Development of the compliance with mechanical ventilator scale in: A validity and reliability study

Abstract

Background

Mechanical ventilation is a widely used treatment modality in cardiac surgery intensive care units (ICUs), especially for patients requiring respiratory support during postoperative management.

Aim

The aim of this study was to define the process of developing a scale to assess adherence to mechanical ventilation in cardiac surgery patients and to assess the validity and reliability of the scale.

Study design

This is a scale development and scale validation study. The study was conducted with 222 patients between April and September 2023. In the first stage, face‐to‐face interviews were conducted with seven nurses. Based on the literature review and content analysis of the interviews, the researchers created an item pool. In the second stage, the validity and reliability of the Mechanical Ventilator Compliance Scale in cardiac surgery patients were assessed. The scale was developed to be used by nurses working in cardiac surgery ICUs.

Results

The first 36‐item scale was subjected to content validity assessment by 14 experts. It was revised according to the experts' feedback and a content validity criterion of 0.83 was obtained. Confirmatory factor analysis showed good fit values for the scale (χ ²/SD = 3.09, GFI = 0.90, AGFI = 0.90, CFI = 0.91, RMSEA = 0.06, and RMR = 0.07). Test‐retest analysis to assess the reliability of the scale showed good agreement between responses over time (ICC: 0.932; p < .001). Cronbach's alpha values for the scale and its subscales, behavioural and physical responses were 0.956, 0.949 and 0.912, respectively.

Conclusions

The study findings indicate that the scale is a reliable and valid tool for assessing mechanical ventilator compliance in cardiac surgery patients.

Relevance for Clinical Practice

This study provides nurses in cardiac surgery intensive care units with a measurement tool to objectively assess patients' compliance with mechanical ventilation therapy. The scale provides more reliable results by preventing subjective judgement and estimation in patient assessment.

What is known about the topic

• Sedation management is a key element in patient follow‐up in mechanical ventilation.

• In the management of the patient on mechanical ventilation, nurses rely on their cumulative knowledge from the past to the present.

• The use of protocols or measurement tools in critical patient care in intensive care units accelerates the correct decision‐making processes of nurses.

What this paper adds

• The Mechanical Ventilator Compliance Scale was found to be a valid and reliable measurement tool in patients undergoing cardiac surgery.

• In intubated patients, urgent and priority physical and behavioural findings were taken into account in the adaptation of patients to the mechanical ventilator before planned intubation.

1. INTRODUCTION

Mechanical ventilation is used to treat respiratory failure caused by various factors. Patients in cardiovascular intensive care units (ICUs) may require respiratory support because of pre‐existing conditions or intraoperative cardiac dysfunction. In such cases, mechanical ventilation support may be used to ensure hemodynamic stability in the postoperative period and achieve optimal oxygenation levels. ¹ , ² The use of mechanical ventilation is a significant and contextual stimulus that affects vital functions. Studies have shown that mechanically ventilated patients often experience cognitive and emotional problems, fear, agitation, discomfort, thirst and immobility because of their inability to communicate their needs, which can be exacerbated by the endotracheal tube. ¹ , ³ , ⁴ , ⁵ , ⁶ , ⁷ , ⁸ , ⁹ Several studies have reported that patients perceive mechanical ventilator treatment as a traumatic experience and fear death during the process. ⁹ , ¹⁰ These negative experiences may affect the individual's adjustment and cause positive or negative reactions in patients. ¹¹ During mechanical ventilation, patients may experience serious complications such as upper respiratory tract injuries, aspiration of gastric contents, hypoxia, respiratory failure, hypotension, bleeding, edema, arrhythmia, the need for reintubation, cardiac arrest and even death. ¹² , ¹³ Mahmood et al.¹³reported that 12% of patients self‐extubated. Indicators of patient‐ventilator incompatibility include patients' attempts to pull or bite the endotracheal tube, self‐extubation and patients' struggle with the ventilator, such as strong or excessive inspiratory or expiratory chest movements at any stage of the respiratory cycle, frequent or continuous cough and/or chest wall muscle tension. ¹⁴ , ¹⁵

Adaptation theory was developed by Sr. Callista Roy in 1970, utilizing von Bertalanffy's General System Theory and Helson's Adaptation Theory. Roy's Adaptation model posits that behaviour is determined by the interaction between humans and their environment. ¹⁶ The model was revised in 1986 to provide guidance for nursing practices. Accordingly, individuals are considered open systems that respond to both internal and external stimuli and attempt to adapt to these stimuli in terms of physiological, psychological, social and environmental aspects. ⁹ Positive responses to stimuli are referred to as adaptation, while negative responses are referred to as maladaptation. The goal of nursing care is to promote positive adaptation. ¹⁷

