Secondary Logo

Journal Logo

Research Reports

Effects of Corrective Exercise for Thoracic Hyperkyphosis on Posture, Balance, and Well-Being in Older Women: A Double-Blind, Group-Matched Design

Jang, Hyun-Jeong PT, PhD1; Hughes, Lynne C. PT, PhD1; Oh, Duck-Won PT, PhD2; Kim, Suhn-Yeop PT, PhD3

Author Information
Journal of Geriatric Physical Therapy: July/September 2019 - Volume 42 - Issue 3 - p E17-E27
doi: 10.1519/JPT.0000000000000146
  • Free



Hyperkyphosis related to aging is characterized by excessive forward curvature of the thoracic spine that can lead to significant deterioration in health status.1 Approximately 20% of older women have hyperkyphosis.2 Hyperkyphosis has been associated with increased mortality,2,3 poor balance,4 and low well-being scores.5 The underlying reasons for increased thoracic kyphosis (TK) with aging are unknown; however, a growing body of evidence suggests that vertebral degeneration such as wedge-shaped changes of the vertebral body and dehydration of the intervertebral disc are primary precipitating factors.6 The general deconditioning and muscle weakness, particularly of the trunk extensors, associated with aging may also contribute to postural malalignment of the spine and reduced segmental mobility.7

Older women with excessive kyphosis often have altered postural alignment and difficulties in using normal strategies for balance control during daily activities.8 Evidence shows that TK increases fall risk.9 Hyperkyphosis shifts the line of gravity in the thoracic spine further anterior than what is found in normal standing.10,11 This shift places greater loads on the anterior vertebral bodies as well as creating a forward shift in the center of gravity, thus pushing the limits of stability needed for functional activities.11–13 Balance and walking are challenged by this changed adaptation to load. Individuals with increased kyphosis most commonly use a hip strategy to maintain balance. They also often demonstrate an increase in sway.14 This change in overall physical performance can result in greater generalized fear of falling in daily activities as well as a decreased self-reported satisfaction in perceptions of health and family relationships.5 In addition to detrimental effects on balance and function, hyperkyphosis has been linked to adverse effects on general health and quality of life in the older women.5,8,9

Recently, a number of clinical trials have established therapeutic guidelines and protocols for the rehabilitation of older women with hyperkyphosis.9,15,16 These trials have typically focused on the use of spinal bracing,17 therapeutic taping,18 manual therapy,16 and therapeutic exercise for postural training.10 The focus of most therapeutic exercise protocols is to facilitate functional adaptations in dysfunctional muscle groups and to improve muscle strength, flexibility, endurance, coordination, and alignment of the spinal column.19,20 Research and clinical practice have shown some efficacy across approaches but no one approach has been shown to be superior.20,21

In addition, given that most studies have used multimodal treatments, it is difficult to draw conclusions about the most effective type of exercise. We propose that aberrant alignment and stiffness of the spinal column contribute to the natural progression of TK and, therefore, should be the target of treatment. A program that focuses on unilateral and bilateral mobility in diagonal patterns using upper limbs may promote more upright thoracic posture and mobility of the rib cage.

Evidence for a brief intensive exercise program, of 8 weeks or less, focusing on the postural correction of the thoracic spine and rib cage is limited. An earlier published study showed that thoracic exercises performed for 8 weeks were effective for older women with hyperkyphosis in improving kyphosis angle, forward head, and chest expansion.22 Accordingly, the aim of the current study was to examine the effectiveness of an 8-week corrective exercise program for thoracic hyperkyphosis (THK) on posture, balance, and well-being in older women with hyperkyphosis.



Fifty women 65 years of age and older were recruited from 2 senior welfare centers in Daejeon City. Participants were assigned to either the experimental group (EG) or the control group (CG) on the basis of convenience of location, 25 in each. All participants met the following inclusion criteria: (1) TK of greater than 40°23 using the dual inclinometer method,24 (2) ability to walk independently without assistive device, (3) no cognitive impairment as determined by a score of greater than 24 points on the Mini-Mental State Examination,7 and (4) no orthopedic, cardiopulmonary, and neurological disorders that would impede the ability to participate in the experimental procedure. The sample size calculations were performed with the G*Power software (version 3.1.7). The minimal sample size calculated was 36 for 90% statistical power, 5% significance level 2-tailed, and medium effect size. In anticipation of a participant dropout rate of 28%, the final target number was 50, 25 for each group. Prior to the beginning of this study, individuals were provided information about experimental procedure and safety and they gave written consent to participation. Ethical approval was granted by Daejeon University Institutional Review Board (Reference number: 1040647-201306-HR-021-01).

