Effects of a Community-Based Progressive Resistance Training Program on Muscle Performance and Physical Function in Adults With Down Syndrome: A Randomized Controlled Trial

Research Design 

We conducted a repeated-measures RCT. Adults with DS were randomly allocated to either an intervention group that received 10 weeks of progressive resistance training or a control group that continued with their usual activities. Both groups were assessed at baseline and after the intervention phase of the study. The trial received ethics approval from the university ethics committee, and all participants and their carers gave written informed consent to take part in the trial.

Participants 

We recruited adults with DS through 2 disability agencies (1 rural, 1 metropolitan) that provide supported employment and day activity programs for adults with intellectual disabilities. The agencies mailed a flyer promoting the trial to a total of 46 families and those who were interested in taking part contacted the researchers directly. Participants were included if they were aged 18 years or more, had the ability to follow simple verbal instructions in English, and were fit and well enough to participate in a progressive resistance training program. The latter was determined by asking the carer of each participant to complete a copy of the 7-item Physical Activity Readiness Questionnaire.14 The level of intellectual disability of each of the participants (described as mild, moderate, or severe as perceived by their parent or carer) was also documented. The exclusion criterion was participation in a strength training program in the 6 months prior to the start of the study. Based on a power analysis of the training effects on lower-limb muscle strength found in people with DS in a previous study (d=1.26),12 it was calculated that 10 subjects would be needed in each group to obtain power of 80%.

Randomization 

Participants were randomly allocated to either the intervention or control group using a concealed allocation, block randomization method that ensured approximately equal numbers in each group with the relatively small sample.15 We considered participants in blocks of 4, with 6 ways or permutations in which to get 2 intervention and 2 control participants in each block. The order of the blocks was generated from a random number table and assignments sealed in sequentially numbered opaque envelopes. Only after the recruiter determined if a person was eligible for the study and they had agreed to participate was random assignment made. Participants were assigned to their allocation by a physiotherapist not involved in the study after the baseline measures were taken.

Intervention 

Participants in the intervention group completed a 10-week progressive resistance training program. Training was completed twice a week at a community gymnasium and followed the principles of progressive resistance training as recommended by the ACSM.16 The program consisted of 6 exercises using weight machines: 3 for the upper limbs (shoulder press, seated chest press, seated row) and 3 for the lower limbs (seated leg press, knee extension, seated calf raise). These exercises formed the basis of the training program but could be modified to suit the needs of the individual. For example, if a participant found it difficult to do the seated calf raise exercise, the trainer could try modifying the exercise as a standing calf raise exercise. Participants completed 2 to 3 sets of between 10 to 12 repetitions of each exercise until they reached fatigue. A 2-minute rest period was given between each set to allow for recovery, and the resistance was increased when 2 sets of 12 repetitions of an exercise could be completed.16 A log book for each participant that detailed each exercise, the weight lifted, the number of repetitions, and number of sets was filled at each session by the exercise trainer. Participants completed the program as a group, supervised by 2 accredited fitness trainers. Each trainer supervised the training of a subgroup of 2 to 3 participants. Exercising in a small subgroup allowed for a close level of supervision.

Participants in the control group continued with their typical daily activities, which included employment, leisure, and sporting activities, but did not include a progressive resistance training program. For ethical reasons, after the 10-week assessment, participants in the control group were invited to complete the strengthening program, but no further follow-up measurements were conducted.

Outcome Measurements 

We assessed all participants at baseline (week 0) and immediately after the intervention period (week 10) at the recreation center where the intervention took place. The outcome measurements were taken by assessors who were blind to group allocation and who had no involvement in the recruitment, the randomization, or the training of participants. There were 2 assessors: 1 assessor completed the muscle performance tests and the second assessor completed the physical function tests. Assessors received training on how to administer the outcome measures.

