Falls in older people: assessing risk and prevention

What the guideline recommends

The guideline recommends strength and balance training as one of the components of a successful multifactorial intervention programme for older people with recurrent falls or at increased risk of falling (recommendation 1.1.3.1). But recommendation 1.1.4.1 of the guideline also states that: "Strength and balance training is recommended. Those most likely to benefit are older people living in the community with a history of recurrent falls and/or balance and gait deficit. A muscle-strengthening and balance programme should be offered. This should be individually prescribed and monitored by an appropriately trained professional. "

Methods

Sherrington et al. 2014 was an Australian randomised controlled trial (RCT) that investigated the effects of a home-based exercise programme on falls and mobility among people aged 60 years and over (mean age of 81.2 years, 6.8 health conditions, 7.5 medications and 70% had fallen in the last year) that had recently been discharged from four public hospitals following admissions in geriatric, rehabilitation and orthopaedic wards (n=340).

Prospective candidates for the trial were approached during their admission in hospital and those that agreed to take part were contacted after discharge. Participants were stratified by hospital site and falls history, and block randomisation was used to allocate participants into an intervention group (n=171) and a control group (n=169). Trial investigators who conducted interviews and assessments, received calendars and questionnaires, made phone calls and entered data were blinded to participants' group allocation.

Both groups received usual care from health and community services and were given an education booklet about falls prevention. The intervention group participants were additionally asked to carry out a 20–30 minute programme of lower limb balance and strengthening exercises at home up to 6 times weekly for 12 months. Exercises were mainly conducted while standing, according to the Weight-bearing Exercise for Better Balance (WEBB) programme. Participants were supported by experienced physiotherapists with a total of 10 visits during the study period with more frequent visits at the beginning to ensure safety and enable tailoring and progression of the programme.

Primary outcomes were rate of falls, performance-based mobility and self-reported mobility. Secondary outcomes included additional measures of mobility, falls and risk of falling, strength, flexibility, quality of life, health system and community service contact, assistance required from others, difficulty with daily tasks, community participation and physical activity levels. The study also reported on adverse events of the exercise programme.

Rate of falls was calculated using records of falls completed by participants on monthly calendars and was assessed over the 12-month duration of the study. Performance-based mobility was measured using the Short Physical Performance Battery (score range 0 to 12, where a higher score is better), but data analysis was conducted using the lower extremity Summary Performance Score version of the tool, at baseline and 3 and 12 months post-randomisation (score range 0 to 3, where a higher score is better). Self-reported mobility was measured via a questionnaire that was completed monthly. Ease in performing eight mobility tasks was recorded using a five-point scale with the scores summed up (score range 0 to 40, where a higher score is better).

Results

Ninety-two per cent of participants (n=312) completed the 12-month assessment; 49% (n=168) reported a total of 300 falls during the study period, with participants in the intervention group reporting significantly more falls than the control group (177 falls versus 123 falls; incidence rate ratio 1.43, 95% confidence interval [CI] 1.07 to 1.93, p=0.017). There were 98 fallers in the intervention group (57%), compared to 70 in the control group (41%), risk ratio 1.38, 95% CI 1.11 to 1.73, p=0.004. However, no participants reported a fall while they were doing the home-based exercises.

There was a small but statistically significant improvement in performance based mobility at 12‑months in the intervention group compared to the control group (difference between groups change in mean score after adjusting for baseline performance 0.13, 95% CI 0.04 to 0.21, p=0.004). There was no significant difference between the intervention and control groups for self-reported ease in undertaking mobility tasks over the study period (difference between groups change in mean score 0.49, 95% CI -0.91 to 1.90, p=0.488).

Strengths and limitations

Impact on guideline

This is a single intervention in a highly specific clinical context – post discharge. In this population of older people recently discharged from hospital, this RCT found that home-based lower limb balance and strengthening exercises improved performance-based mobility but also significantly increased the rate of falls and number of fallers. The findings suggest, therefore, that this form of home exercise cannot be recommended as a standalone or single, 'blanket' intervention for fall prevention for this group. This result, however, may be out of step with other studies on exercise in falls prevention but most are done in healthier older people.

What the guideline recommends

The guideline does not recommend the use of falls risk prediction tools to predict inpatients' risk of falling in hospital.

