Research Report Comparative Effects of 2 Aqua Exercise Programs on Physical Function, Balance, and Perceived Quality of Life in Older Adults With Osteoarthritis Alison L. Fisken, PhD1; Debra L. Waters, PhD2; Wayne A. Hing, PhD3,4; Michael Steele, PhD5,6; Justin W. Keogh, PhD3,1,7 ABSTRACT Background: Osteoarthritis (OA) is a degenerative joint disease, which affects a large number of older adults. Many older adults with OA are physically inactive, which can contribute to reduced functional capability, quality of life, and an increased risk of falls. Although hydrotherapy is often recommended for older adults with OA, less is known about aqua fitness (AF), a widely available form of aqua-based exercise. Purpose: To compare the effect of an AF program and a seated aqua-based exercise program on a range of functional measures and quality of life among older adults with OA. Methods: Thirty-five older adults with OA were allocated to an AF group or an active control group who performed seated exercises in warm water for 12 weeks. The primary outcome measure was the timed up-and-go (TUG) test; other measures included step test, sit-to-stand (STS) test, handgrip strength test, 400-m walk test, Arthritis Impact Measurement ScaleShort Form (AIMS2-SF), and Falls Efficacy Scale-International (FES-I). Results: FES-I scores improved significantly in the AF group compared with the control group (P = 0.04). Withingroup analysis indicated both groups significantly improved their 400-m walk time (P = 0.04) and that the AF group 1Human
Potential Centre, AUT University, New Zealand. of Preventive and Social Medicine, University of Otago, New Zealand. 3Faculty of Health Sciences and Medicine, Bond University, Australia. 4Health and Rehabilitation Research Institute, AUT University, New Zealand. 5Department of Mathematics and Computing, Universiti Brunei Darussalam, Brunei Darussalam. 6Graduate Research School, Griffith University, Australia. 7Faculty of Science, Health, Education and Engineering, University of the Sunshine Coast, Australia. The authors declare no conflicts of interest. Address correspondence to: Alison L. Fisken, PhD, Centre for Physical Activity and Nutrition Research, School of Sport and Recreation, AUT University, Private Bag 92006, Auckland 1142, New Zealand ([email protected]). Bernadette Williams-York was the Decision Editor. DOI: 10.1519/JPT.0000000000000019 2Department
significantly improved its step test right (P = 0.02) and left (P = 0.00) and the AIMS2-SF total score (P = 0.02). No significant change in TUG, STS, or handgrip strength was observed for either group. Conclusions: Aqua fitness may offer a number of positive functional and psychosocial benefits for older adults with OA, such as a reduced fear of falling and increased ability to perform everyday tasks. Key Words: aqua-based exercise, falls, functional capability, osteoarthritis, quality of life (J Geriatr Phys Ther 2015;38:17–27.)
INTRODUCTION Osteoarthritis (OA) is a degenerative joint disease, which affects many older adults. According to the World Health Organization, approximately 45% of women over the age of 65 years1 have OA. In New Zealand, over 30% of women and 20% of men aged 65 to 74 years have OA.2 Symptoms include joint pain and stiffness; the large weightbearing joints of the hips, knees, and spine are most often affected although joints in the hands are also a common site.3 Older adults with OA are less physically active than older adults without OA,4 and the majority of adults with all forms of arthritis are not sufficiently active to meet the public health guidelines relative to physical activity levels.5 Low levels of physical activity exacerbate age-related physiological changes such as loss of muscular strength and endurance as well as reduced balance ability, all of which contribute to an increase in fall risk and a reduced ability to perform activities of daily living.6,7 Consequently, older adults with OA are particularly vulnerable to age-related reductions in muscle strength, functional ability and balance, and related declines in independence and perceived quality of life. Aqua-based exercise is recommended for adults with OA by numerous organizations such as the American College of Rheumatology.8 Performing exercise in an aqua environment reduces joint loading, and water immersion has been associated with decreased pain symptoms because 17
Journal of GERIATRIC Physical Therapy Copyright © 2015 The Academy of Geriatric Physical Therapy of the American Physical Therapy Association. Unauthorized reproduction of this article is prohibited.
