Effects of exercise and nutrition on postural balance
and risk of falling in elderly people with decreased
bone mineral density: randomized controlled trial pilot
Eling D. de Bruin
Clinical Rehabilitation 2007; 21: 523–534
Objective: To compare the effect of calcium/vitamin D supplements with a combination of
calcium/vitamin D supplements and exercise/protein on risk of falling and postural balance.
Design: Randomized clinical trial. Setting: University hospital physiotherapy department. Subjects:
Twenty-four independently living elderly females aged 65 years and older with osteopenia or
osteoporosis, and mean total Hip T-score (SD) of -1.8 (0.8). Interventions: A three-month
programme consisting of exercise/protein including training of muscular strength, co-ordination,
balance and endurance. Calcium/ vitamin D was supplemented in all participants for a 12-month
period. Outcome measures: Assessment took place prior to and following the months 3, 6, 9 and at
the end of the study; primary dependent variables assessed were risk of falling (Berg Balance Test)
and postural balance (forceplate). Secondary measures included body composition, strength,
activity level, number of falls, bone mineral content, biochemical indices, nutritional status and
general health. Results: Significant reductions of risk of falling (repeated measures ANOVA
F=8.90, p=0.008), an increase in muscular strength (ANOVA F=3.0, p=0.03), and an increase in
activity level (ANOVA F=3.38, p=0.02) were found in the experimental group as compared to the
control group. Further on, there was 89% reduction of falls reported in the experimental group
(experimental pre/post 8/1 falls; control group pre/post 5/6 falls). Conclusion: This study provides
support for our intervention programme aimed at reducing the risk of falling in elderly participants
diagnosed with osteopenia or osteoporosis. The data obtained from the pilot study allow the
calculation of the actual sample size needed for a larger randomised trial.
Osteoporosis is a major health problem, characterized by a significant morbidity, mortality and a
high economic burden1. Hip fractures are a frequent consequence of osteoporosis and the costs of
these fractures form a relevant part of the total expenses of osteoporosis. Furthermore, hip fractures
lead to reduced functional independence in the elderly2. In the year 2000, osteoporotic fractures
accounted for >18 000 hospital admissions in Switzerland; subsequently resulting in an average
hospital stay of 16 days at a cost of 234 million euro’s3. There are various factors that increase the
risk of a hip fracture, an increased risk of falling4 and osteoporosis5. A strong predictor of increased
risk of falling in the elderly is the loss of (automatic) balance control6. Indeed, most elderly
experience at least some decline in sensorimotor control with advanced age,7,8 leading to problems
in balance and gait. A Cochrane Database Systematic Review showed that to reduce the risk of
falling in elderly, a home-based muscle-strengthening programme, combined with balance
retraining can be applied9. When starting an exercise programme it is necessary to take into account
that the most of the elderly tend to be of low energy consumption. Hence they may be at risk for
low protein intake, and thus protein status, if they are active10. It is important for this reason to add
protein supplementation when beginning resistance strength training with the elderly11,12. The
promotion of exercise training in the elderly has a number of other implications. Older populations
have reduced flexibility when adapting their motor control to sudden altered changes in the
environment13. For example, a dual-task test consisting of walking on flippers (normally used for
swimming) while simultaneously solving a mental calculation task resulted in significant gait
impairment in the elderly, whereas young adults were minimally hindered by this combined
cognitive motor task14. Furthermore, sensory information processing is compromised by age, so that
the performance of postural tasks becomes increasingly difficult for the elderly15. Elderly people are
much more affected by a discrepancy between visual and proprioceptive information than young
adults16. It also seems possible that vitamin D has an effect on the risk of falling and the risk of
fracturing. First, chronic low calcium intake and vitamin D deficits are important co-factors leading
to secondary hyperparathyroidism and increased bone reabsorption, which are problems associated
with an increased incidence of osteoporosis and fractures in elderly people17. Second, it has been
argued that a vitamin D supplement has an influence on falls. A three-month application of
calcium/vitamin D reduced the number of falls per person by 49% compared with supplementation
with calcium alone18. It is unclear, however, whether this is a causal relationship or not.
