Risk prediction of sleep disturbance in clinical nurses: a nomogram and artificial neural network model

doi:10.21203/rs.3.rs-1977577/v1

Download PDF

Research Article

Risk prediction of sleep disturbance in clinical nurses: a nomogram and artificial neural network model

https://doi.org/10.21203/rs.3.rs-1977577/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 28 Aug, 2023

Read the published version in BMC Nursing →

You are reading this latest preprint version

Background:To develop a Nomogram and a Artificial Neural Network (ANN) model to predict sleep disturbance in clinical nurses.

Methods:A cross-sectional study was conducted from August 2021 to June 2022 ,434 clinical nurses participated in the study and completed questionnaires. They were randomly distributed in a 7:3 ratio between training and validation cohorts.Nomogram and ANN model were developed using predictors of sleep disturbance identified by univariate and multivariate analyses in the training cohort; The 1000 bootstrap resampling and receiver operating characteristic curve (ROC) were used to evaluate the predictive accuracy in the training and validation cohorts.

Results:Sleep disturbance was found in 180 of 304 nurses(59.2%) in the training cohort and 80 of 130 nurses (61.5%) in the validation cohort.Age, chronic diseases, anxiety, depression, burnout, and fatigue were identified as risk factors for sleep disturbance. The calibration curves of the two models are well-fitted. The sensitivity and specificity (95% CI) of the models were calculated, resulting in sensitivity of 83.9%(77.5–88.8%)and 88.8% (79.2–94.4%) and specificity of83.1% (75.0–89.0%) and 74.0% (59.4–84.9%) for the training and validation cohorts, respectively.

Conclusions:The sleep disturbance risk prediction models constructed in this study have good consistency and prediction efficiency, and can effectively predict the occurrence of sleep disturbance in clinical nurses.

clinical nurse

sleep disturbance

nomogram

artificial neural network

prediction model

Sleep is a normal physiological phenomenon, which is spent about 1 / 3 of a person's life.¹Related studies have shown that sleep has the functions of saving and storing body energy, regulating body heat and metabolism, promoting cell tissue repair after injury, enhancing brain cognition, delaying memory decline and so on, which is closely related to our physical and mental health.^2,3However, Sleep disturbance is common among clinical nurses.The incidence of sleep disoturbance among domestic nurses has reached 55.38%.⁴Sleep disturbance can lead to sleepiness, fatigue, lethargy, and other serious work-related consequences, such as work errors.⁵

Due to the particularity of the nursing profession, the nature of high-intensity and high-stress work,⁶ the night shifts⁷ and the continuous learning of advanced medical knowledge and technology, the various needs of nursing work do not adapt to their physical and mental state, resulting in job burnout,⁸affecting the quality of sleep. At the same time, nurses' working environment, work intensity, interpersonal relationship with patients and colleagues, and social role disorders also lead to higher stress than other professional groups, Thus, nurses become a high group of sleep disturbance.And their sleep problems are more easily overlooked.⁹Good sleep quality is the key for nurses to provide the best care for patients. Therefore, it is very important for nurses in the clinical to screen for the risk of sleep disturbance and early intervention. However, at present, the research on nurses' sleep problems in China is limited to the investigation of the current situation, A reliable tool for identifying the risk of nurses' sleep disturbance has not been established. Practical and reliable predictive models are needed to predict the occurrence of sleep disturbance. Nomograms are statistical models specifically designed to maximize predictive accuracy, which can allow for an individualized prediction of outcomes.¹⁰ANN (artificial neural network) as a computer technology has also been shown to outperform traditional discriminant analysis by determining the importance of each variable for a particular event and predicting outcomes based on the variables present. ^11,12However, to our knowledge few studies have focused on using the above models to predict the risk of sleep disorders.

Therefore, we conducted a cross-sectional study to develop a nomogram and ANN model valid for predict sleep disturbance in clinical nurses.