JGPT-D-13-00060R1_LR 17
13/12/14 2:12 AM
Research Report of increased sensory input and decreased joint compression.9 A number of studies investigating the potential health benefits of aqua-based exercise for older adults with OA have focused on hydrotherapy (HT)-based exercise programs10-12 rather than aqua fitness (AF) exercise classes, which are more commonly available in communitybased swimming pools. HT is typically an individualized, therapist-supervised program that focuses on strength and range of motion exercises,13 usually in water temperatures between 92 and 96°F.14 Although HT-based programs have been shown to improve physical function15 and reduce pain11,16 among adults with OA, HT classes are often expensive and not always widely available because of the requirement of specialized staff and facilities. AF classes take place at many community swimming pools, and classes consist of partial weight-bearing aerobic and strengthening exercises performed to music. In a review of literature, Rahmann17 noted that aquabased exercise is often recommended for people with OA, but that there is no evidence to establish whether a physiotherapist-supervised HT program is more effective than a generic aqua exercise program, such as AF, at improving strength and function. The purpose of this randomized controlled trial was to investigate the effects of an AF program—compared with a seated, HT-type program—on strength, function, balance, fear of falling, and perceived quality of life among older adults with OA.
METHODS Participants Volunteers aged 60 years and older with OA were recruited by advertisements on the Arthritis New Zealand website, in local community groups, general practitioners’ rooms, at the orthopedic ward at the local public hospital, and in the local newspaper. Interested participants were provided with an information sheet, which they were asked to read before signing informed consent. Participants with OA were included if they had radiological diagnosis of OA in the hips, knees, spine and/or hands; had current and chronic (>1 year) pain, and were able to obtain medical clearance to participate in the study. Exclusion criteria included joint replacement surgery in the past 12 weeks, physical therapy intervention in the preceding 12 weeks, current participation in an organized exercise program twice a week or more, inability to safely enter and exit pool, and cognitive ability of providing informed consent. Participants were randomly allocated using computer software to either the AF exercise group or the control group. Participants were assessed in the week preceding the first exercise classes and within a week of completing the 12-week exercise program. A 12-week intervention was selected as several previous aqua-based exercise interventions have also investigated a program of this length.15,18,19 The same assessor conducted measurements across all 18
assessment points. The assessor was not blinded to group allocation but was blinded to measurement data from prior assessments.
AF Intervention The AF program consisted of aerobic and strength-based exercises performed twice weekly for 12 weeks (see Appendixes 1 and 2). Each session lasted for approximately 45 to 60 minutes. Intensity and length of sessions increased progressively over the 12-week period. Music was used to motivate and aid synchronization of participants as well as pace the velocity of movements.20 The beats per minute (bpm) of the music tracks were used to help quantify the velocity of the movements, and the instructor encouraged increased movement range as the program progressed. An experienced aqua instructor supervised each session, with a maximum of 10 participants per group. Average water temperature was 87°F; exercises were performed at a depth of 3′11″ to 4′11″ depending on participants’ height, so that the water level was at chest height.
Control Group The rationale for having the control group attend weekly exercise classes was to facilitate social interaction with their peers, similar to that experienced in the AF class.21 This was considered important, as social interaction has been associated with maintenance of function among older adults.22 Seated exercises in warm water were chosen because warm water immersion and gentle movements may improve acute OA symptoms,14 which was considered to be an appropriate motivator to attend these classes. The control group attended a once-weekly HT-type exercise session, during which they performed a random selection of exercises from the Arthritis Foundation Arthritis Water Exercise DVD (see Appendix 3 for exercises) for 12 weeks. Each session lasted for approximately 35 to 40 minutes; there was no progression of exercises and all sessions were performed while seated. The emphasis was on range of motion and relaxation, and an experienced aqua instructor oversaw each session with a maximum of 10 participants per group. Average water temperature was 97.7°F.
