The Attitude Towards Medical Device-Related Pressure Injuries Questionnaire: A Turkish Validity and Reliability Study

Equipment such as endotracheal tubes, tracheostomy tubes, oxygen masks, nasogastric tubes, urinary catheters, plasters, and neck collars are the most commonly used medical devices in hospitals.¹ The formation of a localized wound on the skin or the underlying tissue due to the pressure caused by these devices is called a pressure injury due to a medical device.^2,3 Not only do localized wounds result in increased cost of care and treatment services in general, but the formation of these wounds causes pain, distress, deterioration in body image, increased risk of infection, prolonged hospital stay, and decreased quality of life for the patients.^4-6 

Recently, the subject of pressure injuries due to medical devices has begun to attract the attention of researchers, especially in terms of examining the attitudes of healthcare team members toward such injuries, and several studies have been conducted on the topic.^2,7-10 Kim and Lee² studied the perceived importance and performance of 620 clinical nurses in Korea toward medical device-related pressure injury prevention. They found that nurses’ knowledge and attitudes about pressure ulcer management are very important factors that contribute to the development of effective clinical practice guidelines for prevention. The results of a descriptive study conducted in 2019 that reviewed 20 studies on nurses’ attitudes toward pressure injuries due to medical devices showed that nurses had a positive attitude toward the prevention of these injuries.⁷ The authors of a descriptive study conducted in the United Kingdom stated that 119 of 151 health care team members—including nurses, physicians, speech therapists, hospital managers, and occupational therapists—displayed positive attitudes toward the prevention of medical device-related pressure injuries.⁸ In a study conducted with intensive care nurses by Assis et al., nurses' attitudes towards the prevention of pressure injuries due to the use of nasal catheters, cords for fixing the orotracheal tube, oximeter, and indwelling urinary catheter were found to be positive.⁹ A study conducted in Iran in 2019 that examined the attitudes of 214 nurses toward pressure injuries revealed that participants had relatively positive attitudes toward preventing medical device-related pressure injuries.¹⁰

Prevention of pressure injuries due to medical devices is among the primary responsibilities of all healthcare team members, especially nurses who provide necessary patient care. In addition, among the quality indicators of patient safety management, pressure injury care is a very important aspect of nursing. Pressure injury prevention and management is an issue with important outcomes in healthcare, while providing quality care for the patient and early discharge contributes to the desired attitude of nurses in providing qualified and cost-effective care.¹¹ The attitudes of nurses should be measured using valid and reliable tools to develop effective nursing care strategies for the prevention of medical device-related pressure injuries and to organize necessary training programs. 

Prior to the work of the authors of the present study, no Turkish tool existed for measuring the attitude of nurses toward medical device-related pressure injuries prevention and care in Turkey. Therefore, the main goal of this study was to test the validity and reliability of a Turkish version of the Attitude Towards Medical Device-related Pressure Injuries Questionnaire (MDRPI Questionnaire). In this study, the term 'ulcer' in the original version of the questionnaire was removed from the title and content since 'pressure injury' is used in the scientific literature today.

Research design. This study was a methodological type. This was the Turkish validity and reliability study of the MDRPU/I Questionnaire in nurses, developed by Fereidouni et al⁶ to evaluate nursing students’ attitudes toward the prevention and care of medical device-related pressure injuries. 

Population and study sample. The study was carried out in hospitals of different sizes and characteristics (public hospitals, training and research hospitals, and university hospitals) in Turkey. Nurses working in different health institutions in Turkey formed the study population. The study sample was determined based on the principle of having 5 to 10 times the number of items (11 items) in the scale in the scale validity and reliability studies¹² (11 × 10 = 110). In total, the study sample consisted of 134 nurses who fully completed the forms on the data collection dates and met the inclusion criteria. In addition, a pre-application phase involving 30 nurses was conducted to test the time invariance of the MDRPI Questionnaire (test-retest). Nurses who participated in the pre-application phase of the study were excluded from the study sample. 