Subjects

This cross-sectional study was conducted between August 2021 and June 2022, and 444 male and female clinical nurses were approached.Nurses were selected based on convenience sampling.This study was approved by the ethics committee of the Jin zhou Medical University of traditional Chinese and followed the guidelines of the Declaration of Helsinki. Written consent was obtained from all participants. Ultimately, complete responses were obtained from 434 nurses, for a response rate of 97.7%. They were randomly allocated to either the training cohort or the validation cohort at a ratio of 7:3.The inclusion criteria were: (a) registered nurse with qualification certificate; (b) with at least 1 year of experience.The exclusion criteria were:(a) retired nurses; or (b) trainee nurses. (c)nurses with severe sleep disturbances,severe stressful events, or physical or mental illness.

Instruments

General Demographic Characteristics Questionnaire

This form was used to gather information on the participant's gender,age, education level, marital status,fertility status, chronic illness, way of appointment,night shifts and tea or coffee usage during work.

The Pittsburgh Sleep Quality Index

The PSQI,¹³was designed to assessed sleep quality. The PSQI consists of seven components: sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, daytime dysfunction, and hypnotics, each of which can be rated with a score from 0 to 3. PSQI scores range from 0 to 21 and are obtained by summing the scores of the seven components. 0 points are in all areas. A score of 21 means no difficulty, while a score of 21 indicates severe difficulty; a total PSQI score of 7 or higher is used to distinguish between poor and good sleepers.(Cronbach’s α = .73)

The Maslach Burnout Inventory

The MBI,¹⁴was designed to measure burnout. The questionnaire consists of 22 items in descriptive form related to personal feelings and attitudes and is composed of three subscales. The Emotional Exhaustion subscale (EE) (9 items), the Depersonalization subscale (DP) (5 items), and the Personality Coping subscale (PA) (8 items). 27 or more points for EE, 10 or more points for DP, and 33 or less points for PA are considered "high BO"; 19 to 26 points for EE, 6 to 9 points for DP, and 34 to 39 points for PA are considered "moderate BO"; and 18 points or less, DP of 5 points or less, and PA of 40 points or less were considered "low BO".(Cronbach’s α = .84)

Fatigue Scale

The FS-14,¹⁵ was designed to measure the severity of fatigue symptoms. The scale consists of 14 items that reflect different perspectives on the degree of fatigue, with items 1–8 reflecting physical fatigue and items 9–14 reflecting mental fatigue; scores from items 1–8 are summed to form the physical fatigue score, scores from items 9–14 are summed to form the mental fatigue score, and the sum of the physical fatigue score and mental fatigue score is the total fatigue score. The total of the physical and mental fatigue scores is the overall fatigue score. The highest physical fatigue score is 8, the highest mental fatigue score is 6, and the total score is 14. A total score of 7 or higher indicates a state of fatigue. Higher scores indicate more severe fatigue.(Cronbach’s α = .81)

Hospital Anxiety and Depression Scale

The HADS, ¹⁶ was used to determine anxiety and depression symptoms.The scale consists of 14 items: 7 items assessing anxiety and 7 items assessing depression. Scores can be interpreted on the following scale: 0–7: no depression or anxiety disorder; 8–10: mild depression or anxiety disorder; 11–14: moderate disorder; 15–21: severe disorder.(Cronbach’s α = .88)

Data collection

Before the study began, each participant was informed of the purpose of the study and of his or her right to refuse or discontinue participation in the study at any time, while maintaining the anonymity and confidentiality of all participants. The questionnaires were distributed by a member of the research team. Before distributing the questionnaires, the purpose and importance of the study were fully explained and instructions were given on how to complete the questionnaires so that there would be no ambiguity in the answers to the questions. The questionnaires were answered anonymously, so participants' privacy was not violated. The questionnaires were collected by the research team immediately after completion. If more than 10% of the questionnaire items were unanswered, they were considered invalid.