Study Design

This study was a double-blinded, group-matched clinical trial, in which the assessors who performed the measurements were blinded from group assignments (Figure 1). Participants were also blinded to group allocation. All participants consented to participation in a thoracic correction exercise program but were unaware of the differences in the application of the program. Two senior welfare centers were chosen in our region that were open to participation in our research program. One center was designated as the EG in which the participants performed the intervention. This EG consisted of 25 participants recruited according to selection criteria. Twenty-five participants age- and condition-matched were recruited from another center for the CG. Participants of the CG were paired within 5 years of the group mean age and 5° from group mean angle of thoracic kyphosis (ATK) in participants of the EG. Participants of the EG underwent a thoracic correction exercise for 8 weeks aimed at correcting THK. Participants of the CG were provided a booklet with a detailed description of the same exercise program to be performed as self-regulated exercise at home. Outcome measurements were performed 3 times: (1) before the intervention at baseline, (2) after the completion of the intervention period at 8 weeks (postintervention), and (3) during follow-up check 8 weeks after the end of the intervention. Six assessors, senior physical therapist students, were trained by the same therapist and carried out all measurements. These assessors participated only in data collection without knowledge of the treatment allocation.

Figure 1.
Figure 1.:
Flowchart of the study.


In the first session of the intervention, all participants in both groups learned to properly perform the exercise program. The CG received a booklet and checklist describing in detail the exercise program to guide the participants in self-performance of the home program. The CG group participated in supervised exercise on the first visit, performed their home program independently twice weekly, and returned for measurement at 8 weeks. The exercise was performed at the same frequency of twice per week; the EG participated in the supervised exercise and the CG exercised at home. Participants of the EG performed a thoracic corrective exercise program for 60 minutes, twice weekly for an 8-week period (a total of 16 sessions). The program consisted of exercise tasks to facilitate diaphragmatic breathing, thoracic mobility, thoracic stability, and awareness of thoracic alignment. See Appendix 1 for the complete exercise program. An elastic band (THERABAND, Akton, California) was used for the benefits of resistance exercise and the elastic tension was determined depending on the individual level of muscle performance. Color of elastic band utilized by participants was decided by therapist's observation and participant's reporting of level of ease. Most participants used yellow or red theraband. Exercise guidance was maintained by a physical therapist with a specialty in the field of senior exercise and 3 trained senior physical therapy students. Theraband exercise intensity was assessed using the 20-point Borg Scale, which is a rating of perceived exertion, with a goal range of 11 (light) to 13 (somewhat hard) and resistance or repetitions were adjusted for the individual.25 Participants were monitored for correct and controlled movement during exercise. Meanwhile, there was no intervention for participants of the CG except the initial performance of the exercise with education on the equivalent thoracic corrective exercise program to the program of the EG and a booklet to guide home exercise. Both the EG and the CG were to perform the exercise program twice weekly, the EG supervised and the CG unsupervised.


Condition of posture, balance, and well-being were assessed. Thoracic kyphotic posture was assessed using the ATK, the kyphosis index (KI), the thoracic stiffness index (TSI), and the forward head posture. The ATK was measured with the dual inclinometer (F00550, Acumar, Lafayette Instrument Company, Lafayette, Indiana), which consists of 2 inclinometers with 2 feet on the bottom face. Feet of each inclinometer were placed over spinous processes of T1 and T2 for one site and spinous processes of T12 and L1 for the second site. The ATK was calculated as a sum of angles collected from each inclinometer.24 Data were averaged over triplicate trials performed on the same day by the same assessor. This dual inclinometer method has been reported to have high interrater reliability (intraclass correlation coefficients [ICCs] ranging from 0.89 to 0.94) and validity.24 The KI was measured by using a flexicurve ruler, which was a flexible plastic 60-cm ruler to evaluate a surface contour of thoracic curvature. Flexicurve measurements were conducted in the same manner as described by Kado et al.3 Flexicurve measurements have been found to have concurrent validity and have high inter- and intrarater reliability (ICC = 0.91) to measure TK.26 The KI was calculated by multiplying 100 by the value of thoracic width divided by thoracic length or KI = (thoracic width/length) × 100.3 Furthermore, the TSI was calculated as the value of the KI in relaxed posture divided by the KI in best posture to demonstrate stiffness and elasticity of thoracic region.27 A value of greater than 1 implies more flexible TK, whereas value of less than 1 indicates less flexible TK relatively.27 Data were averaged over triplicate trials. Each participant was measured by the same assessor at the 3 time points on the same day. Forward head posture was measured by evaluating a craniovertebral angle (CVA) and tragus-to-wall distance (TWD) in standing against wall. The CVA is defined as angle between line connecting external acoustic pore to spinous process of C7 and horizontal line passing through the spinous process of C7.28 For correct measurement of the standing CVA, participants were instructed to focus on a yellow circle (diameter: 15 cm) marked on the wall 1 m in front of them and then a picture of the head and the neck was obtained in a side view. The photographic image was used to analyze CVA. The measure of CVA has been found to be a highly reliable measure in test-retest, interrater, and intrarater reliability.28,29 Also, TWD was measured by using a 30-cm ruler in comfortable standing against a wall. During measurement, a wooden block (width: 5 cm) was placed between participant's heel and wall to avoid the tilting of trunk. Data were averaged over triplicate trials. Tragus-to-wall distance has been reported to be reliable (for both intra- and intertester) and valid nonradiologic measure of forward flexed posture (ICC = 0.78-0.89).30

