An approach to the emotional design of female nursing scrubs

Abstract

Nurses often encounter negative emotions such as fatigue, boredom, anxiety and fear in the course of their nursing duties, and one of the efficient approaches to improve their emotions is to wear the proper nursing scrubs during their tasks. This paper presents a work with the overall objective to have an emotionally tailored apparel design of nursing scrubs. To reach this objective, the study first mines Kansei words to recognize nurses’ emotions in nursing scrubs domain. Then, design attributes and their corresponding attribute levels for nursing scrubs are identified and determined by applying the rank sum ratio method. Furthermore, a model mapping nurses’ emotions to design attributes of nursing scrubs is established by employing the entropy-weight technique for order preference by similarity to an ideal solution approach. The study yields two significant contributions: (a) presenting the first ever emotion recognition for nurses, which provides a domain-specific foundation for investigating nurses’ psychological wellbeing; (b) constructing a model to conduct the emotional apparel design of female nursing scrubs, which offers a guide to improve preferred and satisfied female nursing scrubs for nurses, and thus positive emotions will be experienced during their nursing duties.

Keywords

Emotions nursing scrubs apparel design nurses

Nurses continually encounter demanding workload, intricate and repetitive nursing procedures, and the responsibility for ensuring patients’ nursing security, which triggers various emotions, including fatigue, boredom, anxiety, and fear. These negative emotions can have adverse effects on the safety and quality of nursing work, as well as the nurse–patient relationship and the hospital’s overall image.¹ In order to deal with these negative emotions among nurses, researchers have proposed several potential solutions. For instance, music therapy and the emotional freedom technique have been studied as interventions to mitigate nurses’ negative emotions, including anxiety and fear.^2,3 Moreover, researchers have found a correlation between nurses’ emotions and their nursing scrubs, prompting investigations into using different types of nursing scrubs to enhance nurses’ emotional wellbeing. Currently, nursing scrubs in various medical systems follow specific styles and regulations. For instance, in Chinese medical institutions, nursing scrubs are typically designed in white, split style or long skirt style,⁴ while other countries often prefer green or blue scrubs below the knee.^5,6 However, it is evident that the present selection of nursing scrubs lacks diversity. Scholars such as Petronio-Coia and Schwartz-Barcott have successfully promoted positive emotions among nurses by encouraging frequent changes in nursing scrubs, incorporating different colors during their nursing duties.⁷ This approach has proved quite effective in increasing positive emotions among pediatric nurses. In addition, studies by Garrigues et al. have shown that matching specific nursing jobs with corresponding nursing scrubs can effectively prevent nurses from experiencing boredom and fatigue during their shifts.⁸ Furthermore, Nayak et al. have discovered that negative emotions may lead nurses to choose nursing scrubs with darker tones, while positive emotions may influence them to opt for scrubs with more interesting details.⁹ In a similar way, Samur and Seren Intepeler have emphasized the importance of providing nurses with diverse nursing scrubs as a means to enhance their positive emotions.¹⁰ It can be concluded that the significance of appropriate nursing scrubs in fostering positive emotions among nurses is evident, and it should be acknowledged that nurses actively consider their preferences when selecting their nursing scrubs.

To cater to nurses’ preferences for nursing scrubs, and consequently promote positive emotions during their nursing duties, the recognition of nurses’ emotions becomes primary and essential. The current methods for emotion recognition generally are two. The first method involves using hardware devices to recognize emotions from nurses’ physiological signals, including aspects such as their voices and facial expressions. Although this approach can achieve a reasonably accurate emotion recognition, it requires a complex process, and may raise concerns about intruding on personal privacy. This method is also scenario-sensitive and time-sensitive, as the participants should not be disturbed even before the experiment.¹¹ The second method is to recognize emotions represented by using ‘emotion words’ extracted from textual sources. For example, Kansei words encompass over 600 adjectives that describe emotions associated with various products,¹² and they are typically sourced from magazines, books, and other relevant materials. Kansei words have been found to be widely used in different product domains,¹³ including apparel and mechanical product domains,^14,15 and medicine.¹⁶ The accuracy of applying Kansei words to identify human emotions has been proved to be quite good in general product domains.¹⁷ Moreover, Zhao et al. have proposed a limited set of affective words to represent emotions specifically related to generic apparel products.¹⁸ It is noted that emotion words can establish a basis for human–machine interaction in product design and operation management,¹⁹ serving as a language that both humans can express, and computers can comprehend. Thus, the Kansei words approach is generally accessible, while also offering accurate representation of diverse emotions.²⁰

The aforementioned investigations highlight the relevance of nurses’ emotions and the potential benefits of emotion-based nursing scrubs in their nursing duties. However, to the best of our knowledge, there has been no comprehensive study focusing on the recognition of nurses’ emotions, and the subsequent emotional design of nursing scrubs. Thus, this study incorporates design attributes that affect the nursing scrubs design, and then constructs a model that reflects how these attributes affect the design. The innovation of this study is that the developed model can help to conduct an emotional apparel design for nursing scrubs, aiming to enhance positive emotions among nurses during their various nursing work. The paper is organized as follows: the second section is the methodology, detailing the process of emotion recognition for nurses, the determination of design attributes for nursing scrubs, and the development of a model for emotional apparel design of female nursing scrubs; the third section presents the results obtained from the study; finally, the paper concludes with a conclusion, limitation, and future perspectives in the last section.

