WK 8 DIS.DATA

7260  |  J Clin Nurs. 2023;32:7260–7272.wileyonlinelibrary.com/journal/jocn

Received: 12 December 2022  | Revised: 8 March 2023  | Accepted: 29 May 2023

DOI: 10.1111/jocn.16792

E M P I R I C A L R E S E A R C H Q U A N T I T A T I V E

The impact of safety culture, quality of care, missed care and nurse staffing on patient falls: A multisource association study

Faisal Khalaf Alanazi RN, BN, MNP, PhD Candidate1  | Samuel Lapkin RN, BN Hons (1st Class), Grad Cert Tertiary Ed, PhD MACN, Associate Professor, Honorary Associate Professor1,2  | Luke Molloy RN, BN, PhD, Senior Lecturer2  | Jenny Sim RN, BAppSc (Nurs), PhD MACN, Associate Professor, Honorary Associate Professor1,3,4

This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. © 2023 The Authors. Journal of Clinical Nursing published by John Wiley & Sons Ltd.

1School of Nursing, University of Wollongong, Wollongong, New South Wales, Australia 2Discipline of Nursing, Faculty of Health, Southern Cross University, Gold Coast, Queensland, Australia 3World Health Organization Collaborating Centre for Nursing, Midwifery & Health Development, University of Technology Sydney, Ultimo, New South Wales, Australia 4School of Nursing & Midwifery, University of Newcastle, Callaghan, New South Wales, Australia

Correspondence Faisal Khalaf Alanazi, School of Nursing University of Wollongong, Ave Wollongong NSW 2522, Australia. Email: [email protected]

Abstract Aims: To examine the association between nursing unit safety culture, quality of care, missed care and nurse staffing levels, and inpatient falls using two data sources: in- cidence of falls and nurses' perceptions of fall frequency in their units. The study explores the association between the two sources of patient falls and identifies if nurses' perceptions of patient fall frequency reflect the actual patient falls recorded in the incident management system. Background: Inpatient falls are associated with severe complications that result in ex- tended hospitalisation and increased financial consequences for patients and health- care services. Design: A multi- source cross- sectional study guided by the STROBE guidelines. Methods: A purposive sample of 33 nursing units (619 nurses) from five hospitals completed an online survey from August to November 2021. The survey measured safety culture, quality of care, missed care, nurse staffing levels and nurses' percep- tions of patient fall frequency. In addition, secondary data on falls from participating units between 2018 and 2021 were also collected. Generalised linear models were fitted to examine the association between study variables. Results: Nursing units with strong safety climate and working conditions and lower missed care were associated with lower rates of falls using both data sources. Nurses' perceptions of the frequency of falls in their units were reflective of the actual inci- dence rate of falls, but the association was not statistically significant. Conclusion: Nursing units with a strong safety climate and better collaborations be- tween nurses and other professionals, including physicians and pharmacists, were as- sociated with lower incidents of patient falls. Relevance to Clinical Practice: This study provided evidence for healthcare services and hospital managers to minimise patient falls.

    |  7261ALANAZI et al.

1  |  INTRODUC TION

Patient falls represent a significant threat to patient outcomes and should be minimised to keep patients safe. Falls with injuries, such as fractures, are serious health issues and may also cause other ad- verse patient outcomes, such as pressure injuries or hospital- acquired infections (Kim et al., 2022). The World Health Organization (2021) estimates that 684,000 fatal falls occur annually, making falls the sec- ond greatest cause of accidental death globally. Despite the preva- lence of patient falls in healthcare settings, the Centers for Disease Control and Prevention view them as preventable events (Centers for Disease Control and Prevention, 2020). Even though falls can be pre- vented using targeted interventions, preventing inpatient falls in acute care settings remains a significant challenge for healthcare managers (Morris et al., 2022). Nurses are in close contact to patients; therefore, their attitudes towards safety are essential in reducing and preventing patient falls.

2  |  BACKGROUND

Safety culture can be described as the individuals' values and be- liefs regarding what is essential in healthcare settings and their at- titudes and behaviour towards the appropriate and inappropriate actions and the management role in the rewards and punishments for better patient safety (Kakemam et al., 2021). Strong safety cul- ture is associated with a higher quality of care and better patient outcomes (Lee et al., 2019). Considering nurses represent the larg- est proportion of healthcare workers (Sim et al., 2019), examining the safety culture within nursing units and the association with patient falls is essential for patient safety and better healthcare services.

A systematic review by Alanazi et al. (2022) included seven stud- ies that examined patient falls, and only three reported significant as- sociations between nurses' positive safety attitudes and a reduction in patient falls. Two recent studies have also demonstrated a limited association between safety attitudes sub- scales and nurses' percep- tions of patient fall frequency (Kakemam et al., 2021; Yesilyaprak & Demir Korkmaz, 2023). Safety culture is a complex concept, and it is unclear which factors influence patient falls. Therefore, this study examined safety culture and identified which sub- scales of the safety attitudes questionnaire (SAQ) are associated with patient falls in the Saudi Arabian context.

Nurse staffing and nurses' attitudes towards quality and missed care are critical for patient safety. However, evidence of their asso- ciation with falls remains unclear. For example, Hessels et al. (2019) found no association between missed care and falls. A systematic review by Recio- Saucedo et al. (2018) reported that only two out of four studies found significant associations between missed care and patient falls. Tuinman et al. (2021) reported mixed results on the relationship between staffing levels and patient falls. A more re- cent study by Nantsupawat et al. (2022) reported higher numbers of patients per nurse were associated with higher missed care and increased adverse events; however, there was no significant associa- tion between missed care and patient falls. Nhongo et al. (2022) rec- ommended that more research on staffing levels and falls is needed. Therefore, the association between staffing levels, missed care and falls requires further investigation.

Patient falls are commonly examined using two sources of data: incident and survey data (Alanazi et al., 2022). Incident data from incident management systems are widely considered to be the gold standard for identifying patient falls (Brown & Wolosin, 2013; Taylor et al., 2012). However, non- reporting can be an issue with this method and nurses' perceptions of fall frequency in their unit are also considered a reliable measure (Sim et al., 2018). This study will collect patient falls data from both sources and investigate whether nurses' perceptions of fall frequency align with the actual number of reported falls.

Donabedian's (1988) conceptual framework was used to un- derstand the relationship between safety culture, quality of care,

Patient or Public Contribution: Patients who had experienced a fall, which was re- ported in the incident management system, from the included units in the five hospi- tals were part of this study.

K E Y W O R D S missed care, nurse staffing, nursing, patient falls, quality of care, safety culture

What does this paper contribute to the wider global clinical community?

• Nursing units with a strong safety climate and safety be- haviour reported lower incidence rates of falls and less frequent nurses' perceptions of falls occurrence.

• Higher hospital quality of care and lower missed care across nursing units were associated with lower falls rate and nurses' perceptions of falls with injury frequency.

• A positive association between the incidence rate of falls and nurses' perceptions of the frequency of falls, falls with injury and falls without injury was identified, but the association was not statistically significant.

13652702, 2023, 19-20, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1111/jocn.16792, W

iley O nline L

ibrary on [13/04/2025]. See the T erm

s and C onditions (https://onlinelibrary.w

iley.com /term

s-and-conditions) on W iley O

nline L ibrary for rules of use; O

A articles are governed by the applicable C

reative C om

m ons L

icense

7262  |    ALANAZI et al.

missed care, staffing levels and patient falls. This framework as- sesses healthcare quality using a three- part approach: structure, process and outcomes. Structural components, such as quality of care and staffing levels within the unit, may affect the process of nursing care through nursing unit safety culture and missed care. Both structural and process components have the potential to influ- ence patient outcomes, including patient falls.

2.1  |  Aims

1. To describe the incidence of patient falls and nurses' percep- tions of the frequency of patient falls in nursing units in five Ministry of Health (MOH) hospitals in Saudi Arabia.

2. To examine the influence of nursing unit safety culture, quality of care, missed care and nurse staffing levels on patient falls.

3. To explore the relationship between the incidence of patient falls as reported in administrative data and nurses' perceptions of the frequency of patient falls as reported in a cross- sectional survey.

3  |  METHODS

3.1  |  Design

A cross- sectional study was conducted in five hospitals and col- lected existing secondary data on falls from a MOH administrative dataset for the period between 2018 and 2021. The report of the study followed the guidelines recommended by the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE). This study is part of a doctoral project that aims to investigate the relationship between nurses' safety attitudes and nursing- sensitive patient outcomes (Alanazi et al., 2023).

