Can exercise reduce the incidence of falls in the elderly, and, if so, what form of exercise is most effective? (Evidence in Practice).

* Initial database used for search: MEDLINE

The MEDLINE database, created by the US National Library of Medicine, is the leading source for bibliographic references and abstracts from the biomedical literature. We decided to use this database to initiate our search because it gave us access to citations from more than 4,000 biomedical journals published worldwide as well as links to a number of full-text articles online. Although many vendors offer access to MEDLINE, we found PubMed, the free Internet version offered by the National Library of Medicine, to be the most effective way of accessing this database. We conducted our search using PubMed (www.ncbi.nlm.nih.gov/PubMed). This search was done on February 22, 2002.

* Initial keywords: falls, exercise

We started our search by performing 2 queries, one for the keyword falls and one for the keyword exercise. In order to minimize bias in our search, we did not use keywords that focused on a particular outcome (ie, falls and reduction). We felt using the terms thus and exercise would encompass all studies, regardless of outcome. These searches resulted in 14,784 citations for falls and 104,516 citations for exercise. We decided to combine the 2 searches using the Boolean operator "AND" to focus our search on citations that contained both terms. In PubMed, Boolean operators such as "AND," "OR," and "NOT" must be typed in upper case. In order to combine our first 2 search results, we typed #1 AND #2 in the query box, which corresponded to the first 2 search statement numbers in our search history (see "Tips for Using the History Feature in PubMed"). The number of citations containing both keywords was reduced to 559 articles. Because 559 citations was still an overwhelming number of references to consider, we decided to place some limits on our search in order to eliminate less relevant articles from our list.

* Limits: English, Human, Aged: 65+ years, 10 years, Randomized controlled trial

To set the appropriate limits, we clicked on the Limits option on the features bar directly beneath the query box. A screen offering a number of different options to limit our search appeared (Fig. 1). We first decided to focus on articles that were written in the English language by clicking on the Languages dropdown list and selecting the option English. To further narrow our search, we also decided to look only at the articles describing studies that were done on human subjects by selecting Human from the Human or Animal dropdown list. We also looked only at the studies that considered a population with an age range similar to that of the patients in our independent living center by selecting Aged: 65+ years from the Ages dropdown list. Finally, in order to consider the latest findings and the highest level of evidence available on this topic, we limited our search to the randomized controlled trials published in the past 10 years by selecting 10 years from the Entrez Date list, which would pull up citations based on the date they were entered into PubMed, and Randomized Controlled Trial from the Publication Types list.

[FIGURE 1 OMITTED]

Imposing these limits narrowed down the number of citations from 559 to 50 articles. After a quick review of the titles, we were able to eliminate articles that focused on other interventions to reduce falls, such as consultation services, home evaluations, and psychotropic medication withdrawal. We found that only 14 articles were directly related to the effects of exercise on falls in older adults (Fig. 2), and we were able to keep track of the relevant articles by checking the box next to each citation and hitting the Display button underneath the Features bar after we completed our review of the citation list. Next, we decided to further evaluate the relevance of the remaining 14 articles by reading their abstracts. Of these 14 articles, 7 showed that exercise had a positive effect on risk of falls in elderly people, 6 showed that exercise had no effect on risk of falls in elderly people, and 1 did not have an accompanying abstract. To make matters more complicated, each of the 7 studies that showed that exercise had a positive effect on fall risk used varying exercise protocols. Given the conflicting evidence, it was clear to us that we needed additional information. For example, were some of these 14 randomized controlled trials more credible than others? Were some flawed from a scientific standpoint? Why did some studies show a positive effect and others no effect?

Figure 2. Citations Retrieved by Search in PubMed Using the Keywords
"Falls" and "Exercise" With the Following Limits Added--English
Language, Human studies, Aged: 65+ years, Entrez date: 10 years,
Randomized controlled trial.

1. Carter ND, Khan KM, Petit MA, Heinonen A, Waterman
C, Donaldson MG, Janssen PA, Mallinson A, Riddell
L, Kruse K, Prior JC, Flicker L, McKay HA. Results
of a 10 week community based strength and balance
training programme to reduce fall risk factors: a randomized
controlled trial in 65-75 year old women with
osteoporosis. Br J Sports Med. 2001;35(5):348-51.