Thus we considered it important to test an exercise programme within an elderly population with
decreased bone mineral density. The purpose of this pilot study was, therefore, to investigate the
feasibility of a three-month exercise programme on fall-related outcomes, in the context of vitamin
D supplementation. The authors are not aware of any previous study that has investigated such an
The study hypothesis was that exercise with proteins complemented with calcium/vitamin D
supplementation has a larger effect on fall-related outcomes than a calcium/vitamin D only
supplementation in elderly with decreased bone mineral density.
This randomized controlled prospective trial with two treatment arms was performed at the Center
for Osteoporosis of the Department of Rheumatology in close co-operation with the Institute of
Physical Medicine of the University Hospital Zurich. Random assignment to the intervention or
control group was performed on completion of baseline assessment with a stratified randomization
procedure (Figure 1). Stratification is useful in small trials in which it can avert imbalances on
prognostic factors19. To achieve an optimal balance, two important predictive factors: (1) risk of
falling (Berg Balance Test) and (2) bone mineral density level, were selected. Sealed envelopes
with random numbers were used to ensure that both groups were allocated in a balanced way; this
was undertaken by a co-worker independent of the study. The intervention group entered a three-
month exercise programme that included the training of muscular strength, co-ordination, balance,
and endurance and that was accompanied with nutritional (protein) supplementation. These
participants additionally received calcium/vitamin D supplementation20. The control group received
calcium/vitamin D supplementation only. The study lasted for a period of 12 months. Follow-up
measurements took place at 3, 6, 9 and 12 months following baseline.
Participants’ Baseline Characteristics
All Intervention group Control group
Mean Total Hip T-score
Mean Lumbar Spine T-score
Falls 3 months before Baseline
Values are mean (SD)
A total of 24 individuals were initially recruited (all female). All were living independently and
travelled on their own to participate in the exercise programme. Before entering the study, the
participants gave written informed consent after a physical examination by a physician to check
whether they fit the set criteria for inclusion. The study was approved by the local ethics committee.
The characteristics of these participants are presented in Table 1. Inclusion criteria were osteopenia
or osteoporosis diagnosed with dual energy X-ray absorptiometry (DEXA) (Table 1), according to
WHO criteria21, with or without prevalent fractures. Exclusion criteria: patients with any severe
peripheral or central neurological disease known to influence gait, balance or muscle strength. Also
patients with medical contraindications for exercise (e.g. major cardiovascular problems or postural
hypotension) were excluded.
Calcium/vitamin D supplementation
Both control and intervention group received a mineral supplementation according to physician’s
assessment at baseline, to correct any possible vitamin D and calcium deficiency. The
supplementation of the participants was, according to baseline assessment, 1–2 tablets/day for 12
months (Calcimagon D3; 500 mg Ca, 400 IU vitamin D). The calcium supplementation was 500–
1000 mg per day, whereas the vitamin D supplementation was 400–800 IU of cholecalciferol a day
22. A three-month supply was given at each assessment.
Exercise and protein intervention
The participants in the intervention group participated in a 12-week training programme aimed at an
improvement of balance abilities and a reduction of the risk of falling. The initial two weeks were
used to tailor the programme to the individual capacities of the participants. There were three
sessions per week of 70 minutes each. Two sessions consisted of progressive resistance training,
and individual exercises that focused on the improvement of co-ordination, balance and
endurance17. One session consisted of a group exercise focused on balance exercises and games. A
more detailed description of the exercise protocol and an exercise timetable is presented in
Appendix 1. To prevent undesirable training-induced losses in the lean tissue mass of our elderly
osteoporotic sample we gave our study subjects additional protein supplements11. The intervention
group was supplied with Resource Protein Drink (Novartis, Basel, Switzerland), a nutritional
supplement enriched with proteins, albumin and amino acids for 3 months on a daily basis11. The
daily dose per patient was 20 g (18.2 g protein, 0.2 g fat, 250 kcal)23. Each participant received a
seven-day supply ration once weekly. The participants were asked to consume their protein rations
at 10 am to minimize the effect on food intake11. The control group received no protein
supplementation. The participants of the intervention group were encouraged to continue the
exercise programme after the initial three months. The control group received a leaflet about home
exercises but did not attend the in-house training programme.
