ORIGINAL ARTICLE

The Journal of Nursing Research h VOL. 23, NO. 1, MARCH 2015

Predictors of Sleep Quality and Successful Weaning From Mechanical Ventilation Among Patients in Respiratory Care Centers Ching-Ju Chen1 & Ling-Nu Hsu2 & Gretl McHugh3 & Malcolm Campbell4 Ya-Ling Tzeng5* 1

PhD, RN, Head Nurse, Department of Nursing, China Medical University Hospital, Taiwan, ROC & 2MSN, RN, Advisor, Department of Nursing, China Medical University Hospital, Taiwan, ROC & 3PhD, RN, Senior Lecturer, School of Nursing, Midwifery and Social, Work, University of Manchester, Manchester, England & 4PhD, Lecturer in Statistics, School of Nursing, Midwifery and Social Work, University of Manchester, Manchester, England & 5 PhD, RN, Associate Professor, School of Nursing, China Medical University, and Adjunct Advisor, Department of Nursing, China Medical University Hospital, Taiwan, ROC.

ABSTRACT Background: Poor quality of sleep may result in more problems for patients who undergo weaning from mechanical ventilation because it could result in disabled muscle relaxation and affect the function of the respiratory muscles. Few studies have specifically investigated what factors contributed to quality of sleep and weaning outcomes. Purpose: This study investigates the predictors of quality of sleep and successful weaning from mechanical ventilation in patients at respiratory care centers. Methods: We used a cross-sectional design to recruit 94 patients who were in the process of weaning from ventilation at three respiratory care centers in a medical center in central Taiwan. A structured questionnaire was used to collect data. Disease severity during the first 24 hours after commencing the weaning process was assessed using the Acute Physiology and Chronic Health Evaluation II. Level of consciousness was evaluated using the Glasgow Coma Scale, and quality of sleep was measured using the Verran and SnyderYHalpern Sleep Scale. Stepwise multiple regression and logistic regression were used for multivariate analysis. Results: Fifty-three (56.4%) of the 94 participants successfully completed the weaning process. Participants who successfully weaned within 72 hours were younger (p = .038), had a lower level of disease severity (p G .001), and had a better quality of sleep (p = .004) than their counterparts who failed to wean. Factors including disease severity (B = j1.32), current use of hypnotic drugs (B = j10.71), and having three-to-four coexisting chronic diseases (B = j9.91) contributed negatively to quality of sleep. Factors including level of consciousness (odds ratio [OR] = 1.64), quality of sleep (OR = 1.05), disease severity (OR = 0.81), and alcohol consumption history (OR = 0.21) were found to significantly impact weaning success. Conclusions/Implications for Practice: A strong relationship was identified between disease severity and quality of

sleep. Both factors are significant predictors of successful weaning from mechanical ventilation. A better understanding of the related risk factors will help improve the care provided by nurses and medical personnel to patients undergoing the weaning process.

KEY WORDS quality of sleep, weaning, mechanical ventilation, respiratory care center.

Introduction Quality of sleep is an important indicator in patients who undergo weaning from mechanical ventilation (Tracy & Chlan, 2011). Sufficient sleep at night is recognized as a critical factor in the success of the weaning process. Therefore, improving the nighttime sleep quality of patients being weaned from ventilators is a key goal for medical personnel (Astle & Smith, 2007; Branson, 2012). Mechanical ventilation support is designed to help patients with respiratory failure maintain reasonable oxygenation and ventilation (Christine, 2006). The ultimate goal of successful weaning from mechanical ventilators is to enable patients to breathe spontaneously and without the aid of a ventilator for over 72 hours with proper physical monitoring (Dries, McGonigal, Malian, Bor, & Sullivan, 2004). Once the cause of the respiratory failure has abated, patients Accepted for Publication: November 20, 2013 *Address correspondence to: Ya-Ling Tzeng, No. 91, Hsueh-Shih Road, Taichung City 40402, Taiwan, ROC. Tel: +886 (4) 2205-3366 ext. 7110; Fax: +886 (4) 2201-7825; E-mail: [email protected] Cite this article as: Chen, C. J., Hsu, L. N., McHugh, G., Campbell, M., & Tzeng, Y. L. (2015). Predictors of sleep quality and successful weaning from mechanical ventilation among patients in respiratory care centers. The Journal of Nursing Research, 23(1), 65Y74. doi:10.1097/jnr.0000000000000066

65 Copyright © 2015 Taiwan Nurses Association. Unauthorized reproduction of this article is prohibited.

The Journal of Nursing Research

should be weaned as soon as possible to quickly resume spontaneous respiratory and physical functions. For example, Boles et al. (2007) integrated the statements issued by six separate international consensus conferences on weaning from mechanical ventilation and emphasized that prolonged mechanical ventilator causes physical discomfort, increases the risks for pneumonia and airway trauma, and increases the costs of treatment and care. Mechanically ventilated patients may experience fragmented sleep because of the ventilation-related distresses caused by endotracheal tube and ventilator use (Patel, Chipman, Carlin, & Shade, 2008). Therefore, poor subjective sleep-related outcomes in hospitalized patients should not be ignored (Chen, Lin, Tzeng, & Hsu, 2009). A complete sleep cycle is typically 90 minutes for a normal adult who requires 6Y8 hours of sleep time every night (Campbell & Murphy, 2007). However, intensive care patients have experienced frequent interruptions in their sleep cycle and wake up every 20 minutes on average. The intubation period for sleep in these patients is prolonged, and sleeping quality is lowered. It is difficult for patients to achieve deep sleep and to make up for their loss of sleep. Although some studies have shown that 50%Y60% of the patients in hospital units are able to maintain their normal nighttime sleep patterns, maintaining normal patterns does not necessarily correlate with the successful completion of the sleep cycle (Hardin, Seyal, Stewart, & Bonekat, 2006; Honkus, 2003). Weaning patients from a mechanical ventilator is a timeconsuming and complex process that is a unique experience for each patient. Before starting this process, patients should be physically and psychologically prepared and show overall progress (Goodman, 2006) in terms of achieving adequate physical fitness and getting adequate sleep. A review of the literature on sleep quality and weaning outcomes from a mechanical ventilator showed that physiological, psychological, and environmental factors impact sleep quality and may also affect the outcomes of weaning. Various studies have investigated the physiological relationship between sleep quality and several relevant variables, including age (Reid et al., 2006; Yao, Yu, Cheng, & Chen, 2008), gender (Hsu & Lin, 2005; Machado, Barreto, Passos, & Lima-Costa, 2006), and long-term condition and number of existing chronic diseases (Haung & Tsai, 2001; Wu, Su, Fang, & Chang, 2012). For instance, Foley, Ancoli-Israel, Britz, and Walsh (2004) found that, for 1,506 participants aged 55Y84 years in the United States, more than 40% with four or more chronic diseases rated their quality of sleep as fair or poor. By contrast, only 10% of participants who did not have a chronic disease reported their quality of sleep as fair or poor. Other studies indicated that multiple factors such as disease severity, albumin level, time to tracheostomy, sum of chronic comorbidities, and alcohol may affect weaning outcomes (Diaz-Abad, Verceles, Brown, & Scharf, 2012; Klatsky, 2004; Wu, Kao, Hsu, Hsieh, & Tsai, 2009). It is very important to consider whether disease symptoms in patients continue to improve, whether they have optimum physical fitness and sufficient nutrition to take