2. Nowalk MP, Prendergast JM, Bayles CM, D'Amico FJ,
Calvin GC. A randomized trial of exercise programs
among older individuals living in two long-term care
facilities: the FallsFREE program. J Am Geriatr Soc.
2001;49(7):859-65.

3. Robertson MC, Devlin N, Gardner MM, Cambell AJ.
Effectiveness and economic evaluation of a nurse
delivered home exercise programme to prevent falls.
1: Randomised controlled trial. BMJ.
2001;322(7288):697-701.

4. Hauer K, Rost B, Rutschle K, Opitz H, Secht N,
Bartsch P, Oster P, Schlierf G. Exercise training for rehabilitation
and secondary prevention of falls in geriatric
patients with a history of injurious falls. J Am Geriatr
Soc. 2001;49(1):10-20.

5. Rubenstein LZ, Josephson KR, Trueblood PR, Lay S,
Harker Jo, Pietruszka FM, Robbins AS. Effects of a
group exercise program on strength, mobility, and falls
among fall-prone elderly men. J Gerontol A Biol Sci
Med Sci. 2000;55(6):M317-21.

6. Schoenfelder DP. A fall prevention program for
elderly individuals. Exercise in long-term care settings.
J Gerontol Nurs. 2000;26(3):43-51.

7. Cambell AJ, Robertson MC, Gardner MM, Norton
RN, Buchner DM. Falls prevention over 2 years: a randomized
controlled trial in women 80 years and older.
Age Ageing. 1999;28(6):513-8.

8. Campbell AJ, Robertson MC, Gardner MM, Norton
RN, Tilyard MW, Buchner DM. Randomised controlled
trial of a general practice programme of home based
exercise to prevent falls in elderly women. BMJ.
1997;315(7115):1065-9.

9. Buchner DM, Cress ME, de Lateur BJ, Esselman PC,
Margherita AJ, Price R, Wagner EH. The effect of
strength and endurance training on gait, balance, fall
risk, and health service use in community-living older
adults. J Gerontol A Biol Sci Med Sci.
1997;52(4):M218-24.

10. McMurdo ME, Mole PA, Paterson CR. Controlled
trial of weight bearing exercise in older women in
relation to bone density and falls. BMJ.
1997;314(7080):569.

11. Buchner DM, Cress ME, de Lateur BJ, Esselman PC,
Margherita AJ, Price R, Wagner EH. A comparison of
the effects of three types of endurance training on balance
and other fall risk factors in older adults. Aging
(Milano). 1997;9(1-2):112-9.

12. Means KM, Rodell DE, O'Sullivan PS, Cranford LA.
Rehabilitation of elderly failers: pilot study of a low to
moderate intensity exercise program. Arch Phys Med
Rehabil. 1996;77(10):1030-6.

13. Wolf SL, Barnhart HX, Kutner NG, McNeely E,
Coogler C, Xu T. Reducing frailty and falls in older
persons: an investigation of Tai Chi and computerized
balance training. Atlanta FICSIT Group. Frailty and
Injuries: Cooperative Studies of Intervention Techniques.
J Am Geriatr Soc. 1996;44(5):489-97.

14. Lord SR, Ward JA, Williams P, Strudwick M. the effect
of a 12-month exercise trial on balance, strength, and
falls in older women: a randomized controlled trial.
J Am Geriatr Soc. 1995;43(11)1198-206.

Because we did not feel we had the knowledge to adequately critique the scientific merit of these trials, we decided to see if a systematic review had been performed in this area. As defined by Sackett et al, a systematic review is "a summary of the medical literature that uses explicit methods to perform a thorough literature search and critical appraisal of individual studies and that uses appropriate statistical techniques to combine these valid studies." (1(p250)) Although systematic reviews are not immune to reviewer bias and contradictory conclusions, we believed that, if a systematic review in this area did exist, it would help us answer our question.