Muscle performance was assessed using tests for maximal force generation (1-RM) and muscle endurance. Maximal muscle force generation was tested by establishing the amount of weight each participant could lift in a single seated chest press and seated leg press (1-RM). A single 1-RM chest press and leg press have been measured with high levels of retest reliability (r>.89) and no systematic change when measured over 3 weeks in adults with neurologic impairment.17 Muscle endurance was measured by counting the number of repetitions that could be completed when the weight on the seated chest press and seated leg press was lowered to 50% of 1-RM. The baseline weight that was lifted was used for the 10-week measurement.18 There is evidence of construct validity for this measure because hypothesized changes were detected in a group of adults with neurologic impairment.17

Physical function was measured using 2 outcomes measures; the timed stairs test and the grocery shelving task. For the timed up and down stairs test, participants stood at the bottom of a flight of stairs and were asked to ascend and descend the steps as quickly and as safely as possible. Participants were allowed to choose any method of traversing the stairs including alternating steps, running up the stairs, or skipping steps. There were 2 handrails available on the stairs and participants were allowed to use these. The participants were given the cues, ready and go, and the time taken to go up the stairs, turn around, come all the way down until both feet landed at the foot of the stairs was recorded in seconds using a stopwatch. The test was repeated twice with the fastest time used in the analysis.19 The test has been measured reliably (r=.89) in a population of adults with multiple sclerosis.20

We used the grocery shelving task21, 22, 23 to measure upper-body functional activity. For this test, participants were asked to stand up from a chair and carry 2 grocery bags to a bench 2m way. Each grocery bag contained 10 items; each item weighed 410g (total weight of each grocery bag, 4.1kg). The participants then took the items out of the bag and stacked them onto a shelf at shoulder height. Participants were asked to complete the task as quickly as possible and the time taken to complete the test was measured. All the participants were given 1 practice trial at each assessment before they completed 2 timed tests, the average of which was used in the analysis. The grocery shelving task has shown high retest reliability (intraclass correlation coefficient, .98) when tested over 6 weeks in a group of adults with respiratory disability,21 but has not previously been applied to adults with intellectual disability.

Adverse events that occurred during training (including minor events such as delayed onset muscle soreness) were systematically recorded by the fitness trainers on the participant's exercise log book. At the start and at the end of each exercise session the trainer asked each participant if they had experienced any injuries or any other problems.

Data Analysis 

Data were analyzed using SPSS statistical softwarea to determine if there were any significant baseline demographic (sex, age, weight) differences between the intervention and control groups using the independent t test and Fisher exact probability test, as appropriate. Outcomes were analyzed using analysis of covariance on the change scores with the baseline measure of that variable used as the covariate.24 Where baseline variables were significantly different, they too were entered as covariates. The mean difference within each group and the mean difference between the groups and the 95% CIs of the mean differences were also calculated. Effect sizes and 95% CIs were also calculated for the change scores. To avoid bias and to maximize the randomization process, we used intention-to-treat analysis. Where data were missing, the carry forward technique was used, which assumes that missing data remain constant.25

Flow of the Participants Through the Trial 

Twenty adults (13 men, 7 women) with DS volunteered to take part in the study. Participants ranged in age from 20 to 49 years, with a mean age ± SD of 26.8±7.8 years. The mean height of the group was 155.1±9.3cm (range, 134–169cm) and the mean weight was 68.9±13.3kg (range, 51.8–102.0kg). Nine participants were randomly allocated to the intervention group and 11 participants to the control group and their progress through the trial is shown in figure 1. The demographic details of the participants for each group are displayed in table 1 and as the table shows, the only significant difference at baseline was that the intervention group weighed more than the control group. Eight of the 20 participants were employed at least 1 day a week performing manual type work that required them to pack boxes of confectionery, sorting and cutting clothing, and assembling automotive parts.

View Large Image Download to PowerPoint

Fig 1. Number of participants at each stage of the trial (enrollment, allocation, follow-up, analysis).

Table 1. Demographic Data for Intervention and Control Groups

	Characteristic	Intervention (n=9)	Control (n=11)	Statistical Test	P
	Mean age ± SD (y)	25.8±5.4	27.6±9.5	−0.52	0.61
	Sex (male/female)	7/2	6/5	NA	0.37
	Mean height ± SD (cm)	158.8±7.12	152.0±10.0	1.7	0.11
	Mean weight ± SD (kg)	78.4±13.5	61.2±6.7	3.5	0.01
	Mean BMI ± SD (kg/m2)⁎	31.2±6.0	26.8±5.2	1.8	0.09
	Level of perceived ID			NA
	Mild	2	2		0.63
	Moderate or severe	7	9
	Living arrangements
	With parents	8	8	NA	0.38
	SSA or independent	1	3
	Employed	4	4	NA	1.00

Abbreviations: BMI, body mass index; ID, intellectual disability; NA, not applicable; SSA, shared supported accommodation.