Outcome Measures The primary outcome measure was the timed up-and-go (TUG) test, which has been shown to be a valid and reliable measure of functional mobility.11,23 The TUG has been widely used in previous studies investigating aqua-based exercise for older adults with OA.15,18,19 In addition to having a high correlation with lower extremity strength,24 it has been identified as a useful tool for identifying older adults who are at risk for falls.25 Participants were allowed one practice trial; thereafter, the best time from 2 timed trials was used for analysis. Secondary outcome measures included the 15-second step test, which was used as a measure of dynamic balance. The step test has been identified
Volume 38 • Number 1 • January-March 2015 Copyright © 2015 The Academy of Geriatric Physical Therapy of the American Physical Therapy Association. Unauthorized reproduction of this article is prohibited.
JGPT-D-13-00060R1_LR 18
13/12/14 2:12 AM
Research Report as a valid clinical test for dynamic balance with good testretest reliability26 and which can identify previous fallers from healthy older adults.27 Participants were instructed to stand unsupported with feet parallel and 5 cm in front of a 7.5-cm (2.95″) block. Participants were advised which leg to step with and instructed to place the whole foot onto the block and then return it fully down to the floor. This was repeated as many times as possible in 15 seconds. After a short break, participants changed legs and repeated the test with the opposite foot stepping. The 30-second sit-to-stand (STS) test is recommended by the Osteoarthritis Research Society International (OARSI) as a functional measure of lower body strength and balance for older adults with OA.28 The STS has been validated as a reliable measure of these modalities.29 The test was performed on a straight-backed, armless chair, 43 cm (16.93″) in height. Participants were instructed to sit in the chair with their arms crossed across the chest. On the word “go,” participants stood up and sat down as many times as possible in 30 seconds. Handgrip dynamometry has been identified as a useful screening tool for health outcomes in older adults30 and a valid and reliable measure of muscle strength.31 A Jamar dynamometer (0-200 lb, Sammons Preston Inc, Bolingbrook, Illinois) was used to measure grip strength using the methodology employed by Ranganathan et al.32 Participants were seated and were instructed to grip the device with their maximal effort. Grip strength in each hand was assessed 3 times; the highest value from each hand was used for statistical analysis. The 400-m walk test was chosen to assess functional limitations,33 walking endurance, and cardiorespiratory fitness.34 Although the OARSI recommend the 6-minute walk test as a performance outcome measure for older adults with OA,28 the 400-m walk test has been identified as a comparable test that may be easier to administer.35 Furthermore, Simonsick et al34 suggested that a predetermined distance, as given in the 400-m walk test, may be more motivating for older adults than a timed test. Participants were instructed to walk as quickly as possible over 400 m, walking back and forth between 2 cones set 25 m apart. Participants were informed that they may slow down, stop, and rest as necessary but should resume walking as soon as they were able. Time taken to complete 400 m was recorded and used for analysis. The Arthritis Impact Measurement Scale 2-Short Form (AIMS2-SF) questionnaire was chosen to assess perceived quality of life and pain levels of the participants. This questionnaire has been shown to be a valid and reliable tool to assess the quality of life and pain levels of patients with OA.36 As well as providing a disease-specific assessment of quality of life, the questionnaire is easily administered.37 The questionnaire contains 4 domains—physical, symptoms, affects and social—with scores from all domains also added together to give a total score. Fear of falling was assessed using the Falls Efficacy Scale-International (FES-I), which has been found to be a valid and reliable measure of fear of falling
and sensitive to change in older adults.38 This questionnaire is simple and easy to administer.39 Physical activity levels were assessed by the Rapid Assessment of Physical Activity (RAPA) questionnaire. The RAPA is considered an easy-to-use, valid measure of physical activity for use in clinical practice with older adults that is sensitive to change and allows categorization of physical activity levels.40 The questionnaire consists of questions relating to activity levels in 2 separate domains—aerobic exercise and strength and flexibility exercise—a higher score indicates a higher level of physical activity. Individuals’ physical activity levels were categorized as “sedentary,” “insufficiently active,” or “active,” similar to previous research.41 Participants were not excluded from the study if they changed their medication. However, they were asked to keep a log of arthritis-related medication taken throughout the study. Combined, these measures provide insight into the effects of an AF program on physical well-being as well as perceived quality of life and fear of falling, all of which contribute to ongoing independence among older adults with OA.