* Second database used for search: Cochrane Database of Systematic Reviews

The Cochrane Database of Systematic Reviews (CDSR) was created by the Cochrane Collaboration, an international network of review groups directed to prepare, maintain, and disseminate systematic reviews concerning health care issues. The goal of the Cochrane Collaboration is to assemble the highest-quality evidence by reviewing the relevant research in order to offer conclusions on the effects of health care. The CDSR is updated regularly as new research is published, and it is revised if errors are identified. If studies do not meet the requirements of the Collaborative Review Groups, they are disregarded. Once a review is published, it remains in the CDSR and is continuously modified on the basis of new evidence.

The CDSR can be accessed through the Internet at www.cochranelibrary.com. Although access to the abstracts is free, a subscription to the CDSR is required in order to gain access to the full-text articles. Nonregistered users can obtain a full copy of a review via mail or fax for a rise ($15 US). Copies of the articles also may be obtained through organizations that are registered users, such as university libraries and teaching hospitals. We accessed this database using Ovid Online (www.ovid.com) * through a local university library. This search was performed on March 20, 2002.

* Keywords: falls AND exercise

After logging onto Ovid, we started our search by clicking on EBM ([dagger]) Reviews-Cochrane Database of Systematic Reviews from the list of databases. Once the main search page appeared, we typed falls AND exercise in the query box. In Ovid, the query box is located under the Icon bar, a row of small icons offering access to the advanced features of Ovid. Like PubMed, Boolean terms must be capitalized. The result of this query was 17 systematic reviews.

* Selection of articles for review: After reading the titles, we found only one review that was directly related to the prevention of falls in elderly people. We then decided to read the abstract tar more information.

Gillespie, LD; Gillespie, WJ; Robertson, MC; Lamb, SE; Cumming, RG; Rowe, BH. Interventions for preventing falls in elderly people. Cochrane Database of Systematic Reviews. Issue 1, 2002.

BACKGROUND: Approximately 30 percent of people over 65 years of age and living in the community foil each year; the number is higher in institutions. Although less than one fall in 10 results in a fracture, a fifth of fall incidents require medical attention. OBJECTIVES: To assess the effects of interventions designed to reduce the incidence of falls in elderly people (living in the community, or in institutional or hospital care). SEARCH STRATEGY: We searched the Cochrane Musculoskeletal Group specialised register (January 2001), Cochrane Controlled Trials Register (The Cochrane Library, Issue 1, 2001), MEDLINE (1966 to February 2001), EMBASE (1988 to 2001 Week 14), CINAHL (1982 to March 2001), The National Research Register, Issue 1, 2001, Current Controlled Trials (accessed 25 May 2001), and reference lists of articles. We also contacted researchers in the field. SELECTION CRITERIA: Randomised trials of interventions designed to minimise the effect of, or exposure to, risk factors for falling in elderly people. Main outcomes of interest were the number of fallers, or falls. Trials reporting only intermediate outcomes were excluded. DATA COLLECTION AND ANALYSIS: Two reviewers independently assessed trial quality and extracted data. Data were pooled using the fixed effect model where appropriate.

MAIN RESULTS:

Interventions likely to be beneficial:

* A programme of muscle strengthening and balance retraining, individually prescribed at home by a trained health professional (3 trials, 566 participants, pooled relative risk (RR) 0.80, 95% confidence interval (95%Cl) 0.66 to 0.98).

* A 15 week Tai Chi group exercise intervention (1 trial, 200 participants, risk ratio 0.51, 95%Cl 0.36 to 0.73).

* Home hazard assessment and modification that is professionally prescribed for older people with a history of falling (1 trial, 530 participants, RR 0.64, 95% Cl 0.49 to 0.84).

* A reduction in falls was seen both inside and outside the home. Withdrawal of psychotropic medication (1 trial, 93 participants, relative hazard 0.34, 95% Cl 0.16 to 0.74).

* Multidisciplinary, multifactorial, health/environmental risk factor screening/intervention programmes, both for unselected community dwelling older people (data pooled from 3 trials, 1973 participants, pooled RR 0.73, 95%Cl 0.63 to 0.86), and for older people with a history of falling, or selected because of known risk factors (data pooled from 2 trials, 713 participants, pooled RR 0.79, 95%Cl 0.67 to 0.94).