3.2  |  Participants

Five acute public MOH hospitals were purposively selected to participate in this study. Hospitals had a minimum capacity of 200 beds and were from two regions in Saudi Arabia, three from the Riyadh region and two from the Eastern region. Hospital units that reported patient outcomes data on falls to the MOH dataset were included. Each participating unit was classified into one of the four categories: intensive care unit (ICU), medical, surgical or mixed medical- surgical. Nurses working in these units between August 2021 and November 2021 were invited to complete a cross- sectional survey. Nurses were eligible to participate if they had worked for at least 3 months at the current hospital, were able to read and understand English and worked within a participat- ing unit. Emergency departments, maternity units, palliative care units, operating theatres, recovery and outpatient departments were excluded from participating in the study as these units do not report patient falls to the MOH database.

Data from patients who had experienced a fall reported within the incident management system from the ICU, medical, surgical and mixed medical- surgical units in the five hospitals were included in this study. Data on patient fall incidents for the years 2018, 2019, 2020 and 2021 were requested from the MOH dataset.

3.3  |  Ethical considerations

Ethical approval was obtained from the Human Research Ethics Committee (HREC) of the University of Wollongong, Australia, on 29 June 2021 (Approval No. 2021/214). Other ethical approvals from Saudi Arabia were sought from the Ministry of Health (Approval No. 1443– 206207), King Saud Medical City (Approval No. H1RI- 18- Jul21- 01), and King Fahad Medical City (Approval No. 21- 296E).

3.4  |  Data collection

An electronic survey was developed and advertised via a single- page poster with a QR code and placed on the announcement board in each participating unit. Nurses in participating units received an email, with a reminder 2 weeks later, containing a link to the ques- tionnaire with information about the study. Once nurses scanned or accessed the survey, full instructions were given, including pro- cesses for consenting to participate in the study. Nurses had to identify the hospital and ward where they worked on the landing page of the survey, and therefore only eligible nurses were able to participate. In addition, 4 years of existing data on patient falls for the same units were requested from the MOH. Data were analysed, and the incident rates per 1000 inpatient days were calculated for each nursing unit.

3.5  |  Study variables

3.5.1  |  Nurse characteristics

Nurses' demographic data included age, gender, nationality, years of experience, qualifications, shift duration and details about nurses' work patterns.

3.5.2  |  Safety culture

Nursing units' safety culture was measured using the SAQ. The SAQ is a validated instrument developed by Sexton et al. (2006) and is widely used to collect individual and group- level attitudes towards six safety- related sub- scales (Churruca et al., 2021). The internal reliability of each sub- scale in this study ranged from .772 for teamwork to .894 for job satisfaction. The overall SAQ scale had a Cronbach's alpha of .956 in this study. The SAQ sub- scales originally included: teamwork climate, safety climate, job satisfaction, stress recognition, perceptions

13652702, 2023, 19-20, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1111/jocn.16792, W

iley O nline L

ibrary on [13/04/2025]. See the T erm

s and C onditions (https://onlinelibrary.w

iley.com /term

s-and-conditions) on W iley O

nline L ibrary for rules of use; O

A articles are governed by the applicable C

reative C om

m ons L

icense

    |  7263ALANAZI et al.

of management and working conditions. This study categorised the perception of management sub- scale into two different sub- scales, unit and hospital management, and calculated the safety behaviour sub- scale, which has been used in a recent study (Dickens et al., 2021) and the overall SAQ score. The stress recognition sub- scale was not included in the overall score as the subscale negatively correlated with other sub- scales and does not fit into the overall safety attitudes construct measured by the SAQ (Churruca et al., 2021). Therefore, in this study, 39 items from the SAQ were used to report the eight sub- scales. Responses were recorded on a five- point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree), and the average score of each sub- scale was calculated. Responses were converted to a per- centage ([mean of item − 1]*25), and a score of 75 or greater indicated a positive safety culture (Taylor et al., 2012).

3.5.3  |  Quality of care

Nurses' perceptions of quality of care were assessed using questions from RN4CAST and have been used in many nursing studies over the past decade (Lee et al., 2018; Sermeus et al., 2011; Smith et al., 2020). This study used three items to assess the quality of care in nursing units and hospitals. Nurses were asked to rate the overall quality of care provided in their unit and hospital. Responses were recorded on a four- point Likert scale (1 = poor, 2 = fair, 3 = good and 4 = excellent), then were dichotomised into a binary variable unfavourable (0 = poor and fair) and favourable (1 = good and excellent). The proportion of nurses who rated the quality of care as favourable was calculated for each unit.

3.5.4  |  Missed care

Missed care, also referred to as ‘care left undone’, was examined using questions from a validated instrument developed by the Basel Extent of Rationing of Nursing scale (Schubert et al., 2008). The questions have been used extensively in nursing research and were informed by previous studies, including RN4CAST (Ball et al., 2018; Nantsupawat et al., 2022; Sermeus et al., 2011). Nurses in this study were provided with a list of 13 essential activities and asked to select all care that was missed on their most recent shift. The Cronbach's alpha value for the 13 items of missed care in this study was .91. Missed care in each nursing unit was summarised as the proportion of nurses who missed at least one nursing activity.

3.5.5  |  Staffing

Nurse staffing levels were calculated as the patient- to- nurse ratio by collecting the total number of patients and nurses in each unit in the most recent shift. This approach is widely used in nursing research and can be a reliable method for estimating staffing lev- els (Ausserhofer et al., 2013; Nantsupawat et al., 2022; Sermeus et al., 2011). For analysis at the unit level, patient- to- nurse ratios

were calculated for each nurse's response, and the average patient load per nurse was reported for each nursing unit.

3.5.6  |  Falls

The incidence of patient falls was the primary outcome of this study. Patient falls data for the full years 2018, 2019, 2020 and 2021 for the participating units were requested from the MOH dataset. Data were provided in an Excel spreadsheet. Data were then shared with the quality departments in each hospital for validation to ensure data on falls had been correctly attributed to participating units. The final validated data set had a total of 214 fall incidents that occurred during 680,118 inpatient days in 33 nursing units. The incident rate of patient falls was calculated using the total number of falls divided by the total number of in- patient days multiplied by 1000 patient days for each nursing unit (Cina- Tschumi et al., 2009).

3.5.7  |  Nurses' perception of the frequency of falls

Nurses' perception of the frequency of overall patient falls, patient falls with injury and patient falls without injury were estimated using a seven- point Likert scale ranging from ‘1 = never’ to ‘7 = daily’. The responses were dichotomised to form a binary variable for analyses: 0 = ‘never happened’ (never and several times per year) and 1 = ‘had happened’ (at least once per month; several times per month; at least once per week; several times per week and daily) (Van Bogaert et al., 2014). Nurses' perceptions of the frequency of ad- verse events have been used extensively in international studies and reported to be reliable and valid measures (Kakemam et al., 2021; Lee et al., 2018).

3.6  |  Data analysis

Data analysis was performed using IBM SPSS Statistics (version 28.0). Missing value analysis was used to identify the proportion and pattern of missing data across all variables. The percentage of missing data ranged from .2% to 5.0%, and it was found that the data were not missing completely at random. Thus, a multiple im- putation approach was used to replace missing data for SAQ items (2.1%) (Boussat et al., 2021). For nurse staffing questions, missing data were managed using the approach described by Al- Ghraiybah et al. (2023). A variety of descriptive analyses were performed and included frequencies for categorical variables and means and stand- ard deviations for continuous variables. Medians and interquartile ranges (IQR) were used to describe skewed data. Analysis of vari- ance (ANOVA) was used to examine differences in the results of the study across hospitals (Table S1). The results demonstrated sig- nificant associations between hospitals and nursing unit types and variations in safety culture subscales, quality of care, missed care

13652702, 2023, 19-20, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1111/jocn.16792, W

iley O nline L

ibrary on [13/04/2025]. See the T erm

s and C onditions (https://onlinelibrary.w

iley.com /term

s-and-conditions) on W iley O

nline L ibrary for rules of use; O

A articles are governed by the applicable C

reative C om

m ons L

icense

7264  |    ALANAZI et al.

and staffing level; however, the study did not directly investigate the differences in these study variables, as this was beyond the scope of this study but is reported in Tables S1– S9.

Bivariate correlation was used to assess the strength of the correlation between the study variables. Generalised linear models (GLMs) were fitted to examine the association between safety cul- ture, quality of care, missed care, staffing and patient falls. Before fitting the GLM, the assumptions of normality and homoscedasticity were checked, and gamma with log link approach was deemed to be the most appropriate (Ng & Cribbie, 2017). Nursing units with zero falls rate (n = 3) were excluded from models as these values were in- valid for the gamma probability distribution. Models were controlled for nursing unit type, hospital size, years of experience and qualifi- cation of nurses as they are known to impact on nursing- sensitive patient outcomes (Ausserhofer et al., 2013). The association was considered significant at p < .05.