Methodology

Emotion recognition for nurses

As discussed in the aforementioned section, emotion words, particularly, Kansei words, have proved to be accessible and accurate in representing human emotions across various product domains. Thus, the current study decided to employ Kansei words within the context of nursing scrubs to capture the emotions experienced by nurses. In this section, data mining technology is applied to mine Kansei words from different academic journals, authoritative books, websites, product test reports, expert reviews, and user opinions related to the hospital and medical institute. Kansei words are then selected in four steps: (a) the initial selection of Kansei words in terms of the frequency statistics of adjectives; (b) the initial determination of paired Kansei words with opposite attributes; (c) the elimination of paired Kansei words with similar semantics; and (d) the final determination of paired Kansei words by applying the expert panel method. Regarding the expert panel method, after a large amount of initial Kansei words were collected, expert survey interviews were conducted with 27 experienced nurses. These nurses are possessing a minimum of 10 years’ experience in the healthcare sector across diverse departments, in particular, in the departments including internal medicine and surgery, and they have routine nursing work by assisting patients and doctors. In the survey, these nurses were asked to select the most suitable Kansei words, and they were also asked to divide these Kansei words into three emotional dimensions such as functional, expressive and aesthetic (FEA) to identify nurses’ different types of emotions toward nursing scrubs, and to improve the nursing scrubs design accordingly afterwards. It is noted that the FEA is a very well-known consumer need model for apparel development, which has been applied in the field of medical apparel.^21,22 The expert interviews include a survey asking each nurse to rank each initial Kansei word, and then the top ranking Kansei words were discussed by experts to divide into three dimensions, and the unclear semantics, limited relevance, and redundant meanings of Kansei words will be eliminated.

Determination of design attributes for nursing scrubs

Data mining technology was conducted again for collecting the nursing scrubs attributes and attribute levels. Through an extensive mining of the literature concerning healthcare professionals’ scrubs, a set of 12 nursing scrubs attributes and 60 attribute levels under each attribute were initially presented (listed in Table 1).^24
–28,^30
–34 It is noted that the criteria of mining these attributes and attribute levels include: (a) the attributes and attribute levels of nursing scrubs were commonly used in many studies; and (b) the attributes and attribute levels need to be associated with the FEA dimensions.

Table 1.

Initial design attributes and attribute level for nursing scrubs

Attribute	Attribute level	Reference
Style	Dress, coat, apron, shirt and pants, shirt and skirt, dress and pants, coat and pants, jacket and pants	24
Silhouette	X-type, H-type, A-type, O-type	25
Collar	Collarless, stand collar, lapel, contoured collar	27
Sleeve	Sleeveless, short sleeve, half sleeve, long sleeve	28
Placket	No placket, front placket, back placket, shoulder placket, armpit placket	30
Pocket	Patch pocket, insert pocket, hidden pocket, rubber band pocket, hood pocket, zipper pocket, combination pocket	31
Line	Vertical line, horizontal line, diagonal line, curve, comprehensive line type	32
Detail	Insetting, binding, contrasting colors, pleats, waistbands, knots	33
Accessories	Tie, bow tie, badge, hat	34
Color	Cold color, warm color, neutral color	26,33
Pattern	Flower, animal, scenery, character, geometry, casualness, fantasy, text	24
Fabric	Pure cotton, polyester cotton	28

The same group of 27 skilled nurses in the previous section were recruited for this study. Employing the semantic differential method with a five-point Likert scale with score values of ‘−2, −1, 0, 1, 2’ denoting fully disagree, disagree, neither agree nor disagree, fully agree, and agree, respectively, and incorporating insights from in-depth interviews, a questionnaire (as presented in Table 2) was constructed including 12 design attributes and 60 attribute levels.²⁹ Then, nurses were tasked with selecting the most representative design attributes and attribute levels based on their professional perceptions. The data collected from the survey were subjected to reliability analysis to ensure the quality and consistency of the responses. Then, the rank sum ratio (RSR) method was applied to determine the priority of design attributes and their corresponding attribute levels for nursing scrubs, aiming to identify the most crucial features in the nursing apparel design.²³

Table 2.

Examples of the illustrated multiple-choice approach to questions about attribute level for nursing scrubs

Attribute	Attribute level
Style	Dress	Coat	Apron	Dress and pants
Style
Silhouette	X-type	H-type	A-type	O-type
Silhouette
Collar	Collarless	Stand collar	Lapel	Contoured collar
Collar
Sleeve	Sleeveless	Short sleeve	Half sleeve	Long sleeve
Sleeve
Placket	No placket	Front placket	Back placket	Shoulder placket
Placket
Pocket	Patch pocket	Insert pocket	Hidden pocket	Hood pocket
Pocket
Accessories	Tie	Bow tie	Badge	Hat
Accessories
Pattern	Flower	Animal	Geometry	Casualness

	Fantasy	Text	Scenery	Cartoon

The RSR is a widely used approach for evaluating the overall significance of multiple indicators within the healthcare domain, and is applicable to studies with limited sample sizes.^35,36 Given that nursing scrubs form a crucial component of the healthcare field, and their attributes exhibit considerable diversity,^37,38 applying the RSR method to analyze the priority of attributes for nursing scrubs was deemed highly appropriate. This method facilitated a comprehensive assessment of the relative importance of various attributes, thereby assisting in the informed decision-making process for designing nursing scrubs. The RSR is performed as follows: first, constructing a data matrix (n × m) to compile a rank matrix, denoting $R = {(R_{i j})}_{m \times n}$ and calculating it according to

R_{i j} = 1 + (n - 1) \frac{X_{i j} - \min (X_{1 j}, X_{n j}, \dots, X_{n j})}{\max (X_{1 j}, X_{n j}, \dots, X_{n j}) - \min (X_{1 j}, X_{n j}, \dots, X_{n j})}

(1)