Statistical analysis

Categorical variables are described as frequencies and percentages,statistical analyses were performed using SPSS 25.0 software for Windows (IBM, Somers, NY). In the training cohort, possible predictors of sleep disturbance were analyzed by univariate logistic regression analysis and variables with p < 0.05 in univariate analysis were considered candidate variables for multivariate logistic regression analysis. p < 0.05 was considered a sign of statistical significance in multivariate analysis. Nomogram and ANN models were constructed based on the results of multivariate logistic regression analysis of the training cohort. Nomogram was constructed using the strategic regression modeling package in R (version 4.0.2; R Project for Statistical Computing; www.rproject.org). After building the predictive model, the accuracy and discriminative power of the internal (training cohort) and external (validation cohort) models were tested using 1000 bootstrap resampling and ROC curves (receiver operating characteristic).

Participant Characteristics

We first assessed the eligibility of 500 clinical nurses to participate in our study (Figure 1). Of these, 26 nurses refused to participate, 30 nurses did not meet the inclusion criteria, and the remaining 444 nurses participated in the study. After the study was completed,

10 invalid questionnaires were excluded from the study. Finally, the data of 434 nurses were analyzed in the present study. The training cohort composed of 304 nurses. 49.7% were age <30years,14.8% were male and 59.2% had sleep disturbance. The validation cohort consisted of 130 nurses. 53.1% were age <30years,17.7% were male and 61.5% had sleep disturbance.The prevalence of sleep disturbance among all nurses was 59.9%. All characteristics were no statistically significant differences between training cohort and validation cohort (all P > 0.05) (Table 1).

Predictors of sleep disturbances in the training cohort

The results of univariable and multivariable logistic regression analyses for sleep disturbance are presented in Table 2 and Table 3,respectively. On multivariate analysis, with results reported as odds ratio (95% CI), fatigue [2.76(1.40-5.45)], anxiety [3.44(1.60-7.41)], depression[3.73(1.83-7.59)], chronic diseases[4.05(1.91-8.59)], burnout [4.41(1.90-10.24)], age [7.33(1.56-34.41)], were independently associated with sleep disturbance.

Development and validation of the nomogram and ANN

These independently associated risk factors were used to build a sleep disturbance risk prediction model (Figure 2). The resulting model was internally validated using the receiver operating characteristic curve (ROC) and bootstrap validation method.According to the internal validation in the training cohort, the C-statistic of this nomogram was 0.90 (95% CI, 0.86–0.93); Figure 3(a)),while the index of external validation in the validation cohort was 0.89(95% CI, 0.83–0.95); Figure 3(b)).Besides, in the training and validation cohorts ,the calibration plots graphically demonstrated good agreement on the presence of sleep disturbance between the risk estimation by the nomogram and analysis results of actual clinical data (Figure 3(c),3(d)). An ANN model was established based on the results of univariate analyses and multivariate logistic regression analyses. The importance of each predictor is shown in Figure 4. According to the ANN model, the burnout was the predominant predictor of sleep disturbance (Figure 4).

Risk of sleep disturbance based on the nomogram scores

The sensitivity, specificity, positive predictive value, and negative predictive value when used in differentiating the presence from absence of sleep disturbance were 82.8%, 83.9%, 87.8%, and 78.0% in the training cohort, and 88.8%, 78.0%,84.5%, and 80.4% in the validation cohort, respectively (Table 4).

In the current study ,nearly 60% of the clinical nurses were reported sleep disturbance. While compared participants in the current study,the prevalence rate in the studies(59.2%) is higher than both studies in the Australian menopausal women¹⁷and general Asian population.¹⁸Sleep disturbance is a medical disorder of sleep patterns and the most significant cause of multiple psychological disorders and somatic diseases.^19–21The early identification of nurses at higher risk of sleep disturbances is critical to nursing career development. Gender, age, anxiety, night shift, etc have identified multiple predictors to significantly correlated with sleep disturbance.^22–25Nevertheless, our data showed that the gender, night shift had no relevance in the prediction of sleep disturbance. Besides, in this study, fatigue, chronic diseases, depression, burnout were independent risk factors for sleep disturbance in the multivariate logistic regression analysis. To our knowledge, this is the first nomogram and ANN model to predict the risk of sleep disturbance in clinical nurses.