4  |  RESULTS

4.1  |  Demographics

A total of 33 nursing units from five MOH hospitals in two dif- ferent regions in Saudi Arabia participated in this study. Of the 898 potential participants, 619 responded to an online survey and identified their working unit with an average unit response rate of 68.9% (range: 33%– 100%). The average age of nurses was 33.7 years (SD = 5.76), and the average experience as a nurse was 9.2 years (SD = 4.92). The majority of nurses were non- Arabic (N = 463, 74.8%), female (N = 517, 83.5%) and had a bachelor's de- gree or higher (N = 543, 87.7%). Table 1 provides a summary of participant characteristics.

4.2  |  Nursing unit safety culture

The overall SAQ score and the subscale scores for nursing units by type, ICU or ward, are shown in Table 2. The average overall SAQ scores in ICU ranged from 65.8 (SD = 9.12) in hospital B to 57.7 (13.18) in hospital C. The average overall SAQ score in wards ranged from 63.5 (SD = 11.83) in hospital A to 57.7 (11.84) in hos- pital B. The highest- rated score for individual sub- scales in ICUs was observed for teamwork climate (Mean = 78.6, SD = 11.63), job satisfaction (Mean = 75.2, SD = 15.03) and safety behaviour (Mean = 73.1, SD = 14.23) in hospital B. The lowest score was for working conditions (Mean = 46.3, SD = 25.06) in hospital C and hos- pital management (Mean = 46.5, SD = 17.85) and working conditions (Mean = 46.5, SD = 18.65) in hospital A. The highest- rated score for individual subscales in wards was for safety behaviour (Mean = 70.1, SD = 15.15), safety climate (Mean = 69.9, SD = 14.79) and teamwork climate (Mean = 67.4, SD = 15.39) in hospital A. The lowest score in wards was for the working conditions (Mean = 45.6, SD = 24.14) in hospital C and stress recognition (Mean = 46.8, SD = 21.16) in hospital D. Only ICU units in hospital B recorded a positive safety

culture score of 75 or higher in teamwork climate and job satisfac- tion subscale.

4.3  |  Quality of care, missed care and staffing

Data on hospital and unit quality of care, nurse staffing and care left undone are presented in Table 3. Hospital quality of care was rated as favourable by participants, with hospital D being rated most highly in ICU (73.1%) and wards (67.4%). Participants rated the unit quality of care as favourable in the five hospitals. The highest proportions were in ICU units in hospital B (79.3%) and wards in hospital A (79.7%). Quality of care in the unit on the last shift was rated as favourable by participants, with ICU units in hospital D (79.5%) and wards in hospital A (83.6%) having the high- est ratings. Nurses who reported care activities missed on their last shift in ICU ranged from an average of 58.6% in hospital B to 87.3% in hospital A. The percentage of nurses who reported at least one care activity was missed in wards ranged from an average of 65.6% in hospital A to 92.6% in hospital C. For nurse staffing levels, the patient- to- nurse ratio across ICU units was 1.5 (SD = .09) in hospital A to 2.4 (SD = 1.20) in hospital D. The average number of patients per nurse in inpatient units was 4.4 (SD = .54) in hospital A to 7.3 (SD = 1.89) in hospital C.

4.4  |  Patient falls

Incidence rates of patient falls and nurses' perceptions of the fre- quency of patient falls are presented in Table 4. For the primary outcome, 214 falls were reported in 33 nursing units between 2018 and 2021. The number of reported falls per nursing unit var- ied considerably, ranging from 0 to 32 falls. The number of falls across the 4 years was highest in 2018 with 1.77 falls per 1000 inpatient days, and lowest in 2021 with 1.13 falls per 1000 inpa- tient days. Over the 4 years, the fall rate ranged from .00 to 1.14 per 1000 inpatient days, with ICU units recording the lowest fall rate (Median = .04 per 1000 inpatient days) and mixed medical- surgical wards recording the highest rate (Median = .39 per 1000 inpatient days).

Approximately 10% (IQR = 5%– 20%) of nurses reported that falls frequently happened in their ward at least once per month. The fre- quency of falls without injury (Median = 10%, IQR = 0%– 21%) was higher than falls with injury (Median = 8%, IQR = 0%– 15%). Nurses working in surgical units reported a higher frequency of patient falls (Median = 13%, IQR = 0%– 42%).

4.5  |  Association between safety culture, quality of care, missed care, staffing and patient falls

GLMs were fitted between 15 independent variables, including SAQ, quality of care, missed care, nurse staffing and four different out- comes; incidence rate of patient falls, nurses' perceptions of overall

13652702, 2023, 19-20, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1111/jocn.16792, W

iley O nline L

ibrary on [13/04/2025]. See the T erm

s and C onditions (https://onlinelibrary.w

iley.com /term

s-and-conditions) on W iley O

nline L ibrary for rules of use; O

A articles are governed by the applicable C

reative C om

m ons L

icense

    |  7265ALANAZI et al.

falls, falls with injury and falls without injuries (Table 5). Since bivari- ate correlation results indicated multicollinearity between individual SAQ subscales (r = .60– .87), each subscale was modelled separately using the same control variables for each outcome. Of the 15 vari- ables, 9 were associated with the incidence of patient falls, 6 with the nurse's perceptions of falls, 10 with falls with injury and six with falls without injury.

4.5.1  |  Incidence of patient falls

Six of the eight safety attitudes sub- scales and the overall SAQ score were found to have a negative impact on the incidence of patient falls (See Table 5). The top three predictors were the over- all SAQ score, working conditions and perception of hospital man- agement. Higher overall SAQ scores (β = −1.699, CI [−1.699, −.748], p = <.001), improved working conditions (β = −1.429, CI [−2.058, −.801], p = <.001) and positive perceptions of hospital management (β = −1.414, CI [−1.905, −.922], p = <.001) were associated with lower incidence rates of patient falls. Strong hospital quality of care ratings were associated with lower incidence of patient falls (β = −1.456, CI [−2.258, −.655], p = <.001). Higher proportions of missed care were

associated with higher rates of patient falls (β = 1.269, CI [.340, 2.199], p = .007).

4.5.2  |  Nurses' perceptions of patient fall frequency

Only two of the eight safety attitudes sub- scales and the overall SAQ score were statistically significant predictors of nurses' perceptions of the frequency of overall patient falls, and patient falls without in- jury. Higher scores of safety climate were negatively associated with the frequency of overall falls (β = −1.309, [CI −2.541, −.077], p = .037) and falls without injury (β = −1.425, CI [−2.612, −.238], p = .019). Nursing units with strong safety behaviour reported less frequent overall falls (β = −.802, CI [−1.515, −.089], p = .027) and falls without injury (β = −.847, CI [−1.545, −.148], p = .017). Hospital and unit rat- ings for quality and quality of care on the most recent shift were also negatively associated with less frequent overall falls and falls with- out injury. The highest associations were found between the hos- pital quality of care and overall falls (β = −1.789, CI [−3.139, −.439], p = .009) and falls without injury (β = −1.955, CI [−3.438, −.473], p = .010). Nursing units with higher proportions of missed care re- lated to ‘adequate patient surveillance’ reported higher frequency

TA B L E 1  Nurses' characteristics (n = 619).

Variable

All hospitals N = 619

Hospital A N = 214 Hospital B N = 69

Hospital C N = 104

Hospital D N = 129

Hospital E N = 103

Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD)

Age (years) 33.7 (5.76) 34.4 (4.21) 34.1 (5.35) 34.4 (6.75) 33.3 (7.30) 31.7 (5.23)

Experience (years) 9.2 (4.92) 10.4 (3.63) 9.3 (5.55) 9.4 (5.77) 8.4 (5.75) 7.5 (4.21)

Frequency (%)

Gender

Male 84 (13.6) 53 (24.8) 7 (10.1) 9 (8.7) 11 (8.7) 4 (3.9)

Female 517 (83.5) 152 (71.0) 61 (88.4) 91 (87.5) 89.9 (91.3) 97 (94.2)

Background

Arabic 154 (24.9) 18 (8.4) 40 (58.0) 39 (37.5) 39 (30.2) 18 (17.5)

Non- Arabic 463 (74.8) 195 (91.1) 29 (42.0) 65 (62.5) 89 (69.0) 85 (82.5)

Qualification

Diploma 76 (12.3) 2 (.9) 20 (29.0) 19 (18.3) 27 (20.9) 8 (7.8)

Bachelor or higher 543 (87.7) 212 (99.1) 49 (71.0) 85 (81.7) 102 (79.1) 95 (92.2)

Shift duration

8 h 204 (33.0) 15 (7.0) 12 (17.4) 61 (58.7) 51 (39.5) 65 (63.1)

12 h 402 (64.9) 195 (91.1) 56 (81.2) 40 (38.5) 77 (59.7) 34 (33.0)

Job category

In charge\Head nurse

138 (22.3) 63 (29.4) 17 (24.6) 33 (31.7) 13 (10.1) 12 (11.7)

Staff nurse 480 (77.5) 150 (70.1) 52 (75.4) 71 (68.3) 116 (89.9) 91 (88.3)

Job status

Temporary 57 (9.2) 5 (2.3) 6 (8.7) 10 (9.6) 22 (17.1) 14 (13.6)

Permanent 555 (89.7) 207 (96.7) 61 (88.4) 92 (88.5) 107 (82.9) 88 (85.4)

Abbreviation: SD, standard deviation.

