Association between toileting and falls in older adults admitted to the emergency department and hospitalised: a cross-sectional study

Abstract

Objectives

This study aimed to explore the potential risk factors associated with toileting-related falls in community-dwelling older adults who presented to the emergency department and were subsequently hospitalised.

Design

This was a cross-sectional study.

Setting and participants

This study was conducted in two teaching hospitals in Shanghai, China between October 2019 and December 2021 among community-dwelling adults aged ≥60 years.

Methods

In-person interviews, physical assessment and medical record review were performed to collect data on the characteristics and risk factors of falls. Associations of toileting-related falls with demographic characteristics and geriatric syndromes were examined using logistic regression models.

Main outcome measures

Potential risk factors for toileting-related falls.

Results

This study included 419 older patients with a mean age of 73.8±9.7 years. Among 60 (14.3%) patients with toileting-related falls (mean age: 78.8±9.2 years), 63.3% of toileting-related falls, mainly occurred between 00:00 and 05:59 hours, compared with 17.3% of non-toileting-related falls, which primarily occurred during the daytime. The rate of recurrent falls (35%) was significantly higher in the toileting-related falls group than in the non-toileting-related falls group (21.2%) (p=0.02). Logistic regression showed that visual impairment (OR 2.7, 95% CI 1.1 to 7.1), cognitive impairment (OR 3.3, 95% CI 1.3 to 8.4), gait instability (OR 3.1, 95% CI 1.1 to 8.8) and urinary incontinence (OR 3.4, 95% CI 1.2 to 9.9) were strongly associated with toileting-related falls. Twenty-three (38.3%) patients in the toileting-related falls group had moderate and severe injuries, compared with 71.7% in the non-toileting-related falls group (p<0.05).

Conclusions

This study revealed that patients who reported toileting-related falls were more likely to have cognitive impairment, urinary incontinence, gait instability, visual impairment than patients who fell during other activities. Social and healthcare professionals should prioritise the management of toileting activities in older patients and provide targeted interventions to those in the high-risk group.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• This study focused on toileting-related falls and the results provide substantial evidence for identifying patients at high risk of toileting-related falls, which can help tailor relevant interventions.

• This study included community-dwelling older adults who presented to the emergency department and were hospitalised after a fall.

• This study used reliable research tools to extensively assess the overall situation of the older adults.

• This study is limited due to its cross-sectional nature and inability to find causality.

Introduction

As the population ages, the number of older patients who accidentally fall gradually increases. Moreover, older patients who presented to the emergency department (ED) and were hospitalised often have multimorbidity, which significantly increases the likelihood of fall-related injury.^{1 2} However, medical institutions often pay more attention to treating fall-related injuries or illnesses in patients and less on the factors causing the falls. Many studies on fall prevention in older adults living in community settings exist;^{3 4} however, there is conflicting evidence about these interventions when applied to patients presenting to acute hospitals with a fall.^{5 6} This may be attributed to low levels of intervention prescribed by hospital staff or low acceptance of interventions by older adults.⁷ Nonetheless, ED visits and hospitalisation provide good opportunities for fall prevention education in older patients as they may be most receptive to fall prevention measures and hospital-based staff members, such as pharmacists and physical therapists, are available.

Toileting is an important factor that leads to falls and fall-related injuries for older patients.^{8 9} It is defined as a sequence of related tasks, which include getting to and from the toilet facility, transferring on and off the toilet or commode, being able to clean oneself afterward, as well as rearranging one’s garments on completion.¹⁰ Increasing age is associated with decline in function and increased multimorbidity,¹¹ and toileting difficulties are one of the challenges that older adults experience.^{12 13} Toileting-related falls occur more commonly at night among older adults with advancing age,¹⁴ and the risk of fall injuries in older adults increases with various multimorbidity or geriatric syndromes.¹⁵