Age

As shown in the nomogram of this study, the older the age, the higher the sleep disorder risk .Previous studies²⁶have shown that symptoms of insomnia may be exacerbated with age. This may be because older nurses experience higher levels of work responsibilities and family needs, which increase chronic stress levels and make them particularly prone to sleep problems.Therefore, it is suggested that managers should communicate more with nursing members, understand their inner world, and appropriately help them alleviate the pressure on their work and life.Therefore, it is recommended that managers communicate with nursing members more to understand everyone's inner world, and appropriately help everyone relieve the pressure in work and life.

Chronic disease

The results of this study showed that the incidence of sleep disturbance in nurses with chronic diseases was 4.05 times higher than that in nurses without chronic diseases. With the increase of age, the health status of nurses become worse. Due to the heavy nursing workload, the more skilled operations, the poorer physical status will be reduce work efficiency, which in turn leads to prolonged working hours, affects the recovery of body functions, and ultimately affects the quality of sleep.²⁷Therefore, nurses and managers must be aware of this, establish a complete physical examination system and working environment, and improve the quality of sleep for nurses.

Fatigue

In clinical work, fatigue is an uncomfortable feeling of subjectively enduring tiredness, as well as a feeling of weakness when engaged in physical or mental activities, and these feelings are not relieved in the short term.Fatigue is the first influencing factor of sub-health.²⁸This study found that physical fatigue is an independent risk factor for sleep disturbance in nurses, which may be related to the mechanical and repetitive nursing operations of nurses and the disturbance of biological clock caused by long-term night shift.Foreign scholars Tomasz Z ²⁹also proved this point of view.For the problem of nurse fatigue, first of all, hospitals should improve the management mechanism to reduce the workload of nurses.According to the situation of different departments, reasonable scheduling, reduce the frequency of night shift and shift, so that nurses more regular life.

Anxiety-depression

Anxiety depression is one of the common psychological barriers of nurse, is individual subjective experience negative emotional state, anxiety, depression and other negative emotions, the greater the impact on the quality of sleep, sleep rhythm disorder, and poor sleep quality is to cause a decline in cognitive function, increase anxiety, depression, long-term repeatedly to form a vicious circle.The results of this study showed that the incidence of sleep disturbance in nurses with anxiety and depression was 3.44 and 3.73 times higher than those in normal mood.Slightly higher than the research results of foreign scholars Chueh K H ³⁰(OR = 1.07, OR = 1.16).It may be due to the fact that clinical nurses are faced with heavy nursing work and are in a state of mental and physical fatigue and stress for a long time, and the majority of clinical nurses are female, bearing the pressure of housework and children's education at the same time, so psychological conflicts are easy to occur, resulting in psychological imbalance.Park C H ³¹also confirmed this view.Therefore, hospitals should carry out extensive psychological training to help nurses master the correct emotional regulation methods, reasonably regulate their own emotions and improve their sleep quality, so as to effectively relieve work pressure and improve work efficiency.

Burnout

Burnout is a series of symptoms related to the work environment and an individual's slow response to chronic emotional and interpersonal pressure at work.Studies have found that job burnout not only affects the mental and physical of staff, and to cause a decline in its efficiency, absenteeism and turnover increased, the serious influence the stability of the nursing team and nursing career, and sleep disorder is the individual in the most prone to the symptoms of job burnout situations, in this study, serious nurse job burnout rate of sleep disorder is higher;Severe job burnout has the strongest predictive effect on sleep disorders, which is similar to previous studies by Sayilan A A ³².Therefore, the management personnel should reduce the emotional exhaustion of nurses, provide a platform for nurses to communicate with each other, and carry out targeted and professional psychological counseling for female nurses, and provide positive psychological support.Reduce the depersonalization of nurses, strengthen vocational training for nurses, constantly improve their professional quality and work skills, enhance their professional identity.

