Systematic review
Review of systematic reviews on prevention of falls in institutions
Systematic review
|Updated
Multifactorial institutions customised to the particular institution will probably reduce rate of falls in hospitals, but not the number of fallers or hip fractures.
Download
Key message
Falls are the most common cause of injuries among the elderly in and outside of health institutions. Falls can have many negative consequences for the elderly. Between 10 and 20% of falls in nursing homes cause serious injuries.
Effect of various interventions to prevent falls in institutions:
- Multifactorial interventions (interventions that consists of a number of various components like education of staff and offers of exercises to the patients) adapted to the institutions conditions probably reduce rate of fall in hospitals (e.g. number of falls in a person year, this capture that some persons fall multiple times).
- Multifactorial interventions compared to usual practice do possibly not influence the rate of falls, the number of fallers (persons who fall at least once) or hip fractures in care facilities.
- Vitamin D supplementation to patients with low levels of vitamin D in care facilities will probably reduce the rate of falls, but not the number of fallers.
- Use of risk assessment tools compared to nurses’ judgment in care facilities probably result in small differences on the rate of falls or the number of fallers.
- The diagnostic accuracy of the STRATIFY rule is limited and should not be used in isolation for identifying individuals at high risk of falls in clinical practice.
For several of other studied interventions there are large uncertainty about the results. We are uncertain if these interventions have an effect or not. We cannot conclude if the following interventions affect the rate of falls or the number of fallers; exercise, medication review, encourage adoption of best practice strategies, very low bed, fall prevention tool kit software, use of educational materials.
Summary
The Patient safety campaign was converted to a five-year programme in 2014. As part of the work to develop the programme, the Norwegian Knowledge Centre for the Health Services received a commission from the Western Norway Regional Health Authority to update the documentation of systematic reviews published after the initiation of the campaign. We have searched for systematic review of high quality for the target area “Prevention of falls”, but limited to systematic reviews on prevention of falls in institutions, published in 2010 or later.
Why prevent falls?
Falls are the most common cause of injuries among the elderly in and outside of health institutions. Falls can have many negative consequences for the elderly. Between 10 and 20% of falls in nursing homes cause serious injuries. In addition to fractures and head trauma there may be problems of lasting impairment, fear of falling again and loss of independence. The target area of “Fall prevention” in the patient safety programme addresses patients older than 65 years and adults with neurological or cognitive illnesses or large visual impairment in hospitals and nursing homes.
Fall is in the patients safety programme defined as “an accidental event that leads to a person ended on the ground, the floor or any other lower level, regardless of cause and if there is any injury caused by the fall”.
The interventions given on the homepage of the patient safety programme are:
1.Risk assessment
- Standard measures for all high risk patients
- Interdisciplinary evaluation
- Individually customised measures
- Transfer of information about risk assessment and measures
Method
We searched for systematic reviews in Cochrane library for preventing falls in institutions and in “Making Healthcare Safer II” published in 2013 by Agency of Healthcare Research and Quality (AHRQ). The question was covered both places, except the choice of risk assessment tool. For this question, we searched in the following databases:
- Ovid MEDLINE(R) In-Process & Other Non-Indexed Citations and Ovid MEDLINE(R)
- Ovid EMBASE
- Cochrane Database of Systematic Reviews
- Cochrane and CRD: Database of Abstracts of Reviews of Effects (DARE)
- Cochrane and CRD: Health Technology Assessment (HTA)
- CINAHL
- Web of Science
- The Knowledge Centre for the Health Services
Two authors independently evaluated titles and abstracts from the search. For abstracts regarded as relevant, the article was read in full text according to the criteria of inclusion. Independently, we assessed the relevance and quality of identified systematic review with the checklist in the Handbook published by the Knowledge Centre. One author extracted the information, the other controlled that the correct information was extracted. For all effect estimates, we have assessed the documentation with GRADE.
Results
We found one systematic review of high quality that had performed searches for literature in March 2012.
The use of risk assessment tools compared to nurses’ judgment probably leads to small or no difference in the rate of falls (e.g. number of falls per person year) (RR 0.96, 95% CI 0.84-1.10) or number of fallers (person falling at least once) (RR 0.99, 95% CI 0.85-1.16) in care facilities. Vitamin D supplementation to patients with low level of vitamin D in care facilities will probably reduce the rate of falls (RR 0.63, 95% CI 0.46-0.86), but probably not the number of fallers (RR 0.99, 95% CI 0.90-1.08). The diagnostic accuracy of the STRATIFY rule is limited and should not be used in isolation for identifying individuals at high risk of falls in clinical practice.
Multifactorial interventions may probably reduce rate of falls in hospitals (RR 0.69, 95% CI 0.49-0.96). Multifactorial interventions in care facilities may not affect rate of falls (RR 0.78, 95% CI 0.59-1.04), number of persons falling (RR 0.89, 95% CI 0.77-1.02) or fractures (RR 0.56, 95% CI 0.3-1.03). The significance of such interventions is complex because there was a large variety of components, patients, duration and intensity in these interventions.
For several of other studied interventions there are large uncertainty about the results. We are uncertain if these interventions have an effect or not. We cannot conclude if the following interventions influence the rate of falls or the number of fallers or the number of fractures sustained by fall; exercise, medication review, encourage adoption of best practice strategies, very low bed, fall prevention tool kit software, educational materials. Patients in need of high level care might not have any effect of exercise, while other patients might.
Information given us imply that the risk assessment tools STRATIFY and Morse Fall Scale are the two most discussed in Norway. We found no systematic review published in 2010 or later that compared these two.
Discussion
Two examples of the content of multifactorial interventions are 1) educational program for staff on risk factors for falling, checklist over risk factors in the institution, education of the patients with written materials, offer of individual education or exercise in addition to group exercises of balance and strength 75 minutes twice a week, 2) exercise three times per week of 40 minutes each time for three months, education of staff, medication review, evaluation of risk factors in the institution and if necessary a visit to an optician or a foot therapist.
AHRQ have published a tool for institutions that plan to intensify their work on fall prevention. They emphasise that various institutions have different needs and resources and choice of implementation strategy will depend on customising to the specific institution.
The effect of vitamin D for the strength of the bone is well known, but vitamin D seems to improve muscle strength, function and balance in older persons, explaining the effect on falls.
Even though there is relatively much research on fall prevention in institutions, Cameron’s et al systematic review from 2013 state that there are large challenges and weaknesses in the research available. They ask for a large range of research projects.
Conclusion
Multifactorial institutions customised to the particular institution will probably reduce rate of falls in hospitals, but not the number of fallers or hip fractures. Vitamin D supplements for patients with low level of vitamin D will probably reduce rate of fall in care facilities, but not the number of fallers. Patient education by as dedicated nurse in acute care hospital will probably reduce the rate of falls. The difference of using risk assessment tools compared to nurses’ judgment in care facilities is probably small on rate of falls and number of fallers.