Methods

The Healey and Haines 2013 study assessed the predictive values of the Morse falls score (MFS) in an acute general hospital in the UK. Age, ward speciality of admission, most recently completed MFS, and any falls in the subsequent 7 days were collected through case note review and incident reporting systems for analysis.

Six variables make up the MFS. There are: history of falling (scored as 0 or 25), secondary diagnosis (scored as 0 or 15), ambulatory aid (scored as 0, 15 or 30), intravenous therapy (scored as 0 or 20), gait (scored as 0, 10 or 20) and mental status (scored as 0 or 15). A total score of 0–20 denotes no or low risk, ≥25 denotes at least medium risk and ≥55 denotes high risk.

Analyses were carried out for MFS ≥25 and ≥55. Subgroup analyses by age (patients aged <75 years or ≥75 years) and specialty (patients on geriatrician-led wards or not on geriatrician-led wards) were also carried out. The Youden Index (YI) and its 95% confidence intervals (CIs) were used to determine the total predictive value of the MFS for the overall sample as well as for each of the pre-planned subgroups; sensitivities and specificities were also derived.

Results

Records from 467 patients were assessed; 35% (n=162) were on geriatrician-led wards. 116 patients were excluded from the data analysis as they didn't have a recent or completed MFS. Results showed a non-significant result for MFS ≥25 in the total sample (YI 0.15, 95% CI −0.04 to 0.32) or in any of the pre-planned subgroups.

An MFS ≥55 was significantly better than chance for predicting falls in the total sample (YI 0.39, 95% CI 0.20 to 0.58), in patients ≥75 years (YI 0.31, 95% CI 0.10 to 0.51) and in geriatrician-led wards (YI 0.37, 95% CI 0.15 to 0.58), although either sensitivity or specificity fell below 70% in each of these groups.

Strengths and limitations

Impact on guideline

This study concluded that overall, the Morse falls score (MFS) risk scoring tool was not satisfactory for predicting falls in hospital. The study findings are consistent with the guideline which does not recommend the use of falls risk prediction tools to predict inpatients' risk of falling in hospital. Therefore, this new evidence is unlikely to impact on guideline recommendations.

What the guideline recommends

The guideline recommends that for patients at risk of falling in hospital a multifactorial assessment and a multifactorial intervention should be considered. The guideline defines a multifactorial intervention as having multiple components that aim to address the risk factors for falling that are identified in a person's multifactorial assessment.

Methods

Sahota et al. 2014 was a randomised controlled trial (the REFINE trial) of bed and bedside chair pressure sensors compared with standard care in older people (mean age at randomisation 84.6 years) admitted to acute, general medical wards, in Queen's Medical Centre, Nottingham, United Kingdom. Both groups received standard care comprising routine geriatric medical care. The intervention comprised of battery-operated bed and bedside chair pressure sensor linked wirelessly to a handheld battery-operated radio-pager. When a participant left the bed or bedside chair, a radio signal alert was transmitted from a transmitter box attached to the foot of the participant's bed, to the radio-pager carried by a member of the nursing team, which provided the location of the participant. An absence of pressure on the sensor of 5 seconds or more triggered an alert. A central receiver on each ward recorded all alerts, which were collected by the research team.

All patients admitted to the hospital from the medical admissions unit to three acute, general medical elderly care wards within 24 hours were eligible for inclusion into the trial. Patients permanently bed bound prior to admission, moribund/unconscious, on end of life care or previously included in the trial on an earlier admission were excluded.

Participants were randomised to the intervention group (n=918) or the control group (n=921) using a web based service that utilised a permuted-block randomisation schedule for allocation to groups. Blinding of participants or those providing medical or nursing care to the intervention or control group was not carried out as a feasibility study prior to the trial showed that staff were able to identity patients with dummy or active sensors.

The primary outcome measure was the number of in-patient bedside falls per 1,000 bed days from time of randomisation to date of discharge, death or study withdrawal, whichever occurred soonest. A bedside fall was defined as an unexpected event in which the participant came to rest on the ground, floor or lower level in the area around the bedside, with the bedside being defined as the area encompassed by the curtained area surrounding the bed. For patients in side rooms, the bedside was defined as the area of the room.