Ching-Ju Chen et al.

on the weaning process, and whether their sleeping habits and patterns will affect the progress of weaning (Goodman, 2006; Wu et al., 2009). In addition, disease severity and use of hypnotic drugs may affect the sleep quality of patients undergoing the weaning process (Collop, Adkins, & Phillips, 2004; Dines-Kalinowski, 2002; Frisk & Nordstro¨m, 2003). Research also indicates that psychological factors affect the physiological condition of patients during the weaning process (Chen et al., 2009). For example, Twibell, Siela, and Mahmoodi (2003) emphasized that patients tend to worry about successful weaning, especially when they experience changes in physiological condition such as hemodynamic instability or metabolic imbalance. Chen and Ma (2004) found that patients who are unable to cope with weaning-related physiological symptoms experienced lower levels of comfort. As a result, assessing the physiological condition of patients immediately before weaning is important to improve their sleep quality and weaning success. Although factors such as light and noise may affect quality of sleep for hospitalized patients (Collop et al., 2004; Morgan & Closs, 1999; Walder, Francioli, Meyer, Lancon, & Romand, 2000), we found no studies that supported a direct effect of environmental factors on weaning outcomes. Overall, quality of sleep is not only affected by individual characteristics and physical condition but also by the success of the mechanical ventilation weaning process. However, few studies have explored the relationships between quality of sleep and weaning outcomes in Taiwan. Therefore, the overall aim of this study was to understand the associations among patient demographics, disease distribution, sleep quality, and weaning outcomes and to identify the predictors of successful weaning and quality of sleep.

Methods Design and Sample We adopted a cross-sectional design with consecutive sampling to recruit participants in three respiratory care centers (RCCs) with 64 beds at a medical center in Taiwan. All eligible patients were approached. Only those patients who met the inclusion criteria and provided written informed consent were enrolled as participants. The inclusion criteria were (a) aged 17 years or older, (b) currently in the process of being weaned from a ventilator, (c) conscious and able to communicate in Mandarin or Taiwanese, and (d) in the ‘‘low-risk’’ health status (heartbeat rate = 60Y120 times per minute, blood pressure = 90Y150/50Y100 mmHg, a respiratory frequency of G25 times per minute with a stable respiratory pattern, SPO2 9 90%, and oxygen use concentration G 50%) as determined by the attending physicians. The exclusion criteria were (a) brain or spinal nerve damage, (b) central nervous system lesions, or (c) mental impairment. An average of 12 patients per month was weaned from mechanical ventilation at the target RCCs. Therefore, we anticipated recruiting 90Y100 eligible patients over the 10-month

66 Copyright © 2015 Taiwan Nurses Association. Unauthorized reproduction of this article is prohibited.

Sleep Quality and Mechanical Ventilation Weaning

duration of the study. This number is adequate for multiple regression analysis to predict quality of sleep with up to six independent predictors, assuming 5% significance and 80% power and moderate effect sizes (Miles & Shevlin, 2001; Tabachnick & Fidell, 2001). One hundred patients who met the inclusion criteria agreed to participate. Six of these were later excluded because of factors including fatigue, a pulse rate of 9130 times per minute, a breathing rate over 35 times per minute, and restoration of ventilator use because of infection. Data were collected from 94 patients.

Measures All participants completed a four-part structured questionnaire that was delivered in an interview format. The first two parts covered demographic and clinical information, including age, gender, number of coexisting chronic diseases, history of alcohol use, current and prior use of hypnotic drugs, prior tracheotomy surgery, albumin level, and days of ventilator use. Weaning outcome was determined as a dichotomous yes/no answer for data analysis.

The Acute Physiology and Chronic Health Evaluation II and the Glasgow Coma Scale The third part of the questionnaire contained the Acute Physiology and Chronic Health Evaluation II (APACHE II). This was used to record disease severity during the first 24 hours of the weaning process and included 14 items with a total possible range of scores between 0 and 71 (Kahn et al., 1998). Subscales included (a) the 12-item acute physiology score (APS), which included the Glasgow Coma Scale (GCS), scored from 0 to 60Vhigher scores were associated with lower physiological normality; (b) age, scored from 0 to 6 according to age groupVhigher scores were associated with higher age groups; and (c) chronic health evaluation, one item. Patients with a defined chronic disease who had undergone emergency surgery were scored 5, those who had undergone optional surgery were scored 2, and all others were scored 0. The Cronbach’s ! for the full scale was calculated as .81 for this study. The GCS used within the APACHE II APS score was also considered separately. The GCS measures the depth and duration of impaired consciousness and of comas in patients based on changes in consciousness (Teasdale & Jennett, 1974). The scale, which may be recorded by doctors and nursing staff, has been generally applied in critical care units in hospitals. The GCS includes three items: motor response (scored 1Y6), verbal response (scored 1Y5), and eyes opening (scored 1Y4). The total possible range of scores for the GCS is 3Y15, with higher scores indicating greater consciousness. For this study, the Cronbach’s ! for the full GCS was .80.