The inclusion criteria comprised currently working in a unit in which medical devices are used intensively and having worked in the unit for at least 6 months. The exclusion criteria consisted of working in a unit in which nursing care services are not continued for a long period of time and in which patient circulation is fast (e.g., emergency department, operating room, polyclinic).

Data collection tools. The personal information form, which was prepared by the authors of the present study, and the Attitude Towards Medical Device-Related Pressure Ulcers/Injuries Questionnaire developed by Fereidouni et al⁶ were used for data collection. The personal information form consisted of questions on the personal and professional characteristics of the participants in this study: age, sex, education status, years of professional experience, unit, years of work in the unit, institution, work schedule in the unit, and position in the unit.

The MDRPU/I Questionnaire was developed by Fereidouni et al⁶ to evaluate nursing students’ attitudes toward the care and prevention of medical device-related pressure injuries. The questionnaire consists of 2 subdimensions or factors (ie, medical device-related pressure injury prevention, medical device-related pressure injury care) and 11 items. The questionnaire uses a 5-point Likert scale (1 = strongly agree, 5 = strongly disagree) for each item. The lowest possible score is 11 and the highest is 55, with a high score indicating a positive attitude toward the care and prevention of medical device-related pressure injuries. A score of 11 to 25 points indicates a negative attitude toward the care and prevention of medical device-related pressure injuries, 26 to 40 points indicates a neutral attitude, and 41 to 55 points indicates a positive attitude. The Cronbach α was 0.77 for the entire questionnaire, 0.76 for the attitude towards MDRPU/Is prevention subdimension, and 0.70 for the attitude towards MDRPU/Is care subdimension.⁶

Data collection. Study data were collected online between February 2022 and April 2022 by sharing the online form with the participants through social media applications; collecting data in this manner was necessitated by the COVID-19 pandemic. An informed consent form containing information about the study was placed on the first page of the online form. Participants who approved this page were asked again whether they worked in a unit in which medical device-related pressure injuries occur, and those who answered “No” were not allowed to continue to the next step in the form. 

Statistical analysis. Data were analyzed using SPSS version 21.0 (SPSS Inc) and SPSS Amos (Analysis of Moment Structures) version 6.0. The Davis technique was used to evaluate expert opinions for the content validity of the scale. The Kaiser-Meyer-Olkin (KMO) test was used before factor analysis to determine whether the sample size used in the evaluation of the scale was sufficient for factor analysis. To determine the reliability of the scale, the test-retest method was used to evaluate the invariance across time and the Pearson correlation coefficient was calculated. To evaluate internal consistency, the Pearson product-moment correlation was calculated for the item-total correlation coefficient and the Cronbach α reliability coefficient was calculated for the internal consistency coefficient. Exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) methods were used to evaluate construct validity. In all analyses, P <.05 was accepted as the statistical significance value.

Ethical considerations. To adapt the MDRPU/I Questionnaire into Turkish, necessary permission was obtained by email from the researchers who developed the original questionnaire. Written permission from the Social and Human Sciences Ethics Committee of Adıyaman University (18.10.2021-147) and the Ministry of Health (2021-09-29T13_05_49) and verbal permission from the institutions in which the face-to-face data were collected during the pre-test were obtained. Participation in the study was voluntary. Nurses who elected to participate were informed about the study on the first page of the online form, and only the participants who gave consent were allowed to continue to fill out the form. 

Sociodemographic characteristics. The mean age of the participating nurses was 28.37 years ± 5.35 standard deviation (range, 22–49 years); 76.1% were 22 to 30 years of age. One hundred seven nurses were female (79.9%) and 27 were male (20.1%), and 108 (80.6%) had an undergraduate degree. The nurses’ mean years of experience in the profession was 5.44 years ± 5.33 (range, 1–28 years). Ninety-one participants (67.9%) had been working in the profession for 1 to 5 years. The mean number of years worked in the nurses’ current units was 3.03 years ± 2.94 (range, 1–17 years); 87.3% of the participants had been working in their unit for 1 to 5 years. At the time of data collection, 54.5% of the nurses worked in a training and research hospital, 77.6% worked in shifts, 38.8% worked in an intensive care unit, and 94.8% worked as inpatient clinic nurses (Table 1). 