Second, assessing the rank sum ratio. When the weights are the same, the calculation equation is shown in equation (2). When the weights are different, the calculation equation is shown in equation (3), where R_ij is the rank of the jth participant’s score for the ith nursing scrub attribute or attribute level. The larger is the value of RSR _i , the better is the evaluation object:

RS R_{i} = \frac{1}{n \times m} \sum_{j = 1}^{m} R_{i j} (i = 1, \dots, n; j = 1, \dots, m)

(2)

WRS R_{i} = \frac{1}{n} \sum_{j = 1}^{m} W_{j} R_{i j} (i = 1, \dots, n; j = 1, \dots, m)

(3)

Then, reordering the rank to obtain the RSR frequency distribution table, calculating the frequency f, cumulative frequency $\sum f$ , and cumulative relative frequency p to obtain the probability unit probit, and deriving a linear regression equation based on probability units and RSR values. Finally, applying the RSR fitting value obtained from the linear regression equation to analyze the priority of various nursing scrub attributes.

A model development for emotion apparel design of female nursing scrubs

In order to facilitate the emotional apparel design for female nursing scrubs, the development of a model linking nurses’ emotions to design attribute levels for female nursing scrubs was deemed necessary. This model development involved two key aspects: first, the acquisition of data, and second, the application of appropriate modeling strategies. The data served as the foundation for constructing the model, while the chosen modeling strategies provided the framework to establish the relationship between nurses’ emotions and the specific attributes of female nursing scrubs.

For the data acquisition, a test-bed was designed to observe human subjects’ opinions on the perceived strength of the connection between female nursing scrubs’ attribute levels and Kansei words relevant to the female nursing scrubs domain. This test-bed comprised two systems: (a) CLO3D, a platform used to simulate three-dimensional (3D) virtual female nursing scrubs in terms of various design attribute levels; and (b) a seven-point Likert scale survey that integrated 3D virtual female nursing scrubs with Kansei words associated with female nursing scrubs, and the scale’s score values are ‘–3, –2, –1, 0, 1, 2, 3’, which, respectively, represent not at all disagree, fully disagree, disagree, neither agree nor disagree, fully agree, agree, extremely agree. This survey aimed to establish a link between the design attribute levels of female nursing scrubs and the emotions experienced by nurses. The combination of these two systems facilitated a comprehensive survey experience for the participants. It is noted that in case of a large number of design attribute levels involved, orthogonal experimental design may be applied to reduce the required sample size for 3D virtual female nursing scrubs.³⁹ In terms of the size of the experiment design,^40
–42 a total of 155 female nurses from diverse medical institutions and departments in China were recruited as survey participants. Among the participants, 31% were aged 20–24 years, 37% were aged 25–29 years, and 32% were aged 30 years and above. In terms of nursing experience, 26% had 1–3 years, 25% had 4–6 years, 26% had 7–9 years, and 23% had 10 or more years. This balanced distribution allows for further analysis.

For the modeling strategies, the study employed the entropy-weight technique for order preference by similarity to an ideal solution (TOPSIS) approach to establish a model linking the ranks of Kansei words in the female nursing scrubs domain, which represented nurses’ emotions to the attributes of female nursing scrubs. The entropy-weight TOPSIS method is a fusion of the entropy weight method (EWM) and the TOPSIS. This approach offered distinct advantages as it not only calculates the index weight based on a comprehensive assessment of the information provided by Kansei words, but it also facilitated decision analysis in the context of female nursing scrubs design.^43,44 It is noted that the sample size available in this method is generally between 100 and 250, and our sample size of 155 female nurses is falling in this range. The entropy-weight TOPSIS approach was performed as follows:

Step 1: Construct the initial matrix. Assuming there are m female nursing scrubs design samples and n pairs of Kansei word sets for decision scoring, the evaluation initial matrix is represented as

X_{i j} = {[\begin{matrix} x_{11} & x_{12} & \dots & x_{1 n} \\ x_{21} & x_{22} & \dots & x_{2 n} \\ ⋮ & ⋮ & ⋮ & ⋮ \\ x_{m 1} & x_{m 2} & \dots & x_{m n} \end{matrix}]}_{m \times n}

(4)

Step 2: Construct the normalized matrix. To obtain the normalized matrix, calculate the normalized values of individual attributes as

x_{i j} = \frac{x_{i j} - \min (x_{i j})}{\max (x_{i j}) - \min (x_{i j})}, j = 1, 2, \dots n

(5)

Q = {(q_{i j})}_{m \times n}, i = 1, 2, \dots, m, j = 1, 2, \dots, n

(6)

where

q_{i j} = \frac{x_{i j}}{\sum_{i = 1}^{m} x_{i j}^{2}}

Step 3: Calculate the entropy of each attribute j. Calculate the information entropy S_j for indicator j based on the normalized matrix Q = ${(q_{i j})}_{m \times n}$ using

S_{j} = - S \sum_{i = 1}^{n} q_{i j} \cdot l n (q_{i j}), j = 1, 2, \dots, n

(7)

where S is the adjustment coefficient, S =

\frac{1}{lnn}

Step 4: Determine weight. The normalized weighted coefficient w_j is

w_{j} = \frac{1 - S_{j}}{m - \sum_{i = 1}^{m} S_{j}}, j = 1, 2, \dots, n

(8)

Step 5: Construct the weighted normalized matrix B, $B = {(b_{i j})}_{m \times n}$ , $b_{i j} = w_{j} \cdot q_{i j}$ .