2. BACKGROUND FOR STUDY

Patient‐ventilator incompatibility can result in various negative outcomes, including decreased patient comfort, increased anxiety, inadequate gas exchange, dyspnea, increased myocardial oxygen demand, hemodynamic instability and can complicate the process of weaning the patient from the ventilator, ultimately prolonging the patient's recovery process. ³ , ¹⁰ , ¹⁸ , ¹⁹ The quality of nursing care plays a crucial role in this process. Nurses are responsible for maintaining the patency of the intubation tube and ensuring its proper placement. ¹² They must evaluate patient compliance with mechanical ventilation treatment, identify and address physical and emotional factors that may impair compliance, use effective intervention and communication methods, prevent potential problems by promoting patient compliance with treatment and environment and ensure patient recovery. ¹⁹ However, currently there is no objective measurement tool available for nurses and other health care professionals to evaluate patient compliance with mechanical ventilation treatment. Therefore, there is a need for such a tool. ²⁰ The implementation of an objective measurement tool can guide the decision‐making process of health care professionals and optimize patient care. This study aims to fill the gap in the field by developing an objective measurement tool.

3. AIMS AND OBJECTIVES OF STUDY

The aim of this study was to develop a measurement tool that allows an objective assessment of Mechanical Ventilator Compliance of postoperative patients in cardiac surgery ICUs (CMVS‐CSP). With the developed measurement tool, it is thought that standardization will be implemented in the evaluation of mechanical ventilator compliance of patients by nurses and in the overall management of mechanical ventilator compliance. Early detection of mechanical ventilator non‐compliance in patients will contribute positively in terms of patient outcomes sensitive to nursing care by preventing delirium, prolonged mechanical ventilator stay, reintubations and self‐intubation of patients.

4. DESIGN AND METHODS

This is a methodological study aimed at developing a new measurement tool for assessing compliance with mechanical ventilation in patients undergoing cardiac surgery. The scale was developed for use by nurses working in cardiac surgery ICUs. Data collection took place between April and September 2023 in the cardiac surgery ICU of a tertiary education and research hospital in Ankara. The study used the COSMIN checklist to evaluate the methodological quality of studies on health‐related measurement tools. ²¹ The flow chart for the study is presented in Figure 1.

FIGURE 1.

Flow chart.

The research was conducted in two stages.

4.1. Stage 1

The initial stage of the study involved conducting a literature review, observation/interview and item pool preparation. The researchers conducted an independent literature review on health professionals' assessment of patients' compliance with mechanical ventilation. Based on the literature review, they created an interview form with 11 items to determine the factors that affect patients' compliance with mechanical ventilation treatment. Seven intensive care nurses, two of whom were experts with at least 2 years of experience caring for patients on mechanical ventilators, were interviewed in person. The responses were analysed by two members of the research team who are experts in qualitative research. The findings were reviewed by the authors after content analysis and literature review, and an item pool was created.

4.2. Stage 2

At this stage, we conducted an analysis of the validity and reliability of the scale.

4.3. Content validity

A total of 14 experts, including two anaesthesiologists, two intensive care nurses and 10 nurses specialized in surgical diseases, evaluated the clarity and comprehensibility of the items. Each expert scored the items on a scale of 1 to 4, with 1 indicating ‘not appropriate’, 2 indicating ‘the item needs appropriate modification’, 3 indicating ‘appropriate but minor changes are needed’ and 4 indicating ‘very appropriate’. The experts were also asked to provide suggestions in the ‘Explanation’ section. Based on expert opinions, we developed a 28‐item form by eliminating 10 similar expressions, revising two items and dividing two items into four. The Lawshe technique was used to analyse the expert opinions, and a content validity ratio (CVR) value of 0.83 was obtained at a significance level of α = 0.05. ²²

5. SETTING AND SAMPLE

When determining the sample size for scale studies, it is recommended to have a sample size of 100–200 or 5–10 times the number of items in the scale to examine the item‐total score correlation. ²³ In accordance with the literature, the study enrolled 222 voluntary patients aged 18 and above who underwent cardiac surgery and received mechanical ventilator treatment for the first time.