Most studies have focused on urinary incontinence,¹⁶ nocturia,¹⁷ urinary urgency¹⁸ or lower urinary tract symptoms (LUTs) to manage risk factors for falls; however, several studies have reported different conclusions.^{19 20} Older adults may have one or more of these symptoms and studying only a single symptom at a time may affect the perception of the healthcare staff on risk factors of fall. In addition, studies on toileting-related falls are confined to either community²¹ or inpatient settings,²² and few have explored the fall status of patients presenting to the ED and on admission. The ED is typically the main channel wherein older patients enter the acute medical system. Older patients hospitalised due to falls may have more chronic diseases and are more prone to head and facial injuries and hip fractures than older patients who present to the ED but are not hospitalised.² As such, a fall with a fracture is more likely to lead to admission than a fall that causes a minor injury. This results in a significant increase in the use of related medical resources after a fall injury.²³

Understanding risk factors associated with toileting-related falls is the key to providing tailored interventions. This study focused on the characteristics of older adult patients who suffered from toileting-related falls and explored unique risk factors associated with this activity.

Methods

Patient and public involvement

Prior to the development of this study, we conducted a qualitative study among older adults who were hospitalised after a fall. These patients served as advisers on the design and conduct of this study. We found that toileting-related falls were common among this population.

Design and participants

A cross-sectional descriptive study was conducted at two teaching hospitals in Shanghai, China between October 2019 and December 2021. Each hospital has over 240 000 combined annual visits. Older patients aged ≥60 years, who presented to the ED and were hospitalised after a fall, were included in the study. A fall was defined as unintentionally coming to rest on the ground, floor or other lower level, excluding intentional changes in position to rest on furniture, walls or other objects.²⁴ We asked the patients or caregivers the question: ‘Are you in the hospital from the ED due to a fall?’ Patients who answered ‘yes’ were considered to have fallen and were asked additional questions about the time, number, locations and causes of their falls. Patients, who fell due to violent injury, loss of consciousness (fainting/syncope), or seizure; had a terminal illness; did not speak Chinese, or were positive for SARS-CoV-2, were excluded. Those who visited the ED only, were hospitalised but did not attend the ED, and those with a length of stay (LOS) <48 hours, were also excluded.

Instruments

We designed a structured questionnaire to survey the risk factors of falls through panel meetings and literature review.^{2 25–27} The survey questionnaire was divided into three sections: demographic information, fall characteristics and an assessment of comprehensive geriatric syndromes. Demographic information, such as sex, marital status, education level, financial resources, medication, smoking history, alcohol history, physical activity per week and body mass index, was investigated as confounders in the analysis of the association between toileting and falls.

Characteristics of falls included fall frequency, time, location, activity at the time of the fall and fall impact (site of injury on the body). Seasons were classified as spring (March–May), summer (June–August), autumn (September–November) and winter (December–February). Fall characteristics were determined through self-report. The degree of fall injury was divided into five grades according to the National Database of Nursing Quality Indicators,²⁸ as follows: no injury (patients did not show related symptoms and signs of postfall injury); mild injury (need for ice, dressing, local medication, limb elevation or other treatment for postfall injury); moderate injury (need for bandage, splint, debridement, suture or treatment for muscle strain and joint injury); severe injury (patients requiring plaster fixation, traction, surgery, neurological consultation (eg, skull fracture or epidural haematoma), treatment for organ injury (eg, hepatic or splenic laceration) or blood products after a fall) and death (some non-physical causes of fall leading to death).

The Continuity Assessment Record and Evaluation (CARE) is a comprehensive geriatric syndrome assessment tool for older adults developed by Medicare and the Paramedical Services in the USA. Our research team translated CARE into Chinese.²⁹ The internal consistency reliability of CARE had an overall Cronbach’s α coefficient >0.6 and intraclass correlation coefficient >0.8. The Chinese version of CARE has shown good reliability and validity in its application to older adults in acute and postacute institutions. CARE has four versions, namely institutional admission, home health admission, discharge and mortality. Each version uses the same evaluation items for the same concepts to facilitate comparison of the evaluation results. This questionnaire covers 13 geriatric syndromes. Polypharmacy is the use of ≥5 drugs. Cognitive function includes memory, time orientation and recall. Problems with any of these items were considered to indicate cognitive impairment. Visual impairment was defined as having difficulty in seeing in sufficient light, even when wearing glasses. Hearing impairment was defined as experiencing hearing difficulty (need for others to speak loudly) or inability to hear even with a hearing aid. Activities of daily living (ADLs) included five items, namely eating, oral hygiene, toileting, wearing clothes and wearing pants, with scores ranging from 1 (heavy dependency) to 6 (total dependence). Incontinence was defined as inability to voluntarily control urination or faeces. Gait instability was defined as needing help with walking across a small room and/or transferring from bed to chair.³⁰ Detailed explanations of these symptoms have been reported previously.^{29 31} The geriatric syndromes assessment component of the questionnaire was used in this study. Patients who had a low back pain are often complicated by knee pain and demonstrate differences in walking gait compared with back-healthy controls.³² Low back pain or knee pain was self-reported.