⁎ Referent BMI is 18.5–24.9kg/m2, overweight is 25–29.9kg/m2, obese is ≥30kg/m2.30

Compliance With the Trial Method 

There were no withdrawals from the study. Participants attended 167 (92.8%) of 180 scheduled training sessions. Missed sessions of training were due to illness unrelated to the training program or to vacation time. None of the sessions were missed due to soreness or injury due to the training program. No serious adverse events for the intervention were noted. Four participants complained of mild muscle soreness as a result of training, most reports occurring in the early weeks of training, and recovered spontaneously. No other adverse events were recorded.

Over the course of the training program, the intervention group increased the amount of weight lifted during training in the 6 exercises by at least 90% of the initial training resistance. To assess the success of assessor masking, assessors were asked to guess whether a participant had been allocated to the intervention or the control group. Assessors correctly guessed group allocation 56.8% of the time (Fisher exact test, P=.74).

Effect of the Intervention 

The intervention group had a statistically significant improvement in upper-limb muscle endurance compared with the control group (mean difference in the number of repetitions of the chest press at 50% of 1-RM, 16.7; 95% CI, 7.1–26.2; P<.01). There were also trends toward improvement in upper-limb muscle strength (mean difference in chest press 1-RM, 8.6kg; 95% CI −1.3 to 18.5kg; P=.08) and upper-limb function (mean difference in time to complete the grocery shelving task, −20.3s; 95% CI, −45.7 to 5.2s; P=.11) that favored the intervention group (table 2). There were no significant differences between the groups for lower-limb muscle strength, lower-limb muscle endurance, or lower-limb physical function (see table 2).

Table 2. Mean Score, Mean Difference Within Groups, and Mean (95% CI) Difference Between Groups for All Outcomes for the Intervention Group and the Control Group

	Outcome	Score				Difference Within Groups		Difference Between Groups
		Baseline (week 0)		Postintervention (week 10)		Week 10–Week 0		Week 10–Week 0⁎	P⁎	Effect Size (95% CI)
		Int	Con	Int	Con	Int	Con	Int–Con (95% CI)
	Chest press 1-RM (kg)	35.9±15.4	28.0±10.2	44.9±15.2	31.6±13.3	9.0±8.2	3.6±7.1	8.6(−1.3to18.5)	.08	0.68(−0.20to1.59)
	Leg press 1-RM (kg)	80.8±25.9	70.8±20.0	96.2±31.6	82.2±19.7	15.4±20.7	11.4±12.0	−1.3(−22.5to20.0)	.90	0.23(−0.65to1.12)
	Chest press endurance (no. of repetitions)	15.0±4.7	19.0±8.1	25.9±8.3	17.5±9.5	10.9±8.6	−1.6±8.0	16.7(7.1to26.2)	.002	1.51(0.46to2.44)
	Leg press endurance (no. of repetitions)	44.1±25.9	44.5±21.3	46.8±37.1	49.4±27.6	2.7±47.4	4.9±21.2	16.5(−23.1to56.0)	.39	−0.06(−0.94to0.82)
	Timed up and down stairs test (s)	16.6±5.2	20.2±6.3	14.4±3.4	18.7±6.5	−2.1±2.7	−1.5±4.3	−3.4(−7.7to1.0)	.12	0.16(−0.73to1.04)
	Grocery shelving task (s)	85.1±49.1	122.8±84.0	67.5±33.4	110.7±66.4	−17.6±29.0	−12.0±23.2	−20.3(−45.7to5.2)	.11	0.22(−0.68to1.09)

NOTE. Values are mean ± SD or as otherwise indicated. Abbreviations: Con, control group; Int, intervention group.

⁎ Derived from analysis of covariance with dependent variable on admission and baseline weight as covariates.

Study Limitations 

This trial was limited by the relatively small sample size of 20 participants, which required the effects of the intervention to be large in order to detect any changes as a result of the strength training program. The duration and frequency of the program may also have been too short to result in large changes in muscle performance and functional activities. Future trials should investigate if longer duration training programs are indeed more beneficial.