Validity results of the MDRPI Questionnaire. 

Language, content, and face validity. Translation and back-translation techniques were used to ensure the language validity of the questionnaire. Maintaining the structure and content of the MDRPI Questionnaire, 3 nurse academicians (experts in surgical nursing, pediatric health and diseases, and fundamentals of nursing) who were fluent in English and an English instructor whose mother tongue is Turkish translated the questionnaire into Turkish. A common Turkish form was created after all the translations were examined together by the researchers. A teacher of Turkish evaluated the suitability and understandability of the form in Turkish. To ensure that the structure and content of the measurement tool were maintained, the text was back-translated by 3 different nurse academicians who were fluent in English and who were not provided the original form of the questionnaire. The back-translations were also combined into a single form by the researchers, and the correctness and understandability of the back-translated form in English were checked by an English teacher. 

To evaluate the content validity of the finalized questionnaire, opinions were obtained from 11 experts (2 in pediatric nursing, 2 in surgical nursing, 2 in internal medicine nursing, and 5 in fundamentals of nursing). Experts were asked to rate each item from 1 to 4 using the content validity index (CVI). The Davis technique was used to evaluate the results obtained using the CVI, and the CVI ratio was found to be 98.03%. After the content validity of the questionnaire was determined and the questionnaire was finalized by the researchers, data were collected from 10 nurses who were not included in the sample to ensure face validity. 

Construct validity. The factor structure of the MDRPI Questionnaire was evaluated using EFA and CFA. Analyses were performed on 11 items in the questionnaire.

EFA was performed to test the construct validity of the 11-item MDRPI Questionnaire. To begin, the KMO test was used to determine whether the data were suitable for factor analysis, and the Bartlett test of sphericity was used to determine the significance and difference from 0 of the relationships between the variables to be analyzed. The KMO coefficient was 0.929, and the χ-square value of the Bartlett test (χ² = 1080.710, df = 55; P < .001) was significant. The data were deemed suitable and sufficient for factor analysis according to the result of the analysis performed. Factor analysis showed that the questionnaire had a 2-factor structure with an eigenvalue greater than 1 and a total variance of 68.689% (Table 2). In the original questionnaire developed by Fereidouni et al,⁶ items 1 and 7 were included in factor 1, and items 4 and 11 were included in factor 2 in the sample group of nursing students. In the Turkish version administered to the nurses in the sample group, the items were included in factors 1 and 2 as a result of the EFA tests as discussed later in this manuscript (Table 2). Factor 1 comprised 6 items related to medical device-related pressure injury prevention (items 2, 3, 4, 5, 6, and 11). Factor 2 comprised 5 items related to medical device-related pressure injury care (items 1, 7, 8, 9, and 10).

After EFA, the fit indices of the 11-item, 2-factor questionnaire as a result of CFA were as follows: χ-square of 69.663 (P = .006) and df of 43 (χ²/df = 1.620), with comparative fit index as follows: normed fit index of 0.938, Tucker-Lewis index of 0.968, incremental fit index of 0.975, comparative fit index of 0.975, and root mean square error of approximation (RMSEA) of 0.068. The absolute fit indices values were as follows: goodness of fit of 0.925 and adjusted goodness of fit of 0.885. The residual-based fit index comprised a root mean square residual of 0.066 (Table 3). The diagram of the CFA is shown in the Figure. 

Reliability results of the MDRPI Questionnaire.