Data Analysis As each section of the AIMS2-SF questionnaire contains a Guttman Scale, items were scored separately and then the score for each section was standardized to a scale of 0 to 10 using standardized formula.37 After scoring for each dimension of the questionnaire, a total score was calculated by summing the scores from each section of the questionnaire. Raw scores from the FES-I were added together to give a total score. Raw scores from the RAPA were recoded to produce Likert scale values for analysis. A power analysis was conducted using the TUG as the primary outcome measure. This power analysis indicated that a sample size of 30 would give 80% power, with a 5% chance of type I error to detect significant differences between aqua exercise groups. Statistical analysis was initially performed on an intention-to-treat basis, using the last-observation-carriedforward methodology.42 Outcome measures were applied as dependent variables (TUG, step test, STS, handgrip strength, 400-m walk, AIM2-SF, FES-I, and RAPA) for each individual at a particular time point.10 Data were checked for normality and homogeneity of variance before analyses; baseline data were compared between groups with independent t tests. Repeatedmeasures analysis of variance was performed with the group (AF versus control) as the between-subject variable and time (0 and 12 weeks) as the within-subject variable. Because of high rates of dropout from baseline to completion of the intervention among the exercise intervention group (31.6%), further analysis was carried out solely for participants who completed the 12-week intervention, using baseline and postintervention scores. Repeated-measures 19
Journal of GERIATRIC Physical Therapy Copyright © 2015 The Academy of Geriatric Physical Therapy of the American Physical Therapy Association. Unauthorized reproduction of this article is prohibited.
JGPT-D-13-00060R1_LR 19
13/12/14 2:12 AM
Research Report analysis of variance was performed with the group (AF versus control) as the between-subject variable and time (0 and 12 weeks) as the within-subject variable. All statistical analyses were performed on SPSS, version 19.0, with significance set at P ≤ 0.05. To calculate effect size, Cohen’s d was calculated for each outcome measure. Standardized changes of
Potential Centre, AUT University, New Zealand. of Preventive and Social Medicine, University of Otago, New Zealand. 3Faculty of Health Sciences and Medicine, Bond University, Australia. 4Health and Rehabilitation Research Institute, AUT University, New Zealand. 5Department of Mathematics and Computing, Universiti Brunei Darussalam, Brunei Darussalam. 6Graduate Research School, Griffith University, Australia. 7Faculty of Science, Health, Education and Engineering, University of the Sunshine Coast, Australia. The authors declare no conflicts of interest. Address correspondence to: Alison L. Fisken, PhD, Centre for Physical Activity and Nutrition Research, School of Sport and Recreation, AUT University, Private Bag 92006, Auckland 1142, New Zealand ([email protected]). Bernadette Williams-York was the Decision Editor. DOI: 10.1519/JPT.0000000000000019 2Department
significantly improved its step test right (P = 0.02) and left (P = 0.00) and the AIMS2-SF total score (P = 0.02). No significant change in TUG, STS, or handgrip strength was observed for either group. Conclusions: Aqua fitness may offer a number of positive functional and psychosocial benefits for older adults with OA, such as a reduced fear of falling and increased ability to perform everyday tasks. Key Words: aqua-based exercise, falls, functional capability, osteoarthritis, quality of life (J Geriatr Phys Ther 2015;38:17–27.)