6. DATA COLLECTION TOOLS AND METHODS

The study employed the CMVS‐CSP and the patient‐ventilator synchrony domain of the Sedation Assessment Scale (SAS), developed by the American Association of Critical‐Care Nurses (AACN), to collect data. The descriptive information form included six questions about age, gender, education level, marital status, type of surgery and duration of stay on the mechanical ventilator. To ensure comprehensibility, Sing and Masuku ²⁴ recommend pretesting items on a smaller group outside the target sample. Based on the suggestion, the CMVS‐CSP pilot study was conducted with intensive care nurses for the comprehensibility of the scale on the data of 10 patients who met the inclusion criteria. The nurses who completed the form were asked to evaluate the comprehensibility of the items. They were also asked to rate how logical and understandable the measurement tool was. Item loads and the time required to complete the scale were also analysed. The scale took an average of 15 to 25 min to complete in the pilot study. The nurses' comments were evaluated after the pilot study, and no changes were made to the items.

6.1. Construct validity

The study employed the AACN SAS which is conceptually related to the attitude being measured for criterion validity. The SAS comprises five domains related to sedation management: consciousness, agitation, patient‐ventilator synchrony, anxiety and sleep. Universal facial expressions were also used to assess the anxiety and sleep domains. Each domain is scored from 1 to 5, with scores above 2 indicating that the patient requires more sedation and relaxation interventions.

7. DATA ANALYSIS

Reliability was measured using Cronbach's alpha coefficient and test‐retest analysis. The relationship between the CMVS‐CSP and the patient‐ventilator synchrony domain of the AACN SAS was examined using Pearson's correlation analysis. Statistical significance was set at p < .05. ²⁵ Data analysis was performed using SPSS version 22.0 and AMOS statistical software. Kurtosis and skewness were used to determine whether the items of the CMVS‐CSP were normally distributed. Kurtosis and skewness values between +1.5 and −1.5, ²⁶ or between +2.0 and −2.0 ²⁷ are indicative of a normal distribution. The scale items were found to follow a normal distribution. Construct validity was analysed using both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA). Internal consistency was examined through item analysis. Scale discrimination was analysed by comparing the lower and upper 27% groups using independent samples t‐tests.

8. ETHICAL AND RESEARCH APPROVALS

Ethical approval was obtained from the institution's ethics committee where the study was conducted (No. E1‐22‐2540).

9. RESULTS

Among the patients enrolled in the study, 30.2% were female, 69.8% were male, 89.2% were married and 35.1% graduated from a primary school. The mean age was 55.890 ± 12.133 years (min. 22‐max. 79 years) and the mean length of intubation was 8.689 ± 7.181 h (min. 1‐max. 96 h) (Table 1).

TABLE 1.

Descriptive characteristics (n: 222).

Groups	n	%
Gender
Female	67	30.2
Male	155	69.8
Marital status
Single	24	10.8
Married	198	89.2
Education level
Illiterate	3	1.4
Literate	4	1.8
Primary school	78	35.1
Secondary school	36	16.2
High school	56	25.2
University	45	20.3

	Mean	SD
Age	55.890	12.133
Length of intubation	8.689	7.181

9.1. Reliability

Reliability analysis was conducted to assess the internal consistency of the scale. This analysis determines whether the scale items are consistent with each other and with the overall scale, as well as whether participants interpret the scale statements in the same way. Cronbach's alpha coefficient is commonly used to determine the reliability (internal consistency) of a scale. The evaluation criteria for Cronbach's alpha coefficient are as follows: 0.00 ≤ α < 0.40 indicates that the scale is not reliable, 0.40 ≤ α < 0.60 indicates low reliability, 0.60 ≤ α < 0.80 indicates high reliability and 0.80 ≤ α < 1.00 indicates very high reliability. ²⁸ The Cronbach's alpha of the CMVS‐CSP was 0.939. Table 2 presents the item analysis for the effect of items on internal consistency. According to Büyüköztürk ²⁹ and Tavşancıl, ³⁰ an item with a correlation coefficient of 0.3 or higher indicates higher discrimination. To improve internal consistency, we removed items 8, 9, 10, 11, 13 and 14, which had a negative impact. We repeated the reliability analysis of the scale, and the alpha coefficient was found to be 0.956. Table 3 presents the item analysis for the effect of the items on internal consistency. The analysis revealed that none of the items had an item‐total correlation value of 0.3 or less.

TABLE 2.

Item analysis.