Verran and SnyderYHalpern Sleep Scale The fourth part of the questionnaire was the Verran and SnyderYHalpern (VSH) Sleep Scale. This self-report ques-

VOL. 23, NO. 1, MARCH 2015

tionnaire provides a subjective evaluation of the sleep-related satisfaction of institutionalized patients (Snyder-Halpern & Verran, 1987). The VSH scale was originally designed to assess the sleep quality of hospitalized patients (including intensive care unit patients) and is considered appropriate for use in this study to measure quality of sleep. The scale uses a 100-millimeter horizontal visual analog measurement. Contrasting descriptions of perceived feelings toward sleep are found at both ends of the scale. There are 15 items (total scores = 1500), with possible scores for each item ranging from 0 to 100. The subscale includes three major dimensions: sleep disturbance (eight items), effective sleep (four items), and supplementary sleep (three items). However, this scale may be normalized to correct for variation in subscale lengths, resulting in the score range of 0Y100, with higher scores associated with better sleep quality. The Chinese Version of the VSH Sleep Scale has high reliability and validity (Lin & Tsai, 2003 ). For this study, the Cronbach’s ! for the full scale was .89. Values for the subscales were .83 for sleep disturbance, .83 for effective sleep, and .85 for supplementary sleep.

Data Collection Data collection was conducted from March to December 2009. Because participants were using a mechanical ventilator, data were collected through interviews conducted by the first author at the bedside of participants and through electronic records checked by a nurse with 7 years of clinical experience in critical care units. The first author collected demographic and disease-related data from participants at the beginning of the weaning process. The nurse then collected the data necessary for calculating APACHE II scores within 24 hours of the start of the weaning process from electronic medical records and double-checked the calculations made by clinical nurses to ensure consistency. To minimize the distress to participants during the weaning process, data for the VSH Sleep Scale were collected from participants by the first author 72 hours after the start of the weaning process. Although the VSH Sleep Scale was designed as a self-administered questionnaire, it may also be completed by the interview. Interviews were timed so as to not intrude on treatment times or visiting hours. Families were allowed to be present during the interview process in case participants required assistance from their relatives to answer questions.

Ethical Consideration The China Medical University Hospital institutional review board approved this study (DMR-97-IRB-024). Data collection began only after a patient provided written informed consent to participate. Before obtaining this consent, researchers explained the purpose, method, and process of this study to patients and their families and stated clearly that they had the right to withdraw at any time. Furthermore, the researchers clarified that participation in this study would not affect their treatment in terms of therapy or quality of care and that all 67

Copyright © 2015 Taiwan Nurses Association. Unauthorized reproduction of this article is prohibited.

The Journal of Nursing Research

information collected would be treated as anonymous and confidential. All participants were above the age of legal consent in Taiwan and thus able to give their own consent.

Statistical Analysis Data were analyzed using SPSS version 16.0 (SPSS, Inc., Chicago, IL, USA). Demographic characteristics and the disease attributes and distributions of the study sample were summarized descriptively. Statistical tests examined differences in demographic characteristics, disease attributes, and VSH Sleep Scale scores between participants who had and had not successfully weaned within 72 hours. The independent t test, the # 2 test, and the # 2 test for trend were used, respectively, for continuous, dichotomous/nominal, and ordinal variables. The independent t test, one-way analysis of variance, Pearson’s correlation, and Spearman’s correlation were used to examine relationships between demographic characteristics, disease attributes, and VSH Sleep Scale scores. Stepwise logistic regression analysis was used to estimate models for predicting the odds of ventilator weaning success, with odds ratios, confidence intervals, p values, and the Nagelkerke pseudo-R2 estimated. The Nagelkerke R2 is a generalized coefficient of determination that is used to measure the generalized variance explained in a logistic regression model, that is, the equivalent of the coefficient of determination R2 in multiple regression models (Nagelkerke, 1991). Finally, stepwise linear regression analysis was used to estimate the ability of models to predict sleep quality; unstandardized regression coefficients, their 95% confidence intervals, standardized regression coefficients, p values, and adjusted R2 were estimated. As a guide to their likely statistical importance, variables were considered candidates for modeling when the p value of their association with dependent variables, that is, weaning outcomes and global quality of sleep, satisfied a conservative p G .25 (Hosmer & Lemeshow, 2000).

Results Participant Characteristics The participants included 45 men (47.9%) and 49 women (52.1%). The mean age of the 94 participants was 69.5 (SD = 15.9, range = 28Y85) years, and 64 (68.1%) were over 65 years old. Most were married, unemployed, educated at the high school or lower level, and lived with their families. The mean GCS score for participants was 13.7 (SD = 1.5, range = 11Y15). Participants had spent an average of 39.9 (SD = 19.0) days on the ventilator.

Coexisting Chronic Diseases Eighty (85.1%) participants had at least one chronic disease, and 62 (66.0%) had one or two coexisting diseases, with a mean of 1.9 (Table 1). Age and number of chronic diseases were significantly correlated (Spearman’s > = .26, p = .015).

Ching-Ju Chen et al.

TABLE 1.

Coexisting Chronic Diseases in Participants (N = 94) Variable

n

%

Number of chronic diseases None 1Y2 3Y4

14 62 18

14.9 66.0 19.1

Type of chronic diseasea Diabetes Hypertension Heart disease Respiratory tract disease Kidney disease Stroke Others (digestive and immune systems)

35 34 21 19 12 3 17

37.2 36.2 22.3 20.2 12.8 3.2 18.1

Note. aParticipants may have more than one type of disease.

Independent Variables and Weaning Outcomes Fifty-three participants (56.4%) successfully completed the weaning process within 72 hours (the weaned group), whereas 41 (43.6%) did not (the nonweaned group). Age (p = .038), GCS score (p = .005), and APACHE II score (p G .001) all differed significantly between these two groups. Participants in the nonweaned group were older and less conscious and had a higher level of disease severity. The APACHE II subscale scores of the nonweaned group were also significantly worse: APS (p = .043), age score (p = .042), and chronic health score (p = .002).

Verran and SnyderYHalpern Sleep Scale and Independent Variables The normalized mean score for the VSH scale was 41.6 (SD = 15.8) for all participants. This represents a considerably lower value for quality of sleep than the midpoint of the 100-point VSH. There were different levels of association between independent variables and the VSH Sleep Scale and its subscales (Table 2). For the VSH Sleep Scale, the item with the highest significance was disease severity (p G .001). Other significant items included current use of hypnotic drugs (p = .002), prior use of hypnotic drugs (p = .003), number of coexisting chronic diseases (p = .025), and age (p = .004).