INTRODUCTION Osteoarthritis (OA) is a degenerative joint disease, which affects many older adults. According to the World Health Organization, approximately 45% of women over the age of 65 years1 have OA. In New Zealand, over 30% of women and 20% of men aged 65 to 74 years have OA.2 Symptoms include joint pain and stiffness; the large weightbearing joints of the hips, knees, and spine are most often affected although joints in the hands are also a common site.3 Older adults with OA are less physically active than older adults without OA,4 and the majority of adults with all forms of arthritis are not sufficiently active to meet the public health guidelines relative to physical activity levels.5 Low levels of physical activity exacerbate age-related physiological changes such as loss of muscular strength and endurance as well as reduced balance ability, all of which contribute to an increase in fall risk and a reduced ability to perform activities of daily living.6,7 Consequently, older adults with OA are particularly vulnerable to age-related reductions in muscle strength, functional ability and balance, and related declines in independence and perceived quality of life. Aqua-based exercise is recommended for adults with OA by numerous organizations such as the American College of Rheumatology.8 Performing exercise in an aqua environment reduces joint loading, and water immersion has been associated with decreased pain symptoms because 17
Journal of GERIATRIC Physical Therapy Copyright © 2015 The Academy of Geriatric Physical Therapy of the American Physical Therapy Association. Unauthorized reproduction of this article is prohibited.
JGPT-D-13-00060R1_LR 17
13/12/14 2:12 AM
Research Report of increased sensory input and decreased joint compression.9 A number of studies investigating the potential health benefits of aqua-based exercise for older adults with OA have focused on hydrotherapy (HT)-based exercise programs10-12 rather than aqua fitness (AF) exercise classes, which are more commonly available in communitybased swimming pools. HT is typically an individualized, therapist-supervised program that focuses on strength and range of motion exercises,13 usually in water temperatures between 92 and 96°F.14 Although HT-based programs have been shown to improve physical function15 and reduce pain11,16 among adults with OA, HT classes are often expensive and not always widely available because of the requirement of specialized staff and facilities. AF classes take place at many community swimming pools, and classes consist of partial weight-bearing aerobic and strengthening exercises performed to music. In a review of literature, Rahmann17 noted that aquabased exercise is often recommended for people with OA, but that there is no evidence to establish whether a physiotherapist-supervised HT program is more effective than a generic aqua exercise program, such as AF, at improving strength and function. The purpose of this randomized controlled trial was to investigate the effects of an AF program—compared with a seated, HT-type program—on strength, function, balance, fear of falling, and perceived quality of life among older adults with OA.
METHODS Participants Volunteers aged 60 years and older with OA were recruited by advertisements on the Arthritis New Zealand website, in local community groups, general practitioners’ rooms, at the orthopedic ward at the local public hospital, and in the local newspaper. Interested participants were provided with an information sheet, which they were asked to read before signing informed consent. Participants with OA were included if they had radiological diagnosis of OA in the hips, knees, spine and/or hands; had current and chronic (>1 year) pain, and were able to obtain medical clearance to participate in the study. Exclusion criteria included joint replacement surgery in the past 12 weeks, physical therapy intervention in the preceding 12 weeks, current participation in an organized exercise program twice a week or more, inability to safely enter and exit pool, and cognitive ability of providing informed consent. Participants were randomly allocated using computer software to either the AF exercise group or the control group. Participants were assessed in the week preceding the first exercise classes and within a week of completing the 12-week exercise program. A 12-week intervention was selected as several previous aqua-based exercise interventions have also investigated a program of this length.15,18,19 The same assessor conducted measurements across all 18
assessment points. The assessor was not blinded to group allocation but was blinded to measurement data from prior assessments.