13652702, 2023, 19-20, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1111/jocn.16792, W

iley O nline L

ibrary on [13/04/2025]. See the T erm

s and C onditions (https://onlinelibrary.w

iley.com /term

s-and-conditions) on W iley O

nline L ibrary for rules of use; O

A articles are governed by the applicable C

reative C om

m ons L

icense

7266  |    ALANAZI et al.

of overall falls (β = 1.709, CI [.599, 2.820], p = .003) and falls without injury (β = 1.799, CI [.734, 2.864], p = <.001).

Four of the eight safety attitudes sub- scales and the overall SAQ score negatively predicted nurses' perceptions of patient falls with injury. Safety climate was the strongest predictor of nurses report- ing fewer falls with injury (β = −.908, CI [−1.567, −.249], p = .007). Hospital, unit and shift quality of care were also negatively associ- ated with a lower frequency of falls, with shift quality of care being the strongest predictor (β = −1.021, CI [−1.807, −.234], p = .011). Nursing units with higher numbers of missed care activities also had a higher incidence of patient falls with injury (β = 1.019, CI [.343, 1.694], p = .003). In addition, a higher patient load per nurse was positively associated with a higher frequency of falls with injury (β = .200, CI [.087, .314], p = <.001).

4.6  |  Association between the rate of patient falls, and nurses' perception of the frequency of patient falls

GLMs were conducted to examine the association between the two sources of patient fall data while controlling for nursing unit type, hospital size and nursing experience. The association between the incidence of patient falls and nurses' perceptions of the frequency of overall falls (β = .164, CI [−.045, 2.590], p = .058), falls without injury (β = .163, CI [−.062, 2.620], p = .062) and falls with injury (β = .874, CI [−.346, 2.094], p = .160) was not statistically significant (See Table 5). The directions of the association between the incidence rate of pa- tient falls and the three outcomes of nurses' perceptions of fall fre- quency were positive, indicating higher nurses' perceptions of the frequency of falls may likely reflect the actual incidence of patient falls.

5  |  DISCUSSION

The study examined the association between nursing unit safety cul- ture, quality and missed care, nurse staffing levels and patient falls collected from two sources. In addition, the study also explored the relationship between the two sources of falls; the incidence rate of patient falls and nurses' perceptions of the frequency of patient falls. The study supports the framework proposed by Donabedian (1988), which emphasises the importance of structure, process and out- comes. Patient outcomes, such as patient falls, are influenced by the processes of nursing care, such as safety culture. Specifically, this study found that nursing units with a strong safety culture, good working conditions and effective collaboration between healthcare providers were associated with lower rates of patient falls. There was a positive relationship between nurses' perceptions of the fre- quency of falls, falls with injury, and falls without injury and the in- cidence rate of falls; however, the association was not statistically significant.

TA B

LE 2

  Av

er ag

e sc

or es

o f n

ur si

ng u

ni ts

' s af

et y

cu ltu

re p

er h

os pi

ta l.

H os

pi ta

l A H

os pi

ta l B

H os

pi ta

l C H

os pi

ta l D

H os

pi ta

l E

M ea

n (S

D )

M ea

n (S

D )

M ea

n (S

D )

M ea

n (S

D )

M ea

n (S

D )

IC U

W ar

ds IC

U W

ar ds

IC U

W ar

ds IC

U W

ar ds

IC U

W ar

ds

n =  8

3 n =

 1 31

n =  3

1 n =

 3 8

n =  47

n =  5

7 n =

 8 0

n =  4

9 n =

 3 9

n =  6

4

O ve

ra ll

SA Q

s co

re 58

.0 (1

1. 88

) 63

.5 (1

1. 83

) 65

.8 (9

.1 2)

57 .7

(1 1.

84 )

57 .7

(1 3.

18 )

58 .8

(1 2.

99 )

62 .5

(1 4.

55 )

59 .1

(1 6.

16 )

61 .8

(1 2.

24 )

59 .1

(1 3.

22 )

Te am

w or

k cl

im at

e 63

.5 (1

6. 48

) 67

.4 (1

5. 39

) 78

.6 (1

1. 63

) 59

.7 (1

9. 21

) 62

.9 (1

5. 41

) 61

.5 (1

5. 94

) 64

.2 (1

6. 36

) 61

.1 (1

7. 40

) 67

.4 (1

2. 57

) 61

.4 (1

4. 85

)

Sa fe

ty c

lim at

e 64

.0 (1

3. 29

) 69

.9 (1

4. 79

) 72

.0 (1

0. 08

) 59

.0 (1

4. 96

) 59

.7 (1

8. 76

) 62

.3 (1

5. 96

) 65

.8 (1

4. 96

) 64

.1 (1

6. 92

) 65

.7 (1

4. 92

) 65

.1 (1

4. 69

)

Jo b

sa tis

fa ct

io n

60 .1

(2 2.

20 )

67 .9

(1 9.

27 )

75 .2

(1 5.

03 )

62 .3

(1 7.

31 )

67 .2

(2 0.

98 )

67 .0

(1 9.

06 )

68 .4

(2 3.

10 )

62 .6

(2 4.

22 )

64 .4

(1 6.

23 )

64 .0

(1 9.

80 )

U ni

t m an

ag em

en t

57 .4

(1 6.

10 )

63 .6

(1 4.

59 )

64 .4

(1 5.

21 )

59 .9

(1 2.

60 )

54 .9

(1 4.

33 )

58 .7

(2 1.

07 )

62 .9

(1 7.

57 )

59 .4

(2 0.

27 )

62 .8

(1 7.

91 )

58 .1

(1 8.

76 )

H os

pi ta

l m

an ag

em en

t 46

.5 (1

7. 85

) 55

.1 (1

8. 07

) 54

.2 (1

9. 24

) 57

.3 (1

5. 94

) 51

.7 (1

7. 76

) 51

.2 (2

0. 63

) 61

.8 (1

9. 69

) 56

.2 (2

2. 10

) 58

.2 (1

8. 58

) 53

.6 (2

1. 78

)

St re

ss re

co gn

iti on

65 .7

(2 6.

14 )

62 .7

(2 3.

52 )

52 .7

(2 4.

37 )

49 .1

(1 9.

28 )

56 .8

(2 2.

46 )

62 .1

(2 0.

59 )

50 .6

(2 1.

25 )

46 .8

(2 1.

16 )

56 .1

(2 3.

14 )

52 .0

(2 1.

15 )

W or

ki ng

c on

di tio

ns 46

.5 (1

8. 65

) 53

.8 (1

9. 77

) 54

.6 (2

1. 71

) 52

.5 (1

7. 22

) 46

.3 (2

5. 06

) 45

.6 (2

4. 14

) 58

.2 (2

2. 78

) 56

.9 (2

3. 80

) 52

.7 (1

6. 15

) 52

.3 (2

0. 46

)

Sa fe

ty b

eh av

io ur

62 .2

(1 6.

58 )

70 .1

(1 5.

15 )

73 .1

(1 4.

23 )

61 .1

(2 1.

69 )

65 .7

(1 9.

95 )

64 .3

(1 8.

95 )

67 .2

(2 0.

74 )

62 .4

(2 3.

64 )

68 .0

(1 7.

26 )

66 .7

(1 9.

01 )

A bb

re vi

at io

n: S

D , s

ta nd

ar d

de vi

at io

n.

13652702, 2023, 19-20, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1111/jocn.16792, W

iley O nline L

ibrary on [13/04/2025]. See the T erm

s and C onditions (https://onlinelibrary.w

iley.com /term

s-and-conditions) on W iley O

nline L ibrary for rules of use; O

A articles are governed by the applicable C

reative C om

m ons L

icense

    |  7267ALANAZI et al.