Based on the results of univariable and multivariable analyses, a predictive nomogram and ANN model were established to predict sleep disturbances. The training cohort data was used as internal validation and the validation cohort data was used as external validation to validate the accuracy of this nomogram. The C-statistic values provided sufficient predictive accuracy in both cohorts,and the calibration plots shows good consistency in the existence of sleep disorders ,which indicated the high accuracy of the nomogram. The designed three layer feed forward ANN has good prediction performance with a sensitivity of 86.1%. Compared with traditional regression methods, ANN did not require a predefined mathematical relationship between the dependent and independent variables and could model any arbitrarily complicated nonlinear relationship. These advantages enable ANN to be a useful tool in solving the complex challenge of prediction.

However, this study still has some limitations. It is pointed out that, first of all, being limited to Jin zhou nurses limits the universality of the research results and needs to be verified by other cities. Secondly, the risk factors of sleep disorder are complex, and there are many factors that need to be studied and predicted. This improves the accuracy of the prediction model to a certain extent.

Age, chronic diseases, fatigue, anxiety,depression, and burnout were independent risk factors of sleep disturbance among clinical nurses. The nomogram and ANN model could effectively predict the risk of sleep disturbances.

Acknowledgements

The authors are grateful to the clinical nurses involved in the study and for their support from the nursing directors.

Authors’ contributions

XZ completed the study design with LZ. XZ, LZ contributed to data collection and participated in manuscript writing. XZ conducted data analysis and interpretated results. The manuscript was drafted by XZ and edited and approved by LZ. All authors have read and approved the manuscript.

Funding

No funding was obtained for this study.

Availability of data and materials

The data garnered during the current study and the final dataset used for statistical analysis are available from the corresponding author on reasonable request.

Ethics approval and consent to participate

Prior to the investigation, participants were informed of the purpose and significance of this study and signed informed consent. All data were protected throughout the study. All procedures were performed with the 1964 Helsinki declaration, and the study protocol was approved by the Ethics Committee of the First Affiliated Hospital,Jinzhou Medical University(202228).

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Author details

¹Department of Nursing, Jinzhou Medical University, No.40, Section 3, Songlu

Road, Linghe District, Jinzhou 121001, China.