Data collection

After obtaining consent of the relevant departments of the hospital, the investigation was completed by two nurses who had received joint training on the purpose, content, method and interests of the investigation. Two methods were used to help the investigators capture the target population. First, the head nurses in the ward were informed of the inclusion and exclusion criteria of the study population. The assessment of fall risk was required for every patient admitted to the hospital, and if patients were admitted for a fall, the head nurse would be responsible for recording these patients’ bed number in an online word document every day. Second, fall risk assessments for all patients admitted to hospital were immediately made available through the nursing information system. The investigators had access to these records with the consent of the Nursing Information Management Council. Subsequently, the investigators interviewed the patients or family members for detailed information about the patient’s demographic data, fall characteristics, fall-related factors and geriatric syndromes. Information on fall outcomes, diagnosis of diseases, medications and medical history was obtained by reviewing medical records and through physical assessments by doctors.

Data analysis

Continuous values were compared between groups using t-tests or Mann-Whitney U tests, as appropriate, and were presented as the mean±SD, or median (IQR). Categorical values were analysed by using χ² tests or Fisher’s exact test, and expressed as frequency (percentage). The Bonferroni method was used for post hoc analyses. Multivariable logistic regression analysis was used to estimate the ORs and 95% CI, controlling for potential confounders. We used toileting-related falls as the dependent variable, and included nine significant risk factors in the analysis (age, living alone, polypharmacy, endocrine system diseases, cognitive impairment, urinary incontinence, impaired vision, gait instability and walking with a walking aid). The validity of all models was examined using the Hosmer-Lemeshow goodness-of-fit test. All tests were two tailed, and values of p<0.05 were considered statistically significant. All statistical analyses were performed using SPSS V.21.0 for Windows (IBM).

Results

Population characteristics

Of the 1000 older patients who presented to the ED and were hospitalised, 419 (41.9%) patients were included and analysed in this study. The mean age was 73.8±9.7 years, and 53.0% were female. The average LOS was 8.0±3.6 days. The three main diagnoses at admission were as follows: fractures in 220 (42.5%), strokes in 100 (23.9%) and traumatic cerebral haemorrhages in 30 (7.2%) patients.

Timing of falls

Falls related to toileting occurred in 60 patients (14.3%) with a mean age of 78.8±9.2 years. Of toileting-related falls, 63.3% occurred between 00:00 and 05:59 hours, whereas falls unrelated to toileting occurred primarily during the day (p<0.05). Moreover, the incidence of recurrent falls (35%) in the toileting-related falls group was significantly higher than that in the non-toileting-related falls group (21.2%) (p<0.05). Overall, 64.2% of falls occurred during the winter in both groups. Table 1 shows the characteristics of both categories of falls.

Table 1.