Sleep Scale and Weaning Outcome Comparison of the VSH scale and its subscale scores by weaning outcome group shows significantly higher scores for the weaned group for the total sleep scale, the sleep disturbance scale, and the sleep effectiveness scale (all ps G .01; Table 3). In addition, the weaned group earned a slightly but not significantly higher score on the sleep supplementation scale (p = .443). Therefore, in general, the sleep quality of the weaned group was higher than that of the nonweaned group.

68 Copyright © 2015 Taiwan Nurses Association. Unauthorized reproduction of this article is prohibited.

Sleep Quality and Mechanical Ventilation Weaning

VOL. 23, NO. 1, MARCH 2015

TABLE 2.

Relationship of Demographic Characteristics and Disease Attributes in the VSH Sleep Scale (N = 94) Variable

VSH Sleep Scale

SDS

SES

SSS

> = j.317**

> = j.237*

> = j.290**

> = j.329**

Gender

t = 1.21

t = 1.76

t = 0.93

t = j0.36

Number of coexisting chronic diseasesa

F = 3.88*

F = 4.48*

F = 3.50*

F = 0.36

Prior use of hypnotic drugs

t = j3.02**

t = j3.07**

t = j3.62***

t = j0.38

Current use of hypnotic drugs

t = j3.25**

t = j3.34***

t = j3.44***

t = j0.83

Tracheotomy

t = j1.64

t = j1.17

t = j1.27

t = j2.16*

Albumin level

t = 2.27*

t = 2.15*

t = 2.16*

t = 1.26

Days of ventilator use

> = j.04*

> = j.07

> = j.07

> = .04*

Glasgow Coma Scale

> = j.05

> = j.05

> = j.08

> = .03*

Disease severity (APACHE II)

> = j.36***

> = j.29**

> = j.31**

> = j.34***

Age

Note. VSH = Verran and SnyderYHalpern; SDS = Sleep Disturbance Scale; SES = Sleep Effectiveness Scale; SSS = Sleep Supplementation Scale; > = Spearman’s correlation; t = independent t test statistic; F = one-way ANOVA statistic; APACHE II = Acute Physiology and Chronic Health Evaluation II. a Tukey test in a one-way ANOVA: three to four diseases G no disease in VSH and SDS; three-to-four diseases G one-to-two diseases and no disease and in SES. *p G .05. **p G .01. ***p G .001.

Predictive Factors of Successful Weaning Variables entered in the multiple stepwise logistic regression model included age, number of coexisting chronic diseases, history of alcohol use, albumin level, days of ventilator use, GCS score, APACHE II score, and VSH scale score based on the bivariate p value of G.25 (Hosmer & Lemeshow, 2000). For modeling purposes, the number of coexisting chronic diseases was converted into three levels: 0, 1Y2, and 3Y4. However, only four variables remained in the final model (Table 4). Adjusted for other variables in the model, GCS score, APACHE II score, VSH Sleep Scale score, and history of alcohol use were each significant predictors of successful weaning, with a Nagelkerke R2 of 41.1%. For each increase of one point in GCS score, the odds of successful weaning were multiplied by a factor of 1.644, each one-point increase in APACHE II score eroded the odds of successful weaning by a multiple of 0.81 (i.e., reduced by one fifth), and each one-point increase in total sleep scale score increased the odds of successful weaning by a multiple of 1.015. Finally, participants with a history of alcohol use were 0.208 times as likely to achieve successful weaning (i.e., reduced by four fifths).

Predictive Factors of Quality of Sleep Variables entered into the stepwise linear regression model included age, number of coexisting chronic diseases, prior use of hypnotic drugs, current use of hypnotic drugs, albumin level, disease severity (APACHE II), days of ventilator use, and the GCS score, again based on the bivariate p G .25 (Hosmer & Lemeshow, 2000). The number of coexisting chronic diseases was again converted into three levels. Only the APACHE II score, current use of hypnotic drugs, and three-to-four chronic diseases were included in the final model, with an adjusted R2 of 25.3% (Table 5). The total quality of sleep decreased by

1.323 for each increase of one point in the disease severity score. Participants currently on hypnotic drugs scored 10.717 less than those not taking the drugs, and participants with threeto-four coexisting chronic diseases scored 9.905 less than those with fewer than three chronic diseases.

Discussion The main purpose of this study was to elicit the predictors of sleep quality in patients who had successful weaned from mechanical ventilation in Taiwan. Our study generated a number of important findings. Initially, 53 (56.4%) of the 94 participants were successfully weaned within 72 hours of beginning the weaning process. This rate of success is similar to the results of other studies that investigated the weaning rate at RCCs in Taiwan (48%Y68%; Liu et al., 2013; Wu et al., 2009). Our weaning failure rate of 43.6% is slightly higher than the average weaning failure rate found in a review of six nonTaiwanese studies (26%Y42%, mean = 31.2%; Boles et al., 2007). In addition, the mean age in this study was lower than another Taiwanese study by Wu et al. in 2009 (69.5 vs. 73.2 years), although those authors did not report the ratio of older patients included in an earlier study. Furthermore, 64 (68.1%) in our sample were aged Q65 years, and 80 (85.1%) had chronic diseases, with an average of nearly two chronic diseases per participant. These findings compare with those of other studies on chronic diseases in the elderly (Chasens, Sereika, Weaver, & Umlauf, 2007; Wu et al., 2009). We found several factors that appear to predict weaning outcome and quality of sleep. Age, level of consciousness, and disease severity were three factors that showed significant differences between the two groups. The average age in the weaned group was less than that in the nonweaned group, supporting the findings of two previous studies (Lone & Walsh, 2011; 69

Copyright © 2015 Taiwan Nurses Association. Unauthorized reproduction of this article is prohibited.

The Journal of Nursing Research

Ching-Ju Chen et al.

TABLE 3.