Primary outcome measures
Primary outcomes were risk of falling and postural balance. Risk of falling was assessed by means
of the Berg Balance Test. The Berg Balance Test consists of 14 functional subtests with a maximal
4 points (normal performance) and a minimal 0 points (no performance possible) per subtest. A
total of 0 points represents a severely impaired balance whereas 56 points does represent an
excellent balance24. The Berg Balance Test has a high inter-rater reliability (intraclass correlation
coefficient (ICC) =0.98) and a high intra-rater reliability (ICC =0.99)24 and is highly specific in
identifying elderly people who are prone to falling (cut-off score =45)25. Postural balance was
measured with the AccuSway PLUS system (Advanced Mechanical Technology Inc., Watertown,
MA, USA). This system provides centre of pressure coordinates, which allows postural sway and
the maximum displacement from the centroid in x-axis (medial–lateral, ML) and y-axis (anterior–
posterior, AP) to be measured. The balance platform was covered with a non-slip plastic cover. The
participant took a comfortable double-leg stance without shoes on the balance plate and the outlines
of both feet were marked on the plastic cover with a permanent marker. The purpose of the plastic
cover for each individual was to obtain the identical position of the feet across the repeated
measurements. During the actual measurement, the patient was instructed to assume a normal
barefooted, comfortable standing position, with arms at the side in a neutral position and to stand
still for 10 seconds. This was repeated four times with a break of 20 seconds between each
measurement26. The average of the four measurements was taken and recorded as the result of one
trial. After a 2-minute break, the procedure was repeated with eyes closed.
Whole body composition (WBC), as well as segmental body fat mass and lean mass, were
measured upon entry and after 12 months by DEXA on a Hologic QDR 4500 scanner (Hologic
Corporation Inc.,Waltham,MA, USA). Muscle strength of quadriceps femoris was measured with a
MicroFet device (Force Evaluating and Testing, Hoggan Health Industries Inc. West Draper, UT,
USA) in a standardized sitting position with the knee in 30° flexion27,28. A hand-held dynamometer
was placed distally at 80% of the tibia length. A measurement protocol was used to counterbalance
any order (or learning) effects. The test–retest reliability for the protocol used has been reported at
ICC 0.94 29. The patient physical activity level was determined with the self administered
Freiburger Questionnaire of Physical Activity. This questionnaire covers occupational, household
and leisure activities during a one-week period and takes 5–15 minutes to complete. Reliability has
been reported between 0.55 and 0.97 (P<0.001), whereas validity of the total activity was r =0.42,
P<0.01 30. The general health of the participants was determined through the Short Form-36
questionnaire (SF-36), which shows validity when used by elderly patients31. Falls in our study
were defined as ‘unintentionally coming to rest on the ground, floor, or other lower level’ 32.
Neither ‘coming to rest against furniture, a wall nor other structure’, nor ‘high-trauma falls (e.g,
falling from a ladder) and falling as a consequence of sustaining a violent blow’ were included as
falls in this study. Falls were assessed by interview at each assessment. Bone mineral content
(BMC) was measured upon entry and after 12 months at both anteroposterior projection of the
lumbar spine (L2–L4) and the non-dominant hip (total hip) using DEXA. The pre-post test
measures were performed by the same technician. All results were expressed in grams. Biochemical
markers were measured upon entry and after 12 months. These markers reflect bone metabolism
and could provide evidence that the intervention induced the hypothesized biological changes. Bone
metabolism was assessed in each patient, measured in morning serum and morning urine after
fasting. Assessments took place at the start of the programme and at 3, 6, 9 and 12-month follow-
ups. Adherence to the exercise programme was assessed, using the individual progressive loading
scheme of the patient, which was monitored following each training session. The assessors were
blinded for all measurements, except for the Berg Balance Test.
The comparability of both groups on prognostic and outcome variables at baseline was analysed
with two-sample t-tests. The normality of the distribution was checked with the Kolmogorov–
Smirnov one-sample test. Data on postural balance and risk of falling were analysed using ANOVA
(repeated measures) to look for significant interaction (group × time) effects (α =0.05). The
reliability coefficient of the measurement is determined by calculating an ICC33. The secondary
outcome variables were analysed similarly. The percentage of change of fallers was used to
compare the two groups. Missing measurements were filled in with the series mean of the patient’s
other four measurement results. The exercise compliance was defined as the number of exercise
sessions reported divided by the number of maximum exercise sessions possible. Data were entered,
stored and analysed using the Statistical Package for Social Sciences (SPSS, version 11.5).