	Scale total score if item deleted	Scale variance if item deleted	Item‐total correlation	Cronbach's alpha if item deleted	Mean	Std. deviation
I1	64.48	327.925	0.692	0.936	2.48	1019
I2	64.39	324.474	0.767	0.935	2.58	1047
I3	64.40	326.006	0.718	0.936	2.56	1056
I4	64.51	328.341	0.643	0.936	2.45	1074
I5	64.80	321.599	0.739	0.935	2.16	1188
I6	64.45	324.981	0.714	0.935	2.52	1100
I7	64.30	326.183	0.644	0.936	2.67	1160
I8	64.46	338.068	0.278	0.931	2.50	1096
I9	64.47	335.427	0.254	0.942	2.50	1334
I10	64.82	342.269	0.265	0.942	2.15	1122
I11	64.79	339.568	0.229	0.942	2.17	1129
I12	64.35	327.505	0.660	0.936	2.61	1082
I13	63.69	361.030	−0.189	0.946	3.27	1068
I14	64.64	336.438	0.285	0.942	2.32	1181
I15	64.43	328.463	0.691	0.936	2.54	1000
I16	63.99	326.769	0.643	0.936	2.97	1137
I17	63.90	329.909	0.606	0.937	3.06	1066
I18	64.86	324.178	0.757	0.935	2.11	1071
I19	64.60	326.956	0.679	0.936	2.36	1075
I20	64.71	326.579	0.719	0.936	2.26	1034
I21	65.03	327.949	0.688	0.936	1.93	1025
I22	64.68	326.553	0.617	0.937	2.28	1190
I23	65.00	327.864	0.672	0.936	1.96	1050
I24	65.04	330.171	0.621	0.937	1.93	1031
I25	64.87	329.272	0.649	0.936	2.09	1027
I26	65.03	329.379	0.614	0.937	1.93	1076
I27	65.08	328.895	0.652	0.936	1.89	1038
I28	64.27	328.062	0.661	0.936	2.69	1058

TABLE 3.

Item analysis (Repeated).

	Scale total score if item deleted	Scale variance if item deleted	Item‐total correlation	Cronbach's alpha if item deleted	Mean	Std. deviation
I1	49.57	265.197	0.718	0.954	2.48	1019
I2	49.47	262.721	0.773	0.953	2.58	1047
I3	49.49	263.210	0.750	0.953	2.56	1056
I4	49.60	265.544	0.667	0.954	2.45	1074
I5	49.89	259.440	0.763	0.953	2.16	1188
I6	49.53	263.589	0.707	0.954	2.52	1100
I7	49.38	263.975	0.656	0.954	2.67	1160
I12	49.44	265.523	0.662	0.954	2.61	1082
I15	49.51	265.817	0.712	.954	2.54	1000
I16	49.08	264.333	0.660	0.954	2.97	1137
I17	48.99	266.783	0.635	0.955	3.06	1066
I18	49.94	262.236	0.769	0.953	2.11	1071
I19	49.68	264.380	0.701	0.954	2.36	1075
I20	49.79	264.065	0.741	0.953	2.26	1034
I21	50.12	265.860	0.693	0.954	1.93	1025
I22	49.77	264.072	0.635	0.955	2.28	1190
I23	50.09	265.934	0.672	0.954	1.96	1050
I24	50.12	267.854	0.626	0.955	1.93	1031
I25	49.95	267.066	0.653	0.954	2.09	1027
I26	50.12	267.271	0.614	0.955	1.93	1076
I27	50.16	266.906	0.650	0.954	1.89	1038
I28	49.36	264.556	0.708	0.954	2.69	1058

9.2. Exploratory factor analysis (EFA)

The scale's construct validity was established through the EFA. The Bartlett's test indicated a significant relationship between the variables included in the factor analysis (p = .000 < .05). Additionally, the Kaiser–Meyer–Olkin test confirmed that the sample size was adequate for factor analysis (KMO = 0.942 > 0.60). The varimax method grouped the items into two factors, explaining 62.303% of the total variance. Table 4 presents the factor structure of the scale, with the first factor labelled as ‘Behavioural Reactions’ and the second factor as ‘Physical Reactions’. Figure 2 illustrates the scree plot.

TABLE 4.

Factor structure of the CMVS.