Characteristics of toileting-related falls versus non-toileting-related falls

Variable	Overall (n=419)	Toileting-related falls (n, %)		χ2	P value
		Yes (N=60)	No (N=359)
Time
6:00–11:59 hours	100	4 (6.7)	96 (26.7)	61.7	<0.001
12:00–17:59 hours	117	9 (15.0)	108 (30.1)
18:00–24:00 hours	77	8 (13.3)	69 (19.2)
00:00–05:59 hours	100	38 (63.3)*	62 (17.3)*
Unknown	25	1 (1.7)	24 (6.7)
Season of fall occurrence
Spring	34	10 (16.7)	24 (6.7)	17.8	<0.001
Summer	47	12 (20.0)	35 (9.8)
Autumn	69	13 (21.7)	56 (15.6)
Winter	269	25 (41.7)*	244 (68.0)*
Light during fall occurrence
Bright	348	30 (50)	318 (90.6)	65.1	<0.001
Dim	63	30 (50)	33 (9.4)
Location†
Indoors	237	60 (100)	177 (49.3)	54.3	<0.001
Outdoors	182	0	182 (50.7)
No of falls
≥2	97	21 (35.0)	76 (21.2)	5.5	0.02
1	322	39 (65.0)	283 (78.8)
Degree of fall injury
Uninjured	90	23 (38.3)	67 (18.7)	28.8	<0.001
Mild	48	14 (23.3)	34 (9.5)
Moderate	15	3 (5.0)	12 (3.3)
Serious	266	20 (33.3)*	246 (68.5)*
Fracture
Yes	245	12 (20.0)	233 (64.9)	43.3	<0.001
No	174	48 (80.0)	126 (35.1)
Hip fracture
Yes	98	6 (10.0)	92 (25.6)	7.0	0.01
No	321	54 (90.0)	267 (74.4)
Traumatic cerebral haemorrhage
Yes	25	8 (13.3)	17 (4.7)	6.8	0.01
No	394	52 (86.7)	342 (95.3)
Length of stay	419	8.2±3.2	7.9±3.7	0.6	0.44

*Indicates that in the post hoc analysis, the difference between the two groups is statistically significant.

†Fisher’s exact test.

Location and injury of toileting-related falls

For both groups, most falls occurred indoors (n=237, 56.6%). Twelve patients (20%) in the toileting-related falls group had fractures in, compared with 64.9% (n=233) in the non-toileting-related falls group (p<0.05). However, the rate of traumatic cerebral haemorrhage was higher (8 (13.3%)) in the toileting-related falls group than that in the non-toileting-related falls group (17 (4.7%); p<0.05). There were no significant differences in the LOS between the two groups (table 1).

Figure 1 shows the sites of injury of the 20 patients with severe injury in the toileting-related falls group. The three main injuries were femoral neck fracture (20.0%), rib fracture (23.3%) and subarachnoid haemorrhage (26.7%). The incidence of cerebral hernia, orbital fracture, nasal bone fracture and occipital bone fracture accounts for 3.3% each; haemothorax for 6.7% and subdural haemorrhage for 10%. In addition, 35% of the patients were injured in two or more sites.

Figure 1.

Types of injury in patients with toileting-related fall injuries.

Table 2 shows the demographic characteristics associated with toileting-related falls versus non-toileting-related falls. The two groups differed significantly in terms of age, the proportion living alone and the prevalence of endocrine system disease. Toileting-related falls occurred mainly in patients aged ≥80 years (56.7%), whereas the incidence of non-toileting-related falls was highest in patients aged 60–69 years (39.6%). The results of multiple comparison of age showed that there was significant difference between patients aged ≥80 years and other age groups (p<0.001).

Table 2.

Demographic characteristics of older adults associated with toileting-related falls versus non-toileting-related falls

Variable	Overall (N=419)	Toileting-related fall (n, %)		χ2	P value
		Yes (%) (N=60)	No (%) (N=359)
Sex
Male	225	29 (48.3)	169 (47.1)	0.9	0.86
Female	221	31 (51.7)	190 (52.9)
Age
60–69	156	14 (23.3)	142 (39.6)	20.1	<0.001
70–79	124	12 (20.0)	112 (31.2)
≥80	139	34 (56.7)*	105 (29.2)*
Education
Primary school or below	124	19 (31.7)	105 (29.2)	5.1	0.17
Junior high school	136	13 (21.7)	123 (34.3)
Senior high school	86	13 (21.7)	73 (20.3)
Graduate and above	73	15 (25.0)	58 (16.2)
Living alone
Yes	46	12 (20.0)	34 (9.5)	8.7	0.00
No	373	48 (80.0)	325 (90.5)
Drinking
Yes	102	19 (31.7)	83 (23.1)	2.2	0.15
No	317	41 (68.3)	276 (76.9)
Nervous system disease
Yes	149	23 (38.3)	126 (35.1)	0.3	0.63
No	270	37 (61.7)	233 (64.9)
Circulation system disease
Yes	239	40 (66.7)	199 (55.4)	2.7	0.10
No	180	20 (33.3)	160 (44.6)
Endocrine system disease
Yes	107	23 (38.3)	84 (23.4)	6.0	0.02
No	310	37 (61.7)	275 (76.6)
Low back pain or knee pain
Yes	131	14 (23.3)	117 (32.6)	2.2	0.15
No	287	46 (76.7)	242 (67.4)

*Significantly different from each other group.