Balance was assessed using the Short Physical Performance Battery (SPPB) and limit of stability (LOS). The SPPB is a geriatric assessment tool comprising 3 subitems: repeated timed chair stands, timed balance tests, and a 4-m timed walk that corresponded to a score of 0 to 4.31,32 Total score was determined by a sum of scores obtained from each subitem, with higher score indicating higher level of performance.32,33 The total SPPB and each subitem have been reported to have excellent test-retest reliability (ICC = 0.89 and 0.83 in 2 samples) and construct validity.33 The LOS was measured by using a square force plate (length: 610 mm, width: 610 mm, and height: 60 mm) (HUR BT4, HURLABS, Tampere, Finland), which consisted of 4 pressure sensors connected to a computer system with software (Smart-suit Balance 1.4) for data analysis, to visually demonstrate the sway patterns in the center of pressure.34 Test-retest reliability of this force plate has been found to be high (ICC = 0.77-0.94) for assessing postural stability.35 The LOS was measured to clarify dynamic balance functions of participants.36 The LOS was defined as the greatest range over which participants can transfer their center of gravity from a midline upright standing within their base of support without falling, stepping, or reaching for support. To measure the LOS, participants were asked to stand over the plate with feet placed at shoulder width and arms at sides with eyes focused on a yellow circle (diameter: 15 cm) marked at wall 1 m in front of them. Participants were then instructed to lean their body in 4 directions (forward, backward, and to each side) as much as possible and held each position for 8 seconds. Measurement was carried out for 30 seconds in eye-open condition. Data were averaged over 2 trials.

Well-being was assessed using Geriatric Depression Scale Short Form—Korean version (GDSSF-K) and 36-Item Short Form Health Survey instrument (SF-36). The GDSSF-K was used to assess the level of depression of participants. The GDSSF-K is a 15-question assessment tool to identify the level of depression of older adults. A higher score indicated more severe depression.37 The GDSSF has been shown to have good internal consistency (Cronbach α = 0.81) and validity (Spearman ρ = 0.82).37 The GDSSF-K has also been shown to have good internal consistency (Cronbach α = 0.88).38 The SF-36 is a set of readily administered quality of life measures, consisting of 36 self-report questions with responses on a 5-point Likert scale.39 The SF-36 consists of 8 subscales: “Physical functioning,” “Role limitations due to physical health,” “Role limitations due to emotional problems,” “Energy/fatigue,” “Emotional well-being,” “Social functioning,” “Pain,” and “General health.” Data are calculated by adding scores of all subscales together, a higher score suggests higher quality of life. The SF-36 has been reported to be highly reliable for clinical use (Cronbach α > 0.85, reliability coefficient >0.75 for all dimensions except social functioning) and demonstrated construct validity.39 The SF-36 in the Korean language developed by Koh et al40 was used in this study. The Korean version of the SF-36 has been shown to have excellent reliability (Cronbach α = 0.94).40

Statistical Analysis

Statistical Package for the Social Sciences version 18.0 (SPSS Inc, Chicago, Illinois) was used to analyze data collected. Data were described as mean and standard deviation in this study. Homogeneity of data collected from participants in each group was tested by a Shapiro-Wilk test, ensuring equal variation and normality of data. Independent t tests were used to determine any differences between groups (participant characteristics) at the onset of the study.

For the main outcome of the study, a 2 × 3 repeated measures analysis of variance, with time factor (baseline, postintervention, and follow-up) and group factor (EG and CG), was carried out to identify the main effect and the interaction effect of each parameter. When statistical significance was found, post hoc pairwise comparison was performed with Bonferroni adjustment to determine where the differences occurred.

To compare the 2 groups, percent change from baseline to postintervention and from baseline to follow-up was calculated for each measure. Multiple independent t tests were conducted to appreciate significance between groups. Significance level was set at P < .05.

The minimal detectable change (MDC95%) for ATK was calculated on the basis of an assumed ICC of 0.95 as reported by Lewis and Valentine.41 The value of ICC was then used to determine the standard error of measurement (SEM = SDadmission ATK × √1 − ICC),42 where ATK is angle of TK, and the MDC95% (SEM × 1.96 × √2).


General Characteristics of Participants

See Table 1 for summarized general characteristics of participants. Final analysis included 22 participants in each group, 3 in the EG failed to attend regularly and 3 in the CG were excluded for nonattendance at posttesting. No significant differences were found in all participant characteristics of age, height, weight, and the score of the Mini-Mental State Examination—Korean version (P > .05).

Table 1. - General Characteristics of Participantsa
Characteristics Experimental Group (n = 22) Control Group (n = 22) t P
Age, y 74.6 (4.6) 76.8 (4.9) −1.51 .14
Height, cm 151.8 (5.5) 150.5 (7.5) 0.17 .87
Weight, kg 58.1 (7.6) 55.3 (7.2) 1.29 .20
MMSE-K, scores 25.7 (1.3) 25.6 (0.7) 0.29 .77
Abbreviation: MMSE-K, Mini-Mental State Examination—Korean version.
aData are presented as mean (SD).

