In-hospital falls are common and can result in significant morbidity and mortality and additional hospital costs. The average rate of in-hospital falls is 4.8 falls per 1000 patient-days of observation based on incident reports.

The lower urinary tract symptoms of urinary urgency, incontinence, frequency and nocturia, have been identified as risk factors for in-hospital falls among older adults [1]. Older people with sleep disturbed due to nocturia have a markedly increased risk of injurious falls and death as compared to people who sleep through the night [2].

In a previous study, 50% of in-hospital falls was related to elimination, but the distribution between day and night was not described [3]. Taking into account the recommendations from the 6th international consultation on incontinence to perform research on the mechanisms, prevalence, incidence and remission rates of lower urinary tract symptoms in acutely hospitalised older people, the aim of this pilot study is to quantify the prevalence and fall-related injuries of in-hospital incident falls associated with nocturnal toileting.

This single-centre retrospective pilot study was conducted at a tertiary referral hospital. All patients that had a registered in-hospital incident fall between 1 january 2015 and 31 december 2015 were included in the study. Patient falls were reported by nurses into the hospital incident reporting and learning system (irl). 

Incident falls reported from paediatric, outpatient, diagnostic imaging, and emergency services were excluded as well as patients with incident falls during physical therapy sessions.

This database includes several variables (date, time interval, hospital service, fall mechanism and injury severity) and a description of the fall and injury. A fall was defined as an unexpected event in which the patient comes to rest on the ground, the floor or a lower level. Incident-falls associated with toileting were defined as a fall during an activity related to toileting, such as going to the toilet (or bedside commode) or using a urinal, dressing and cleaning. 

A pearson chi-square test was used to compare characteristics of incident falls and day- and night time. P-values were two-sided and considered statistically significant at a value of <.05.

In 2015 the rate of reported falls was 1.7 falls per 1000 patient-days of observation, which means 484 incident falls. Thirty-seven incident falls were excluded (no time registration (n=7); paediatric, outpatient, diagnostic imaging or emergency services (n=22) and physical therapy (n=8)). 

The retrospective evaluation was performed for 447 incident falls in 32 hospital services. Services with most reported incident-falls were geriatrics (n= 74), rehabilitation (n= 55), gastroenterology (n=34), haematology (n=30), and neurology (n=30).

Figure 1 shows characteristics of the incident falls. About half of the incident falls (n=242; 54%) occurred in the evening or during the night (8:00 pm-7:59 am). The most reported mechanisms of fall were loss of balance (26%) and slipping (25%). More than one third of the incident falls were related with going to the toilet (n=164, 37%). Incident falls associated with going to the toilet were more likely to occur during night time (n=125) than during daytime (n=39) (p<.001). 

Most of the patients sustained no injuries (n=259, 58%). Severity of injury is presented in table 1 and no associations were found between fall characteristics (time interval and activity) and presence of injury.

The fall rate of 1.7 falls per 1000 patient days and the proportion of incident falls associated with nocturnal toileting (28%) based on incident reports is probably an underestimation. First, approximately 20-25% of the incident falls are not reported in incident reports. However,
the rate of inpatient falls that were not reported in this incident reporting system is unknown.
Second, we couldn’t link cases to patients, because of data de-identification in the irl. To determine which incident falls were related to going to the toilet and the severity of injury we relied on the location and description of the fall and injuries by the reporter in the irl. 

In healthy people urine production during night time is lower than during daytime, hence one would expect the prevalence of incident falls associated with going to the toilet to be lower during night time than during daytime. In our study most of the incident falls associated with going to the toilet occurred during night time (76% vs 24%). 
However, the results from this single centre, retrospective study based on incident reports limits the interpretation of our results. In the light of our findings, performing a multi-centre prospective study to identify factors contributing to the fall risk of nocturnal toileting in hospital care is recommended.

More than one quarter of all falls reported in-hospital is associated with nocturnal toileting. Minimizing the onset or worsening of nocturia could help prevent some in-hospital falls.

Vaughan cp et al. (2010) The association of nocturia with incident falls in an elderly community-dwelling cohort. Int j clin pract; 64(5): 577-583.
Nakagawa et al. (2010) Impact of nocturia on bone fracture and mortality in older individuals: a japanese longitudinal cohort study. J urol; 184(4): 1413-1418.
Hitcho e., et al. (2004) Characteristics and circumstances of falls in a hospital setting. J gen itern med; 19(7): 732-739.