The results of the univariate analysis of factors associated with toileting-related falls are shown in table 3. The incidence of toileting-related falls was significantly higher in patients using multiple medications, and in those with cognitive dysfunction, urinary incontinence, visual impairment, gait instability or who used a walking aid. The number of medications used by patients with toileting-related falls was significantly higher than that in patients with non-toileting-related falls.

Table 3.

Geriatric syndromes of older adults associated with toileting-related versus non-toileting-related falls

Variable	Overall (N=419)	Toileting-related fall (n, %)		χ2	P value
		Yes (%) (N=60)	No (%) (N=359)
Polypharmacy
Yes	181	44 (73.3)	137 (38.2)	27.6	<0.001
No	238	16 (26.7)	222 (61.8)
No of medications	419	5.5±3.4	3.4±3.2	20.5	<0.001
Hypoglycaemic drugs
Yes	76	14 (23.3)	62 (17.3)	20.5	0.30
No	343	46 (76.7)	297 (82.7)
Hypotensive drugs
Yes	212	29 (49.2)	183 (51.0)	0.1	0.80
No	207	31 (51.7)	176 (49.0)
Hypnotics
Yes	18	1 (1.7)	17 (4.7)	1.2	0.28
No	401	59 (98.3)	342 (95.3)
Cognition impairment
Yes	74	30 (50.0)	44 (12.3)	52.0	<0.001
No	288	22 (36.7)	266 (74.1)
Missing	57	8 (13.3)	49 (13.6)
Urinary incontinence
Yes	37	19 (31.7)	18 (5.0)	45.6	<0.001
No	337	35 (58.3)	302 (84.1)
Missing	45	6 (10.0)	39 (10.9)
Visual impairment
Yes	125	36 (60.0)	89 (24.8)	31.6	<0.001
No	239	17 (28.3)	222 (61.8)
Missing	55	7 (11.7)	48 (13.4)
Gait instability
Yes	169	43 (71.7)	126 (35.1)	29.1	<0.001
No	250	17 (28.3)	233 (64.9)
Walking aid
Yes	76	21 (35.0)	55 (15.3)	13.7	<0.001
No	284	31 (51.7)	253 (70.5)
Missing	59	8 (13.3)	51 (14.2)
Activities of daily living*
Independence	247	30 (50.0)	217 (60.4)	3.7	0.24
Partial dependency	111	22 (36.7)	89 (24.9)
Heavy dependency	4	0	4 (1.1)
Missing	57	8 (13.3)	49 (13.6)

*Fisher’s exact test.

In the logistic regression analysis, toileting-related fall was the dependent variable, and the statistically significant factors in the univariate analysis were included as independent variables. The following nine factors were included: age, living alone, polypharmacy, endocrine system diseases, cognitive impairment, urinary incontinence, impaired vision, gait instability and need for walking aid. Table 4 shows that visual impairment (OR 2.7, 95% CI 1.1 to 7.1), cognitive impairment (OR 3.3, 95% CI 1.3 to 8.4), gait instability (OR 3.1, 95% CI 1.1 to 8.8) and urinary incontinence (OR 3.4, 95% CI 1.2 to 9.9) were risk factors for toileting-related falls in older adults.

Table 4.

Multivariable logistic regression analysis of risk factors for toileting-related falls

	Estimate	SE	Wald test		OR	95% CI
			Wald statistic	P value
Constant	−3.5	0.5	48.7	<0.00	0.03	··
Visual impairment	1.0	0.5	4.2	0.04	2.7	1.1 to 7.1
Cognitive impairment	1.2	0.5	6.1	0.01	3.3	1.3 to 8.4
Gait instability	1.1	0.5	4.7	0.03	3.1	1.1 to 8.8
Urinary incontinence	1.2	0.6	4.9	0.03	3.4	1.2 to 9.9

Hosmer-Lemeshow test showed p=0.37.