Comparison of Demographic Characteristics, Disease Attributes, and Quality of Sleep by Weaning Outcome (N = 94) Weaned (n = 53) Variable Age: M (SD), years e64 Q65 Gender Male Female

Nonweaned (n = 41)

n

%

n

%

66.5 21 32

17.6 39.6 60.4

73.3 9 32

12.4 22.0 78.0

26 27

49.1 50.9

19 22

t/# 2 t = j2.10

p .038*

# 2 = 0.07

.794

= 3.29

.070

46.3 53.7 # 2for

Number of coexisting chronic diseases None 1Y2 3Y4

11 32 10

20.7 60.4 18.9

3 30 8

7.3 73.2 19.5

History of alcohol use

11

2.8

15

36.6

# 2 = 2.90

.089

trend

Prior use of hypnotic drugs

10

18.9

6

14.6

# = 2.93

.588

Current use of hypnotic drugs

20

37.7

16

39.0

# 2 = 0.02

.899

46.3

# = 1.06

.304

# 2 = 2.15

.142

Tracheotomy

19

Albumin level, mg/dl Q3.0 e2.9

16 37

35.8 3.2 69.8

M

SD

Days on ventilator

37.9

Glasgow Coma Scale

14.0

Disease severity (APACHE II) Total VSH Sleep Scale Sleep Disturbance Scale Sleep Effectiveness Scale Sleep Supplementation Scale

19 7 34

2

2

17.1 82.9

M

SD

t

17.8

42.5

20.4

0.12

.240

1.4

13.2

1.6

2.91

.005**

7.9

3.7

11.3

3.7

j4.37

45.9 44.1 48.3 47.5

15.3 16.9 19.1 61.0

36.1 33.0 36.2 44.1

16.5 16.5 20.5 22.3

2.98 3.18 2.95 0.77

p

G.001*** .004** .002** .004** .443

Note. APACHE II = Acute Physiology and Chronic Health Evaluation II; VSH = Verran and SnyderYHalpern; # 2for trend: the chi-squared test for trend can be used to evaluate the association between an ordinal categorical variable and a nominal variable representing two groups. *p G .05. **p G .01. ***p G .001.

Wu et al., 2009). Patients in our study required a level of consciousness sufficient for them to agree to participate and complete the VSH. As a result, the GCS score for participants ranged from 11 to 15. The consciousness level in the

weaned group was higher than that in the nonweaned group, and the GCS was identified as a predictor of successful weaning (odds ratio [OR] = 1.644). This result is similar to that of another study (Wu et al., 2009). This study found that the

TABLE 4.

Logistic Regression Analyses to Predict Successful Weaning (N = 94) OR

95% CI

p

Pseudo-R 2

Disease severity (APACHE II)

1.644

[1.150, 2.351]

.006**

41.0%

Glasgow Coma Scale

0.810

[0.695, 0.944]

.007**

Total VSH Sleep Scale

1.003

[1.001, 1.006]

.009**

History of alcohol use

0.208

[0.063, 0.689]

.009**

Variable

Note. OR = odds ratio; CI = confidence interval; APACHE II = Acute Physiology and Chronic Health Evaluation II; VSH = Verran and SnyderYHalpern Sleep Scale. **p G .01.

70 Copyright © 2015 Taiwan Nurses Association. Unauthorized reproduction of this article is prohibited.

Sleep Quality and Mechanical Ventilation Weaning

VOL. 23, NO. 1, MARCH 2015

TABLE 5.

Multiple Linear Regression Analyses to Predict Quality of Sleep (N = 94) Variable

B

Constant

"

95% CI

p

Adjusted R 2

[ 52.078, 67.568]

G.001***

25.3%

j.325

[ j2.052, j0.593]

.001**

59.823

Disease severity (APACHE II)

j1.323

Current use of hypnotic drugs

j10.707

j.317

[j16.718, j4.696]

.001**

j9.905

j.227

[j17.734, j2.077]

.014*

3Y4 chronic diseases

Note. CI = confidence interval; APACHE II = Acute Physiology and Chronic Health Evaluation II; B is the unstandardized regression coefficient, and " is the standardized regression coefficient. *p G .05. **p G .01. ***p G .001.

severity of disease had a negative impact on weaning outcomes, making disease severity a strong predictor of successful weaning (OR = 0.81). We found an average APACHE II disease severity score of 7.9 for those successfully weaned from the ventilator. This is lower than the average of 13Y18 points found in other RCC-related studies (Chen et al., 2004; Wu et al., 2009). This may be because we gathered the APACHE II data at the beginning of the weaning process to ensure an acceptable tolerance in participants, whereas all other studies gathered APACHE II data on the day of admission to the RCC. In addition, it is noteworthy that, although only patients who were conscious and whose disease was stable were included into this study, our results support the findings of previous studies that disease severity in the weaned group was higher than that in the nonweaned group. When a disease cannot be stabilized, it delays the initial stage of weaning or prolongs the duration of weaning, resulting in a frequently interrupted weaning process. Consequently, patients with a higher degree of disease severity have more difficulty in successfully weaning from the ventilator (Lone & Walsh, 2011; Wu et al., 2009). Thus, in our study, disease severity was a key predictor of successful weaning. History of alcohol use was another predictor of successful weaning identified in this study. A long-term history of drinking may cause cardiopulmonary hypertension, pulmonary alveolus damage, and oxygenation dysfunction (Klatsky, 2004). Patients with a history of drinking may thus face greater challenges in being successfully weaned from a mechanical ventilator. We found that, when considered on its own, there was a negative relationship between age and the variables of total quality of sleep, sleep disturbance, sleep effectiveness, and sleep supplementation. These findings support other sleeprelated studies on aging, which show that the sleep cycle of elderly people undergoes changes such as decreased total sleep time, increased time spent falling asleep at sleep onset, sleep fragmentation, and daytime sleep, and that the elderly have poorer sleep efficiency than younger people (Kryger, Roth, & Dement, 2011). Poor quality of sleep not only occurs in older people who reside at home (Yao et al., 2008) but also occurs in hospitalized older patients (Shiung, Chang, Chou, & Chao, 2009) such as those in weaning units (Chen et al., 2009). Theoretically, in older people, the first stage of the