Factor	Load
Behavioural reactions (Eigenvalue = 11.501; Variance explained = 36.963; Alpha = 0.949)
I3	0.779
I7	0.772
I2	0.767
I17	0.758
I15	0.737
I6	0.722
I12	0.719
I16	0.712
I1	0.710
I4	0.704
I18	0.674
I19	0.660
I5	0.648
I28	0.629
I20	0.590
Physical reactions (Eigenvalue = 2.205; Variance explained = 25.340; Alpha = 0.912)
I26	0.860
I27	0.847
I25	0.832
I24	0.824
I23	0.736
I21	0.627
I22	0.569
Total variance = %62.303; Total reliability (Alpha) = 0.956

FIGURE 2.

Scree plot.

9.3. Confirmatory factor analysis (CFA)

CFA is a type of structural equation model (SEM) that measures the relationship between observed and latent variables. ³¹ The study used commonly used goodness‐of‐fit indices in the literature. Figure 3 displays the diagram of the CFA and Table 5 provides the goodness‐of‐fit criteria for CFA. The results of the analysis suggest that the fit statistics calculated by the CFA were at an acceptable level. Table 6 presents the standardized factor loadings and t‐values. The coefficients demonstrate high factor loadings, low standard error values and significant t‐values. These results confirm the construct validity of the factor structure.

FIGURE 3.

Confirmatory factor analysis.

TABLE 5.

Confirmatory factor analysis index values.

Index	Normal value*	Acceptable value**	Value
χ 2/SD	<2	<5	3.09
GFI	>0.95	>0.90	0.90
AGFI	>0.95	>0.90	0.90
CFI	>0.95	>0.90	0.91
RMSEA	<0.05	<0.08	0.06
RMR	<0.05	<0.08	0.07

Note: *, ** Sources: (Şimşek, 2007; Hooper and Mullen, 2008; Schumacker and Lomax, 2010; Waltz, Strcikland and Lenz, 2010; Wang and Wang, 2012; Sümer, 2000; Tabachnick and Fidel, 2007).

TABLE 6.

Factor loads.

Items and factors			β	Std. β	Std. error	t	p
I20	←	F1	1.000	0.736
I28	←	F1	1.036	0.745	0.093	11.155	p < .001
I5	←	F1	1.220	0.781	.104	11.729	p < .001
I19	←	F1	1.012	0.716	0.095	10.687	p < .001
I18	←	F1	1.106	0.786	0.094	11.814	p < .001
I4	←	F1	1.007	0.713	0.095	10.641	p < .001
I6	←	F1	1.042	0.721	0.097	10.762	p < .001
I16	←	F1	1.018	0.681	0.101	10.122	p < .001
I1	←	F1	1.028	0.768	0.089	11.497	p < .001
I17	←	F1	0.925	0.660	0.094	9.797	p < .001
I2	←	F1	1.147	0.833	0.091	12.586	p < .001
I3	←	F1	1.135	0.818	0.092	12.332	p < .001
I26	←	F2	1.000	0.847
I27	←	F2	0.956	0.839	0.061	15.584	p < .001
I24	←	F2	0.918	0.812	0.062	14.772	p < .001
I25	←	F2	0.954	0.847	0.060	15.806	p < .001
I23	←	F2	0.889	0.772	0.065	13.649	p < .001
I21	←	F2	0.780	0.694	0.067	11.702	p < .001
I22	←	F2	0.825	0.632	0.080	10.324	p < .001

9.4. Test‐retest

In the context of test‐retest reliability, the scale was initially administered to 50 patients from the sample group before extubation. Retest was administered 15 min after the administration of the initial test, taking into account the average duration of mechanical ventilator treatment and the duration of awakening after sedation during this period. The intraclass correlation coefficient (ICC) correlation values for the agreement between test‐retest measurements were high, and there was no significant difference between the test‐retest values (p < .05) (Table 7). This finding indicates that the scale provides reliable measurements over a short period of time.

TABLE 7.

Test‐retest.

	Test		Retest		N	t	p a	ICC b	p
	Mean	SD	Mean	SD
CMVS‐CSP total	2.153	0.944	2.156	0.933	50	−0.433	.667	0.932	.000
Behavioural reactions	2.255	0.965	2.267	0.961	50	−1.268	.211	0.918	.000
Physical reactions	1.934	0.969	1.920	0.942	50	1.151	.255	0.926	.000

^a Dependent sample t‐test.

^b Intraclass correlation coefficient.

9.5. Distinctiveness

The scale is expected to clearly differentiate between two extreme groups, the lower and upper 27% groups. ³² If there is no difference between the two groups, it suggests that the range of scores is small. The analysis showed a statistically significant difference between the lower 27% and upper 27% groups, indicating that it is capable of making sensitive measurements for distinctiveness (p < .05) (Table 8).