Step 6: Determine the positive ideal solution $U^{+}$ and negative ideal solution $U^{-},$ as

U^{+} = \{\max_{1 \leq i \leq m} b_{i j}\} = [U_{1}^{+}, U_{2}^{+}, \dots, U_{n}^{+}]

(9)

U^{-} = \{\min_{1 \leq i \leq m} b_{i j}\} = [U_{1}^{-}, U_{2}^{-}, \dots, U_{n}^{-}]

(10)

Step 7: Calculate the distance measures for each alternative, as

D_{i}^{+} = \sqrt{\sum_{j = 1}^{n} {(b_{i j} - U_{j}^{+})}^{2}, i = 1, 2, \dots, m}

(11)

D_{i}^{-} = \sqrt{\sum_{j = 1}^{n} {(b_{i j} - U_{j}^{-})}^{2}, i = 1, 2, \dots, m}

(12)

Step 8: Calculate the relative closeness to ideal solution $v_{j}$ , as

v_{j} = \frac{D_{i}^{-}}{D_{i}^{-} + D_{i}^{+}}, 0 \leq v_{j} \leq 1, i = 1, 2, \dots, m

(13)

Results and discussion

Kansei words in nursing scrubs domain

The initial Kansei words derived from data mining were totally 90 (45 pairs), as presented in Table 3. A total of 1215 surveys (45 pairs of Kansei words × 27 participants) were distributed to nurses. In terms of the top-ranking results and following expert interviews, Kansei words with unclear semantics, limited relevance, and redundant meanings were eliminated. Fourteen pairs of Kansei words remained, and they were regarded as quite suitable for describing nurses’ emotions during the nursing works. The three dimensions of Kansei words are given by Table 4. Functional dimension refers to utility including protection, ease, fit, and convenient for movement. For instance, Kansei words lsuch as ‘comfortable–uncomfortable’ and ‘durable–damageable’ assess whether the nursing scrubs offer a functional experience for the nurse.⁴⁵ Expressive dimension involves communication and is symbolic of nursing scrubs, which are crucial for projecting the professional image of nurses. For instance, Kansei words such as ‘young–mature’ and ‘warm–cold’ evaluate whether the nursing scrubs contribute to a youthful appearance, friendliness, or compassion for the nurse.⁴⁶ Aesthetic dimensions focus on design elements for nursing scrubs, including lines, colors, and textures that contribute to pleasure emotions. Examples such as ‘exquisite–shabby’ and ‘elegant–rough’ indicate the visual appeal conveyed by the lines, colors, and textures of the scrubs.⁴⁷

Table 3.

Pairs of Kansei words in nursing scrubs related domain

Kansei words
light–heavy	interesting–boring	exciting–unexciting	sophisticated–minimalistic	unique–regular
perceptual–rational	luxurious–cheap	strong–weak	fashionable–ancient	young–mature
individualistic–mass	calm–disturbing	concise–cumbersome	elegant–rough	rounded–upright
neat–messy	gentle–powerful	friendly–cold	noble–vulgar	neat–careless
quiet–dynamic	happy–sad	feminine–masculine	desired–undesired	light–dark
organic–geometric	warm–cold	soft–tough	active–inactive	exquisite–shabby
Square–streamline	soft–flowery	modern–conservative	concise–complex	relaxed–restless
durable–damageable	cheap–premium	fashion–traditional	modern–ancient	innovated–dulled
lively–rigid	formal–casual	attractive–unattractive	comfortable–uncomfortable	simple–complex

Table 4.

Three most representative aspects and pairs of Kansei words in nursing scrubs

Dimension	Kansei words
Functional	comfortable–uncomfortable, durable–damageable, simple–complex
Expressive	young–mature, warm–cold, active–inactive, formal–casual, feminine–masculine
Aesthetic	modern–conservative, innovated–dulled, neat–messy, unique–regular, exquisite–shabby, elegant–rough

Design attributes and attribute levels for nursing scrubs

By eliminating the missing data, the valid datasets are 1107 in total, which include scores of design attributes and attribute levels. The data were normalized. The overall Cronbach’s alpha coefficient of the attribute levels for nursing scrubs was 0.901, indicating a good reliability and feasibility for further analysis. The results of RSR used to calculate and classify the evaluation data of nursing scrub attributes were with R² = 0.830, F (1,10) = 47.640, and P < 0.001. The six attribute levels at levels 3 and 4 exert a significant impact on nursing scrubs, whereas the six levels at levels 1 and 2 have a smaller impact (as presented in Table 5). The classification of design attributes is given by Table 5, and it can be seen that placket, line, accessories, pattern, silhouette and detail were at a low level and were given a low priority for nursing scrubs design, while sleeve, pocket, color, collar, fabric, and style were at a high level and were given a higher priority. Thus, the latter six design attributes were extracted for nursing scrubs design. It is noted that the eliminated attributes are not that these attributes are not important for nursing scrubs, but that their influence on nursing scrubs is relatively small compared with others. Moreover, tne texture of fabric was challenging to discern, particularly in the subsequent 3D virtual simulation of nursing scrubs, leading to its elimination from the analysis. In the end, five design attributes including style, collar, sleeve, pocket, and color were determined. For the attribute levels of nursing scrubs selection, RSR was once again applied, leading to the identification of 15 types of attribute levels, as presented in Table 6.

Table 5.

Nursing scrubs attributes classification

Level	Probit	Fitting value	Nursing scrubs attribute
1–2	<5	<0.530	placket(0.530), line(0.515), accessories(0.499), pattern(0.482), silhouette(0.461), detail(0.432)
3–4	>7	>0.530	sleeve(0.598), pocket(0.560), color(0.628), collar(0.545), fabric(0.577), style(0.674)

Table 6.