nonrapid eye movement is prolonged, deep sleep will not persist into the third and fourth stages, and the duration of the rapid eye movement stage decreases in older adults. Therefore, sleep tends to have shallower and shorter cycles in the older adults, resulting in interruptions in nighttime sleep and/or napping during the day. In addition, sleep quality worsens (Campbell & Murphy, 2007; Kryger et al., 2011). Our study also shows a relationship between disease severity and sleep quality. Lack of control over disease severity exacerbates negative physical symptoms such as physical pain and breathing difficulties. This may delay the onset of sleep or shorten sleep times, resulting in frequently interrupted sleep. Having a larger number of chronic diseases also interferes with the quality of sleep. Our findings are consistent with a national survey that found that over 40% of people aged 55Y84 years with more than three-to-four chronic diseases in the United States rated their quality of sleep as either fair or poor (Foley et al., 2004). This study found a strong and negative association between both the prior and current uses of hypnotic drugs with sleep quality. During the data collection period, medical personnel in the RCCs often heard patients complaining about not being able to sleep at night and asking for a sleeping pill. Frequently encountered situations include (a) patients falling asleep right away but not being able to go back to sleep after waking up after 1Y2 hours, (b) patients who were unable to fall asleep immediately, (c) patients who continued to have problems falling asleep and who had to try a different hypnotic for a few days until they found the one that worked satisfactorily, and (d) patients living in a cycle of staying awake for 2 consecutive days and sleeping for another 2 consecutive days. We found that, even when patients took hypnotic drugs, their quality of sleep did not appear to improve. Unlike a previous study (Pasero & McCaffery, 2002), this study shows that using hypnotics yields poorer sleep quality during the weaning stage. When the predictors were considered together in a linear regression model, only disease severity, current use of hypnotic drugs, and having three-to-four chronic diseases showed significant effects. The predictive powers of age and prior use of hypnotic drugs lost significance when they were aggregated with the other variables in the model. 71

Copyright © 2015 Taiwan Nurses Association. Unauthorized reproduction of this article is prohibited.

The Journal of Nursing Research

We found that quality of sleep, measured using the VSH Sleep Scale, was significantly and positively associated with successful weaning from a mechanical ventilator when considered on its own in bivariate analysis and in combination with other predictors in multivariate analysis (OR = 1.051). The mean VSH scores for the two groups were lower than 50 (45.9 in weaned vs. 36.1 in nonweaned) but were similar to the mean of 40.31 highlighted in another sleep-related investigation of Taiwanese critical care units (Feng & Wang, 2007). Authors of a quality-of-sleep study that targeted patients receiving respiratory care commented that an incomplete deep-sleep cycle disabled muscle relaxation, which, in mild cases, caused sore muscles and joint pains and, in extreme cases, led to general fatigue and affected the function of respiratory muscles (Chen et al., 2009). The researchers pointed out the clinical significance of the relationship between sleep quality and successful weaning outcomes, emphasizing that only patients with sufficient sleep possess the energy necessary to successfully accomplish the weaning process. Insufficient quality of sleep in patients undergoing the weaning process will negatively impact weaning outcomes.

Ching-Ju Chen et al.

process and on the key factors affecting quality of sleep and weaning outcomes. Hospital managers should provide the relevant education to nurses. Furthermore, there must be increased awareness among clinical nurses of the relationship between quality of sleep and weaning outcomes. In addition to the use of sleep medication, nurses may offer nonpharmacological treatments such as listening to relaxing music and massage to improve sleep quality. Hypnotics are frequently prescribed to patients to treat sleeplessness and to relieve respiratory tract distress or improve oxygenation. The results of this study suggest that hypnotics do not negatively affect weaning outcomes and should continue to be administered. In addition, clinical nurses and medical personnel must evaluate the physical changes in patients and provide safe and effective nonhypnotic measures or hypnotics to help patients sleep and improve their sleep quality. However, identifying the key factors that affect the success of weaning from a mechanical ventilator is useful to the medical team in improving care and management for patients undergoing the weaning process.

References Astle, S., & Smith, D. (2007). Taking your patient off a ventilator. Registered Nurse, 70(5), 34Y39; quiz, 40.

Limitations There are a number of limitations to this study. Because of limited time, resources, and manpower, recruitment was restricted to patients in central Taiwan. Although results may be generalized to groups with characteristics similar to the study sample, generalization to other populations should be considered with caution. In addition, quality of sleep was measured using a questionnaire that was filled out on the morning of the fourth day after weaning training had commenced. This is a retrospective method, and patient responses may have been influenced by patient memory error or the results of weaning. Because of limited funding, only the sleep quality scale was used to measure quality of sleep. In the future, using multiple evaluation methods with higher reliability such as sleep monitors to analyze quality of sleep may further enhance the objectivity of this line of research. Finally, this study considered the effect of hypnotic drugs use only and did not consider the potential effect of other prescribed medications. Future studies are required to confirm the findings and establish a definitive cause-and-effect relationship. It is recommended that a large-scale study be conducted to measure physical, psychological, pharmacological, and environmental factors and their interactions.

Conclusions/Implications for Practice Few studies have investigated the association between quality of sleep and weaning from mechanical ventilation in patients. This study thus helps fill an important gap in related research. We identified a close relationship between quality of sleep and disease severity among patients in RCCs and estimated the impact of these factors on the success of mechanical ventilation weaning. Initially, to reduce disease severity, clinical nurses must provide appropriate quality healthcare to patients. Nurses must be better trained on how to properly execute the weaning

