Sixteen studies (including two RCTs with 204 participants in total) were included in the assessment of intervention effectiveness. However, one hundred studies in total were reported including two RCTs, four non-randomised controlled trials, seven cohort studies, 20 case-control studies, 17 self-controlled studies, two uncontrolled clinical trials, 44 descriptive studies and four other studies.
Results from RCTs (n=2):
One RCT evaluated the effectiveness of a pressure sensitive alarm and the other the effectiveness of identification bracelets in a rehabilitation hospital. However, because of the lack or rigorous studies this evidence was graded poorly (i.e. grade IV).
Results according to intervention:
1. Alarm systems.
One RCT assessed the effectiveness of a bed alarm in an acute care setting. This study had a small sample size and failed to show any benefit of the intervention. The other study was uncontrolled and evaluated ambularms over a one month period. This system reduced the number of patient falls, but included a limited number of participants and the methodology was also flawed. Fall alarms were also evaluated as part of four multiple intervention studies. In conclusion there was no rigorous evidence on which to base recommendations.
2. Identification bracelets.
The RCT was set in a rehabilitation hospital and found that bracelets were of no benefit among patients at high risk of falling. Identification bracelets, and coloured stickers on the patient's chart, bed or door were also evaluated as part of ten multiple intervention studies. In conclusion there was no rigorous evidence on which to base recommendations.
3. Evaluation of patient fall data.
The study recommended evaluating incident form data to better understand falls (no further information reported).
4. Assessment of risk.
Nine studies were included in the review. No one risk assessment tool can be recommended, and the usefulness of these assessments in clinical practice has yet to be demonstrated.
4. Multiple interventions.
These studies typically included an assessment of a persons risk of falling, followed by the implementation of an intervention aimed at reducing these risks. All of the falls risk assessment tools used were self-developed. Some studies utilised different levels of intervention, in that as a patient's assessed risk of falling increased, so did the number of interventions employed. Rigorous research methods were not used to evaluate the use of multiple intervention programmes. A systematic review of falls in the elderly found that significant protection against falling was achieved by interventions which targeted multiple identified risk factors in individual patients in non-hospital setting. This, however, was not demonstrated in the acute care studies, which reported contradictory results. Due to the lack of rigour and contradictory findings no recommendations can be made about multiple interventions.
Overall quality of fall prevention studies:
The majority of studies involved reports from a single centre or institution, their time frame was limited and many involved only small numbers of participants. The reporting of research methods was poor including what interventions were used and how they were implemented. Commonly, when methods were reported they were not rigorous. The results were often incomplete and some studies failed to report any data at all.