The study found that the rate of patient falls varied significantly between nursing units, with a mean falls rate of .25 falls per 1000 patient days. While this rate is lower than that reported in previous research (Bouldin et al., 2013; Toyabe, 2015), there are several pos- sible reasons why this study found a lower falls rate than the other studies. One possible explanation is the study sample. Our study in- cluded a lower proportion of medical wards, and medical wards are known to have a higher falls rate than surgical and critical care units. Bouldin et al. (2013) have reported a higher falls rate of 4.03 per 1000 patient days in medical units compared to 2.76 per 1000 pa- tient days in surgical units. Under- reporting of falls may also explain the low incidence rate of falls in our study. Under- reporting of falls is a known issue in healthcare settings and can lead to an underes- timation of falls rates (Cina- Tschumi et al., 2009; Toyabe, 2015). A 3- month prevalence study in a university hospital in Saudi Arabia reported a falls rate of 2.4% of admitted patients (n = 27 falls) in medical and surgical wards (Innab, 2022). The findings of this study suggest that patient falls may be under- reported in MOH hospitals in Saudi Arabia, which is supported by the nurse perception data indicating that some nurses reported falls occurring ‘at least once per month’ to ‘daily’ in their wards. To address the under- reporting issue, multiple data sources, including real- time observational data, may be needed (Toyabe, 2015). Accurate reporting of falls rates and developing effective prevention strategies are crucial to improve pa- tient safety.

The present study contributes to the literature on the rela- tionship between safety culture and patient falls in nursing units. Specifically, the findings suggest that nursing units with a strong safety climate, job satisfaction, unit and hospital management sup- port, working conditions and safety behaviour experienced fewer patient falls. Similarly, nursing units with a strong safety climate and safety behaviour were associated with lower nurses' perceptions of the frequency of falls. However, previous studies have reported in- consistent findings regarding the association between nursing unit safety culture and patient falls (Ausserhofer et al., 2013; Brown & Wolosin, 2013; Hessels et al., 2019; Taylor et al., 2012). The discrep- ancies in the results could be attributed to differences in the settings and the sample size of nursing units, the nature of the data and the analysis used in their study.

One of the noteworthy findings of the present study is that nursing units with strong collaboration between nurses, staff phy- sicians and pharmacists were associated with fewer falls incidents and lower nurses' perceptions of the frequency of falls. Given that nurses are the primary providers of hospital care, their perception of safety culture is crucial in preventing patient falls. Therefore, better collaboration between healthcare providers could play a vital role in enhancing patient safety and reducing the incidence of falls in nurs- ing units.

In this study, nurses' perceptions about quality and missed care were found to be associated with the incidence of falls and their per- ceptions of the frequency of falls. While Nantsupawat et al. (2022) did not find a significant association between higher levels of missed care and nurse- reported patient falls, our study showed that nursing TA

B LE

3  

D es

cr ip

tiv e

su m

m ar

y of

th e

qu al

ity o

f c ar

e, m

is se

d ca

re a

nd s

ta ff

in g

ac ro

ss n

ur si

ng u

ni ts

p er

h os

pi ta

l.

H os

pi ta

l A H

os pi

ta l B

H os

pi ta

l C H

os pi

ta l D

H os

pi ta

l E

IC U

W ar

ds IC

U W

ar ds

IC U

W ar

ds IC

U W

ar ds

IC U

W ar

ds

n =  7

8 n =

 1 28

n =  2

9 n =

 3 3

n =  4

5 n =

 5 4

n =  7

8 n =

 4 3

n =  3

6 n =

 6 3

Fr eq

ue nc

y (%

)

H os

pi ta

l q ua

lit y

of c

ar e

(fa vo

ur ab

le )

33 (4

2. 3)

73 (5

7. 0)

20 (6

9. 0)

16 (4

8. 5)

22 (4

8. 9)

32 (5

9. 3)

57 (7

3. 1)

29 (6

7. 4)

25 (6

9. 4)

40 (6

3. 5)

U ni

t q ua

lit y

of c

ar e

(fa vo

ur ab

le )

41 (5

1. 6)

10 2

(7 9.

7) 23

(7 9.

3) 18

(5 4.

5) 26

(5 7.

8) 35

(6 4.

8) 58

(7 4.

4) 31

(7 2.

1) 26

(7 2.

2) 44

(6 9.

8)

U ni

t l as

t s hi

ft q

ua lit

y of

c ar

e (fa

vo ur

ab le

) 45

(5 7.

7) 10

7 (8

3. 6)

22 (7

5. 9)

25 (7

5. 8)

31 (6

8. 9)

37 (6

8. 5)

62 (7

9. 5)

31 (7

2. 1)

27 (7

5. 0)

47 (7

4. 6)

M is

se d

ca re

69 (8

7. 3)

84 (6

5. 6)

17 (5

8. 6)

28 (8

2. 4)

38 (8

2. 6)

50 (9

2. 6)

49 (6

1. 3)

29 (6

5. 9)

26 (7

0. 3)

44 (6

9. 8)

M ea

n (S

D )

St af

fin g

(p at

ie nt

s pe

r n ur

se )

1. 5

(.0 9)

4. 4

(.5 4)

1. 7

(.1 0)

6. 0

(1 .0

1) 1.

8 (.1

0) 7.

3 (1

.8 9)

2. 4

(1 .2

0) 5.

0 (.7

4) 2.

1 (.0

5) 4.

7 (.7

8)

A bb

re vi

at io

n: S

D , s

ta nd

ar d

de vi

at io

n.

13652702, 2023, 19-20, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1111/jocn.16792, W

iley O nline L

ibrary on [13/04/2025]. See the T erm

s and C onditions (https://onlinelibrary.w

iley.com /term

s-and-conditions) on W iley O

nline L ibrary for rules of use; O

A articles are governed by the applicable C

reative C om

m ons L

icense

7268  |    ALANAZI et al.

units with higher ratings for hospital quality of care and lower lev- els of missed care were associated with lower incidence of falls and reduced nurses' perceptions of the frequency of falls with injury in their unit. Furthermore, nursing units with higher levels of missed ‘adequate patient surveillance’ were associated with higher percent- ages of nurses reporting falls with and without injury occurring in their unit during the last year. A similar study by Ball et al. (2018) also reported higher ratings of nursing missed care in nursing units with higher incidence rates of falls. These findings suggest missed care has a significant impact on patient outcomes, and the reasons why nurses left care undone could be related to structural elements, such as staffing levels, where nurses handle more patients and are, therefore, unable to complete all tasks. Staffing levels have previ- ously been associated with higher missed care and lower quality of care (Nantsupawat et al., 2022).

Nurse staffing has been identified as a crucial factor in reduc- ing the incidence of patient falls in healthcare facilities. While our study did not find a significant relationship between staffing levels and patient falls recorded in the administrative data, we observed that a higher number of patients per nurse was associated with a higher percentage of nurses reporting patient falls with injury had occurred ‘at least once per month’ to ‘daily’ in their unit. This finding is supported by previous research indicating that an increase in the number of nurses is associated with a reduction in falls among hospi- talised patients (Bowden et al., 2019; Kalisch et al., 2012). However, the association between nursing staffing and falls is not always con- sistent (Griffiths et al., 2016; Shin et al., 2019; Tuinman et al., 2021). One possible explanation for this inconsistency is the varied meth- ods used to measure nurse staffing and patient falls. Nurse staffing can be measured using different sources, such as nursing hours per patient day, nurses' reported data and hospital administrative data (Shin et al., 2019; Twigg et al., 2021). Moreover, the unit type and the study setting, such as long- term facilities or acute care hospi- tals, may also influence the relationship between nurse staffing and patient falls.

In this study, there was no statistically significant relationship ob- served between incidence data and nurses' perceptions of falls fre- quency. In contrast, a previous study by Cina- Tschumi et al. (2009) reported a significant relationship between nurse- estimated falls and the incidence rate of falls. The discrepancy between the findings

may be attributed to differences in the methods and analyses em- ployed in both studies. Specifically, Cina- Tschumi et al. (2009) used the mean of ordinal variables in each unit to examine the correlation, while in our study, we used binary variables of falls by dichotomising based on whether falls occurred or not in each unit, as recommended by previous research (Han et al., 2020; Kakemam et al., 2021; Van Bogaert et al., 2014). In addition, Cina- Tschumi et al. (2009) analysed the rate of fall for the period of 1 year, while in our study, we used 4 years of fall data due to the impact of the COVID- 19 pandemic on healthcare during the study period.

Despite the lack of a statistically significant association between nurses' perceptions of falls frequency and incidence data in our study, we found a positive association between the two sources of this data. The positive association between the two sources of patient falls was supported by the relationship between safety at- titudes sub- scales, quality of care and falls incidence. Nursing units with a strong safety climate, safety behaviour, and high hospital and unit quality of care were more likely to have decreased nurses' per- ceptions of falls frequency and incidence of falls. Therefore, nurses' perceptions of the frequency of patient falls may accurately reflect the actual incidence of patient falls. Considering the impact of falls on patients and the underreporting of falls, further research is needed to examine the accuracy of the relationship between nurses' perceptions of falls frequency and the incidence of falls.