TABLE 8.

Distinctiveness between the upper and lower 27% groups.

	Lower 27% (n = 60)		Upper 27% (n = 60)		t	SD	p
	Mean	SD	Mean	SD
CMVS‐CSP total	1.474	0.251	3.363	0.478	−27.106	118	.000
Behavioural reactions	1.593	0.313	3.546	0.492	−25.948	118	.000
Physical reactions	1.219	0.363	2.971	0.822	−15.112	118	.000
Independent sample t‐test

9.6. Criterion validity

Patient‐ventilator synchrony domain of the SAS, which was developed by the AACN and adapted into Turkish by Akın Korhan et al., ³³ was used to ensure the convergent validity of the CMVS‐CSP. A positive correlation of 51.1% was found between the CMVS‐CSP and the patient‐ventilator synchrony domain of the SAS (Table 9). The CMVS‐CSP scale provides a more comprehensive assessment of the patient's compliance with mechanical ventilation, with physical indicators of anxiety and agitation beyond sedation compliance, blood gas results of respiratory parameters and response to painful stimuli such as intubation tube and aspiration procedures.

TABLE 9.

Correlation between the CMVS‐CSP and AACN patient‐ventilator synchrony domain of the SAS.

		AACN patient‐ventilator synchrony domain of the SAS
CMVS‐CSP total	r	0.511**
	p	.000
Behavioural reactions	r	0.512**
	p	.000
Physical reactions	r	0.394**
	p	.000

^** p < 0.001

9.7. Convergent validity and divergent validity

The results in Table 10 show that the scale is strong in terms of reliability and validity. The composite reliability (CR) values are 0.938 for F1 and 0.916 for F2, both values indicating high internal consistency and reliability. The average variance extraction (AVE) values are 0.560 for F1 and 0.611 for F2, indicating that the constructs have good validity. The maximum shared variance (MSV) is 0.508 and the average shared variance (ASV) is 0.169 for both factors, and these values are lower than the AVE, indicating that discriminant validity is achieved. Furthermore, the square roots of the AVE on the diagonal (0.748 for F1 and 0.781 for F2) are greater than the correlation coefficient between the factors (0.713), indicating that the constructs are independent of each other and have good discriminant validity. These results confirm that the scale is generally reliable and valid.

TABLE 10.

Composite reliability, average variance extraction, maximum shared variance, average shared variance and correlation analysis.

	MSV	ASV	CR	AVE	F1	F2
F1	0.508	0.169	0.938	0.560	0.748
F2	0.508	0.169	0.916	0.611	0.713	0.781

Note: Values in the diagonal represent the square root of the AVE value.

9.8. Mean values

The CMVS‐CSP items were scored on a scale of 1 to 5. To calculate the distribution range, we used the following formula:

$$\text{Distribution range}=\left(\text{Maximum value}-\text{Minimum value}\right)/\text{Number of ratings}$$

The resulting range has a width of four points, which was divided into five equal parts. The limit values were then determined. The term ‘very low’ refers to scores between 1.00 and 1.79, ‘low’ to scores between 1.80 and 2.59, ‘medium’ to scores between 2.60 and 3.39, ‘high’ to scores between 3.40 and 4.19 and ‘very high’ to scores between 4.20 and 5.00. ³⁴ Table 11 shows that the mean scores obtained from the CMVS‐CSP and its subscales of ‘behavioural reactions’ and ‘physical reactions’ were 2.366 ± 0.774 (Min = 1; Max = 4.55), 2.535 ± 0.825 (Min = 1; Max = 4.67) and 2.003 ± 0.861 (Min = 1; Max = 4.71), respectively, indicating a low level of compliance with the mechanical ventilator.

TABLE 11.

Mean CMVS‐CSP scores.

	N	Mean	SD	Min.	Max.	Scale Min‐Max
CMVS‐CSP total	222	2.366	0.774	1.000	4.550	1–5
Behavioural reactions	222	2.535	0.825	1.000	4.670	1–5
Physical reactions	222	2.003	0.861	1.000	4.710	1–5

9.9. Instrument

The CMVS‐CSP is a 22‐item scale consisting of two subscales: behavioural reactions (items 1–15) and physical reactions (items 16–22) (Table 12). All items were scored on a five‐point Likert scale, with no items being reverse scored. Responses indicate the level of nurses' agreement with the items; (1 = never, 2 = rarely, 3 = sometimes, 4 = often, 5 = always). The possible scores ranged from 22 to 110, with lower scores indicating greater patient compliance with mechanical ventilation treatment. The total score of the scale is 110, indicating the maximum level of non‐compliance with the mechanical ventilator. The lowest score was 22, representing the highest level of compliance with the mechanical ventilator. Summaries as the scores obtained from the scale increase, whereas compliance with the mechanical ventilator decreases.