The final nursing scrubs attributes and attribute levels

Nursing scrubs attribute	Fitting value	Attribute level
Style	>0.590	dress(0.596), coat(0.687), coat and pants(0.633)
Collar	>0.520	collarless(0.566), stand collar(0.604), lapel(0.641), contoured collar(0.689)
Sleeve	>0.594	short sleeve(0.717), long sleeve(0.648)
Pocket	>0.555	patch pocket(0.668), insert pocket(0.622), combination pocket(0.591)
Color	>0.592	cold color(0.633), warm color(0.666), neutral color(0.702)

The model for emotion apparel design of female nursing scrubs

Based on the determined five design attributes and 15 attribute levels of female nursing scrubs, a large number of female nursing scrubs design samples were generated, and it was unfavorable to subsequent evaluation experiments. Thus, orthogonal experimental design was conducted to reduce the sample size. Sixteen samples of female nursing scrubs design (as presented in Table 7) were obtained after orthogonal experiments, and the 3D virtual female nursing scrubs in the survey were generated in terms of these samples. Figure 1 shows a scenario of the survey.

Table 7.

Orthogonal experimental samples of female nursing scrubs

Sample	Style	Collar	Pocket	Sleeve	Color
1	dress	collarless	patch pocket	short sleeve	cold
2	dress	stand collar	insert pocket	long sleeve	warm
3	dress	lapel	combination pocket	short sleeve	neutral
4	dress	contoured collar	patch pocket	long sleeve	cold
5	coat	collarless	insert pocket	short sleeve	warm
6	coat	stand collar	patch pocket	long sleeve	neutral
7	coat	lapel	insert pocket	short sleeve	warm
8	coat	contoured collar	combination pocket	long sleeve	cold
9	coat and pants	collarless	combination pocket	long sleeve	warm
10	coat and pants	stand collar	combination pocket	short sleeve	cold
11	coat and pants	lapel	patch pocket	long sleeve	neutral
12	coat and pants	contoured collar	insert pocket	short sleeve	neutral
13	dress	collarless	patch pocket	long sleeve	neutral
14	coat	stand collar	combination pocket	short sleeve	cold
15	coat and pants	lapel	insert pocket	long sleeve	cold
16	dress	contoured collar	patch pocket	short sleeve	warm

Figure 1.

A scenario from the survey.

By eliminating the missing data from the survey, the valid datasets are 2100 in total, which include scores of emotional evaluation of female nursing scrubs on samples. The data were normalized. The reliability check of these data shows that the Cronbach’s alpha coefficients of each dynamic model evaluation scale were between 0.930 and 0.942, indicating good reliability and feasibility for further analysis. The weight coefficients of each attribute level for female nursing scrubs under Kansei words of functional dimension, expressive dimension, and aesthetic dimension, respectively, by applying the entropy method are presented in Tables 8, 9 and 10. A higher weight coefficient indicated a stronger correlation between the attribute level and the associated Kansei word. For instance, the weight coefficients for the dress attribute level and the Kansei words ‘simple–complex’ of the functional dimension given by Table 8, ‘formal–casual’ of the expressive dimension given by Table 9, ‘modern–conservative’ and ‘innovated–dulled’ of the aesthetic dimension given by Table 10 are relatively high, with values of 0.097, 0.098, 0.092, and 0.094, respectively, indicating a close relationship between the dress attribute level and the mentioned Kansei words. On the other hand, lower weight coefficients suggested a less apparent relationship between the dress attribute level and Kansei words such as ‘comfortable–uncomfortable’ of the functional dimension, ‘young–mature’ of the expressive dimension, and ‘elegant–rough’ of the aesthetic dimension. This kind of weight coefficient can be referred to by hospital managers or designers to guide the improvement of female nursing scrubs design in different clinical scenarios or patient-care contexts. Regarding the above relationship between attribute levels and Kansei words for the dress design of nursing scrubs, it is quite useful for hospital managers or designers working in the department of dermatology or dentistry that do not require significant physical movements by wearing the dress.⁴⁸

Table 8.

Weight coefficients of each attribute level for female nursing scrubs under Kansei words of functional dimension

Attribute	Attribute level	comfortable–uncomfortable	durable–damageable	simple–complex
Style	coat	0.063	0.056	0.063
	coat and pants	0.096	0.064	0.061
	dress	0.050	0.062	0.097
Sleeve	long sleeve	0.059	0.067	0.091
	short sleeve	0.068	0.050	0.088
Pocket	insert pocket	0.058	0.075	0.112
	patch pocket	0.051	0.113	0.075
	combination pocket	0.117	0.097	0.059
Color	cold color	0.069	0.086	0.094
	warm color	0.063	0.058	0.098
	neutral color	0.051	0.051	0.078
Collar	collarless	0.049	0.049	0.078
	stand collar	0.054	0.057	0.059
	lapel	0.098	0.067	0.088
	contoured collar	0.059	0.054	0.074

Table 9.

Weight coefficients of each attribute level for female nursing scrubs under Kansei words of expressive dimension

Attribute	Attribute level	young–mature	warm–cold	active–inactive	formal–casual	feminine–masculine
Style	coat	0.057	0.046	0.062	0.075	0.081
	coat and pants	0.110	0.079	0.050	0.064	0.090
	dress	0.047	0.062	0.069	0.098	0.071
Sleeve	long sleeve	0.058	0.041	0.065	0.135	0.093
	short sleeve	0.085	0.040	0.071	0.077	0.081
Pocket	insert pocket	0.067	0.061	0.046	0.050	0.098
	patch pocket	0.055	0.071	0.083	0.098	0.099
	Combination pocket	0.055	0.053	0.061	0.114	0.059
Color	cold color	0.061	0.049	0.071	0.104	0.083
	warm color	0.102	0.047	0.054	0.066	0.085
	neutral color	0.056	0.041	0.043	0.130	0.113
Collar	collarless	0.088	0.114	0.076	0.048	0.085
	stand collar	0.063	0.060	0.084	0.117	0.081
	lapel	0.075	0.066	0.050	0.113	0.087
	contoured collar	0.067	0.090	0.080	0.106	0.117

Table 10.