Boles, J. M., Bion, J., Connors, A., Herridge, M., Marsh, B., Melot, C., I Welte, T. (2007). Weaning from mechanical ventilation. European Respiratory Journal, 29(5), 1033Y1056. doi:10.1183/09031936.00010206 Branson, R. D. (2012). Modes to facilitate ventilator weaning. Respiratory Care, 57(10), 1635Y1648. doi:10.4187/respcare.02081 Campbell, S. S., & Murphy, P. J. (2007). The nature of spontaneous sleep across adulthood. Journal of Sleep Research, 16(1), 24Y32. doi:10.1111/j.1365-2869.2007.00567.x Chasens, E. R., Sereika, S. M., Weaver, T. E., & Umlauf, M. G. (2007). Daytime sleepiness, exercise, and physical function in older adults. Journal of Sleep Research, 16(1), 60Y65. doi:10.1111/j.1365-2869 .2007.00576.x Chen, C. H., Lin, W. C., Lee, C. H., Chen, C. Z., Chu, Y. C., Chang, H. Y., & Hsiue, T. R. (2004). Determining factors for successful weaning of patients in a respiratory care centerVA one-year experience. Thoracic Medicine, 19(4), 236Y242. (Original work published in Chinese) Chen, C. J., Lin, C. J., Tzeng, Y. L., & Hsu, L. N. (2009). Successful mechanical ventilation weaning experiences at respiratory care centers. The Journal of Nursing Research, 17(2), 93Y101. doi:10.1097/JNR.0b013e3181a6a601 Chen, M. F., & Ma, F. C. (2004). Symptom distresses and coping strategies in breast cancer women with mastectomy. The Journal of Nursing, 51(4), 37Y44. doi:10.6224/JN.51.4.37 (Original work published in Chinese) Christine, N. (2006). Caring for the mechanically ventilated patient: Part one. Nursing Standard, 20(17), 55Y64; quiz, 66. doi: 10.7748/ns2006.01.20.17.55.c6450 Collop, N. A., Adkins, D., & Phillips, B. A. (2004). Gender differences in sleep and sleep-disordered breathing. Clinical Chest Medicine, 25(2), 257Y268. doi:10.1016/j.ccm.2004.01.002 Diaz-Abad, M., Verceles, A. C., Brown, J. E., & Scharf, S. M. (2012). Sleep-disordered breathing may be under-recognized in patients who wean from prolonged mechanical ventilation. Respiratory Care, 57(2), 229Y237. doi:10.4187/respcare.01260

72 Copyright © 2015 Taiwan Nurses Association. Unauthorized reproduction of this article is prohibited.

Sleep Quality and Mechanical Ventilation Weaning

VOL. 23, NO. 1, MARCH 2015

Dines-Kalinowski, C. M. (2002). Nature’s nurse: Promoting sleep in the ICU. Dimensions of Critical Care Nursing, 21(1), 32Y34. doi:10.1097/00003465-200201000-00010

(2006). Health status indicators among community-dwelling elders with arthritis: The Bambui health and aging study. Journal of Rheumatology, 33(2), 342Y347.

Dries, D. J., McGonigal, M. D., Malian, M. S., Bor, B. J., & Sullivan, C. (2004). Protocol-driven ventilator weaning reduces use of mechanical ventilation, rate of early reintubation, and ventilatorassociated pneumonia. Journal of Trauma-Injury Infection and Critical Care, 56(5), 943Y951. doi:10.1097/01.Ta.0000124462.61495.45

Miles, J., & Shevlin, M. (2001). Applying regression and correlation: A guide for students and researchers. London, England: Sage.

Feng, C. S., & Wang, R. H. (2007). Sleep quality and its associated factors among surgical intensive care unit patients. Journal of Evidence-Based Nursing, 3(1), 54Y63. doi:10.6225/JEBN.3.1.54 (Original work published in Chinese) Foley, D., Ancoli-Israel, S., Britz, P., & Walsh, J. (2004). Sleep disturbances and chronic disease in older adults: Results of the 2003 National Sleep Foundation sleep in America survey. Journal of Psychosomatic Research, 56(5), 497Y502. doi:10.1016/ j.jpsychores.2004.02.010 Frisk, U., & Nordstro¨m, G. (2003). Patients’ sleep in an intensive care unitVPatients’ and nurses’ perception. Intensive and Critical Care Nursing, 19(6), 342Y349. doi:10.1016/S0964-3397(03)00076-4 Goodman, S. (2006). Implementing a protocol for weaning patients off mechanical ventilation. Nursing in Critical Care, 11(1), 23Y32. doi:10.1111/j.1362-1017.2006.00146.x

Morgan, K., & Closs, J. S. (1999). Sleep management in nursing practice (1st ed.). London, England: Churchill Livingstone. Nagelkerke, N. J. D. (1991). A note on a general definition of the coefficient of determination. Biometrika, 78(3), 691Y692. doi:10 .1093/biomet/78.3.691 Pasero, C., & McCaffery, M. (2002). Monitoring sedation. American Journal of Nursing, 102(2), 67Y69. doi:10.1097/00000446200202000-00026 Patel, M., Chipman, J., Carlin, B. W., & Shade, D. (2008). Sleep in the intensive care unit setting. Critical Care Nursing Quarterly, 31(4), 309Y318; quiz, 319Y320. doi:10.1097/01.CNQ.0000336816 .89300.41 Reid, K. J., Martinovich, Z., Finkel, S., Statsinger, J., Golden, R., Harter, K., & Zee, P. C. (2006). Sleep: A marker of physical and mental health in the elderly. American Journal of Geriatric Psychiatry, 14(10), 860Y866. doi:10.1097/01.JGP.0000206164 .56404.ba

Hardin, K. A., Seyal, M., Stewart, T., & Bonekat, H. W. (2006). Sleep in critically ill chemically paralyzed patients requiring mechanical ventilation. Chest, 129(6), 1468Y1477. doi:10.1378/chest.129.6.1468

Shiung, T. F., Chang, Y. Y., Chou, S. S., & Chao, Y. F. (2009). An exploration on the related factors of sleep quality in the intensive care. VGH Nursing, 26(3), 274Y281. (Original work published in Chinese)

Haung, D. S., & Tsai, L. H. (2001). The problems and strategies of management for older people in long-term care setting. Hospital, 34(6), 22Y33. (Original work published in Chinese)

Snyder-Halpern, R., & Verran, J. A. (1987). Instrumentation to describe subjective sleep characteristics in healthy subjects. Research of Nursing Health, 10(3), 155Y163. doi:10.1002/nur.4770100307

Honkus, V. L. (2003). Sleep deprivation in critical care units. Critical Care Nursing Quarterly, 26(3), 179Y189; quiz, 190Y191. doi:10.1097/00002727-200307000-00003

Tabachnick, B. G., & Fidell, L. S. (2001). Using multivariate statistics. Needham Heights, MA: Allyn and Bacon.