Interventions of unknown effectiveness:

* Group-delivered exercise interventions (9 trials, 2177 participants).

* Nutritional supplementation (1 trial, 50 participants).

* Vitamin D supplementation, with or without calcium (3 trials, 679 participants).

* Home hazard modification in association with advice on optimising medication (1 trial, 658 participants), or in association with an education package on exercise and reducing fall risk (1 trial, 3182 participants).

* Pharmacological therapy (raubasine-dihydroergocristine, 1 trial, 95 participants).

* Fall prevention programmes in institutional settings.

* Interventions using a cognitive/behavioural approach alone (2 trials, 145 participants).

* Home hazard modification for older-people without a history of falling (1 trial, 530 participants).

* Hormone replacement therapy (1 trial, 116 participants).

Interventions unlikely to be beneficial:

* Brisk walking in women with an upper limb fracture in the previous two years (1 trial, 165 participants).

CONCLUSIONS: Interventions to prevent falls that are likely to be effective are now available; less is known about their effectiveness in preventing fall-related injuries. Costs per fall prevented have been established for four of the interventions and careful economic modeling in the context of the local healthcare system is important. Some potential interventions are of unknown effectiveness and further research is indicated.

[[c] 2001 Update Software Ltd/Cochrane Collaboration. Abstract reprinted from The Cochrane Library with permission of Update Software Ltd.]

Of the trials included in this systematic review, 13 dealt with the effects of exercise in reducing fall risk in elderly people. Of those 13 trials, 4 used interventions that were deemed "likely to be beneficial," whereas 9 of the trials used interventions that were "of unknown effectiveness." Of the 4 trials that used interventions "likely to be beneficial," 3 used exercise programs that were individually prescribed, whereas the fourth trial evaluated the effects of a Tai Chi exercise intervention. Interestingly, of the 9 trials that were found to demonstrate "unknown effectiveness," all used group-delivered exercise interventions. Because it was apparent that there was at least some evidence to support the concept that exercise could reduce falls in elderly people, we decided to obtain the full-text version of this systematic review by clicking on the Topic Review option next to the citation in the results list. We proceeded to read the review, but the details of the exercise protocols that were determined to be successful were not described. However, because the citations for the original articles were included in the text, we were able to refer to the original source to attain the details of these studies. We decided to focus our attention on the articles that dealt with individualized exercise programs because the outcomes for these trials were more compelling. We obtained these articles from the local university library and began to review their contents. The abstracts from these studies as they appear in MEDLINE are presented below, as is a brief discussion of each article.

Campbell AJ, Robertson MC, Gardner MM, Norton RN, Buchner, DM. Falls prevention over 2 years: a randomized controlled trial in women 80 years and older. Age and Ageing 1999;28:513-518.

BACKGROUND: After 1 year, a home-based programme of strength and balance retraining exercises was effective in reducing falls and injuries in women aged 80 years and older. The exercise programme had been individually prescribed by a physiotherapist during the first 2 months of a randomized controlled trial. OBJECTIVE: We aimed to assess the effectiveness of the programme over 2 years. SUBJECTS: Women from both the control group and the exercise group completing a 1-year trial (213 out of the original 233) were invited to continue for a further year. METHODS: Falls and compliance to the exercise programme were monitored for 2 years. RESULTS: 81 (74%) in the control group and 71 (69%) in the exercise group agreed to continue in the study. After 2 years, the rate of falls remained significantly lower in the exercise group than in the control group. The relative hazard for all falls for the exercise group was 0.69 (95% confidence interval 0.49-0.97). The relative hazard for a fall resulting in a moderate or severe injury was 0.63 (95% confidence interval 0.42-0.95). Those complying with the exercise programme at 2 years had a higher level of physical activity at baseline, were more likely to have reported falling in the year before the study and had remained more confident in the first year about not falling compared with the rest of the exercise group. CONCLUSIONS: Falls and injuries can be reduced by an individually tailored exercise programme in the home. For those who keep exercising, the benefit continues over a 2-year period.

[[c] 1999 British Geriatrics Society. Abstract reprinted with permission of Oxford University Press.]