Discussion

To the best of our knowledge, this study is the first to present a comprehensive analysis of risk factors for toilet-related falls focusing on older adults who presented to the ED and were hospitalised. We found that 14.3% of the participants had toileting-related falls, toileting-related falls occurred mainly at night and were more commonly experienced by patients aged ≥80 years. Fractures and traumatic intracerebral haemorrhage were the main injuries reported. Visual impairment, cognition impairment, gait instability and urinary incontinence were significantly associated with toileting-related falls. Few studies have assessed the risk factors for toileting-related falls, and thus the findings of our study have far-reaching implications.

Individuals with toileting-related falls accounted for the highest percentage of the participants in this study. The incidence of toileting-related falls is high among older patients, with an incidence ranging from 18% to 53%.^{22 33} The variance in the rate of incidence reported in the literature is a result of different sample populations and inconsistency in the definition of toileting-related falls. Most studies on toileting-related falls focused on hospitalised patients.^{22 34 35}Furthermore, to minimise sample heterogeneity, we excluded individuals with syncope-related falls whose assessment and management differ remarkably from those of individuals with non-syncope-related falls.³⁶ While there is an increasing number of studies on toileting-related falls; there is no consensus on the definition of a toileting-related fall.14 22 33 37 In addition, owing to the outbreak of COVID-19 during the 27-month study period, the investigators were sometimes restricted from collecting data in the wards due to hospital restrictions. Differences in the eligibility criteria may also have led to between the results of this study and other studies.

In this study, 56.7% of toileting-related falls occurred in older patients aged ≥80 years, consistent with previous findings.^{14 22 33} The incidence rate of LUTs increases with age,^{38 39} and older adults with more severe LUTs are more likely to be phenotypically frail⁴⁰ and to report difficulties with mobility and ADLs.⁴¹ Frequent toilet use combined with decreased mobility increases the risk for falls in older adults.⁴² However, age was not a risk factor for toileting-related falls in this study. This could be attributed to the relatively small number of patients with toileting-related falls, and thus the sample size was insufficient to assess the effect of age on falls.

There were 20 cases of serious injuries from toileting-related falls, and they mainly occurred at night (00:00–05:59 hours) during winter, which is consistent with the findings of other studies.^{22 37 43} At night, it cannot be assumed that older adults are sleeping soundly; they require sufficient support for their toileting needs in a timely manner. Interventions for high-risk older adults should include toileting schedules to facilitate safe use of the bathroom or provision of a bedside commode. Moreover, our findings revealed that toileting-related falls were frequent during winter. Cold stress increases sympathetic activity, which stimulates smooth muscle contraction in the prostate, and induces detrusor overactivity, which decreases the voiding interval and volume.⁴⁴ This, in turn, can increase the frequency and unpredictability of older adults’ toilet trips. Another possibility is that the data collection period encompassed more winter months than summer months. The results of this study suggest that social and healthcare professionals should pay attention to the groups at high risk of falls at night, especially during winter. Moreover, this study can help devise policies focus on fall prevention at night.

Our results showed that visual impairment is associated with toileting-related falls, which is consistent with other studies.^{24 45} Vision provides information about the surrounding environment, helping people to maintain a state of balance. Visual impairment may elicit fear of falling among older adults and restrict their activities, thus further increasing their risk of falling.²⁴

Cognitive impairment associated with falls has been reported in previous studies.^{46 47} In our study, 50% of individuals with toileting-related falls had impaired cognitive function, and this was significantly higher than that in the non-toileting-related falls group (12.3%). Older patients with cognitive impairment are at increased risk of reduced mobility and gait instability.⁴⁶ The incidence of toileting difficulty is higher in older adults with cognitive impairment than in those with normal cognitive function.⁴⁸ Severe cognitive impairment is associated with impaired ability to recognise or transfer to the toilet, difficulties dressing and reduced cognitive ability to respond to the sensation of a full bladder. Another possible reason is that older adults with cognitive impairment may be more reluctant to ask for help while performing toileting activities. These findings have implications for policy-makers regarding designing safer indoor environments and setting-up automatic alarm systems to prompt affected older adults to get up to perform toileting activities at night.