Comparison of Means of Each Parameter Between Groups

There were no significant differences between the EG and the CG in all baseline data including ATK, relaxed- and best KI, TSI, CVA, TWD, SPPB, LOS, GDSSF-K, and SF-36 (P > .05). Data for each group were found to have a normal distribution and homogeneity of variance. Table 2 outlines results of the 2 × 3 repeated measures analysis of variance, showing the comparison of outcomes of each parameter in baseline, postintervention, and follow-up measurements between the EG and the CG. Main effects are statistically significant over time (baseline, postintervention, and follow-up) within group for all parameters, and between groups (EG and CG), the parameters of TSI and TWD showed statistical significance (P < .05). Also, there were significant time-by-group interactions in all parameters, showing that EG exhibited significant improvement over the CG (P < .05). In the results of post hoc tests, all parameters of the EG are significantly better in the postintervention and follow-up values than the baseline value. Limit of stability was the one exception in the EG follow-up values that did not show a significant difference over the baseline. On the contrary, participants of the CG showed significant differences in only the ATK, TWD, and TSI (P < .05). The ATK and the TSI showed higher values at postintervention than follow-up and the TWD showed higher value at postintervention than baseline in the post hoc test (P < .05).

Table 2. - Comparison of Means in Baseline, Postintervention, and Follow-up Data of Each Parameter Between the 2 Groupsa
Outcomes Experimental Group (n = 22) Control Group (n = 22) Main Effects (F) Interaction Effects (F)
Baseline Postintervention (8 wk) Follow-up (16 wk) Baseline Postintervention (8 wk) Follow-up (16 wk) Time Group Timeb Group
Kyphotic posture
Angle of thoracic kyphosis, ° 57.0 (2.9) 54.8 (2.8) 54.9 (2.7) 55.7 (4.9) 56.5 (5.5) 55.6 (4.8) 20.68c 0.10 17.28c
Kyphosis index
Relaxed 14.2 (1.3) 13.9 (1.3) 13.9 (1.3) 13.9 (1.2) 13.8 (1.3) 13.8 (1.2) 19.80c 0.14 15.50c
Best 13.7 (1.3) 13.2 (1.2) 13.4 (1.2) 13.5 (1.2) 13.5 (1.2) 13.4 (1.2) 27.44c 0.07 36.46c
Thoracic stiffness index 1.0 (0.0) 1.1 (0.0) 1.1 (0.0) 1.0 (0.0) 1.0 (0.0) 1.0 (0.0) 9.48c 13.06c 17.39c
Forward head posture
CVA, ° 43.5 (6.1) 45.6 (6.6) 45.9 (6.8) 45.0 (7.3) 45.2 (7.4) 45.3 (7.5) 10.75c 0.01 16.62c
TWD, cm 17.9 (1.9) 15.6 (1.9) 15.8 (1.9) 18.2 (2.2) 17.3 (2.5) 17.7 (2.1) 45.18c 4.44b 14.25c
SPPB, scores
Balance 3.0 (0.6) 3.7 (0.5) 3.5 (0.5) 3.3 (0.7) 3.1 (0.5) 3.1 (0.5) 5.65b 2.39 12.25c
Strength 2.3 (0.9) 2.7 (0.9) 2.7 (0.8) 2.4 (1.1) 2.2 (1.1) 2.4 (1.1) 2.36 0.43 4.34b
Gait 3.1 (0.7) 3.6 (0.4) 3.6 (0.5) 3.3 (0.6) 3.5 (0.6) 3.4 (0.6) 4.86b 1.53 0.56
Total 8.7 (1.2) 10.1 (1.2) 9.8 (1.1) 8.9 (1.7) 8.9 (1.5) 8.9 (1.5) 21.39c 2.25 23.64c
Limit of stability, ° 13.7 (2.1) 14.8 (1.9) 13.9 (2.3) 13.7 (3.5) 14.0 (3.1) 14.0 (3.1) 9.59c 0.05 4.12b
GDSSF-K, score 8.6 (2.5) 7.7 (2.2) 7.9 (2.1) 7.6 (2.8) 7.7 (2.5) 7.8 (2.6) 15.47c 0.02 20.32c
SF-36, score 57.3 (7.5) 61.8 (9.9) 61.9 (5.9) 60.0 (10.2) 62.2 (9.9) 61.7 (9.2) 16.91c 0.16 6.22b
Abbreviations: CVA, craniovertebral angle; GDSSF-K, Geriatric Depression Scale Short Form—Korea version; SF-36, 36-Item Short Form Health Survey instrument; SPPB, Short physical performance battery; TWD, Tragus-to-wall distance.
aData were presented as mean (SD).
bP < .05.
cP < .01.

Percent Change Between the 2 Groups

Table 3 outlines the comparison of the percent change of postintervention and follow-up values to the baseline value for all parameters between the 2 groups. In all parameters, percent change between baseline and postintervention data was significantly higher (P < .05) for the EG than that for the CG, except the LOS and SF-36. Percent change between baseline and follow-up data was also significantly higher (P < .05) for the EG than that for the CG, except the LOS (P < .05).