Flow diagram for the study.
Twenty patients completed the study. One participant stopped because of hip replacement surgery
and one participant due to knee fracture. Two participants left the study due to unwillingness to
participate or without giving a specific reason. Figure 1 shows the patient flow diagram.
Due to administrative reasons, two measurements from the intervention group, one at six months
and one at 12 months, were missing and two measurements at nine months were missing from the
N=4 no low bone
Registered but not randomised
N=1 incompatible with vitamin D
n=24 registered participants
Exercise + Proteine
Baseline (N= 12)
Baseline (N= 12)
Losses N= 1
Losses N= 1
No reason given
Losses N= 1
Losses N= 1
9 and 12months
9 and 12 months
control group. The exercise compliance of the intervention group was 93%. A change in home
exercise by the control group was monitored with the Freiburger Questionnaire of Physical Activity.
All data for both groups appeared to be normally distributed. There were no statistical differences
for the prognostic variables at baseline (Table 2).
Primary outcome measures
Risk of falling
The intervention group showed a significant decrease in the Berg Balance Test (ANOVA F= 8.90,
P= 0.008, with a lower bound adjustment) (Table 2). The reliability (ICC) of this measurement in
our study was 0.90. The change is expressed in Figure 2a. Two individuals of the intervention group
scored under 45 points (43 points and 44 points) at baseline. At the end of the study they scored 50
and 56 points respectively. These two participants suffered a total of five falls before the start of the
study. At the end of the study only one of these individuals suffered a fall during the 12- month
study period. The change of scores between baseline and the follow-ups are shown in Table 3.
Maximal sway in AP and ML showed no interaction effect between the groups (Table 2). The ICC
of this measurement was 0.92 for AP and 0.95 for ML direction. The results are expressed in Figure
2b,c. The change of scores between baseline and the follow-ups are shown in Table 3.
The Freiburg Questionnaire of Physical Activity (F=3.38, P=0.02) and the quadriceps muscle
strength (F=3.0, P=0.03) showed a significant interaction effect between the groups, ANOVA with
sphericity assumed (Table 2)
Results from each measure at each time point in the two groups (n = 10 for each group)
BBT = Berg Balance Test; IG = Intervention Group; CG = Control Group; ICC, intraclass correlations coefficient; NA = Not Applicable; Values are given as mean
(SD) or range= min/max. *Significant p< 0.05.
Baseline 3 months 6 months 9 months 12 months ANOVA ICC
IG CG IG CG IG CG IG CG F
Berg Balance Test score
0.3 55.3 (1.5)
*8.90* *0.008* *0.90
0.4 0.62 (0.21)
*1.32 *0.27* *0.92
0.2 0.42 (0.20)
Activity (FAS) (kcal/week)
There was no significant difference between the groups at month 9 (F=0.76, P=0.40) and 12
(F=1.18, P=0.29) as determined with one-way ANOVA. The ICC of the quadriceps muscle strength
measurement in our study was 0.94 (Table 2) and the change of scores are shown in Table 3.
There was no change of lean and fat mass (WBC) in both groups. No change in bone mineral
content and important bone markers in both groups were found. Only the total hip bone mineral
content showed a significant increase for the control group (Table 4).