Mahalakshmi AM, Ray B, Tuladhar S, et al. Sleep, brain vascular health and ageing. Geroscience.2020;42(5):1257–1283.
Liew SC, Aung T. Sleep deprivation and its association with diseases- a review. Sleep Med. 2021;77:192–204.
Billings ME, Hale L, Johnson DA. Physical and Social Environment Relationship With Sleep Health and Disorders. Chest. 2020;157(5):1304–1312.
Yang H,Ding YX,Tang S,et al. Progress on adverse outcomes of circadian rhythm disorders in nursing staff.Chinese Nursing Research2019(5):816–821.
Shelton AR, Malow B. Neurodevelopmental Disorders Commonly Presenting with Sleep Disturbances. Neurotherapeutics. 2021;18(1):156–169.
Stimpfel AW, Fatehi F, Kovner C. Nurses' sleep, work hours, and patient care quality, and safety. Sleep Health. 2020;6(3):314–320.
Kang J, Noh W, Lee Y. Sleep quality among shift-work nurses: A systematic review and meta-analysis. Appl Nurs Res. 2020;52:151227.
Brooks Carthon JM, Hatfield L, Brom H, et al. System-Level Improvements in Work Environments Lead to Lower Nurse Burnout and Higher Patient Satisfaction. J Nurs Care Qual. 2021;36(1):7–13.
Dong H, Zhang Q, Sun Z, Sang F, Xu Y. Sleep disturbances among Chinese clinical nurses in general hospitals and its influencing factors. BMC Psychiatry. 2017;17(1):241. Published 2017 Jul 3.
Yang S, Zhang Q, Xu Y, et al. Development and Validation of Nomogram Prediction Model for Postoperative Sleep Disturbance in Patients Undergoing Non-Cardiac Surgery: A Prospective Cohort Study. Nat Sci Sleep. 2021;13:1473–1483. Published 2021 Aug 24.
Cross SS, Harrison RF, Kennedy RL. Introduction to neural networks. Lancet. 1995;346(8982):1075–1079.
Baxt WG. Application of artificial neural networks to clinical medicine. Lancet. 1995;346(8983):1135–1138.
Buysse DJ, Reynolds CF 3rd, Monk TH, Berman SR, Kupfer DJ. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res. 1989;28(2):193–213.
Maslach C, Jackson S E. The measurement of experienced burnout. J Journal of Organizational Behavior, 1981, 2(2):99–113.
Chalder T, Berelowitz G, Pawlikowska T, Watts L, Wallace EP. Development of a fatigue scale. Journal of Psychosomatic Research, 37(2), 147–153.
Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand. 1983;67(6):361–370.
Choi H, Kim S, Kim B, Kim I. Psychometric Properties of the Korean Versions of Three Sleep Evaluation Questionnaires. Clin Nurs Res. 2015;24(5):526–538.
Seib C, Anderson D, Lee K. Prevalence and correlates of sleep disturbance in postmenopausal women: the Australian Healthy Aging of Women (HOW) Study. J Womens Health (Larchmt). 2014;23(2):151–158.
Devine MF, St Louis EK. Sleep Disturbances Associated with Neurological Autoimmunity [published correction appears in Neurotherapeutics. 2021 May 3;:]. Neurotherapeutics. 2021;18(1):181–201.
Riemann D, Krone LB, Wulff K, Nissen C. Sleep, insomnia, and depression. Neuropsychopharmacology.2020;45(1):74–89.
Korkmaz S, Kazgan A, Çekiç S, Tartar AS, Balcı HN, Atmaca M. The anxiety levels, quality of sleep and life and problem-solving skills in healthcare workers employed in COVID-19 services. J Clin Neurosci. 2020;80:131–136.
D'Ettorre G, Pellicani V, Caroli A, Greco M. Shift work sleep disorder and job stress in shift nurses: implications for preventive interventions. Med Lav. 2020;111(3):195–202. Published 2020 Jun 26.
Van Nguyen T, Liu HE. A cross-sectional study on sleep disturbances and associated factors among nurses. BMC Psychiatry. 2022;22(1):119. Published 2022 Feb 15.
Simonetti V, Durante A, Ambrosca R, et al. Anxiety, sleep disorders and self-efficacy among nurses during COVID-19 pandemic: A large cross-sectional study. J Clin Nurs. 2021;30(9–10):1360–1371.
Cappadona R, De Giorgi A, Boari B, et al. Nurses, sleep disturbances, desynchronization of circadian rhythms, and performance: a dangerous liaison? A narrative mini-review. Eur Rev Med Pharmacol Sci. 2021;25(22):6924–6933.
Walker JL, Slavish DC, Dolan M, et al. Age-dependent associations among insomnia, depression, and inflammation in nurses. Psychol Health. 2021;36(8):967–984.
Shen SH, Yen M, Yang SL, Lee CY. Insomnia, anxiety, and heart rate variability among nurses working different shift systems in Taiwan. Nurs Health Sci. 2016;18(2):223–229.
Chrobak AA, Nowakowski J, Zwolińska-Wcisło M, et al. Associations between chronotype, sleep disturbances and seasonality with fatigue and inflammatory bowel disease symptoms. Chronobiol Int. 2018;35(8):1142–1152.
Zdanowicz T, Turowski K, Celej-Szuster J, Lorencowicz R, Przychodzka E. Insomnia, Sleepiness, and Fatigue Among Polish Nurses. Workplace Health Saf. 2020;68(6):272–278.
Chueh KH, Chen KR, Lin YH. Psychological Distress and Sleep Disturbance Among Female Nurses: Anxiety or Depression?. J Transcult Nurs. 2021;32(1):14–20.
Park CH, Bang M, Ahn KJ, Kim WJ, Shin NY. Sleep disturbance-related depressive symptom and brain volume reduction in shift-working nurses. Sci Rep. 2020;10(1):9100. Published 2020 Jun 4.
Aydin Sayilan A, Kulakaç N, Uzun S. Burnout levels and sleep quality of COVID-19 heroes. Perspect Psychiatr Care. 2021;57(3):1231–1236.