In this study, 71.7% of patients in the toileting-related falls group had gait instability, which was significantly higher than that in the non-toileting-related falls group. Gait instability can directly or indirectly lead to an increased risk of falling among patients going to the toilet on an uneven ground, such as stairs.⁴⁹ In addition, an urgent need to urinate increases gait speed and stride length, thus increasing the risk of falling.⁵⁰ This effect is closely related to the severity of urinary incontinence, and gait is closely related to complex cognitive function. When cognitive function declines, patients’ executive function is affected, which often leads to an abnormal gait.⁴⁹

An increasing number of studies have confirmed that urinary incontinence is closely related to falls.^{20 51} The risk of toileting-related falls in this study was 3.4 times higher in older patients with urinary incontinence than in those without, which is relatively higher than in previous reports.^{51 52} This may be due to the increased severity of illness of patients presenting to the ED and requiring hospitalisation. The strong urge to avoid urination may have made them rush to the bathroom and increased their risk of falling. This study did not examine the effect of the frequency of urinary incontinence on falls. Schluter et al⁵¹ showed that in men and women occasional urinary incontinence increased the risk of fall by 1.5 and 1.3 times, respectively, and frequent urinary incontinence increased the risk by 1.7 or 1.4 times, respectively. In addition, LUTs, such as urgency,⁵⁰ nocturia⁵³ and overactive bladder syndrome,⁵⁴ have been confirmed to be closely related to falls; however, they were not assessed in this study. The mechanism of falls caused by urinary incontinence remains unclear. In future studies, LUTs in older patients should be assessed in more detail.

Strengths and limitations

This study focused on toileting-related falls, which could help policy-makers in devising risk assessment for toileting-related falls as well as help healthcare providers develop targeted interventions. In addition, the ED is the main channel to hospital admission in older adults with a fall injury. Collecting appropriate ED and hospitalisation information is of considerable importance for evaluating patients’ situation and planning intervention and preventive strategies accordingly. This study used reliable research tools to extensively assess the overall situation of older adults. However, this study has some limitations that need to be addressed. First, the concept of toileting-related falls remains relatively vague and requires further clarification by investigating the toileting status of the target population. Second, only urinary incontinence was included in the study; other LUT symptoms were not evaluated. Third, the participants were only included if they presented to the ED and were hospitalised after a fall. These patients whose falls lead to the greatest healthcare utilisation are unlikely to be representative of all patients attending the ED with a fall. Moreover, patients with syncope-related falls were not included in this study. Syncope is defined as an abrupt, complete, transient loss of consciousness, associated with the loss of motor control and with rapid and spontaneous recovery. Patients with non-syncope-related falls may be affected by internal (eg, gait, balance, cognition) and external factors (eg, environment, dress). The assessment and management of these two types of falls differ greatly.³⁶ Overall, 23.9% of the patients who were included were assessed as having a fall because of stroke onset. There may be differences between falls caused by impaired physical function because of sudden onset of disease and falls with non-disease causes. This study did not investigate these differences. In addition, to prevent the spread of COVID-19 within the hospital, the investigators were sometimes prevented from collecting data in the wards due to hospital restrictions. Therefore, some eligible patients may not have been included.

Considering these limitations, generalisation to community-dwelling older people should be avoided. Moreover, due to ageing and poor health, some participants were unable to provide clear histories during the interview, resulting in some missing information. This may have led to bias in the results.

Conclusion

This study showed that patients with toileting-related falls more tended to have cognitive impairment, urinary incontinence, gait instability, visual impairment than patients who fell during other activities. Social and healthcare professionals should pay attention to the management of toileting of high-risk older adults and provide targeted interventions accordingly.

Data availability statement

All data relevant to the study are included in the article.

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

This study involves human participants and was approved by Ethics Committee of Xin Hua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine (grant number: XHEC-C-2022-110). Participants gave informed consent to participate in the study before taking part.