The reported falls were reduced within the intervention group by 100% and within the control group
by 80% after the first three months. After six months the decrease was 100% for the intervention
group and 40% control group; at months 9 and 12 reduction was 75% for the intervention group, but
an increase of 20% in falls was seen in the control group. The reported falls over the 12-month
period are shown in Figure 3. One participant in the intervention group sustained a knee fracture
after a fall during a holiday at month 9 from baseline. No other fractures occurred during the study
period in either group. Both groups had no change in the SF-36 questionnaire during the entire
Change scores between baseline and follow up 3, 6, 9 and 12 months
Δ 0/3 months Δ 0/6 months Δ 0/9 months Δ 0/12 months
IG CG IG CG IG CG IG CG
+3.6 (3.2) -0.3 (1.2) +3.8 (3.6) -0.5 (1.1) +3.6 (3.5) -0.5 (1.4) +3.9 (4.0) -1.3 (2.7)
AP (SD) +0.02 (0.13) +0.04 (0.12) -0.06 (0.11) +0.06 (0.21) -0.03 (0.16) +0.01 (0.14) -0.08 (0.08) -0.02 (0.13)
ML (SD) +0.05 (0.12) +0.02 (0.13) +0.08 (0.15) +0.06 (0.21) +0.08 (0.12) +0.06 (0.10) +0.02 (0.07) -0.01 (0.09)
+1029 (675) +80 (720) +938 (973) +382 (746) +358 (767) +317 (601) +265 (338) +182 (553)
+18.3 (15.0) -10.2 (21.6) +20.2 (13.6) -13.2 (19.2) +8.6 (25.2) +19.4 (70.3) +27.5 (28.0) +8.7 (24.9)
Δ, change from baseline; SD, standard deviation; BBT, Berg Balance Test; IG, intervention group; CG,
control group; AP,anterior–posterior; ML, medial–lateral.Values are given as mean, (+) increase and (-) a
(a) Outcome of the risk of falling (Berg Balance Test) and (b) outcome of the postural balance in
anterior–posterior (AP) direction and (c) outcome of the postural balance in medial–lateral (ML)
Error Bars show Mean +/- 1.0 SD
Dot/Lines show Means
Berg Balance Test
Error Bars show Mean +/- 1.0 SD
Dot/Lines show Means
Max Sway in AP Direction (cm)
Error Bars show Mean +/- 1.0 SD
Dot/Lines show Means
Max Sway in ML Direction (cm)
Biology and densitometry
T0 T12 T0 T12 F
WBC Lean Mass (g) (SD) 31606 (4189) 31209 (4201) 35451 (4052) 35269 (3850) 0.05 0.83
WBC Fat Mass (g) (SD) 20654 (6381) 20643 (6208) 19879 (5780) 20159 (5880) 0.39 0.54
WBC BMC (g) (SD) 1662 (246) 1659 (235) 1940 (307) 1959 (306) 1.65 0.22
Total Hip BMC (g) (SD) 25.7 (5.3) 25.8 (5.1) 27.3 (4.4) 28.2 (4.4) 7.52 0.01*
Total Spine BMC (g) (SD) 32.0 (5.7) 32.7 (5.7) 32.5 (8.4) 34.0 (7.7) 1.36 0.26
Body Mass Index (SD) 23.9 (3.0) 23.8 (3.0) 22.6 (3.0) 22.5 (2.9) 2.23 0.15
D-Pyr/ creat. (SD)
reference value 3.0-9.5 (nmol/ mmol)
7.0 (1.6) 7.1 (1.9) 6.1 (2.0) 6.7 (2.6) 0.19 0.70
25-OH Vit. D (SD)
reference value 10-42 (ųg/l)
31.10 (9.5) 30.05 (6.9) 31.80 (7.7) 38.38 (10.8) 4.00 0.06
Total Alc. phosphatase (SD)
reference value 3.4-19.8 (U/l)
11.7 (4.7) 9.3 (3.4) 11.5 (3.0) 11.1 (4.5) 0.08 0.80
WBC, whole body composition; BMC, bone mineral content; D-Pyr/creat., deoxypyridinium/creatinine; 25-
OH Vit. D, 25-OH vitamin D; SD, standard deviation; T0, baseline; T12, follow-up at 12 months. *P <0.05.
Dot/Lines show Sums
We investigated whether the calcium/vitamin D supplementation intervention programme plus
exercise/protein would have a larger effect on fall-related outcomes than a calcium/vitamin D
supplementation only in elderly with low bone mineral density. Our results showed a significant
decrease in the risk of falling as a consequence of the three-month intervention programme. These
changes were not observed in the control group.
However the Berg Balance Test showed a ceiling effect in the intervention group after the three-
month intervention34. This indicates that the test chosen was too easy for most of the participants.
Interesting are the two individuals with a score of less than 45 points at baseline. Subjects scoring
under 45 points have a higher risk of falling35. These two participants showed a large improvement
on the Berg Balance Test score after the intervention (from 43 to 50 and 44 to 56). They had had
five falls before the study, which represented more than half of the total falls of the intervention
group. However, these two individuals improved so much that only one suffered a single fall event
within the 12-month follow-up period. No similar individual improvements were seen in the control
group (Table 2).