Table 1 Participant Characteristics

Variables

Total(n=434)

Training (n=304)

Validation(n=130)

c²

P-Value

Sex

Male

Female

68(15.7)

366(84.3)

45(14.8)

259(85.2)

23(17.7)

107(82.3)

0.575

0.448

Age [years]

＜30

30-40

＞40

220(50.7)

150(34.6)

64(14.7)

151(49.7)

107(35.2)

46(15.1)

69(53.1)

43(33.1)

18(13.8)

0.429

0.807

Education level

College degree and below

Undergraduate

Master degree and above

82(18.9)

308(71.0)

44(10.1)

54(17.8)

219(72)

31(10.2)

28(21.5)

89(68.5)

13(10.0)

0.855

0.652

Marital status

Single

Married

218(50.2)

216(49.8)

156(51.3)

148(4.78)

62(47.7)

68(52.3)

0.478

0.489

Fertility

Not giving birth

pregnancy

Have given birth

290(66.8)

50(11.5)

94(21.7)

203(66.8)

36(11.8)

65(21.4)

87(66.9)

14(10.8)

29(22.3)

0.127

0.938

Chronic diseases

None

Yes

240(55.3)

194(44.7)

168(55.3)

136(44.7)

72(55.4)

58(44.6)

0.001

0.981

Way of appointment

Formally editing

Contract system

other

98(22.6)

256(59.0)

80(18.4)

64(21.1)

187(61.5)

53(17.4)

34(26.2)

69(53.1)

27(20.8)

2.698

0.260

Drink tea or coffe at work

Yes

154(35.5)

280(64.5)

105(34.5)

199(65.5)

49(37.7)

81(62.3)

0.395

0.529

Shift work

Yes

210(48.4)

224(51.6)

154(50.7)

150(49.3)

56(43.1)

74(56.9)

2.096

0.148

Fatigue

None

Yes

161(37.1)

273(62.9)

114(37.5)

190(62.5)

47(36.2)

83(63.8)

0.071

0.790

Anxiety

None

Yes

246(56.7)

188(43.3)

172(56.6)

132(43.4)

74(56.9)

56(43.1)

0.004

0.947

Depression

None

Yes

228(52.5)

206(47.5)

161(53.0)

143(47.0)

67(51.5)

63(48.5)

0.074

0.786

Burnout

Low

Mederate

High

132(30.4)

156(35.9)

146(33.6)

94(30.9)

109(35.9)

101(33.2)

38(29.2)

47(36.2)

45(34.6)

0.140

0.932

Table 2 Risk Factors for Sleep Disturbance According to Univariate Analysis

Variable

Sleep disturbance

Normal sleep

(n=124)

Sleep disturbance

(n=180)

c²/Z

P-value

Sex

Male

Female

22(48.9%)

102(39.4%)

23(51.1%)

157(60.6%)

1.435^a

0.231

Age [years]

＜30

30-40

＞40

83(55.0%)

38(35.5%)

3(6.5%)

68(45.0%)

69(64.5%)

43(93.5%)

36.165^a

0.000

Education level

College degree and below

Undergraduate

Master degree and above

16(29.6%)

93(42.5%)

15(48.4%)

38(70.4%)

126(57.5%)

16(51.6%)

3.780^a

0.151

Marital status

Single

Married

86(55.1%)

38(25.7%)

70(44.9%)

110(74.3%)

27.278^a

0.000

Fertility

Not giving birth

pregnancy

Have given birth

106(52.2%)

8(22.2%)

10(15.4%)

97(47.8%)

28(77.8%)

55(84.6%)