Weight coefficients of each attribute level for female nursing scrubs under Kansei words of aesthetic dimension

Attribute	Attribute level	modern–conservative	innovated–dulled	neat–messy	unique–regular	exquisite–shabby	elegant–rough
Style	coat	0.039	0.069	0.044	0.114	0.077	0.154
	coat and pants	0.066	0.055	0.063	0.073	0.077	0.052
	dress	0.092	0.094	0.068	0.088	0.057	0.043
Sleeve	long sleeve	0.054	0.090	0.050	0.048	0.087	0.062
	short sleeve	0.078	0.074	0.073	0.042	0.081	0.094
Pocket	insert pocket	0.075	0.090	0.051	0.083	0.069	0.066
	patch pocket	0.057	0.068	0.049	0.051	0.083	0.047
	combination pocket	0.084	0.077	0.072	0.050	0.052	0.051
Color	cold color	0.103	0.062	0.051	0.048	0.067	0.052
	warm color	0.055	0.082	0.111	0.069	0.057	0.053
	neutral color	0.106	0.071	0.056	0.094	0.071	0.040
Collar	collarless	0.052	0.059	0.120	0.061	0.070	0.052
	stand collar	0.058	0.061	0.070	0.066	0.069	0.103
	lapel	0.073	0.057	0.051	0.050	0.067	0.057
	contoured collar	0.054	0.081	0.055	0.053	0.057	0.054

Moreover, some coefficients for an attribute with different attribute levels are quite similar, which shows a certain level of coefficients, and through this, the importance of an attribute to certain Kansei words can be captured. For example, the coefficients of attribute levels of style and Kansei words of ‘durable–damageable’ about the functional dimension given by Table 8 are quite similar, ranging from 0.056 to 0.064, so is the style and ‘comfortable–uncomfortable’ ranging from 0.050 to 0.096. By comparing the coefficients of two pairs of Kansei words, the former pair has an overall higher coefficient, indicating a closer relationship between the attribute of style and ‘durable–damageable’. However, other attributes with attribute levels having dissimilar coefficients cannot be compared for the weight between an attribute and Kansei words.

By further reasoning the relative closeness U_i in the TOPSIS method, the optimal sample of female nursing scrubs designed based on nurses’ emotions of three dimensions (functional, expressive, and aesthetic) can be determined (as presented in Table 11). For instance, in the optimal sample design of no. 15 (C) presented in Table 11, representing the sample numbers of female nursing scrubs in the orthogonal design, which mainly focused on the attribute level of coat and pants, including attribute levels of coat and pants, lapel, insert pocket, long sleeve, and cold as well (as indicated by Table 7). As presented in Tables 8 –10, these attributes were most closely related to the Kansei words ‘comfortable–uncomfortable’, ‘young–mature’, ‘exquisite–shabby’, ‘formal–casual’, ‘modern–conservative’, ‘simple–complex’, ‘feminine–masculine’, and ‘innovated–dulled’. Thus, in the emotion-based female nursing scrubs design, if nurses experience the above ‘comfortable–uncomfortable’, ‘simple–complex’, ‘elegant–rough’, ‘unique–regular’, ‘formal–casual’ and ‘active–inactive’ emotions, the attribute levels in sample no. 15 (given by Table 7) can be referred to when designing coat and pants style female nursing scrubs. Regarding the above relationship between attribute levels and Kansei words for coat and pants design of nursing scrubs, it is quite useful for hospital managers or designers working in any departments as this style almost allows all work movements in different departments in hospitals.⁴⁹ They can focus the coat and pants design on ‘comfortable–uncomfortable’, ‘young–mature’, ‘exquisite–shabby’, ‘formal–casual’, ‘modern–conservative’, ‘simple–complex’, ‘feminine–masculine’, ‘innovated–dulled’ and to make a coat and pants design with clean and simple curves, using soft and comfortable fabrics, striking a balance between a youthful and formally professional look, integrating an elegant and feminine collar or sleeve, and exploring contemporary and innovative color combinations. Similarly, if nurses experience emotions such as ‘simple–complex’, ‘elegant–rough’, ‘unique–regular’, ‘formal–casual’ and ‘active–inactive’, the attribute levels in sample no. 7 with coat, lapel, insert pocket, short sleeve, and warm can be referred to when designing coat style female nursing scrubs.

Table 11.

The optimum sample of female nursing scrubs

Attribute	Attribute level (C)
Style	coat (7), coat and pants (15), dress (2)
Sleeve	long sleeve (15), short sleeve (7)
Pocket	insert pocket (7), patch pocket (1), combination pocket (10)
Color	cold color (15), warm color (2), neutral color (13)
Collar	collarless (1), stand collar (10), lapel (15), contoured collar (12)

Conclusions, limitations, and future perspectives

In this paper, 14 pairs of Kansei words and their three dimensions including functional, expressive, and aesthetic, were extracted to recognize the nurses’ emotions in the female nursing scrubs domain, and five design attributes and 15 attribute levels of female nursing scrubs have been identified for mainly constructing a female nursing scrubs design by employing the RSR method. Based on these Kansei words and attribute levels, a model connecting nurses’ emotions and female nursing scrubs design has been proposed by applying the entropy-weight TOPSIS method. Through the model, the various emotions of nurses can be mapped to an accordingly female nursing scrubs design with the optimal design attributes and attribute levels. The following conclusion can be drawn from the results: (a) the clarification of nurses’ emotions during their nursing duties provides a solid foundation for helping address contemporary studies concerning nurses’ psychological wellbeing; (b) the establishment of an emotional apparel design model for female nursing scrubs enables the development of female nursing scrubs that resonate with nurses’ emotions, resulting in greater preference and satisfaction when wearing the scrubs during their work, and the significance of these emotionally tailored female nursing scrubs has been demonstrated in improving the safety and efficiency of nursing practices.⁵⁰

