improvement in static balance and a nonsignificant increase in fall incidents was observed between the postprogram and 3-month follow-up assessments (p > .05). However, significant improvements were observed in the waitlist arm between the postprogram and 3-month follow-up evaluations, with an increase in static balance (z = 2.14, p = .03) and a reduction in fall incidents (z = 2.07, p = .04). These improvements in the waitlist arm were consistent across all cohorts. At the same time, minimal changes in the waitlist arm were noted in other variables, such as the timed up-and-go test or the physical component of the SF-36, between the postprogram and 3-month follow-up (p > .05). Cognitive function, which was not a planned outcome measure, showed no significant changes as measured by the Short Portable Mental Status Questionnaire across the study’s time points.

Discussion

This was the first study to investigate the effects of an innovative 10-week TET program on the physical functioning, HR-QoL, and emotional well-being of SSH residents using an RCT design. The TET program, based on Stanislavski’s acting principles, significantly improved static balance and reduced fall incidents at the postprogram evaluation. However, these significant differences were not maintained at 3-month follow-up. The program did not significantly affect emotional stress, HR-QoL, functional mobility, or lower body strength.

The improvements in static balance likely contributed to the reduction in falls. The TET program’s structured activities, which included physical warmup and acting exercises, may have enhanced participants’ balance and coordination. The TET program appeared to have a carry-over effect, as evidenced by the significant improvement in static balance at 3-month follow-up compared with baseline (p = .004), which occurred in the absence of continued theatre training. However, the unexpected improvements in the waitlist arm between the postprogram and 3-month follow-up evaluations highlight the potential influence of external factors or the natural variability in balance and fall incidents among older adults. Although the timing of data collection and cohort location varied, this improvement was seen across all 5 cohorts. To our knowledge, no newly organized physical activity programs were introduced during the 3-month follow-up period that had not already been in place during the baseline to postprogram interval in any of the 5 cohorts. This currently unexplainable improvement in static balance and fall incidents in the waitlist arm led to no significant net effect of the TET on these 2 measures at the 3-month follow-up. Future studies should consider more rigorous control measures and larger sample sizes to clarify these findings.

Previous research indicates that balance training in older adults can improve static balance performance more quickly and with less training than dynamic balance, as measured by the timed up-and-go test (35,36). Furthermore, one study showed that a twice-weekly 20-week group exercise program improved static balance, but not dynamic balance, in this population (37). It has also been observed that older adults can achieve significant improvements in static balance even with light activity, such as that done in the TET program (38).

The number of fall incidents was based on recall alone. We did not ask residents in advance to use any system to track the number of falls they experienced in the past 3 months. Strategies for collecting a more valid and accurate number of falls should include providing residents with electronic devices that they can use to track fall incidents at the time they occur. In addition, to further validate the impact of TET on fall reduction, future studies should collect information about injurious falls that lead to physician or emergency department visits or hospitalization related to hip fracture.

Other research has shown that lower body strength among older adults improved significantly after 4 weeks of resistance training (35); however, the TET program in the current study did not include resistance exercises. The TET program’s lack of impact on the physical component of the SF-36 and lower body strength may be due to the relatively low intensity and duration of physical activity involved. Incorporating more physically demanding exercises or extending the program duration could potentially yield more significant improvements in these areas.

The mixed emotions (ie, anxiety, stress, and happiness) that occurred during theatre training in the present study could not be captured by analyzing hair cortisol levels at postprogram. Instead, hair cortisol levels could serve only as an indicator of stress accumulated across the study period, not a one-time snapshot, which may explain why there were minimal net changes in hair cortisol levels at postprogram.

Limitations

It is important to point out that the need for participants to walk twice a week from their apartments to the SSH common room where the TET was held may have contributed to their improved balance. Therefore, to isolate the effect of the TET program on improved balance, future studies should include an attention arm (ie, theatre appreciation) that is held in the same place as the TET. The theatre appreciation arm may focus on enriching understanding and appreciation of theatrical events through discussion but not the experience of acting and physical movement per se.

Another limitation of the present study was not being able to verify the accuracy of falls reported by the participants. Overall, one-third of the participants fell in the 3 months prior to participating in the TET program. According to several population and national survey studies (39,40), a third or more of those aged 75 years and older reported a fall in the previous year. Given that more than 45% of participants in the present study were aged 75 and older, the number of self-reported fall incidents seems within the trustworthy range.

Variables that were not controlled for in this study were the participation by a few participants (about 3) in a regular exercise program offered in the SSH or in a gym class outside the SSH. In addition, some may have participated in other in-house programs that included a movement component. Some of these programs were offered around the time of the TET program by other community organizations, a factor our research team could not control for and was not aware of when planning the implementation of the present study. To avoid potential confounders that affect physical functioning, future investigators may need to include more stringent study eligibility criteria, such as not planning to join other in-house programs with a movement component during the study period.

Finally, 12 hair samples from 9 participants at different study time points did not include enough cortisol for valid analysis, mainly because hair on the posterior vertex of these participants’ head was too sparse. In addition, out of 15 male participants, 5 were bald. As a result, no hair samples could be collected from these participants. This can introduce bias.

Conclusion

While prior research has highlighted the cognitive and socio-emotional benefits of TET for older adults, its impact on physical function is not well-explored. The innovative 10-week TET program was effective in improving static balance and reducing falls for residents of SSH communities. Some carry-over effect was observed at 3-month follow-up. However, we found no net benefit of the TET program on lower body strength, functional mobility/dynamic balance, stress (cortisol) levels, or overall HR-QoL. Incorporating Stanislavsky’s method of physical action in theatre training can provide SSH residents with additional benefits, such as improved static balance and reduced fall incidents, beyond the cognitive, social, and emotional well-being benefits of traditional theatre training. Most residents who attended the initial program information session enrolled in the TET program, with a 69% attendance rate. This indicates that this innovative TET program is a feasible intervention and was of interest to SSH residents. Continuing to offer a meaningful program such as TET in SSH may help residents maintain a lower rate of fall incidents.

Funding

This research was supported by the National Endowment for the Arts (Grant number: 1863445-38-20).

Conflict of Interest

No potential conflict of interest was reported by the authors.

Data Availability

This study was registered at ClinicalTrials.gov (https://clinicaltrials.gov/study/NCT04582370). The data that support the findings of this study are available from the corresponding author (H. Yuen) upon reasonable request.