Table 3. - Percent Change Between Baseline Data and Postintervention and Follow-up Data of Each Parameter Between the 2 Groupsa
Outcomes Δ Baseline to Postintervention t Δ Baseline to Follow-up t
EG (n = 22) CG (n = 22) EG (n = 22) CG (n = 22)
Kyphotic posture
Angle of thoracic kyphosis −3.8 (2.2) 1.4 (3.8) −5.57b −3.8 (2.4) −0.2 (2.1) −5.29b
Kyphosis index
Relaxed −1.8 (1.2) −0.14 (0.9) −5.17b −1.9 (1.2) −0.1 (1.6) −4.43b
Best −3.5 (1.7) 0.38 (1.2) −8.91b −3.6 (1.8) 0.1 (1.4) −6.96b
Thoracic stiffness index 1.8 (1.9) −0.5 (0.9) 4.95b 1.7 (0.0) 1.9 (1.1) 3.58b
Forward head posture
CVA 4.8 (4.9) 0.4 (2.8) 3.65b 5.4 (5.9) 0.5 (2.3) 3.61b
TWD −12.7 (6.9) −5.2 (0.7) −3.92b −11.8 (6.3) −2.14 (5.1) −5.58b
SPPB total 16.3 (9.6) 0.3 (9.7) 5.51b 12.7 (8.3) 0.7 (9.1) 4.57b
Limit of stability 8.7 (6.8) 3.9 (10.8) 1.75 1.5 (9.9) 3.5 (7.5) −0.73
GDSSF-K −10.1 (7.5) 4.2 (12.6) −4.57b −8.2 (8.7) 4.9 (12.4) −4.06b
SF-36 8.3 (7.1) 3.9 (9.6) 1.70 8.7 (7.9) 3.3 (7.7) 2.32c
Abbreviations: CG, control group; CVA, craniovertebral angle; EG, experimental group; GDSSF-K, Geriatric Depression Scale Short Form—Korea version; SF-36, 36-Item Short Form Health Survey Instrument; SPPB, Short physical performance battery; TWD, Tragus-to-wall distance.
aData were presented as mean (SD).
bP < .01.
cP < .05.

Minimal Detectable Change

The MDC value for ATK was calculated to determine whether the change in score on a measure for an individual patient has reached a real improvement or is due to the measurement error. The difference of means, or actual change score, within the EG for ATK baseline to postintervention was 2.2°. On the basis of an ICC of 0.95, the associated MDC95% for ATK calculated was 2.51°. The change of 2.2° in this study is near the previously calculated MDC95% of 2.51°.

The SPPB was reported to have an MDC90% value of 2.9 points as calculated with a population of older African American adults.43 These authors recognize that a 3-point change would require large changes clinically. More recently, a study with older adults in a primary care setting found the MDC90% of the SPPB total score to be 1.66 points.44 Another study specifically examining 424 sedentary older adults showed the minimal clinically important difference on the SPPB to be 0.3 to 0.8 points or less than 1 point on the score.45 In the current study, the difference of means within the EG for SPPB total score baseline to postintervention was 1.4 point, which is more than the minimal clinically important difference but less than the MDC previously calculated for the SPPB.

The MDC95% value for the SF-36 previously reported was 5 points on a 100-point scale from a normative sample.46 The difference of means found within the EG for SF-36 baseline to postintervention was 4.5 points, again near the previously reported MDC95%.


Thoracic hyperkyphosis can lead to a variety of musculoskeletal problems related to aging.1,3,6,7 This study sought to determine the effect of corrective exercise for THK on posture, balance, and well-being in older community-living women. The findings indicate that our exercise program may be beneficial for improving spinal posture, balance, and well-being of older women with hyperkyphosis.

The intervention in this study emphasized thoracic rotation, extension, rib cage expansion, scapular retraction, and arm elevation with external rotation. This exercise program was an effective method for modifying both static and dynamic postures as evinced by improved kyphotic posture (ATK, KI, and TSI) and forward head posture (CVA and TWD) in the EG. Several studies have reported similar results in regard to improved posture. Greendale et al9 used a 6-month Hatha yoga program (3 days per week for 1-hour sessions) in older adults with hyperkyphosis. Their study showed a significant pre-/postimprovement of 5.17% in the Debrunner kyphometer angle in the median and a 3.64% change in the median flexicurve KI. Greendale et al9 achieved a slightly better improvement in the angle of kyphosis (5.17% vs 3.8%) and similar results in KI (3.64% vs 3.5%) but their intervention was 3 times as long. Similarly, a 6° increase in ATK (Debrunner kyphometer) was reported by Katzman and colleagues,47 who enrolled older women with THK in a 12-week (2 days per week) group exercise program. Bautmans et al16 reported small but significant improvements (mean of 3.4°) in the TK angle as measured with a wireless inclinometer in older women with osteoporosis. Their intervention consisted of manual therapy, taping, and exercise over a 12-week period. These findings are comparable in kyphotic angle improvement in comparison with the current study. However, Bergström et al48 reported no significant changes in ATK for older women following 16 weeks of a back extensor training. It is possible that significant changes in ATK are associated with multidimensional programs that include postural activities rather than programs focused on strengthening the back extensors alone. The exercise program used in the current study included theraband-resisted scapular retraction, overhead arm raising, diagonal patterns, and active trunk extension. These exercises are multidimensional in targeting strengthening, flexibility, and postural control and, thereby, may correct thoracic posture. In addition, the current intervention was delivered over a shorter period of time of 8 weeks for a total of 16 sessions. Therefore, this brief intense thoracic correction exercise program appears to be effective and efficient in positively influencing static and dynamic postural change.