TABLE 12.

Compliance with Mechanical Ventilation Scale in cardiac surgery patients.

Time from awakening until extubation	Never	Rarely	Sometimes	Frequently	Always
Items	1	2	3	4	5
Behavioural reactions
Restless, aimless behaviour and sudden motor movements
2Tense, grimaced face expression like crying
3Agitation
4Retching
5Moving hands to invasive materials, trying to remove tubes or IV lines
6Difficulty breathing/feeling of suffocation
7Elevation in peak pressure
8Alarming ventilator
9Overlap with ventilator
10Negative reaction to aspiration
11Painful expression during aspiration
12Chewing the tube
13Sweating
14Tachypnea
15Painful facial expression while on mechanical ventilator
Physical reactions
16Dyspnea
17Tachycardia
18Increase in PaCO2 level
19Decrease in PaO2 level
20Changes in pH level
21Respiratory acidosis
22Respiratory alkalosis

The literature was reviewed for studies evaluating the adaptation process in different patient groups. Although one study did not use a specific scale such as stoma compliance, it did assess patient compliance with procedures such as stoma diameter, skin care, and insertion of a new stoma bag in relation to patients' adaptation to life after stoma. ³⁵ In our study, we included questions about patients' response to the intubation tube, compliance with ventilation modes, and response to aspiration for oral care.

10. DISCUSSION

The results of the study were discussed in terms of scale development process, analyses and results.

This study introduces the CMVS‐CSP, a novel tool for measuring the compliance of intubated cardiac surgery patients in the ICU with mechanical ventilation. The validity and reliability of the instrument were also evaluated. Our literature review indicates that the CMVS‐CSP is the first comprehensive scale for determining patient compliance with mechanical ventilation. The 22‐item scale consists of two subscales that evaluate the physical and behavioural reactions of the patients. The scale items have the desired characteristics for the evaluation of patients' compliance with the mechanical ventilator. ¹² In the study in which the competencies of nurses were evaluated, it can be said that the multiple factors, physical and behavioural reactions affecting the patient in compliance with the mechanical ventilator are similar to the literature. ¹⁹ , ³⁶ The application of the scale with the data of a single group (cardiovascular surgery patient population) and the use of the scale only by nurses ensured the strength of the study by eliminating confounding factors. The nurses who completed the form were asked to evaluate the comprehensibility of the items. They were also asked to rate how logical and understandable the measurement tool was. Thus, the face validity of the scale was also evaluated. The psychometric properties of the scale were tested after the development of the item pool. ³⁷ The item pool was created by conducting a thorough literature review, interviewing experienced nurses who had worked in the cardiovascular surgery ICU for at least 2 years and obtaining expert opinions.

Validity and reliability analyses were conducted to test the psychometric properties of the CMVS‐CSP. Factor analyses were performed to evaluate the scale's structural validity. The sample size adequacy was determined using the KMO value, which was found to be 0.942, indicating that the sample size was sufficient for factor analysis. The Bartlett test revealed a significant relationship between the variables analysed in the factor analysis (p = .000 < .05). The EFA yielded a two‐factor structure. To ensure accurate measurement, factor loadings above 0.30–0.40 are required. We evaluated the compliance of the items with mechanical ventilation by analysing their factor loadings. In a multifactor structure, if an item has high loading values in more than one factor, the difference between the loading values of the item should be at least 0.10. According to Çokluk et al. ³⁸ an item that has high loading on multiple factors is considered an overlapping item and should be removed from the scale. In this study, six items with factor loadings below 0.30 were removed from the scale following the EFA.

The varimax vertical rotation technique was used to identify factors with higher or lower loading values, as it is known for its precision in distinguishing between factors. The process was repeated after eliminating overlapping items with a difference of 0.10 or less between the two loadings. Following the EFA and rotation, six items with factor loadings below 0.30 were removed from the scale.