Secondary outcome measures were: number of injurious in-patient bedside falls per 1,000 bed days, activities of daily living, fear of falling, length of hospital stay, residential status on discharge and health related quality of life. Mean costs of the intervention/control and QALY gains per patient were also reported.

Results

Of the eligible patients, 918 patients were randomised to the intervention group and 921 to the control (n=1839). Results for the primary outcome were as follows:

There was no significant difference between the two groups with respect to time to first bedside fall (adjusted hazard ratio (HR), 0.95, 95% CI 0.67 to 1.34, P=0.12). None of the secondary outcomes differed significantly between the two groups.

The mean cost per patient in the intervention group was £7199 compared with £6400 in the control group, mean difference in QALYs per patient was 0.0001; 95% confidence interval -0.0006 to 0.0004, P=0.67.

Strengths and limitations

Impact on guideline

This UK RCT of assistive technology found that bed and bedside pressure sensors using radio-paging as part of a single intervention do not reduce falls rates. This study helps raise awareness of the limitations of such interventions. These are important negative findings which lend support to the existing recommendation. The guideline recommends that for patients at risk of falling in hospital a multifactorial assessment and a multifactorial intervention should be considered. Therefore, this new evidence is unlikely to impact on guideline recommendations.

What the guideline recommends

The guideline endorses the need to understand which improvements to the inpatient environment are the most effective and cost-effective for preventing falls and injuries in hospital, and the factors that architects should take into account when designing new hospitals. Therefore, a research recommendation with the following wording, "What environmental adaptations can be made in existing inpatient units, and should be considered when inpatient units are built, to reduce the risk of falls and injuries in older inpatients?", was made in the guideline.

Methods

Drahota et al. 2013 conducted a non-blinded pilot cluster RCT to assess the feasibility, potential benefits and harms and to guide further research on the use of shock-absorbing flooring for fall-related injury prevention in elderly care wards.

Participants were identified and recruited in eight hospitals in England. One bay at each site was assigned as the 'study area' for 1 year. Block randomisation, using a computer-generated random list, was used to allocate hospital sites into intervention and control groups. The study investigators were blinded to the randomisation blocks and sequence until the sites had been allocated.

All adults admitted to a bed in the study area were eligible. Hospital wards predominantly for elderly care in England were eligible for inclusion, with no other location restrictions. No exclusion criteria were applied.

Intervention sites received an 8.3-mm vinyl floor over fibreglass mat with PVC foam backing which was installed in bedroom areas only. Control sites received no change in flooring. All sites had concrete subfloors. Data were collected for 2 to 5 months (median=4 months) before the floors were laid. Then, data were collected for a further 12–13 months.

The primary outcome was the fall-related injury rate per 1,000 occupied bed days (OBD). Secondary outcomes were: injury severity; fall rate per 1,000 OBD and adverse events.

Results

During the intervention period, 225 participants were recruited to the intervention group and 223 to the control group. Baseline characteristics showed similar medication usage across groups but the control group had more comorbidities associated with falls risk.

Of 35 falls (31 fallers) in the intervention group, 22.9% were injurious, compared with 42.4% of 33 falls (22 fallers) in the control group (injury incident rate ratio (IRR) 0.58, 95% CI 0.18 to 1.91). There were no moderate or major injuries in the intervention group and six in the control group. There were more falls in the intervention group compared to the control group (fall IRR 1.07 (95% CI 0.64 to 1.81). It is worth noting that both of these results were not statistically significant for the reported outcomes.

No adverse events were reported for participants but staff at intervention sites raised concerns about pushing equipment, with one documented case of a back injury. No adverse events related to flooring were reported from control sites.

Strengths and limitations

Impact on guideline

This UK pilot cluster RCT of shock-absorbing flooring to reduce injuries from falls in wards for older people found a higher rate of falls in the intervention compared to the control group, although this was not significant. There was a lower rate of injurious falls in the intervention group and higher rate of moderate or major injuries in the control group, although, again, the results were not statistically significant. The authors concluded that further research is required to assess to the risk of increasing fall rates with a shock-absorbing floor. Hence although this pilot study adds to the evidence base on preventing falls in older people during a hospital stay, further research is needed to answer the research recommendation in the guideline.