Balance was also significantly improved following the current exercise intervention in persons with increased kyphosis. Balance and posture are intimately related; a shift in the line of gravity outside the base of support results in decreased postural stability, and ultimately, an increased fall risk.20 A physical activity program for THK has been shown to improve balance in previous studies. Benedetti et al7 investigated the effects of a structured exercise program for improving the flexibility of the pelvic and shoulder girdle as well as strengthening of the trunk extensors in persons with flexed posture for a 3-month period. They reported a significant improvement of 7.16% in SPPB total score after the intervention. The current study showed greater gains (16.3% SPPB total score) in a shorter time. The improvements in the balance test are particularly important to reduce fall risk and to maintain physical function in persons with hyperkyphosis.11 The findings of the current study reinforce the evidence that thoracic corrective exercise may improve dynamic balance and level of function in older women with THK.

Measures of depression and quality of life were included in this study to capture psychosocial changes that may be impacted by change in self-image, physical functioning, social engagement, and physical activity. The current study found improvements in GDSSF-K and SF-36 scores that are assumed to be associated with the application of the intervention and improved posture. Previous findings showed a relationship between the severity of flexed posture and greater depression in older women.9 Another study looked at quality of life measures in older persons with trunk deformities including kyphosis, lordosis, and flat back. Participants with abnormal postures reported lower satisfaction with “subjective health condition,” “human relationships to family,” “economic conditions,” and their “present life.”5 Nair et al49 reported that adults who were placed and taped in an upright posture had higher self-esteem and used more positive emotion words than those positioned in a flexed posture. The improvement of kyphotic posture through exercise based on thoracic correction may positively affect psychological status. Unlike the results of the current study, Benedetti et al7 reported that the exercise program did not improve the level of depression in women with kyphosis. However, participants showed relatively low levels of depression at baseline (5.33 out of 30). In addition, both groups performed exercise and both groups showed improved scores on the depression scale, just not at a statistically significant level.

There is a strong relationship between physical activity and depression, showing that more active older adults have a lower incidence of depression.50,51 One means of increasing physical activity in older adults has been community-based group exercise programs that make use of peer social support as a motivator.52 Therefore, therapeutic strategies to increase activity and improve posture should be established with a social component when working with older women. Although this study found an improvement in depression, we are unsure of the relationship between exercise, posture, and the multidimensional aspects of depression.

Despite favorable effects of our exercise program, there were several limitations that can be improved in future studies. This study is a group-matched design and is not a randomized controlled trial. However, we maximized homogeneity in each group, to be as similar as possible, especially age (related subject homogeneity) and ATK (related key condition). The exercise program also included the same exercises and was performed at the same intensity—but the EG received corrective supervision while the CG performed the program at home. The greater time spent in supervision of the EG versus the CG does create a confounding factor that may have influenced the favorable EG outcome. However, both groups were aware that they were participating in a study and were being monitored, so this may have helped equalize the effect. The MDC for ATK, SPPB, and SF-36 found was less than the previously reported MDC or calculated MDC. However, the MDCs were approaching the level needed and may be attainable with the current intervention protocol using a larger sample. Also, our program was relatively short-term and it is unclear whether greater differences in outcomes would exist with a longer program, such as 6 or 12 months. Nevertheless, our results show, even within a relatively short training period, that corrective thoracic exercise appears to have a beneficial effect on posture, balance, and well-being. This is an important practical outcome for those involved in the exercise training of older women with hyperkyphosis. Therefore, research efforts in the future need to establish a more controlled experimental design by addressing limitations of the present study, thus leading to a more solid conclusion regarding the benefits of the exercise program in older women with THK.


An 8-week exercise program for older women with hyperkyphosis may improve the measures of posture, balance, and well-being. These results, of the current study, suggest that our therapeutic exercise program designed specifically to mobilize the rib cage and improve TK is beneficial to improve spinal posture, balance, and well-being in older women with THK. We recommend the use of the therapeutic strategies utilized in this study to enhance thoracic posture and function of older women with thoracic THK. Future research is needed to apply this thoracic corrective exercise protocol to a larger and more diverse population.


The authors thank Drs Mansoo Ko, PT, PhD, and Steven R. Fisher, PT, PhD, GCS, for their guidance and support in the development of this manuscript.