We did not see any significant change within the postural balance measures. There was a reduction
in AP direction between three and six months. This reduction remained when compared with the
baseline values (Table 3). We expected that the patients of the intervention group would stay active
after the three month intervention, so that a positive effect on balance would remain even after the
intervention it self finished. This expectation seemed to be confirmed by the Freiburg Questionnaire
of Physical Activity data. The questionnaire showed a considerable increase in activity for the
intervention group measured three months and six months following baseline assessment. A wear-
off in questionnaire-determined activity was observed at 9 and 12 months. However, this did not
result in significant differences between the intervention and control groups.
No change of postural control was found in maximal displacement of ML direction. One
explanation for this finding is that lateral control is accomplished by the hip abductors/adductors
while the forward–backward direction is controlled by the ankle plantar and dorsiflexors36. In our
programme the exercise was focused on the anterior/posterior direction. Therefore quadriceps
training was a major exercise in our programme.
Previous recommendations have stated that it is important to add protein supplementation when
starting resistance strength training with the elderly12. The anabolic effect of starting resistance
strength training in the elderly did not seem to take place in our intervention group, possibly due to
the protein supplementation during the exercise period. This result is supported by the unchanged
body mass index of our participants.
The number of falls dropped dramatically in both groups within the first three months (Figure 3).
One reason for the reduction of falls within the control group could be the educational lecture given
at the beginning of the study. Similar short-term patient education effects have been seen in
rheumatoid arthritis patients37. However, these effects generally vanish over time. This was also
observed in the present study. Within six months, the control group returned to the number of
reported falls at baseline, whereas the intervention group stayed at a lower level.
The progression of osteoporosis slowed down in all participants in both groups. The significant
change in total hip bone mineral content within the control group can be explained by the small size
of the group and the fact that all participants received optimal individual osteoporosis treatment.
Bone metabolism results showed that all the patients were within the reference value at baseline and
at the end of the programme. Bone reabsorption activity decreased, but none of the changes were
significant. It is possible that the groups were too small to detect a significant change. Future
research should substantiate this assumption.
A limitation of our pilot study is the small number of participants. However, the collected data
allow for the calculation of the sample size needed for a larger study that has enough power33,34. To
avoid a type I or II error in a future study we need, based on our observed values (Table 2), an
estimated sample size of 116 participants in total for a two group pretest–posttest design (mean SD).
This would result in 80% power at an α-level of 0.05.
Another limitation of the present study is that it is not possible to determine whether all parts of the
calcium/vitamin D plus exercise/protein programme were needed in order to see a change or
whether one individual part of the programme could be responsible for the change. Future studies,
therefore, should compare different components of the programme. Also, a more defined fall
assessment (e.g. fall calendar) would reduce imprecise fall data.
Furthermore, the generalizability of our study was limited as only women who were independent in
the community with relatively well-preserved balance took part. A future study should strive to
include men in order to be able to assess the effects of the interventions and generalizability. This
issue should be addressed in a future study.
In conclusion, this study provides support for our intervention programme aimed at reducing the
risk of falling in elderly participants diagnosed with osteopenia or osteoporosis. The data obtained
from the pilot study will allow the power calculation of the actual sample size needed for a larger
randomized long-term controlled trial aimed at studying potential beneficial effects.
1. Nichols DL, Bonnick SL, Sanborn CF. Bone health and osteoporosis. Clin Sport Med 2000; 19: 233-
2. Dolan P, Torgerson DJ. The cost of treating osteoporotic fractures in the United Kingdom female
population. Osteoporosis Int 1998; 8: 611-17.
3. Lippuner K, Golder M, Greiner R. Epidemiology and direct medical costs of osteoporotic fractures
in men and women in Switzerland. Osteoporosis Int 2004; 16: 8-1
4. Parkkari J, Kannus P, Palvanen M, Natri A, Vainio J, Aho H, Vuori I, Jarvinen M. Majority of hip
fractures occur as a result of a fall and impact on the greater trochanter of the femur: a prospective
controlled hip fracture study with 206 consecutive patients. Calcif Tissue Int. 1999 Sep;65(3):183-7.