Test-retest analysis. The time invariance of the Turkish version of the MDRPI Questionnaire was tested face-to-face with 30 nurses who were working in a medical (n = 10), surgical (n = 10), or intensive care unit (n = 10) of 2 training and research hospitals in different cities in the southeast region of Turkey. Two measurements were performed 15 days apart. Test-retest measurements carried out at a 2-week questionnaire administration interval were evaluated using Pearson product-moment correlation and t test in the dependent groups. When the relationship between the scores obtained from test-retest of the MDRPI Questionnaire was examined using Pearson correlation analysis, the reliability coefficient between the 2 measurement scores was found to be 0.874; this relationship between the scores was determined to be positive and statistically significant at a strong level (P < .005). Comparison of participants’ mean questionnaire scores at 2-week intervals showed a statistically similar distribution between the mean scores (t = 2.560, P = .015) (Table 4). 

Item analysis. When the item-total correlations of 11 items were examined for the reliability study of the MDRPI Questionnaire, none of the items was excluded from the analysis because the coefficients of all items were greater than 0.30. The reliability coefficients of the 11-item MDRPI Questionnaire were between 0.35 and 0.85, and there was a positive, statistically significant relationship between item scores and total questionnaire scores (P < .01) (Table 5). The item-total correlations of each of the subdimensions were examined as well. The reliability coefficients (Pearson correlation) of the 6 items in the subdimension medical device-related pressure injury prevention (factor 1) were between 0.42 and 0.66. The reliability coefficients of the 5 items in the subdimension medical device-related pressure injury care (factor 2) dimension were between 0.77 and 0.92, and the correlation coefficients of all 11 items were positive and statistically significant (P < .001) (Table 5). 

Internal consistency reliability coefficient. The Cronbach α reliability coefficient in the analysis of the internal consistency of the MDRPI Questionnaire was 0.806 for the prevention subdimension (factor 1), 0.950 for the care subdimension (factor 2), and 0.926 for the entire questionnaire (Table 5).

This study discussed the findings regarding the validity and reliability of the MDRPI Questionnaire, which was developed to evaluate the attitudes of nurses and nursing students toward the prevention and care of medical device-related pressure injuries. The validity and reliability of the questionnaire were tested using language validity, content validity, face validity, construct validity, and reliability analyses. 

The back translation method was used to ensure the language validity of the MDRPI Questionnaire. For this method, the literature indicates that translators should be able to translate independent of each other, should have a good command of the languages involved in the translation, and should have expertise in the field.^13-16 The translations for the present study, which were done by 3 nursing faculty members and an English instructor, as well as the analysis for the language validity of the MDRPI Questionnaire showed that the Turkish version had a suitable and understandable structure. 

Content validity is performed to evaluate whether a measurement tool measures the feature it aims to measure. One of the most commonly used methods to ensure content validity is to obtain expert opinions using the Davis technique.^17-19 According to the literature, the opinions of between 3 and 20 experts should be obtained to ensure content validity. If the CVI of each item in the measurement tool is greater than or equal to 80%, the CVI score is interpreted as good.^13,20 To ensure the content validity of the MDRPI Questionnaire, opinions were obtained from 11 experts, with a resulting CVI score of 98.03% for the items in the questionnaire; this finding indicates sufficient content validity of the questionnaire.

In scale adaptation studies, after ensuring the language and content validity of the scale, it is recommended that a pre-application be conducted in a small group that meets the inclusion criteria of the study to evaluate the understandability of the scale items and ensure face validity.^14,21-23 Accordingly, the MDRPI Questionnaire was applied to 10 nurses who were not included in the study sample, and their feedback about the measurement tool was obtained. The researchers finalized the questionnaire by making necessary adjustments in line with these participants’ suggestions about the format and understandability of the tool. 

The factor analysis method was used to test the construct validity of the MDRPI Questionnaire, after first using the KMO and Bartlett test of sphericity to check whether the sample size to which the measurement tool was applied was suitable for factor analysis. The KMO values vary between 0 and 1; a value greater than 0.90 indicates perfect sampling adequacy, while a value less than 0.50 is unacceptable for factor analysis.¹³ The Bartlett test of sphericity should also be significant (P < .05).^13,17,21 In the present study, the KMO value was greater than 0.90 and the Bartlett test of sphericity was significant. These values indicate that the sample and data were related and were sufficient for factor analysis.