1. Katzman WB, Wanek L, Shepherd JA, Sellmeyer DE. Age-related hyperkyphosis: its causes, consequences, and management. J Orthop Sports Phys Ther. 2010;40(6):352–360.
2. Kado DM, Huang MH, Karlamangla AS, Barrett-Connor E, Greendale GA. Hyperkyphotic posture predicts mortality in older community-dwelling men and women: a prospective study. J Am Geriatr Soc. 2004;52(10):1662–1667.
3. Kado DM, Lui LY, Ensrud KE, et al. Hyperkyphosis predicts mortality independent of vertebral osteoporosis in older women. Ann Intern Med. 2009;150(10):681–687.
4. Sinaki M, Brey RH, Hughes CA, Larson DR, Kaufman KR. Significant reduction in risk of falls and back pain in osteoporotic-kyphotic women through a spinal proprioceptive extension exercise dynamic (SPEED) program. Mayo Clin Proc. 2005;80(7):849–855.
5. Takahashi T, Ishida K, Hirose D, et al. Trunk deformity is associated with a reduction in outdoor activities of daily living and life satisfaction in community-dwelling older people. Osteoporos Int. 2005;16(3):273–279.
6. Ferguson SJ, Steffen T. Biomechanics of the aging spine. Eur Spine J. 2003;12(2):97–103.
7. Benedetti MG, Berti L, Presti C, Frizziero A, Giannini S. Effects of an adapted physical activity program in a group of elderly subjects with flexed posture: clinical and instrumental assessment. J Neuroeng Rehabil. 2008;5(1):32.
8. Schenkman M, Morey M, Kuchibhatla M. Spinal flexibility and balance control among community-dwelling adults with and without Parkinson's disease. J Gerontol A Biol Sci Med Sci. 2000;55(8):441–445.
9. Greendale GA, Huang MH, Karlamangla AS, Seeger L, Crawford S. Yoga decreases kyphosis in senior women and men with adult-onset hyperkyphosis: results of a randomized controlled trial. J Am Geriatr Soc. 2009;57(9):1569–1579.
10. Roaf R. Vertebral growth and its mechanical control. J Bone Joint Surg Br. 1960;42-B(1):40–59.
11. Sinaki M, Brey RH, Hughes CA, Larson DR, Kaufman KR. Balance disorder and increased risk of falls in osteoporosis and kyphosis: significance of kyphotic posture and muscle strength. Osteoporos Int. 2005;16(8):1004–1010.
12. Horak FB, Shupert CL, Mirka A. Components of postural dyscontrol in the elderly: a review. Neurobiol Aging. 1989;10(6):727–738.
13. Lynn SG, Sinaki M, Westerlin KC. Balance characteristics of persons with osteoporosis. Arch Phys Med Rehabil. 1997;78(3):273–277.
14. Balzini L, Vannucchi L, Benvenuti F, et al. Clinical characteristics of flexed posture in elderly women. J Am Geriatr Soc. 2003;51(10):1419–1426.
15. Katzman WB, Vittinghoff E, Kado DM, et al. Study of Hyperkyphosis, Exercise and Function (SHEAF) protocol of a randomized controlled trial of multimodal spine-strengthening exercise in older adults with hyperkyphosis. Phys Ther. 2016;96(3):371–381.
16. Bautmans I, Van Arken J, Van Mackelenberg M, Mets T. Rehabilitation using manual mobilization for thoracic kyphosis in elderly postmenopausal patients with osteoporosis. J Rehabil Med. 2010;42(2):129–135.
17. Pfeifer M, Begerow B, Minne HW. Effects of a new spinal orthosis on posture, trunk strength, and quality of life in women with postmenopausal osteoporosis: a randomized trial. Am J Phys Med Rehabil. 2004;83(3):177–186.
18. Greig AM, Bennell KL, Briggs AM, Hodges PW. Postural taping decreases thoracic kyphosis but does not influence trunk muscle electromyographic activity or balance in women with osteoporosis. Man Ther. 2008;13(3):249–257.
19. Bansal S, Katzman WB, Giangregorio LM. Exercise for improving age-related hyperkyphotic posture: a systematic review. Arch Phys Med Rehabil. 2014;95(1):129–140.
20. Kado DM. The rehabilitation of hyperkyphotic posture in the elderly. Eur J Phys Rehabil Med. 2009;45(4):583–593.
21. Ailon T, Shaffrey CI, Lenke LG, Harrop JS, Smith JS. Progressive spinal kyphosis in the aging population. Neurosurgery. 2015;77(1):164–172.
22. Jang HJ, Kim MJ, Kim SY. Effect of thorax correction exercises on flexed posture and chest function in older women with age-related hyperkyphosis. J Phys Ther Sci. 2015;27(4):1161–1164.
23. Fon GT, Pitt MJ, Thies AC Jr. Thoracic kyphosis: range in normal subjects. Am J Roentgenol. 1980;134(5):979–983.
24. Bush KW, Collins N, Portman L, Tillett N. Validity and intertester reliability of cervical range of motion using inclinometer measurements. J Man Manip Ther. 2000;8(2):52–61.
25. Frankel JE, Bean JF, Frontera WR. Exercise in the elderly: research and clinical practice. Clin Geriatr Med. 2006;22(2):239–256.
26. Teixeira FA, Carvalho GA. Reliability and validity of thoracic kyphosis measurements using flexicurve method. Braz J Phys Ther. 2007;11(3):199–204.
27. Hinman MR. Comparison of thoracic kyphosis and postural stiffness in younger and older women. Spine J. 2004;4(4):413–417.
28. Weber P, Corrêa ECR, Milanesi JM, Soares JC, Trevisan ME. Craniocervical posture: cephalometric and biophotogrammetric analysis. Braz J Oral Sci. 2012;11(3):416–421.