Hosmer, D. W., & Lemeshow, S. (2000). Applied logistic regression (2nd ed.). New York, NY: Wiley. Hsu, H. C., & Lin, M. H. (2005). Exploring quality of sleep and its related factors among menopausal women. The Journal of Nursing Research, 13(2), 153Y164. doi:10.1097/01.JNR.0000387536.60760.4e Kahn, D. M., Cook, T. E., Carlisle, C. C., Nelson, D. L., Kramer, N. R., & Millman, R. P. (1998). Identification and modification of environmental noise in an ICU setting. Chest, 114(2), 535Y540. doi:10.1378/ chest.114.2.535 Klatsky, A. L. (2004). Alcohol and cardiovascular health. Integrative and Comparative Biology, 44(4), 324Y328. doi:10.1093/icb/44.4.324 Kryger, M. H., Roth, T., & Dement, W. C. (2011). Principles and practice of sleep medicine. St. Louis, MO: Elsevier Saunders. Lin, S. L., & Tsai, S. L. (2003). The reliability and validity of the Chinese version of Verran and Snyder-Halpernu Sleep Scale. VGH Nursing, 20(1), 105Y106. (Original work published in Chinese) Liu, C. J., Chu, C. C., Chen, W., Cheng, W. E., Shih, C. M., Tsai, Y. S., I Chen, P. C. (2013). Impact of Taiwan’s integrated prospective payment program on prolonged mechanical ventilation: A 6-year nationwide study. Respiratory Care, 58(4), 676Y682. doi:10 .4187/respcare.01242 Lone, N. I., & Walsh, T. S. (2011). Prolonged mechanical ventilation in critically ill patients: Epidemiology, outcomes and modelling the potential cost consequences of establishing a regional weaning unit. Critical Care, 15(2), R102. doi:10.1186/cc10117 Machado, G. M., Barreto, S. M., Passos, V. M., & Lima-Costa, M. F.

Teasdale, G., & Jennett, B. (1974). Assessment of coma and impaired consciousness: A practical scale. The Lancet, 2(7872), 81Y84. doi:10.1016/S0140-6736(74)91639-0 Tracy, M. F., & Chlan, L. (2011). Nonpharmacological interventions to manage common symptoms in patients receiving mechanical ventilation. Critical Care Nurse, 31(3), 19Y28. doi:10.4037/ ccn2011653 Twibell, R., Siela, D., & Mahmoodi, M. (2003). Subjective perceptions and physiological variables during weaning from mechanical ventilation. American Journal of Critical Care, 12(2), 101Y112. Walder, B., Francioli, D., Meyer, J. J., Lancon, M., & Romand, J. A. (2000). Effects of guidelines implementation in a surgical in tensive care unit to control nighttime light and noise levels. Critical Care Medicine, 28(7), 2242Y2247. doi:10.1097/00003246200007000-00010 Wu, C. Y., Su, T. P., Fang, C. L., & Chang, M. (2012). Sleep quality among community-dwelling elderly people and its demographic, mental, and physical correlates. Journal of Chinical Medicine Association, 75(2), 75Y80. doi:10.1016/j.jcma.2011.12.011 Wu, Y. K., Kao, K. C., Hsu, K. H., Hsieh, M. J., & Tsai, Y. H. (2009). Predictors of successful weaning from prolonged mechanical ventilation in Taiwan. Respiratory Medicine, 103(8), 1189Y1195. doi:10.1016/j.rmed.2009.02.005 Yao, K. W., Yu, S., Cheng, S. P., & Chen, I. J. (2008). Relationships between personal, depression and social network factors and sleep quality in community-dwelling older adults. The Journal of Nursing Research, 16(2), 131Y139. doi:10.1097/01.JNR.0000387298 .37419.ff

73 Copyright © 2015 Taiwan Nurses Association. Unauthorized reproduction of this article is prohibited.

The Journal of Nursing Research

睡眠品質和呼吸器脫離

VOL. 23, NO. 1, MARCH 2015

呼吸照護中心病患睡眠品質及成功脫離呼吸器之預測因素 陳靜如1 許玲女2 Gretl McHugh3 Malcolm Campbell4 曾雅玲5* 1

中國醫藥大學附設醫院護理部護理長 2中國醫藥大學附設醫院護理部顧問 3英國曼徹斯特大學 護理助產暨社工學院資深講師 4英國曼徹斯特大學護理助產暨社工學院資深統計 講師 5中國醫藥大學護理系副教授兼任中國醫藥大學附設醫院護理部顧問

背 景

睡眠品質不佳可能導致呼吸肌無法放鬆而影響呼吸器成功脫離。然而少有研究探討影響 呼吸器脫離病患的睡眠品質與呼吸器脫離結果的關係。

目 的

本研究旨在探討影響呼吸器脫離病患睡眠品質與成功脫離呼吸器之預測因素。

方 法

採橫斷面研究設計。以結構式問卷收集中部某醫學中心呼吸照護單位 94 位執行呼吸器 脫離訓練的病患資料。問卷內容包括人口學和健康狀況 、疾病嚴重度分類系統量表 、 意識清醒程度評量表 、以及維辛式睡眠量表。採逐步複迴歸和羅吉斯迴歸進行多變量 分析。

結 果

呼吸器脫離成功率為 56.4%(n = 53)。與未脫離呼吸器的病患比較 ,脫離病患年齡較輕 (p = .038)、病情嚴重度較低(p < .001)、以及有較好的睡眠品質(p = .004)。睡眠總分 之預測因子為疾病嚴重度(B = -1.32)、目前有使用安眠藥(B = -10.71)、以及 3–4 種慢 性病(B = -9.91)。呼吸器脫離成功的預測因子為意識程度(OR = 1.64)、睡眠品質(OR = 1.05)、疾病嚴重度(OR = 0.81)、以及飲酒史(OR = 0.21)。

結 論 本研究顯示 ,疾病嚴重度 、睡眠品質影響呼吸器脫離。增進對這些危險因素的瞭解 , 實務應用 有助於護理人員和臨床專業照護者提昇呼吸器脫離過程病患之照護品質。

關鍵詞:睡眠品質、脫離、呼吸器、呼吸照護中心。

接受刊載:102年11月20日 *通訊作者地址:曾雅玲  40402臺中市北區學士路91號 電話:(04)22053366-7100  E-mail: [email protected]

74 Copyright © 2015 Taiwan Nurses Association. Unauthorized reproduction of this article is prohibited.

Predictors of sleep quality and successful weaning from mechanical ventilation among patients in respiratory care centers.

Poor quality of sleep may result in more problems for patients who undergo weaning from mechanical ventilation because it could result in disabled mus...
780KB Sizes 0 Downloads 7 Views