Our work nevertheless has several limitations that deserve further discussion. The first limitation lies in the relatively narrow scope of emotions considered for nurses. With only 14 pairs and 28 Kansei words representing nurses’ common emotions extracted in this research, it is essential to acknowledge that nurses often encounter intricate and demanding nursing tasks, which may lead to a broader spectrum of emotions beyond the ones captured here. Therefore, future research endeavors should aim to encompass a more comprehensive range of emotions experienced by nurses, to provide a more holistic understanding of their emotional experiences in the context of nursing work.

The second limitation is the omission of the design attribute of fabric, which was not considered in the section of design attributes and attribute levels for female nursing scrubs. The fabric constitutes a crucial design element in apparel, including female nursing scrubs design, and its texture can significantly influence comfort and overall feelings.⁵¹ Future work is warranted to overcome this limitation.

There are several extensions to our work, which need to be done in the future. The first work is to decouple the attribute levels into specific design parameters for female nursing scrubs. This needs to consider factors such as the color of hue, saturation and value, as well as the silhouette attributes of the shoulder, waist and hem. By breaking down the attribute levels into finer design parameters, the emotional female nursing scrubs design can be tailored with greater precision, ensuring a more targeted and accurate representation of nurses’ emotions. The second work is to apply hardware devices to double check the emotion identification for nurses. The third work is to consider more variables in nursing scrubs design, including the comments from patients and doctors. The fourth work is to consider different nursing scrubs designs in different occasions and departments, and figure out the different design attributes for a specific occasion or department.

Footnotes

Declaration of conflicting interests

The author(s) declare that there is no potential conflicts of interest or personal relationships that could have influenced the work reported in this paper.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study was supported by the Humanities and Social Science Funds of the Ministry of Education (grant no. 21YJCZH239), and the Innovation Research 2035 Pilot Plan of Southwest University (grant no. SWUPilotPlan027).

ORCID iDs

Yu Zhao

Xixi Wei

References

Xie

Liu

Zhong

, et al. Relationships among character strengths, self-efficacy, social support, depression, and psychological well-being of hospital nurses. Asian Nurs Res 2020; 14: 150–157.

Son

Kim

Effects of aromatherapy combined with music therapy on anxiety, stress, and fundamental nursing skills in nursing students: a randomized controlled trial. Int J Environ Res Public Health 2019; 16: 1–10.

Patterson

SL.

The effect of emotional freedom technique on stress and anxiety in nursing students: a pilot study. Nurse Educ Today 2016; 40: 104–110.

Zhang

, et al. Occupational clothing for surgeons and nurses. Med Text Nat Res 2022; 22: 697–715.

West

Wantz

Campbell

, et al. Contributing to a quality patient experience: applying evidence based practice to support changes in nursing dress code policies. Online J Issues Nurs 2016; 21: 1–11.

Küçük

Çömez

Kaçar

, et al. Psychiatric patients’ perspective: nursing uniforms. Arch Psychiatr Nurs 2015; 29: 383–387.

Petronio-Coia

Schwartz-Barcott

A description of approachable nurses: an exploratory study, the voice of the hospitalized child. J Pediatr Nurs 2020; 54: 18–23.

Garrigues

Soulé

Vermeesch

AL.

Nursing students and the human body: application of an ethics pilot project. Int J Environ Res Public Health 2022; 19: 1–8.

Nayak

Padhye

Wang

How to dress at work. Manage Leadership – A Guide for Clin Profess 2015; 17: 241–255.

10.

Samur

Seren Intepeler

Nurses’ view of their work environment, health and safety: a qualitative study. J Nurs Manage 2019; 27: 1400–1408.

11.

McStay

Emotional AI, soft biometrics and the surveillance of emotional life: an unusual consensus on privacy. Big Data Soc 2020; 7: 1–12.

12.

Ishihara

Nagamachi

Schütte

, et al. Affective meaning: the kansei engineering approach. Product Exp 2008; 20: 477–496.

13.

Lin

Zhang

WJ.

Integrated design of function, usability, and aesthetics for automobile interiors: state of the art, challenges, and solutions. Proc Inst Mech Engrs. Part I: J Syst Control Engng 2006; 220: 697–708.

14.

Zhao

Sun

Gupta

, et al. Developing a mapping from affective words to design parameters for affective design of apparel products. Text Res J 2017; 87: 2224–2232.

15.

Zhao

Zhu

Zhou

, et al. An axiomatic design theory for design of apparel products. J Engd Fibers Fabrics 2022; 17: 15589250221134350.

16.

Yang

Qian

Lin

, et al. Cognitive fatigue effect on rehabilitation task performance in a haptic virtual environment system. J Rehabil Assist Technol Engng 2017; 4: 1–9.

17.

Vaidya

Kalita

PC.

Emotion assessment methods in the field of product design: a review. Int J Design Educ 2021; 16: 1.

18.

Zhao

Song

Montazeri

, et al. Mining affective words to capture customer’s affective response to apparel products. Text Res J 2018; 88: 1426–1436.

19.

Lin

Zhang

WJ.

Towards a novel interface design framework: function–behavior–state paradigm. Int J Human-Comput Stud 2004; 61: 259–297.

20.

Uban

Chulvi

Rosso

An emotion and cognitive based analysis of mental health disorders from social media data. Fut Gen Comput Syst 2021; 124: 480–494.