AF Intervention The AF program consisted of aerobic and strength-based exercises performed twice weekly for 12 weeks (see Appendixes 1 and 2). Each session lasted for approximately 45 to 60 minutes. Intensity and length of sessions increased progressively over the 12-week period. Music was used to motivate and aid synchronization of participants as well as pace the velocity of movements.20 The beats per minute (bpm) of the music tracks were used to help quantify the velocity of the movements, and the instructor encouraged increased movement range as the program progressed. An experienced aqua instructor supervised each session, with a maximum of 10 participants per group. Average water temperature was 87°F; exercises were performed at a depth of 3′11″ to 4′11″ depending on participants’ height, so that the water level was at chest height.
Control Group The rationale for having the control group attend weekly exercise classes was to facilitate social interaction with their peers, similar to that experienced in the AF class.21 This was considered important, as social interaction has been associated with maintenance of function among older adults.22 Seated exercises in warm water were chosen because warm water immersion and gentle movements may improve acute OA symptoms,14 which was considered to be an appropriate motivator to attend these classes. The control group attended a once-weekly HT-type exercise session, during which they performed a random selection of exercises from the Arthritis Foundation Arthritis Water Exercise DVD (see Appendix 3 for exercises) for 12 weeks. Each session lasted for approximately 35 to 40 minutes; there was no progression of exercises and all sessions were performed while seated. The emphasis was on range of motion and relaxation, and an experienced aqua instructor oversaw each session with a maximum of 10 participants per group. Average water temperature was 97.7°F.
Outcome Measures The primary outcome measure was the timed up-and-go (TUG) test, which has been shown to be a valid and reliable measure of functional mobility.11,23 The TUG has been widely used in previous studies investigating aqua-based exercise for older adults with OA.15,18,19 In addition to having a high correlation with lower extremity strength,24 it has been identified as a useful tool for identifying older adults who are at risk for falls.25 Participants were allowed one practice trial; thereafter, the best time from 2 timed trials was used for analysis. Secondary outcome measures included the 15-second step test, which was used as a measure of dynamic balance. The step test has been identified
Volume 38 • Number 1 • January-March 2015 Copyright © 2015 The Academy of Geriatric Physical Therapy of the American Physical Therapy Association. Unauthorized reproduction of this article is prohibited.
JGPT-D-13-00060R1_LR 18
13/12/14 2:12 AM
Research Report as a valid clinical test for dynamic balance with good testretest reliability26 and which can identify previous fallers from healthy older adults.27 Participants were instructed to stand unsupported with feet parallel and 5 cm in front of a 7.5-cm (2.95″) block. Participants were advised which leg to step with and instructed to place the whole foot onto the block and then return it fully down to the floor. This was repeated as many times as possible in 15 seconds. After a short break, participants changed legs and repeated the test with the opposite foot stepping. The 30-second sit-to-stand (STS) test is recommended by the Osteoarthritis Research Society International (OARSI) as a functional measure of lower body strength and balance for older adults with OA.28 The STS has been validated as a reliable measure of these modalities.29 The test was performed on a straight-backed, armless chair, 43 cm (16.93″) in height. Participants were instructed to sit in the chair with their arms crossed across the chest. On the word “go,” participants stood up and sat down as many times as possible in 30 seconds. Handgrip dynamometry has been identified as a useful screening tool for health outcomes in older adults30 and a valid and reliable measure of muscle strength.31 A Jamar dynamometer (0-200 lb, Sammons Preston Inc, Bolingbrook, Illinois) was used to measure grip strength using the methodology employed by Ranganathan et al.32 Participants were seated and were instructed to grip the device with their maximal effort. Grip strength in each hand was assessed 3 times; the highest value from each hand was used for statistical analysis. The 400-m walk test was chosen to assess functional limitations,33 walking endurance, and cardiorespiratory fitness.34 Although the OARSI recommend the 6-minute walk test as a performance outcome measure for older adults with OA,28 