This randomized controlled trial found that a personalized home balance and exercise training program significantly reduced the risk of falls in elderly people. The efficacy of this program was maintained over a 2-year period; however, the benefit was sustained only as long as the subjects adhered to the exercise program. The conclusions offered by this article seemed appropriate to answer our initial question regarding the efficacy of exercise in reducing falls in elderly people. Unfortunately, the article provided limited information regarding the details of the exercise program used, and readers were referred to another study by the same authors for the complete protocol (Campbell et al, 1997; see abstract below). The 2 factors that were deemed essential to the success of this exercise regimen were: (1) all of the exercises were individually prescribed by a physical therapist and (2) the physical therapist remained in telephone contact with the subjects for the purposes of giving advice as well as maintaining their motivation.

Campbell AJ, Robertson MC, Gardner MM, Norton RN, Tilyard MW, Buchner DM. Randomised controlled trial of a general practice programme of home based exercise to prevent fails in elderly women. BMJ 1997;315:1065-1069.

OBJECTIVE: To assess the effectiveness of a home exercise program of strength and balance retraining exercises in reducing falls and injuries in elderly women. DESIGNS: Randomized controlled trail of an individually tailored program of physical therapy in the home (exercise group, n=116) compared with the usual care and an equal number of social visits (control group, n=117). SETTING: 17 general practices in Dunedin, New Zealand. MAIN OUTCOME MEASURES: Number of falls and injuries related to falls and time between falls during one year of follow up; changes in muscle strength and balance measures after six months. RESULTS: After one year there were 152 falls in the control group and 88 falls in the exercise group. The mean (SD) rate of falls was lower in the exercise than control group (0.87 (1.29) v 1.34 (1.93) falls per year respectively; difference 0.47; 95% confidence interval 0.04 to 0.90). The relative hazard for a first fall with injury in the exercise group compared with the control group was 0.61 (0.39 to 0.97). After six months, balance had improved in the exercise group (difference between groups in change in balance score 0.43 (0.21 to 0.65). CONCLUSIONS: An individual program of strength and balance retraining exercises improved physical function and was effective in reducing falls and injuries in women 80 years and older.

[[c] 1997 British Medical Association. Abstract reprinted with the permission of the BMJ Medical Group.]

This randomized clinical trial concluded that an individually based exercise program consisting of physical therapy in the home was effective in reducing falls and improving function in women who were 80 years of age and older. Tests and measures that were used included an instrumental activities of daily living scale, the Physical Self-Maintenance Scale, fear of falling, and the Physical Activity Scale for the Elderly.

Participants in the control group received regular "social visits" during the first 2 months and were then telephoned regularly during the year as a follow-up. Participants in the exercise group were prescribed a selection of strengthening and balance exercises from a program at appropriate and increasing levels of difficulty. Strengthening exercises included moderate intensity exercises with ankle cuff weights for the following muscles: hip extensor and abductors, knee flexors and extensors, and ankle plantar flexors and dorsiflexors. Balance exercises consisted of: walking and standing with one foot in front of the other, walking on toes and heels, walking backwards and sideways, and turning around. Stepping over objects, climbing stairs, squats, and sit-to-stand movements from a chair also were performed to increase strength and balance. The exercise protocol was performed for 30 minutes 3 times a week. In addition, subjects were encouraged to walk 3 times a week on their own.

We were pleased with the results of this study, which showed that, after 1 year of training, balance had improved significantly and the number of falls had decreased in the exercise group compared with the control group. Furthermore, the control group became less active and their fear of falling increased. This study clearly supports the notion that implementing an individualized exercise program would be beneficial for elderly people living in independent living centers.

Robertson MC, Devlin N, Gardner MM, Campbell AJ. Effectiveness and economic evaluation of a nurse delivered home exercise programme to prevent falls. 1: Randomised controlled trial. BMJ 2001;322(7288):697-701.