5.1  |  Limitations

Despite our study's strengths, several limitations have been identi- fied. This study was a cross- sectional survey; thus, causal relation- ships cannot be determined. In addition, the sample and setting of the study were limited to five MOH hospitals in two regions of Saudi Arabia and, therefore, may not be generalisable to other set- tings or regions. The study used an existing source from the admin- istrative dataset to gather data on the incidence of patient falls, and thus, we do not have any way of quantifying missing data or under- reported falls. In addition, no risk adjustments have been conducted. However, to the best of our knowledge, this is the first study that reported the incidence of falls over a 4- year period in Saudi Arabia. Further research should be undertaken to investigate

All units (N = 33 units) ICUs (N = 6 units)

Wards (N = 27 units)

Median (IQR) Median (IQR) Median (IQR)

Falls rate (per 1000 IPDs*) .25 (.13– .47) .04 (.00– .09) .28 (.21– .48)

(n = 587 nurses) (n = 266 nurses) (n = 321 nurses)

Nurses reported falls

Overall 10% (2%– 21%) 11% (6%– 14%) 10% (0%– 25%)

With injury 8% (0%– 15%) 8% (6%– 13%) 8% (0%– 17%)

Without injury 10% (0%– 21%) 9% (6%– 15%) 10% (0%– 25%)

Abbreviations: IPDs, inpatient days; IQR, interquartile ranges.

TA B L E 4  Distribution of patient falls across nursing units.

13652702, 2023, 19-20, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1111/jocn.16792, W

iley O nline L

ibrary on [13/04/2025]. See the T erm

s and C onditions (https://onlinelibrary.w

iley.com /term

s-and-conditions) on W iley O

nline L ibrary for rules of use; O

A articles are governed by the applicable C

reative C om

m ons L

icense

    |  7269ALANAZI et al.

TA B

LE 5

  A

ss oc

ia tio

n be

tw ee

n in

de pe

nd en

t p re

di ct

or s

an d

pa tie

nt fa

lls -

G LM

s.

Va ria

bl es

In ci

de nc

e da

ta (p

rim ar

y ou

tc om

e) N

ur se

- r ep

or te

d fr

eq ue

nc y

of fa

lls (s

ec on

da ry

o ut

co m

e)

In ci

de nc

e of

fa lls

(n  =

 3 0 

un its

) O

ve ra

ll fa

lls (n

 =  3

3  un

its )

Fa lls

w ith

in ju

ry (n

 =  3

3  un

its )

Fa lls

w ith

ou t i

nj ur

y (n

 =  3

3  un

its )

β 95

% C

I p

β 95

% C

I p

β 95

% C

I p

β 95

% C

I p

O ve

ra ll

SA Q

−1 .6

99 −1

.6 99

, − .7

48 <

.0 01

−. 97

2 −2

.0 78

, . 26

6 .1

24 −.

63 9

−1 .3

33 , .

05 6

.0 71

−1 .0

86 −2

.3 54

, . 18

3 .0

93

Te am

w or

k cl

im at

e −.

60 0

−1 .4

33 , .

23 3

.1 58

−. 79

0 −1

.6 12

, . 03

3 .0

60 −.

53 6

−. 98

4, −

.0 89

.0 19

−. 82

5 −1

.6 98

, . 04

8 .0

64

Sa fe

ty c

lim at

e −1

.1 63

−2 .2

93 , −

.0 33

.0 44

−1 .3

09 −2

.5 41

, − .0

77 .0

37 −.

90 8

−1 .5

67 , −

.2 49

.0 07

−1 .4

25 −2

.6 12

, − .2

38 .0

19

Jo b

sa tis

fa ct

io n

−. 66

3 −1

.2 58

, − .0

68 .0

29 −.

33 0

−1 .1

16 , .

45 6

.4 10

−. 12

6 −.

55 9,

.3 07

.5 68

−. 44

3 −1

.2 16

, . 32

9 .2

61

Pe rc

ep tio

ns o

f u ni

t m

an ag

em en

t −1

.0 26

−1 .9

99 , −

.0 53

.0 39

−. 07

8 −1

.2 82

, 1 .1

26 .8

99 −.

78 6

−1 .0

13 , .

40 3

.3 98

−. 18

4 −1

.4 43

, 1 .0

75 .7

74

Pe rc

ep tio

n of

h os

pi ta

l m

an ag

em en

t −1

.4 14

−1 .9

05 , −

.9 22

< .0

01 −.

18 4

−1 .1

92 , .

82 3

.7 20

−. 28

7 −.

79 5,

.2 20

.2 13

.3 71

−. 92

5, 1

.6 67

.5 75

St re

ss re

co gn

iti on

.3 68

−. 39

0, 1

.1 25

.3 42

−. 06

4 −1

.1 16

, . 98

7 .9

05 .4

22 −.

11 1,

.9 55

.1 21

−. 68

2 −1

.6 42

, . 27

8 .1

64

W or

ki ng

c on

di tio

n −1

.4 29

−2 .0

58 , −

.8 01

< .0

01 −.

55 4

−1 .3

65 , .

25 7

.1 81

−. 50

6 −.

99 0,

.0 22

.0 40

−. 51

6 −1

.4 58

, . 42

7 .2

83

Sa fe

ty b

eh av

io ur

−1 .0

33 −1

.5 80

, − .8

46 <

.0 01

−. 80

2 −1

.5 15

, − .0

89 .0

27 −.

42 0

−. 84

1, −

.0 01

.0 50

−. 84

7 −1

.5 45

, − .1

48 .0

17

H os

pi ta

l q ua

lit y

of c

ar e

−1 .4

56 −2

.2 58

, − .6

55 <

.0 01

−1 .7

89 −3

.1 39

, − .4

39 .0

09 −.

83 5

−1 .5

73 , −

.0 97

.0 27

−1 .9

55 −3

.4 38

, − .4

73 .0

10

U ni

t q ua

lit y

of c

ar e

−. 60

1 −1

.5 62

, . 35

9 .2

20 −1

.5 55

−2 .6

18 , −

.4 91

.0 04

−1 .0

04 −1

.6 13

, − .3

95 .0

01 −1

.6 66

−2 .7

64 , −

.5 69

.0 03

U ni

t q ua

lit y

of c

ar e

(la st

s hi

ft )

−. 61

9 −1

.8 73

, . 63

6 .3

34 −1

.7 27

−3 .0

81 , −

.3 74

.0 12

−1 .0

21 −1

.8 07

, − .2

34 .0

11 −1

.5 98

−3 .0

43 , −

.1 54

.0 30

O ve

ra ll

m is

se d

ca re

1. 26

9 .3

40 , 2

.1 99

.0 07

1. 39

3 −.

09 1,

2 .8

77 .0

66 1.

01 9

.3 43

, 1 .6

94 .0

03 1.

38 2

−. 28

6, 3

.0 15

.1 04

M is

se d

ad eq

ua te

p at

ie nt

su

rv ei

lla nc

e .3

42 −.

90 1,

1 .5

86 .5

89 1.

70 9

.5 99

, 2 .8

20 .0

03 .8

87 .2

38 , 1

.5 37

.0 07

1. 79

9 .7

34 , 2

.8 64

< .0

01

St af

fin g

.0 08

−. 15

4, .1

70 .9

25 .0

95 −.

14 6,

.3 36

.4 39

.2 00

.0 87

, . 31

4 <

.0 01

.1 21

−. 14

3, .3

85 .3

68

In ci

de nc

e of

p at

ie nt

fa lls

– –

– .1

64 −.

04 5,

2 .5

90 .0

58 .8

74 −.

34 6,

2 .0

94 .1

60 .1

63 −.

06 2,

2 .6

20 .0

62

A bb

re vi

at io

ns : β

, b et

a; C

I, co

nf id

en ce

in te

rv al

; a dj

us te

d fo

r - in

ci de

nc e

ra te

(n ur

si ng

u ni

t t yp

e, h

os pi

ta l s

iz e

an d

ex pe

rie nc

e) a

nd n

ur se

s re

po rt

ed fa

lls (n

ur si

ng u

ni t t

yp e,

h os

pi ta

l s iz

e an

d qu

al ifi

ca tio

n) .

13652702, 2023, 19-20, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1111/jocn.16792, W

iley O nline L

ibrary on [13/04/2025]. See the T erm

s and C onditions (https://onlinelibrary.w

iley.com /term

s-and-conditions) on W iley O

nline L ibrary for rules of use; O

A articles are governed by the applicable C

reative C om

m ons L

icense

7270  |    ALANAZI et al.

the prevalence of patient falls, such as retrospective chart review, in acute care hospitals in Saudi Arabia and explore the barriers to reporting falls among nurses.