the 400-m walk test has been identified as a comparable test that may be easier to administer.35 Furthermore, Simonsick et al34 suggested that a predetermined distance, as given in the 400-m walk test, may be more motivating for older adults than a timed test. Participants were instructed to walk as quickly as possible over 400 m, walking back and forth between 2 cones set 25 m apart. Participants were informed that they may slow down, stop, and rest as necessary but should resume walking as soon as they were able. Time taken to complete 400 m was recorded and used for analysis. The Arthritis Impact Measurement Scale 2-Short Form (AIMS2-SF) questionnaire was chosen to assess perceived quality of life and pain levels of the participants. This questionnaire has been shown to be a valid and reliable tool to assess the quality of life and pain levels of patients with OA.36 As well as providing a disease-specific assessment of quality of life, the questionnaire is easily administered.37 The questionnaire contains 4 domains—physical, symptoms, affects and social—with scores from all domains also added together to give a total score. Fear of falling was assessed using the Falls Efficacy Scale-International (FES-I), which has been found to be a valid and reliable measure of fear of falling
and sensitive to change in older adults.38 This questionnaire is simple and easy to administer.39 Physical activity levels were assessed by the Rapid Assessment of Physical Activity (RAPA) questionnaire. The RAPA is considered an easy-to-use, valid measure of physical activity for use in clinical practice with older adults that is sensitive to change and allows categorization of physical activity levels.40 The questionnaire consists of questions relating to activity levels in 2 separate domains—aerobic exercise and strength and flexibility exercise—a higher score indicates a higher level of physical activity. Individuals’ physical activity levels were categorized as “sedentary,” “insufficiently active,” or “active,” similar to previous research.41 Participants were not excluded from the study if they changed their medication. However, they were asked to keep a log of arthritis-related medication taken throughout the study. Combined, these measures provide insight into the effects of an AF program on physical well-being as well as perceived quality of life and fear of falling, all of which contribute to ongoing independence among older adults with OA.
Data Analysis As each section of the AIMS2-SF questionnaire contains a Guttman Scale, items were scored separately and then the score for each section was standardized to a scale of 0 to 10 using standardized formula.37 After scoring for each dimension of the questionnaire, a total score was calculated by summing the scores from each section of the questionnaire. Raw scores from the FES-I were added together to give a total score. Raw scores from the RAPA were recoded to produce Likert scale values for analysis. A power analysis was conducted using the TUG as the primary outcome measure. This power analysis indicated that a sample size of 30 would give 80% power, with a 5% chance of type I error to detect significant differences between aqua exercise groups. Statistical analysis was initially performed on an intention-to-treat basis, using the last-observation-carriedforward methodology.42 Outcome measures were applied as dependent variables (TUG, step test, STS, handgrip strength, 400-m walk, AIM2-SF, FES-I, and RAPA) for each individual at a particular time point.10 Data were checked for normality and homogeneity of variance before analyses; baseline data were compared between groups with independent t tests. Repeatedmeasures analysis of variance was performed with the group (AF versus control) as the between-subject variable and time (0 and 12 weeks) as the within-subject variable. Because of high rates of dropout from baseline to completion of the intervention among the exercise intervention group (31.6%), further analysis was carried out solely for participants who completed the 12-week intervention, using baseline and postintervention scores. Repeated-measures 19
Journal of GERIATRIC Physical Therapy Copyright © 2015 The Academy of Geriatric Physical Therapy of the American Physical Therapy Association. Unauthorized reproduction of this article is prohibited.
JGPT-D-13-00060R1_LR 19
13/12/14 2:12 AM
Research Report analysis of variance was performed with the group (AF versus control) as the between-subject variable and time (0 and 12 weeks) as the within-subject variable. All statistical analyses were performed on SPSS, version 19.0, with significance set at P ≤ 0.05. To calculate effect size, Cohen’s d was calculated for each outcome measure. Standardized changes of