33.484^a

0.000

Chronic diseases

None

Yes

103(61.3%)

21(15.4%)

65(38.7%)

115(84.6%)

65.472^a

0.000

Way of appointment

Formally editing

Contract system

other

27(42.2%)

71(38.0%)

26(49.1%)

37(57.8%)

116(62.0%)

27(50.9%)

2.168^a

0.338

Drink tea or coffe at work

Yes

48(45.7%)

76(38.2%)

57(54.3%)

123(61.8%)

1.611^a

0.204

Shift work

Yes

79(51.3%)

45(30.0%)

75(48.7%)

105(70.0%)

14.272^a

0.000

Fatigue

None

Yes

68(62.2%)

17(12.9%)

65(37.8%)

115(87.1%)

72.251^a

0.000

Anxiety

None

Yes

107(60.9%)

26(18.2%)

63(39.1%)

117(81.8%)

57.141^a

0.000

Depression

None

Yes

98(60.9%)

26(18.2%)

63(39.1%)

117(81.8%)

57.141^a

0.000

Burnout

Low

Mederate

High

50(53.2%)

49(45.0%)

25(24.8%)

44(46.8%)

60(55.0%)

76(75.2%)

-4.604^b

0.000

Notes: Data are number (%).^aChi-square test. ^bMann-Whitney

Table 3 Logistic Regression for the Predictors of Sleep Disturbance

Variable	Β Coefficient	OR(95%CI)	P-Value
Age(＜30/30-40/≥40)	1.992	7.33(1.56-34.41)	0.012
Chronic diseases（Y/N）	1.400	4.05(1.91-8.59)	＜0.001
Anxiety（Y/N）	1.235	3.44(1.60-7.41)	0.002
Depression（Y/N）	1.315	3.73(1.83-7.59)	＜0.001
Burnout（Y/N）	1.485	4.41(1.90-10.24)	0.001
Fatigue（Y/N）	1.016	2.76(1.40-5.45)	0.003

Table 4 Accuracy of the Prediction Score of the Nomogram for Estimating the Risk of Sleep Disturbance Presence

Variable	Value(95%CI)
Variable	Training Cohort	Validation Cohort
Area under ROC curve, concordance index	0.90(0.86-0.93)	0.89(0.83-0.95)
Cutoff value	0.53	0.55
Sensitivity, %	83.9(77.5,88.8)	88.8(79.2,94.4)
Specificity, %	83.1(75.0,89.0)	74.0(59.4,84.9)
Positive predictive value, %	87.8(81.7,92.1)	84.5(74.6,91.2)
Negative predictive value, %	78.0(69.8,84.6)	80.4(65.6,90.1)
Positive likelihood ratio	4.95	3.41
Negative likelihood ratio	0.19	0.15

No competing interests reported.

Download PDF

Journal Publication

published 28 Aug, 2023

Read the published version in BMC Nursing →

Editorial decision: Major revision
10 Apr, 2023
Reviews received at journal
02 Apr, 2023
Reviewers agreed at journal
21 Mar, 2023
Reviews received at journal
30 Jan, 2023
Reviewers agreed at journal
20 Jan, 2023
Reviewers invited by journal
20 Jan, 2023
Editor assigned by journal
20 Jan, 2023
Editor invited by journal
07 Sep, 2022
Submission checks completed at journal
07 Sep, 2022
First submitted to journal
19 Aug, 2022

You are reading this latest preprint version

Risk prediction of sleep disturbance in clinical nurses: a nomogram and artificial neural network model

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Subjects And Methods

Subjects

Instruments

General Demographic Characteristics Questionnaire

The Pittsburgh Sleep Quality Index

The Maslach Burnout Inventory

Fatigue Scale

Hospital Anxiety and Depression Scale

Data collection

Statistical analysis

Results

Discussion

Age

Chronic disease

Fatigue

Anxiety-depression

Burnout

Limitations

Conclusion

Declarations

References

Tables

Additional Declarations

Status:

Journal Publication

Version 1