6  |  CONCLUSION

The current findings provide insight into the complex relationship between safety culture, missed care, nursing staffing and patient falls in nursing units. The results suggest that nursing units with a strong safety climate, working conditions and safety behaviours and higher ratings of hospital quality of care are associated with a lower incidence rate of falls and reduced nurses' perceptions of the frequency of falls in their unit over the last year. Higher levels of missed care and increased patient- to- nurse ratios were associated with higher nurses' perceptions of the frequency of patient falls with injury in their units. Although the positive re- lationship between nurses' perceptions of falls frequency and falls incidence rates was not statistically significant, it highlights the potential value of nurses' perceptions in identifying the oc- currence of patient falls. Future research using longitudinal and quasi- experimental designs is needed to establish causal relation- ships between safety culture factors such as teamwork, staffing levels and falls. This will enable healthcare organisations to de- velop evidence- based interventions to improve patient safety and reduce the incidence of falls in healthcare settings.

7  |  RELE VANCE TO CLINIC AL PR AC TICE

The findings of this study have significant implications for clini- cal practice. Hospitals and healthcare managers can use this in- formation to improve patient safety and reduce falls incidents. Strategies that promote a strong safety climate, working condi- tions, safety behaviours and collaborations between interdiscipli- nary team members in nursing units may reduce the incidence of patient falls. Nursing units with higher levels of job satisfaction and support from both unit and hospital management were more likely to record a lower incidence rate of patient falls. Improving nurse staffing levels and fostering collaboration between health- care workers could lead to a reduction in the workload, which in turn can alleviate pressure on nursing staff and minimise missed care. These measures can ultimately contribute to a safer and more effective healthcare environment and improve the quality of care for patients.

AUTHOR CONTRIBUTIONS All authors: made substantial contributions to the conception and design, or acquisition of data or analysis and interpretation of data. FKA: involved in data collection and analysis, and initial draft writ- ing. FKA, JS, SL and LM: involved in drafting the manuscript or revis- ing it critically for important intellectual content. FKA, JS, SL and LM: approved the final version to be submitted for publication. All

authors: have agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the article are appropriately investigated and resolved.

ACKNO WLE DG E MENTS The authors thank Dr Saleh Alghamdi from the General Directorate of Clinical Excellence at MOH for his support in collecting secondary data and Dr Bradley Wakefield from the University of Wollongong for his statistical support. Open access publishing facilitated by University of Wollongong, as part of the Wiley - University of Wollongong agreement via the Council of Australian University Librarians.

FUNDING INFORMATION The first author holds a PhD scholarship from the Ministry of Education in Saudi Arabia. The funding body was not involved in the writing of the report and the decision to submit the manuscript for publication.

CONFLIC T OF INTERE S T S TATEMENT No conflict of interest has been declared by the author(s).

DATA AVAIL ABILIT Y S TATEMENT The primary dataset collected and analysed during the current study is available from the corresponding author on request. Secondary data are not available due to privacy/ethical restrictions imposed by the data custodian.

ORCID Faisal Khalaf Alanazi https://orcid.org/0000-0003-4266-4588 Samuel Lapkin https://orcid.org/0000-0002-1618-3812 Luke Molloy https://orcid.org/0000-0002-6120-9380 Jenny Sim https://orcid.org/0000-0001-6863-0541

T WIT TER Faisal Khalaf Alanazi @Falanazi45 Samuel Lapkin @DrLapkin Luke Molloy @lukemolloy Jenny Sim @jennysim_1

R E FE R E N C E S Alanazi, F. K., Lapkin, S., Molloy, L., & Sim, J. (2023). Safety culture, qual-

ity of care, missed care, nurse staffing and their impact on pressure injuries: A cross- sectional multi- source study. International Journal of Nursing Studies Advances, 5, 100125. https://doi.org/10.1016/j. ijnsa.2023.100125

Alanazi, F. K., Sim, J., & Lapkin, S. (2022). Systematic review: Nurses' safety attitudes and their impact on patient outcomes in acute- care hospitals. Nursing Open, 9(1), 30– 43. https://doi.org/10.1002/ nop2.1063

Al- Ghraiybah, T., Sim, J., Fernandez, R., & Lago, L. (2023). Managing miss- ing and erroneous data in nurse staffing surveys. Nurse Researcher. https://doi.org/10.7748/nr.2023.e1878

Ausserhofer, D., Schubert, M., Desmedt, M., Blegen, M. A., De Geest, S., & Schwendimann, R. (2013). The association of patient safety climate and nurse- related organizational factors with selected

13652702, 2023, 19-20, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1111/jocn.16792, W

iley O nline L

ibrary on [13/04/2025]. See the T erm

s and C onditions (https://onlinelibrary.w

iley.com /term

s-and-conditions) on W iley O

nline L ibrary for rules of use; O

A articles are governed by the applicable C

reative C om

m ons L

icense

    |  7271ALANAZI et al.

patient outcomes: A cross- sectional survey. International Journal of Nursing Studies, 50(2), 240– 252. https://doi.org/10.1016/j.ijnur stu.2012.04.007

Ball, J. E., Bruyneel, L., Aiken, L. H., Sermeus, W., Sloane, D. M., Rafferty, A. M., Lindqvist, R., Tishelman, C., & Griffiths, P. (2018). Post- operative mortality, missed care and nurse staffing in nine coun- tries: A cross- sectional study. International Journal of Nursing Studies, 78(8), 10– 15. https://doi.org/10.1016/j.ijnur stu.2017.08.004

Bouldin, E. L. D., Andresen, E. M., Dunton, N. E., Simon, M., Waters, T. M., Liu, M., Daniels, M. J., Mion, L. C., & Shorr, R. I. (2013). Falls among adult patients hospitalized in the United States. Journal of Patient Safety, 9(1), 13– 17. https://doi.org/10.1097/PTS.0b013 e3182 699b64

Boussat, B., François, O., Viotti, J., Seigneurin, A., Giai, J., François, P., & Labarère, J. (2021). Managing missing data in the hospital survey on patient safety culture: A simulation study. Journal of Patient Safety, 17(2), e98– e106. https://doi.org/10.1097/PTS.00000 00000 000595

Bowden, V., Bradas, C., & McNett, M. (2019). Impact of level of nurse experience on falls in medical surgical units. Journal of Nursing Management, 27(4), 833– 839. https://doi.org/10.1111/jonm.12742

Brown, D. S., & Wolosin, R. (2013). Safety culture relationships with hos- pital nursing sensitive metrics. Journal for Healthcare Quality, 35(4), 61– 74. https://doi.org/10.1111/jhq.12016

Centers for Disease Control and Prevention. (2020). Keep on your feet— Preventing older adult falls. https://www.cdc.gov/injur y/featu res/ older - adult - falls/ index.html

Churruca, K., Ellis, L. A., Pomare, C., Hogden, A., Bierbaum, M., Long, J. C., Olekalns, A., & Braithwaite, J. (2021). Dimensions of safety culture: A systematic review of quantitative, qualitative and mixed methods for assessing safety culture in hospitals. BMJ Open, 11(7), e043982. https://doi.org/10.1136/bmjop en-2020-043982

Cina- Tschumi, B., Schubert, M., Kressig, R. W., De Geest, S., & Schwendimann, R. (2009). Frequencies of falls in Swiss hospitals: Concordance between nurses' estimates and fall incident reports. International Journal of Nursing Studies, 46(2), 164– 171. https://doi. org/10.1016/j.ijnur stu.2008.09.008

Dickens, G. L., Salamonson, Y., Johnson, A., Ramjan, L., Steel, K., Taylor, M., & Everett, B. (2021). Longitudinal evaluation of a programme for safety culture change in a mental health service. Journal of Nursing Management, 29(4), 690– 698. https://doi.org/10.1111/jonm.13205

Donabedian, A. (1988). The quality of care. JAMA, 260(12), 1743. https:// doi.org/10.1001/jama.1988.03410 12008 9033

Griffiths, P., Ball, J., Drennan, J., Dall'Ora, C., Jones, J., Maruotti, A., Pope, C., Recio Saucedo, A., & Simon, M. (2016). Nurse staffing and patient outcomes: Strengths and limitations of the evidence to inform policy and practice. A review and discussion paper based on evidence reviewed for the National Institute for health and care excellence safe staffing guideline develo. International Journal of Nursing Studies, 63, 213– 225. https://doi.org/10.1016/j.ijnur stu.2016.03.012

Han, Y., Kim, J.- S., & Seo, Y. (2020). Cross- sectional study on patient safety culture, patient safety competency, and adverse events. Western Journal of Nursing Research, 42(1), 32– 40. https://doi. org/10.1177/01939 45919 838990

Hessels, A. J., Paliwal, M., Weaver, S. H., Siddiqui, D., & Wurmser, T. A. (2019). Impact of patient safety culture on missed nursing care and adverse patient events. Journal of Nursing Care Quality, 34(4), 287– 294. https://doi.org/10.1097/NCQ.00000 00000 000378