5. De Laet CE, van Hout BA, Burger H, Hofman A, Pols HA. Bone density and risk of hip fracture in
men and women: cross sectional analysis. Brit Med J 1997; 315: 221-25.
6. Nguyen T, Sambrook P, Kelly P, et al. Prediction of osteoporotic fractures by postural instability and
bone density. Brit Med J 1993; 307: 1111-15.
7. Kerschan-Schindl K, Uher E, Kainberger F, Kaider A, Ghanem AH, Preisinger E. Long-term home
exercise programme: effect in women at high risk of fracture. Arch Phys Med Rehab 2000; 81: 319-
8. Wolfson LI, Whipple R, Amerman P, Kaplan L, Kleinberg A. Gait and balance in the elderly. Two
functional capacities that link sensory and motor ability to falls. Clin Geriatr Med 1985; 1: 649-59.
9. Gillespie LD, Gillespie WJ, Robertson MC, Lamb SE, Cumming RG, Rowe BH. Interventions for
preventing falls in elderly people. Cochrane Database of Systematic Reviews 2003; 4: CD000340.
Rankin JW. Role of protein in exercise. Clin Sport Med 1999; 18: 499-511.
Meredith CN, Frontera WR, O'Reilly KP, Evans WJ. Body composition in elderly men: effect of
dietary modification during strength training. J Am Geriatr Soc 1992; 40: 155-62.
ACSM Position Stand on Exercise and Physical Activity for Older Adults. Med. Sci. Sports. Exerc
1998; 6: 992-1008.
ACSM Position Stand on Exercise and Physical Activity for Older Adults. Med Sci Sports
Exerc1998; 6: 992–1008.
13. Canavan AGM, Bayer N, Lutzenberger W. The decline of spatial and temporal abilities with age. In
GE Stelmach, V Homberg, (eds.): Sensorimotor Impairment in the Elderly. Kluwer Academic
Publishers, Dordrecht, 1999, pp. 57–72.
14. Mulder T, Berndt H, Pauwels J, Nienhuis B. Sensorimotor adaptability in the elderly and disabled. In
GE Stelmach, V Homberg, (eds.): Sensorimotor Impairment in the Elderly. Kluwer Academic
Publishers, The Hague, 1993, pp. 413–26.
15. Teasdale N, Bard C, LaRue J, Fleury M. On the cognitive penetrability of postural control. Exp
Aging Res 1992; 19: 1-13.
16. Woollacott MH, Shumway-Cook A, Nashner LM. Aging and posture control: changes in sensory
organization and muscular coordination. Int J Aging Hum Dev 1986; 23: 97-114.
17. Oakley A, Dawson MF, Holland J, et al. Preventing falls and subsequent injury in older people. Qual
Saf Health Care 1996; 5: 243-9.
18. Bischoff HA, Stahelin HB, Dick W, et al. Effects of vitamin D and calcium supplementation on falls:
a randomized controlled trial. J Bone Miner Res. 2003 Feb;18(2):343-51.
Schulz KF, Grimes DA. Unequal group sizes in randomised trials: guarding against guessing. Lancet
2002; 359: 966-70.
20. Schurch MA, Rizzoli R, Slosman D, Vadas L, Vergnaud P, Bonjour JP. Protein supplements increase
serum insulin-like growth factor-I levels and attenuate proximal femur bone loss in patients with
recent hip fracture. A randomized, double-blind, placebo-controlled trial. Ann Intern Med
21. WHO Study Group. Assessment of fracture risk and its application to screening for postmenopausal
osteoporosis. WHO Technical Report Series 843. World Health Organization, Geneva, 1994.
22. Rizzoli R, Bonjour JP. Undernutrition and osteoporosis. In WO Seiler, HB Stähelin, (eds.):
Malnutrition in the Elderly. Dr. D. Steinkopff Verlag GmbH & Co and Springer, Darmstadt, 1999,
23. Seiler WO, Stähelin HB. Malnutrition in the Elderly. Dr. D. Steinkopff Verlag GmbH & Co. and
Springer, Darmstadt, 1999.
24. Berg KO, Wood-Dauphinee SL, Williams JI, Maki B. Measuring balance in the elderly: preliminary
development of an instrument. Physiother Can 1989; 41: 304-11.