OBJECTIVES: To assess the effectiveness of a trained district nurse individually prescribing a home based exercise program to reduce falls and injuries in elderly people and to estimate the cost effectiveness of the program. DESIGN: Randomized controlled trial with one year follow up. SETTING: Community health service at a New Zealand hospital. PARTICIPANTS: 240 women and men aged 75 years and older. INTERVENTIONS: 121 participants received the exercise program (exercise group) and 119 received usual care (control group); 90% (211 of 233) completed the trial. MAIN OUTCOME MEASURES: Number of falls, number of injuries resulting from falls, costs of implementing the program, and hospital costs as a result of falls. RESULTS: Falls were reduced by 46% (incidence rate ratio 0.54, 95% confidence interval 0.32 to 0.90). Five hospital admissions were due to injuries caused by falls in the control group and none in the exercise group. The program cost $NZ1803 (523 [pounds sterling]) (at 1998 prices) per fall prevented for delivering the program and $NZ155 per fall prevented when hospital costs averted were considered. CONCLUSION: A home exercise program, previously shown to be successful when delivered by a physiotherapist, was also effective in reducing falls when delivered by a trained nurse from within a home health service. Serious injuries and hospital admissions due to falls were also reduced. The program was cost effective in participants aged 80 years and older compared with younger participants.

[[c] 2001 British Medical Association. Abstract reprinted with the permission of the BMJ Medical Group.]

This randomized controlled trial reported that a home exercise program to prevent falls delivered by a nurse could be effective and cost-efficient in people who were 75 years of age and older. Like the first article reviewed (Campbell et al, 1999), this article gave limited information on the specifics of the home exercise program and referred the reader to their 1997 article for details (see previous abstract). The difference between this study and the previous 2 studies that we reviewed was the fact that the exercise program was prescribed by a nurse and not a physical therapist. The nurse was trained by a qualified physical therapist in a 1-week course; however, the specifics of this training were not mentioned.

Overall, a significant 46% reduction in falls was observed in the exercise group. This study confirmed the effectiveness of strength and balance training in reducing falls in elderly people, and further suggests that the success of an exercise program depends on whether it has been individualized.

* Clinical decision: When we saw the center administrator again, we told him that there was scientific evidence that exercise could reduce falls in elderly people. Based on the conclusions of the Cochrane systematic review, we also suggested that an individually prescribed strength and balance program would likely be more effective than a group exercise program. In our opinion, a fall prevention program could be implemented rather inexpensively, because the 3 studies we reviewed all used home exercise programs, thereby negating the need for expensive equipment and space requirements. It also was clear to us that the initial program development for each participant would need to be based on identified impairments and functional limitations and that the services of a physical therapist should be employed. Based on the 1999 Campbell et al study, it was apparent that the physical therapist also would have to participate in occasional follow-ups to aid exercise progression, answer questions, and provide motivation.

As a result of our search, we felt confident in advocating for the development of a fall prevention program. The findings of the trials we discussed were based on a relatively large number of subjects (566 combined), and the average age and demographics of the residents of our facility were consistent with the study populations. Therefore, the decision to initiate such a program was justified in our opinion.

References

(1) Sackett DL, Straus SE, Richardson WS, et al. Evidence-Based Medicine: How to Practice and Teach EBM. 2nd ed. New York, NY: Churchill Livingstone Inc; 2000.

RELATED ARTICLE: Tips for using the history feature in PubMed.

PubMed's History function--located on the Features bar directly beneath the query box--automatically holds and displays up to 100 search statements (a numbered line containing a completed search strategy, the time of the search, and the number of citations produced). These search statement numbers can be combined or used with new search queries.

* To combine 2 searches, type the pound (#) sign and the search statement number for each search, separate them by adding AND (ie, #1 AND #4), and click Go.

* To combine a search with a new keyword, type the pound sign and the search statement number, then type AND in all capitals, then type the new keyword, and finally click Go.

* To refine a search, type the pound sign and the search statement number, then click on the Limits feature on the Features bar and select the appropriate limits,

The Search History is automatically lost after 1 hour of inactivity.

* Ovid Technologies, 333 Seventh Ave, 4th Floor, New York, NY 10001.

([dagger]) EBM=evidence-based medicine.

Christopher M Powers, PT, PhD, is Assistant Professor, Department of Biokinesiology and Physical Therapy, University of Southern California, Los Angeles, Calif.

Shawn Farrokhi and Jeff Moreno are DPT students, Department of Biokinesiology and Physical Therapy.