Eigen values of the two‐factor structure of the CMVS‐CSP ranged from 2.205 to 11.501. An eigen value represents the sum of squared loadings for a factor. As the eigen value increases, so does the variance explained by the factor. Generally, factors with an eigen value of 1 or higher are considered important. ³⁹ The scale's factors subscales explain 62.30% of the total variance, which falls within the 40%–60% variance ratio commonly seen in social sciences. ⁴⁰ The analysis of the absolute and progressive fit indices of the CFA revealed that the error variances in four items were correlated, indicating a partially low overall fit. According to Harrington, ⁴¹ the high covariance between error terms may be because of the fact that the statements have directive features or the same meaning. In our study, we found a correlation between the error variances of items 8 and 11 and 9 and 10, which may indicate that the statements were similarly interpreted. To evaluate criterion validity, we used the patient‐ventilator synchrony domain of the SAS. The SAS items were scored on a five‐point Likert scale, similar to our scale. The reliability of the SAS and the 51.1% positive correlation between the SAS and our scale demonstrate the reliability of our scale.

Cronbach's alpha reliability coefficient is a commonly used tool to determine the reliability of measurements. In this study, the CMVS‐CSP demonstrated high reliability with a Cronbach's alpha value of 0.956. The behavioural and physical reactions subscales also showed high levels of reliability, with Cronbach's coefficients of 0.949 and 0.912, respectively. This finding may be attributed to the limited number of items in the CMVS‐CSP. ⁴²

The test‐retest is a commonly used method to measure the reliability of a construct over time. This method aims to evaluate changes in a scale over time and the relationship between two administrations. To determine the reliability of the test‐retest method, researchers calculate the correlation coefficient between two applications, which should be at least 0.70. ⁴³ The study analysed the ICC value in the test‐retest results and found a strong positive correlation with the correlation values. However, because the scale is designed to assess compliance with the mechanical ventilator, the retest was conducted 15 min after the initial test.

The discrimination indices for all items and subscales were analysed separately. The discrimination index measures an item's ability to distinguish between high and low achievers and is related to the item's validity. The index ranges between −1 and +1. An index approaching 0 indicates low discrimination of the lower and upper groups for the item, while an approach to 1 indicates high discrimination. ⁴⁴ The results of the independent samples t‐test show that the mean scores for the upper and lower groups were significantly different for all items on the scale (p < .05). Therefore, it can be concluded that all items have ideal item discrimination indices for both groups.

In conclusion, the CMVS‐CSP is a valid and reliable tool for measuring compliance with mechanical ventilation therapy in cardiac surgery patients. The instrument is particularly useful for evaluating patients who have undergone prolonged major surgeries or are receiving long‐term sedation, as mechanical ventilation can be a painful procedure that prevents verbal communication and causes agitation. The CMVS‐CSP can aid nurses in evaluating patients and producing evidence to protect and defend patient rights. Early assessment and monitoring of patient compliance with mechanical ventilation can prevent premature self‐extubation and aid in managing sedation requirements and mechanical ventilator‐associated pneumonia. In the mechanical ventilator compliance scale we developed in our study, questions about reaction to the intubation tube, compliance with ventilator modes and reaction to aspiration for oral care were included. In our scale, the compliance process in the case of reintubation was not evaluated, and this situation was included in the exclusion criteria. Further studies should be conducted to evaluate the validity and reliability of the scale in other cultures and patient groups.

10.1. Limitations

The study has a significant limitation as the data were collected solely from cardiovascular ICU patients at a single health centre. Therefore, future research should aim to test the validity and reliability of the CMVS‐CSP on cardiac surgery patients receiving mechanical ventilation therapy in other hospitals in Turkey and abroad.

10.2. Implications and recommendations for practice

This study presents a measurement tool for nurses in cardiac surgery ICUs to objectively assess patient compliance with mechanical ventilation treatment. The tool avoids subjective judgements and predictions in patient assessment, resulting in more reliable results.

ETHICS STATEMENT

Ethical approval was obtained from the ethics committee of the training and research hospital (No. E1‐22‐2540). The study was conducted in accordance with the principles of the Declaration of Helsinki.

PATIENT CONSENT STATEMENT

The study was explained to the patients and consent was obtained from those who agreed to participate.

Pazar B, Savas H, Turgut S, Dal HC. Development of the compliance with mechanical ventilator scale in: A validity and reliability study. Nurs Crit Care. 2025;30(3):e13191. doi: 10.1111/nicc.13191

DATA AVAILABILITY STATEMENT

The data that supports the findings of this study are available in the supplementary material of this article.