25. Bogle Thorbahn LD, Newton RA. Use of the Berg Balance Test to predict falls in elderly persons.
Phys Ther 1996; 76: 576-83.
26. Corriveau H, Hebert R, Prince F, Raiche M. Intrasession reliability of the "center of pressure minus
center of mass" variable of postural control in the healthy elderly. Arch Phys Med Rehab 2000; 81:
27. Kwoh CK, Petrick MA, Munin MC. Inter-rater reliability for function and strength measurements in
the acute care hospital after elective hip and knee arthroplasty. Ar C Res 1997; 10: 128-34.
28. van Essen D, Aufdemkampe G, Houvast M, Nijeboer I. Reliability study of the microFET for
pressure pain thresholds of myofascial trigger points in patients who have been operated on for
herniated nucleus pulposi. Physiother Theory Pract 1995; 11: 81-8.
29. Swanenburg J, Stappaerts KH, Tirez B, Uebelhart D, Aufdemkampe G. Development and reliability
of a measurement device for flexion force of the first metatarsophalangeal joint. J Appl Biomech
2003; 19: 271-8.
30. Frey I, Berg A, Grathwohl D, Keul J. Freiburger Fragenbogen zur koerperlichen Aktivitaet –
Entwicklung, Pruefung und Anwendung. [Freiburg Questionnaire of physical activity - design,
evaluation and application]. Soz Praventivmed 1999; 44: 55-64.
31. Ware JE Jr, Kosinski M, Bayliss MS, McHorney CA, Rogers WH, Raczek A. Comparison of
methods for the scoring and statistical analysis of SF-36 health profile and summary measures:
summary of results from the Medical Outcomes Study. Med Care 1995; 33: AS264-279.
32. Buchner DM, Hornbrook MC, Kutner NG, et al. Development of the common data base for the
FICSIT trials. J Am Geriatr Soc. 1993;41:297-30
Norman GR, Streiner DL. Biostatistics; The Bare Essentials. Decker Inc, Hamilton, BC, 2000.
Portney LG, Watkins MP. Foundations of Clinical Research. Applications to Practice. Appleton &
Lange, Norwalk, 1993.
35. Riddle DL. Stratford PW: Interpreting validity indexes for diagnostic tests: an illustration using the
Berg balance test. Phys Ther 1999; 79: 939-48.
36. Winter DA. ABC (Anatomy, Biomechanics, Control) of balance during standing and walking.
Graphics Services, Waterloo, 1995.
37. Niedermann K, Fransen J, Knols R, Uebelhart D. Gap between short- and long-term effects of
patient education in rheumatoid arthritis patients: a systematic review. Arth Rheum 2004; 51: 388-
Appendix 1 – The exercise protocol and timetable
The progressive resistance training consisted of dynamic exercises and isometric exercises with
fixed weights, within a certain range of motion. Barbells and dumbbells were used for training
because of the co-ordinative aspectsA1. The individual training load was determined with eight
repetitions maximum. The training sessions comprised the completion of three sets of eight
repetitions, with recovery periods of 1–2 minutes between sets. A personal training log, supervised
by a physical therapist, was used to ensure a progressive loading (eight repetitions maximum) and
to ensure individuality and securityA1,A3. The aim of the co-ordination/balance training was to
stimulate the sensory information system (visual, vestibular and somatosensoric) and enhance the
pace and adequacy of adaptive reactions. Thus the participants were trained to react in a more
suitable way in dynamic situations with and without external disturbance. Examples are exercises
with slow and fast movement and/or with eyes open and closed and/or with cognitive distraction
whilst performing exerciseA4.
van Wingerden BAM. Connective tissue in rehabilitation. Scripo, Vaduz, 1995.
Graves JE, Franklin BA. Resistance training for health and rehabilitation. Human Kinetics,
A3 Latham N, Anderson C, Bennett D, Stretton C. Progressive resistance strength training for physical
disability in older people. Cochrane Database of Systematic Reviews 2003; 2: CD002759.
A4 Shumway-Cook A, Woollacott MH. Motor Control; Theory and Practical Applications. Lippincott
Williams and Wilkins, Baltimore, 2001.
Exercise Timetable Download full-text
Monday Wednesday Friday
Steps, Squats, Back-extension,
Steps, Squats, Back-extension,
Coordination / Balance
Coordination / Balance