Falls in individuals with stroke during inpatient rehabilitation at a tertiary care hospital in Saudi Arabia

Discussion

Twenty four point seven percent of patients in our study had a fall, which is nearly average of what is previously reported in the international literature. The incidence of falls in acute care setting is comparatively less (8-22%) as compared to inpatient rehabilitation which ranges from 10-47%.13-16 The higher incidence of falls during inpatient rehabilitation compared to acute hospitalization could be due to increased mobilization, regained abilities with recovery, and longer LOS in the rehabilitation hospital than in acute care. Even though incidence of falls tends to decline over years among stroke survivors, the risk of falls after stroke is much higher as compared to general population of elderly people. Interestingly, system reviews and meta-analyses show that there is little evidence to support interventions, such as exercise, for preventing falls post stroke; however vitamin D and bisphosphonates have shown significant reduction in falls.17,18 The number of studies examining post stroke falls are limited or specific to a subgroup, rendering the need of further research to replicate these results before these recommendations are adapted in clinical practice.

It is reported that stroke survivors have twice the risk of fall as compare to other population.1 The mean age of patients in our study group is around 60 years which is consistent with other studies in Kingdom of Saudi Arabia; however, it is lower than other Middle Eastern and Western countries.19 There was no mean age difference between fallers and non-fallers in our study; however, analysis showed that the risk of fall in patients above 65 years of age was 3 times as compared to patients younger than 65 years. This is consistent with previous studies which report that risk of falls increases with increasing age.20 Results from the Behavioral risk factor surveillance survey, conducted by the Centers for Disease Control and Prevention, indicated that, 15.9% of adults aged 65 years or older reported falling in the first 3 months.21 Males are at higher risk of falls as compared to females; however, Jørgensen et al,20 reported that the incidence is equal in both gender. In our study, males had nearly 3/4 th of the total falls. Out of 50 females included in our study, 8 females sustained a fall during inpatient rehabilitation. It is interesting to note that none of the females had any injuries during falls. This may be attributed to the fact that, even if a bedside caregiver is not required, females prefer to be accompanied by a care giver during their inpatient rehabilitation, as a cultural norm, whereas males may or may not have bedside caregiver at all times. Laterality of stroke was strongly associated with outcomes of falls (injury or no injury) in our study. Fifty five point five percent of the fallers had left hemispheric stroke. Similar findings were reported by Alemdaroğlu et al,22 who found that patients with left hemispheric stroke were more likely to fall within 6 months of discharge from a rehabilitation hospital than were those with right hemispheric stroke. One possible explanation is that right hemispheric lesions tend to be better supervised given their impaired judgment and lack of insight into their problems, compared with persons with left hemispheric lesions who usually have communication deficits but visuomotor perceptual deficits are not considerable. Our study did not show any significant association between fall status and neglect, visual deficits or communication disorder. This is in contrast to previous studies which showed a higher incidence of falls in patients who had an inability to follow commands or had visual impairment.23,24 Visual deficits are of particular significance as they increase the likelihood of falls due to loss of depth perception and impaired visual spatial judgments.

The type of stroke is not known to be associated with fall risk; however, it had strong association with fall status (fallers or non-fallers) in our study.16 Hemorrhagic stroke was identified as an independent risk factor, and it was 3.5 times more likely to affect patients’ fall risk during inpatient rehabilitation. Anti-epileptics are commonly used prophylactically in patients with hemorrhagic stroke, and are associated with falls due their sedative effects; however, definite conclusions cannot be established from our study due to lack of statistical significance, even though nearly one third of patients using anti-epileptics had a fall during their rehabilitation course. Similarly, almost 25% of the patients using benzodiazepines, antihypertensive, diuretics and antidepressants fell during their rehabilitation course; however, medication use was not found to be significantly associated with fall status. Also, diabetes and hypertension had no significant association with fall status. There are several types of drugs which are associated with significant risk of falls including sedatives and hypnotics, antidepressants and benzodiazepines.25 Antidepressants have shown to have the strongest association with falls.26 An increased likelihood of falls was reportedly associated with neuroleptics, antipsychotics, and nonsteroidal anti-inflammatory drug. Campbell et al,27 reported a 66% reduction in falls when psychotropic drugs were gradually withdrawn compared with those who continued on them. Drug prescriptions for stroke survivors require close attention as drug metabolism, efficacy and adverse drug reactions vary between different patients. Additionally, side effects of drugs may increase the risk of falls due to associated neurological and cognitive impairment as a sequel of stroke. In our study, history of previous stroke was identified as an independent risk factor of patients’ fall status. This brings attention to various scoring systems used for determining risk of falls; most of which do not include history of previous stroke. As per the American Heart Association Guidelines for stroke rehabilitation, the most commonly used assessment instrument in the 43 prevention studies reviewed was the Morse fall scale.28 Morse score showed statistical significance with patients’ fall outcomes in our study. Sixty one percent of fallers had high Morse score (>45), out of which 22% sustained injury. Morse fall scale is validated for use in hospitalized patients and is a common screening instrument for screening falls in elderly patients.29 In our study, Morse scale was used as it is approved by the institution and conventionally used across the hospital; however, there are numerous other scales which can be used in fall risk assessment specifically in stroke patients, including Falls efficacy scale (FES), Stroke assessment of fall risk, Fall risk assessment tool (FRAT), or Tinetti performance oriented mobility assessment.30-33 Falls efficacy scale has been found to be a simple and valuable scale because it predicts falling risks within the first 6 months after discharge.34

Most of the falls in our study were unwitnessed and they took place during day time and inside patients’ room from the bed. There are limited number of studies regarding timing of falls. In one study, it was reported that 85% of the falls are reported to be during day time.32 However, in other types of brain injuries, falls are usually during night time which may be attributed to altered sleep wake cycle.35 There is no bed alarm system at our facility but they are commonly used in hospitalized patients. Falls were reduced by 20-60% in hospitals where alarm systems were used as primary intervention.36 One of the 2 cluster randomized studies included alarms as a part of fall prevention program. One trial was effective but, the other did not reduce falls.37,38 In our study 20% of the falls had injury. This is consistent with previous reported studies which report that the injury rates related to falls among patients undergoing inpatient rehabilitation range from 13-29%.9,39 The details of injuries was beyond the scope of our research; however, stroke patients have a 4-fold increased risk of hip fracture associated with a high incidence of falls.7

Literature review indicates that the falls increase the length of stay regardless of the degree of harm.40,41 This is also similar to our study, where the average LOS of fallers was around 55 days as compared to non-fallers who had an average LOS of 49 days, but was not found to be statistically significant. Among those who sustained a fall, there was a signficant statistical signficance of LOS between injured and non-injured. The length of hospital stay of stroke survivors who sustained injuries was 19 days on average as compared to the ones who did not sustain any injuries. This may be attributed to more time required for their treatment of injury, recovery and regaining the decline in function post falls.

In conclusion, age, hemorrhagic stroke, laterality of stroke and previous history of stroke were identified as risk factors associated with falls. Injured patients tend to stay longer for inpatient rehabilitation. Large multicentered studies are required on patients with stroke undergoing inpatient rehabilitation to determine the impact of falls in the Saudi population. Due to different culturual, environmental and health system variations, development of regional guidelines for fall prevention in stroke survivors can help to reduce the risk of falls.