Innab, A. M. (2022). Nurses' perceptions of fall risk factors and fall pre- vention strategies in acute care settings in Saudi Arabia. Nursing Open, 9(2), 1362– 1369. https://doi.org/10.1002/nop2.1182

Kakemam, E., Gharaee, H., Rajabi, M. R., Nadernejad, M., Khakdel, Z., Raeissi, P., & Kalhor, R. (2021). Nurses' perception of patient safety culture and its relationship with adverse events: A national

questionnaire survey in Iran. BMC Nursing, 20(1), 1– 11. https://doi. org/10.1186/s12912-021-00571-w

Kalisch, B. J., Tschannen, D., & Lee, K. H. (2012). Missed nursing care, staffing, and patient falls. Journal of Nursing Care Quality, 27(1), 6– 12. https://doi.org/10.1097/NCQ.0b013 e3182 25aa23

Kim, J., Lee, E., Jung, Y., Kwon, H., & Lee, S. (2022). Patient- level and organizational- level factors influencing in- hospital falls. Journal of Advanced Nursing, 78(11), 3641– 3651. https://doi.org/10.1111/ jan.15254

Lee, S. E., Scott, L. D., Dahinten, V. S., Vincent, C., Lopez, K. D., & Park, C. G. (2019). Safety culture, patient safety, and quality of care out- comes: A literature review. Western Journal of Nursing Research, 41(2), 279– 304. https://doi.org/10.1177/01939 45917 747416

Lee, S. E., Vincent, C., Dahinten, V. S., Scott, L. D., Park, C. G., & Dunn Lopez, K. (2018). Effects of individual nurse and hospital charac- teristics on patient adverse events and quality of care: A multilevel analysis. Journal of Nursing Scholarship, 50(4), 432– 440. https://doi. org/10.1111/jnu.12396

Morris, M. E., Webster, K., Jones, C., Hill, A. M., Haines, T., McPhail, S., Kiegaldie, D., Slade, S., Jazayeri, D., Heng, H., Shorr, R., Carey, L., Barker, A., & Cameron, I. (2022). Interventions to reduce falls in hospitals: A systematic review and meta- analysis. Age and Ageing, 51(5), afac077. https://doi.org/10.1093/agein g/afac077

Nantsupawat, A., Poghosyan, L., Wichaikhum, O. A., Kunaviktikul, W., Fang, Y., Kueakomoldej, S., Thienthong, H., & Turale, S. (2022). Nurse staffing, missed care, quality of care and adverse events: A cross- sectional study. Journal of Nursing Management, 30(2), 447– 454. https://doi.org/10.1111/jonm.13501

Ng, V. K. Y., & Cribbie, R. A. (2017). Using the gamma generalized linear model for modeling continuous, skewed and heteroscedastic out- comes in psychology. Current Psychology, 36(2), 225– 235.

Nhongo, D., Holt, A., Flenady, T., Rebar, A., & Bail, K. (2022). Nurse staff- ing and adverse events in residential aged care: Retrospective multi- site analysis. Collegian, 30, 343– 349. https://doi.org/10.1016/j. colegn.2022.09.017

Recio- Saucedo, A., Dall'Ora, C., Maruotti, A., Ball, J., Briggs, J., Meredith, P., Redfern, O. C., Kovacs, C., Prytherch, D., Smith, G. B., & Griffiths, P. (2018). What impact does nursing care left undone have on pa- tient outcomes? Review of the literature. Journal of Clinical Nursing, 27(11– 12), 2248– 2259. https://doi.org/10.1111/jocn.14058

Schubert, M., Glass, T. R., Clarke, S. P., Aiken, L. H., Schaffert- Witvliet, B., Sloane, D. M., & De Geest, S. (2008). Rationing of nursing care and its relationship to patient outcomes: The Swiss extension of the in- ternational hospital outcomes study. International Journal for Quality in Health Care, 20(4), 227– 237. https://doi.org/10.1093/intqh c/mzn017

Sermeus, W., Aiken, L. H., Van den Heede, K., Rafferty, A. M., Griffiths, P., Moreno- Casbas, M. T., Busse, R., Lindqvist, R., Scott, A. P., Bruyneel, L., Brzostek, T., Kinnunen, J., Schubert, M., Schoonhoven, L., & Zikos, D. (2011). Nurse forecasting in Europe (RN4CAST): Rationale, design and methodology. BMC Nursing, 10(1), 6. https:// doi.org/10.1186/1472-6955-10-6

Sexton, J. B., Helmreich, R. L., Neilands, T. B., Rowan, K., Vella, K., Boyden, J., Roberts, P. R., & Thomas, E. J. (2006). The safety atti- tudes questionnaire: Psychometric properties, benchmarking data, and emerging research. BMC Health Services Research, 6(1), 44. https://doi.org/10.1186/1472-6963-6-44

Shin, S., Park, J. H., & Bae, S. H. (2019). Nurse staffing and hospital- acquired conditions: A systematic review. Journal of Clinical Nursing, 28(23– 24), 4264– 4275. https://doi.org/10.1111/jocn.15046

Sim, J., Crookes, P., Walsh, K., & Halcomb, E. (2018). Measuring the outcomes of nursing practice: A Delphi study. Journal of Clinical Nursing, 27(1– 2), e368– e378. https://doi.org/10.1111/jocn.13971

Sim, J., Joyce- McCoach, J., Gordon, R., & Kobel, C. (2019). Development of a data registry to evaluate the quality and safety of nursing prac- tice. Journal of Advanced Nursing, 75(9), 1877– 1888. https://doi. org/10.1111/jan.13967

13652702, 2023, 19-20, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1111/jocn.16792, W

iley O nline L

ibrary on [13/04/2025]. See the T erm

s and C onditions (https://onlinelibrary.w

iley.com /term

s-and-conditions) on W iley O

nline L ibrary for rules of use; O

A articles are governed by the applicable C

reative C om

m ons L

icense

7272  |    ALANAZI et al.

Smith, S., Lapkin, S., Sim, J., & Halcomb, E. (2020). Nursing care left un- done, practice environment and perceived quality of care in small rural hospitals. Journal of Nursing Management, 28(8), 2166– 2173. https://doi.org/10.1111/jonm.12975

Taylor, J. A., Dominici, F., Agnew, J., Gerwin, D., Morlock, L., & Miller, M. R. (2012). Do nurse and patient injuries share common anteced- ents? An analysis of associations with safety climate and working conditions. BMJ Quality and Safety, 21(2), 101– 111. https://doi. org/10.1136/bmjqs-2011-000082

Toyabe, S. (2015). Characteristics of inpatient falls not reported in an incident reporting system. Global Journal of Health Science, 8(3), 17– 25. https://doi.org/10.5539/gjhs.v8n3p17

Tuinman, A., De Greef, M. H. G., Finnema, E. J., & Roodbol, P. F. (2021). A systematic review of the association between nursing staff and nursing- sensitive outcomes in long- term institutional care. Journal of Advanced Nursing, 77(8), 3303– 3316. https://doi.org/10.1111/ jan.14840

Twigg, D. E., Whitehead, L., Doleman, G., & El- Zaemey, S. (2021). The impact of nurse staffing methodologies on nurse and patient out- comes: A systematic review. Journal of Advanced Nursing, 77(12), 4599– 4611. https://doi.org/10.1111/jan.14909

Van Bogaert, P., Timmermans, O., Weeks, S. M., van Heusden, D., Wouters, K., & Franck, E. (2014). Nursing unit teams matter: Impact of unit- level nurse practice environment, nurse work character- istics, and burnout on nurse reported job outcomes, and qual- ity of care, and patient adverse events- A cross- sectional survey.

International Journal of Nursing Studies, 51(8), 1123– 1134. https:// doi.org/10.1016/j.ijnur stu.2013.12.009

World Health Organization. (2021). Falls: Key facts. https://www.who. int/news-room/fact-sheet s/detai l/falls

Yesilyaprak, T., & Demir Korkmaz, F. (2023). The relationship between surgical intensive care unit nurses' patient safety culture and ad- verse events. Nursing in Critical Care, 28(1), 63– 71. https://doi. org/10.1111/nicc.12611

SUPPORTING INFORMATION Additional supporting information can be found online in the Supporting Information section at the end of this article.

How to cite this article: Alanazi, F. K., Lapkin, S., Molloy, L., & Sim, J. (2023). The impact of safety culture, quality of care, missed care and nurse staffing on patient falls: A multisource association study. Journal of Clinical Nursing, 32, 7260–7272. https://doi.org/10.1111/jocn.16792

13652702, 2023, 19-20, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1111/jocn.16792, W

iley O nline L

ibrary on [13/04/2025]. See the T erm

s and C onditions (https://onlinelibrary.w

iley.com /term

s-and-conditions) on W iley O

nline L ibrary for rules of use; O

A articles are governed by the applicable C

reative C om

m ons L

icense