21.

Jang

Lee

Yoo

JJ.

Analysis of male and female nurses’ attitudes toward nurse uniforms in South Korea: the functional, expressive, and aesthetic (FEA) framework. Int J Costume Fashion 2021; 21: 25–37.

22.

Rao

Review on application of “functional, expressive, and aesthetic consumer needs model” in designing patient gowns. J Text Apparel Technol Manage (JTATM) 2022; 12: 1–15.

23.

Ran

Tan

, et al. The application of subjective and objective method in the evaluation of healthy cities: a case study in Central China. Sustain Cities Soc 2021; 65: 102581.

24.

Van den Berg

Vonk

The new discomforts of precarious workers: wardrobe matter, insecurity and the temporality of calibration in dress work. Sociol Rev 2020; 68: 574–589.

25.

Cao

Cloud

RM.

Balancing comfort and function in textiles worn by medical personnel. Improv Comfort Cloth 2011; 14: 370–384.

26.

Aagard

EA.

A pre-design study of patient and medical professional attitudes and reactions towards the colors of medical scrubs. PhD Thesis, The Ohio State University, USA, 2008.

27.

Zhou

Conjoint analysis of consumer preferences for dress design. Int J Cloth Sci Technol 2019; 32: 73–84.

28.

Kim

Brandewie

Kim

MO.

Analysis of user satisfaction for unisex medical uniforms. Res J Text Apparel 2017; 21: 162–177.

29.

Cai

Lin

Zhang

WJ.

Study of the optimal number of rating bars in the Likert scale. In: Proceedings of the 18th international Conference on Information Integration and Web-based Applications and Services. Singapore, 28–30 November 2016, pp. 193–198.

30.

Hennink

Kaiser

BN.

Sample sizes for saturation in qualitative research: a systematic review of empirical tests. Soc Sci Med 2022; 292: 1–10.

31.

Smith

O’Driscoll

Lamb

AJ.

Gender influences bacterial contamination of reusable cleanroom operators’ garments following wear. Eur J Parenter Pharmaceut Sci 2020; 25: 3–13.

32.

Wiener-Well

Galuty

Rudensky

, et al. Nursing and physician attire as possible source of nosocomial infections. Am J Infect Control 2011; 39: 555–559.

33.

Suprun

Vlasenko

Ostrovetchkhaya

Some aspects of medical clothing manufacturing. Int J Cloth Sci Technol 2003; 15: 224–230.

34.

Kaser

Bugle

Jackson

Dress code debate. Nurs Manage 2009; 40: 33–38.

35.

Litak

Colour coding scrubs as a means of improving perioperative communication. J Periop Pract 2011; 21: 177–178.

36.

Wang

, et al. The potential evaluation of groundwater by integrating rank sum ratio (RSR) and machine learning algorithms in the Qaidam Basin. Environ Sci Pollut Res 2023; 30: 63991–64005.

37.

Zhang

A study on RSR evaluation of health service capacity of county-level administrative areas based on entropy weight TOPSIS method in Fujian Province in China. medRxiv 2022; 10: 1–15.

38.

Bhattacharjee

Joshi

Chughtai

, et al. Graphene modified multifunctional personal protective clothing. Adv Mater Interf 2019; 6: 1–27.

39.

Zhu

Chen

, et al. Interactive genetic algorithm based on typical style for clothing customization. J Engd Fibers Fabrics 2020; 15: 1558925020920035.

40.

Sihombing

Barus

LS.

Comparison of the effectiveness of the learning process of the community nursing nurse profession conventionally before the Covid-19 pandemic and online during the Covid-19 pandemic. Proceedings 2022; 3: 16–28.

41.

Mitra

Goswami

SS.

Selection of the desktop computer model by AHP-TOPSIS hybrid MCDM methodology. Int J Res Analyt Rev 2019; 6: 784–790.

42.

Ocampo

Tanaid

Tiu

, et al. Classifying the degree of exposure of customers to COVID-19 in the restaurant industry: a novel intuitionistic fuzzy set extension of the TOPSIS-Sort. Appl Soft Comput 2021; 113: 107906.

43.

Yoo

Design elements and consumer characteristics relating to design preferences of working females. Cloth Text Res J 2003; 21: 49–62.

44.

Yao

Wang

, et al. Research on the selection of logistics mode for clothing e-commerce returns and exchanges based on entropy weight TOPSIS method. Int Soc Opt Engng 2023; 12591: 737–745.

45.

McAndrews

Brooks

One size does not fit all: a qualitative study exploring the apparel wants and needs of people on dialysis. Disabil Rehabil 2022; 44: 2660–2669.

46.

Ling

Hong

Pan

Development of a dress design knowledge base (DDKB) based on sensory evaluation and fuzzy logic. Int J Cloth Sci Technol 2021; 33: 137–149.

47.

Zhou

Chen

Research on optimization design of men’s suit considering users’ perception. Int J Cloth Sci Technol 2022; 34: 379–390.

48.

Laird

Owen

The role of protective clothing in healthcare and its decontamination. Decontam Hosp Healthcare 2020; 10: 209–226.

49.

Leleck

Boles

Brewster

, et al. Contamination of medical scrubs worn in a non-healthcare environment: a pilot study. Am J Infect Control 2020; 48: S42.

50.

Ulenaers

Grosemans

Schrooten

, et al. Clinical placement experience of nursing students during the COVID-19 pandemic: a cross-sectional study. Nurse Educ Today 2021; 99: 104746.

51.

Mohammadi

Shirazi

Moaref

, et al. Protective smart textiles for sportswear. Protect Text Nat Resources 2022; 11: 317–345.