Factor analysis is a method that makes it easier to understand and interpret the relationship between many variables that are thought to be related by reducing them to a smaller number of dimensions.²⁴ Whereas EFA is used to determine a factor model or structure within a group of variables, CFA is performed to determine the scale’s consistency with the original factor structure.^17,21 In the present study, the MDRPI Questionnaire had a 2-factor structure with an eigenvalue greater than 1 and a total variance of 68.689%. According to the literature, the total variance explained in multifactorial scales should be between 40% and 60%, and the factor eigenvalue should be greater than 1.^{17,21,22,25,26} Therefore, the 2-factor structure of the MDRPI Questionnaire can be said to be valid for nurses in Turkey. Furthermore, examination of the fit indices values of the CFA performed on the questionnaire showed the χ²/df, P, normed fit index, incremental fit index, comparative fit index, goodness of fit, adjusted goodness of fit, and root mean square residual values to be at perfect levels and the RMSEA value to be at an acceptable level, thus confirming the 2-factor structure of the questionnaire.

Invariance across time of the measurement tool used in this study was assessed using the test-retest method. Test-retest is carried out to determine the degree of reliability by reapplying the measurement tool to the same individuals within a certain interval of time (2–4 weeks) and examining the correlation coefficients of the scores obtained.^22,27-29 The correlation coefficient of the scores ranges from −1 to 1, and reliability increases as the score approaches 1.^13,22,30 In the present study, the questionnaire was applied to 30 nurses who were not included in the sample, with a questionnaire administration interval of 2 weeks, and the correlation coefficient between both applications was calculated. A strong positive significant relationship was found between the scores obtained from the test-retest applications of the MDRPI Questionnaire, and the Turkish version of the questionnaire was determined to have invariance across time. 

In this study, correlation-based item analysis and internal reliability coefficient analysis were used to test the reliability of the MDRPI Questionnaire. Correlation-based item analysis is used to evaluate the contribution of each item in the scale to the measurement of the desired feature; the higher the correlation coefficient of the item, the higher its contribution to the scale.^14,17,21 Although some studies indicate that the total correlation value of the item must be at least 0.25 for the item to have an acceptable contribution to the measurement tool,¹⁷ others state that the total correlation value of the item should not be less than 0.30.¹³ In the present study, the item-total score correlation coefficient of the MDRPI Questionnaire was positive and was greater than 0.30 for all items, and the items were related to each other. Internal consistency is an indicator of whether the items in a scale are homogeneous among themselves (ie, whether all of the items move in the same direction) and often is evaluated using the Cronbach α coefficient.^14,17 A Cronbach α value between 0.80 and 1 indicates high reliability.^17,30 In the present study, the Cronbach α reliability coefficient of the MDRPI Questionnaire was between 0.80 and 1 for the subdimensions and the entire questionnaire; thus, the questionnaire has high reliability.

This study has limitations. Because the research planning process occurred during a period of intense COVID-19 pandemic measures, with the exception of the preliminary application the data collection process was carried out online. This may have affected participation in the research because the COVID-19 pandemic process is one where nurses’ workload is high and unit changes are made frequently. Another limitation is that the survey used in this study is a self-reported data collection tool; thus, the research findings are limited to the sample group.

In line with the findings obtained from the validity and reliability study of the MDRPI Questionnaire in nurses, it was determined that the Turkish version of the questionnaire is a valid and reliable tool suitable for use in Turkey. The final version of the questionnaire demonstrated a CVI of 98.03%, sufficient construct validity, and a Cronbach α value for the total questionnaire of 0.92. The validity and reliability of the MDRPI Questionnaire were high. As a result of this study, the Turkish version of the MDRPI Questionnaire has 11 items and 2 subdimensions. It is recommended that the MDRPI Questionnaire be used by nurses working in different health institutions and caring for patients at risk of developing or who have developed medical device-related pressure injuries. Future validity and reliability studies of the MDRPI Questionnaire in larger samples and including nurses from different countries would contribute to the limited literature on the subject.