29. Dimitriadis Z, Podogyros G, Polyviou D, Tasopoulos I, Passa K. The reliability of lateral photography for the assessment of the forward head posture through four different angle-based analysis methods in healthy individuals. Musculoskeletal Care. 2015;30(13):179–186.
30. Nair P, Bohannon RW, Devaney L, Maloney C, Romano A. Reliability and validity of nonradiologic measures of forward flexed posture in Parkinson disease. Arch Phys Med Rehabil. 2017;98(3):508–516.
31. Guralnik JM, Simonsick EM, Ferrucci L, et al. A short physical performance battery assessing lower extremity function: association with self-reported disability and prediction of mortality and nursing home admission. J Gerontol. 1994;49(2):85–94.
32. Guralnik JM. Training Resources. Short Physical Performance Battery (SPPB) Guide. Accessed January 15, 2017.
33. Freire AN, Guerra RO, Alvarado B, Guralnik JM, Zunzunegui MV. Validity and reliability of the Short Physical Performance Battery in two diverse older adult populations in Quebec and Brazil. J Aging Health. 2012;24(5):863–878.
34. Guralnik JM, Ferrucci L, Simonsick EM, Salive ME, Wallace RB. Lower-extremity function in persons over the age of 70 years as a predictor of subsequent disability. N Engl J Med. 1995;332(9):556–651.
35. Granacher U, Bridenbaugh SA, Muehlbauer T, Wehrle A, Kressig RW. Age-related effects on postural control under multi-task conditions. Gerontology. 2011;57(3):247–255.
36. Borg FG, Laxaback G. Entropy of balance-some recent results. J Neuroeng Rehabil. 2010;7(1):38.
37. Almeida OP, Almeida SA. Short versions of the Geriatric Depression Scale: a study of their validity for the diagnosis of a major depressive episode according to ICD-10 and DSM-IV. Int J Geriatr Psychiatry. 1999;14(10):858–865.
38. Gi BS, Lee CW. A Preliminary study for the standardization of Geriatric Depression Scale Short Form-Korea version. J Korean Neuropsychiatr Assoc. 1996;35(2):298–307.
39. Brazier JE, Harper R, Jones NM, et al. Validating the SF-36 Health Survey questionnaire: new outcome measure for primary care. BMJ. 1992;305(6846):160–164.
40. Koh SB, Chang SJ, Kang MG, Cha BS, Park JK. Reliability and validity on measurement instrument for health status assessment in occupational workers. Korean J Prev Med. 1997;30(2):251–266.
41. Lewis JS, Valentine RE. Clinical measurement of the thoracic kyphosis. A study of the intra-rater reliability in subjects with and without shoulder pain. BMC Musculoskelet Disord. 2010;11(1):39.
42. Beaton DE. Understanding the relevance of measured change through studies of responsiveness. Spine. 2000;25(24):3192–3199.
43. Mangione KK, Craik RL, McCormick AA, et al. Detectable changes in physical performance measures in elderly African Americans. Phys Ther. 2010;90(6):921–927.
44. Beauchamp MK, Jette AM, Ward RE, Kurlinski LA. Predictive validity and responsiveness of patient-reported and performance-based measures of function in the Boston RISE study. J Gerontol A Biol Sci Med Sci. 2015;70(5):616–622.
45. Kwon S, Perera S, Pahor M, et al. What is a meaningful change in physical performance? Findings from a clinical trial in older adults (the LIFE-P study). J Nutr Health Aging. 2009;13(6):538–544.
46. Ware JE, Kosinski MA, Gandek B. SF-36 Health Survey: manual and interpretation guide. Lincoln, RI: Quality Metric; 2005.
47. Katzman WB, Sellmeyer DE, Stewart AL, Wanek L, Hamel KA. Changes in flexed posture, musculoskeletal impairments, and physical performance after group exercise in community-dwelling older women. Arch Phys Med Rehabil. 2007;88(2):192–199.
48. Bergström I, Bergström K, Kronhed AG, Karlsson S, Brinck J. Back extensor training increases muscle strength in postmenopausal women with osteoporosis, kyphosis and vertebral fractures. Adv Physiother. 2011;13(2):110–117.
49. Nair S, Sagar M, Sollers J, Consedine N, Broadbent E. Do slumped and upright postures affect stress responses? A randomized trial. Health Psychol. 2015;34(6):632–641.
50. Chodzko-Zajko WJ, Proctor DN, Fiatarone Singh MA, et al. American College of Sports Medicine position stand. Exercise and physical activity for older adults. Med Sci Sports Exerc. 2009;41(7):1510–1530.
51. Nelson ME, Rejeski WJ, Blair SN, et al. Physical activity and public health in older adults: recommendation from the American College of Sports Medicine and the American Heart Association. Med Sci Sports Exerc. 2007;39(8):1435–1445.
52. Hughes SL, Seymour RB, Campbell RT, Whitelaw N, Bazzarre T. Best-practice physical activity programs for older adults: findings from the National Impact Study. Am J Public Health. 2009;99(2):362–368.
Appendix 1-A.
Appendix 1-A.:
Thoracic Corrective Exercise Program
Appendix 1-B.
Appendix 1-B.:
Thoracic Corrective Exercise Program

aged; exercise; kyphosis; posture

© 2017 Academy of Geriatric Physical Therapy, APTA.