Cultural Adaptation and Reliability of the Compliance with Standard Precautions Scale (CSPS) for Nurses in Brazil 1

ABSTRACT Objective: this study aimed to carry of the cultural adaptation and to evaluate the reliability of the Compliance with Standard Precautions Scale (CSPS) for nurses in Brazil. Method: the adaptation process entailed translation, consensus among judges, back-translation, semantic validation and pretest. The reliability was evaluated by internal consistency (Cronbach alpha) and stability (test-retest). The instrument was administered to a sample group of 300 nurses who worked in a large hospital located in the city of São Paulo/SP, Brazil. Results: through the semantic validation, the items from the scale were considered understandable and deemed important for the nurse´s clinical practice. The CSPS Brazilian Portuguese version (CSPS-PB) revealed excellent interpretability. The Cronbach`s alpha was 0.61 and the intraclass correlation coefficient was 0.85. Conclusion: the initial study showed that CSPS-PB is appropriate to assess compliance with standard precautions among nurses in Brazil. The reliability was considered acceptable. Furhter study is necessary to evaluate its comprehensive psychometric properties.


Introduction
Standard precautions (SP) are measures for healthcare professionals to minimize the risk of occupational exposure and to ensure patient safety (1)(2)(3) . Therefore, these measures should be aligned with the task in hand while considering potential exposure to blood and organic fluids (2) . Direct contact with patients, combined with the type and frequency of nursing procedures performed, increase occupational exposure risk of nursing staff to infectious material (4) . However, research studies demonstrated that the use of personal protective equipment (PPE) was still limited among health professionals, especially among nursing staff (5)(6)(7)(8) . Inappropriate use of gloves indicates poor adherence to hand hygiene (8) and may increase cross infection (9) . Although hand hygiene is an essential component of SP, studies reveal that health professionals minimally adhere to this practice (8)(9)(10)(11)(12) . Consequently, it is important to evaluate the compliance of professionals with SP in their clinical practice. However, evaluation is difficult because instruments for comprehensively measuring compliance with SP are mostly in English and have not been translated to or validated to Brazilian Portuguese.
In order to study compliance with SP, some authors in Brazil (11,13) and other countries (14)(15) have worked on the construction and validation of instruments designed to measure whether professionals are adhering to preventive measures during patient care.
In 2002, a 15-item universal precautions scale was constructed for the purpose of studying the practice of nurses in relation to these measures in Hong Kong (14) .
On the basis of this scale, the Compliance with Standard Precautions Scale (CSPS) in English was developed in Hong Kong in 2011 by modifying most items and adding several new ones (15) . This 20-item scale provides a means to assess the compliance of nursing professionals with infection control practices. The CSPS addresses issues related to daily clinical practice, such as the use of protective equipment, disposal of sharps, handling of articles, and prevention of cross infection. After its development, CSPS went through a comprehensive psychometric testing in a group of 453 participants, including nursing staff and students. Results showed that CSPS had satisfactory reliability (i.e., internal consistency and stability), construct validity (i.e., satisfactory results of known-group method and hypothesis testing) and concurrent validity (16) . More importantly, CSPS was subject to a cross-cultural pilot test involving 19 experts from 16 countries (16) . It preliminarily indicated that CSPS is relevant and applicable to most developed and developing regions (16) . Indeed, up to our knowledge, CSPS has been or will be translated to several languages, including but not limited to Arabic, Korean, Mainland, Chinese, Italian, Spanish and Turkish (17)(18) .
The process of adapting an existing measure has been widely used in different fields because it provides certain benefits, such as saving time and funds, and allowing direct comparisons, which facilitate multicenter research collaboration (19) . Adaptation of CSPS for Brazil is imperative to check the compliance of nurses with SP and facilitate cross-cultural comparisons further on. Several reasons indicate that CSPS is potentially appropriate to be adapted to Brazil. First, CSPS is the only instrument for which cross-cultural pilot testing has been conducted in 16 countries (16) . Such result increases the evidence that the CSPS items will be relevant to Brazil. Second, CSPS is one of instruments that has been adapted to other countries (17)(18) . Hence, adaption of CSPS would facilitate cross-cultural comparisons and contribute significantly to this research field. Third, the CSPS is developed based on the SP guideline published by WHO and CDC (15) , which is in line with the Brazilian guideline (20) . The CSPS includes indispensable aspects related to compliance with SP regarding the use of protective equipment, recap of needles, prevention of cross-infection from person to person, decontamination of used articles and disposal of waste and sharps. These are also important aspects mentioned in the Brazilian governmental regulatory standard of safety and health (20) .
Unlike many instruments that only briefly described the development process, the fourth is that the developer of CSPS provides a detailed instrument development process, a clear explanation of the compliance concept, explicit justification on each developed item (15) , which allows the other researchers to adapt the CSPS to their own cultures. Last, the psychometric properties of CSPS have been comprehensively examined with satisfactory results (16) , which suggest that CSPS is reliable and valid.
Thus, it is justified to carry out the cultural adaptation of CSPS and to evaluate its reliability for nurses in Brazil.

Method
This study consisted of two phases: translation/ adaptation and reliability assessment of the CSPS.
Adaptation included the following stages: translation, consensus among judges, back-translation, semantic validation, and pretest (21) . Reliability included analysis of the reliability (internal consistency) and stability (testretest). The participants were nurses who worked in a large hospital located in the city of São Paulo/SP. Pretest: After the cultural adaptation, a pre-test was carried out. The translated and adapted instrument was applied to hospital nurses. It is considered that 30 to 40 is the ideal number for this test (21) . The CSPS-PB was applied to 50 nurses who worked at a hospital.
Field research -Reliability: The CSPS-PB was applied to 300 nurses working at a large Brazilian public hospital. The inclusion criteria were: being a nurse, acting directly in patient care. The exclusion criteria were: nurse who performed exclusively administrative functions.
These criteria were strictly according to the definition of clinical nurse (15)(16) . Reliability was evaluated by internal consistency and stability. Internal consistency was checked by Cronbach's alpha coefficient (α), ranging from 0 to 1, appropriate coefficients being superior to 0.60 for preliminary investigation (22)(23) , while coefficients of 0.9 to 0.95 are considered excellent (24) . For the stability, test-retest was used by means of the Intra-class correlation coefficient (ICC). This method is used to verify the correlation between the scores resulting from the first application of the instrument and the second application to the same participants two weeks later. The values used as a reference for such analysis are described as ICC <0.40, indicating a weak correlation; 0.41 <ICC <0.60, moderate correlation; 0.61 <ICC <0.80, good or substantial correlation, and ICC> 0.81 almost perfect or very good (25)(26) .

Results
The adaptation process of the CSPS involved the following stages: translation, consensus among judges, back-translation, semantic validation and pretest.
Committee of judges: The committee suggested modifying ten items in the instrument. Their suggestions were accepted when there was 80% agreement among the judges (Figure 1).
It was suggested that the verb "lavar" be changed to "higienizar".
It was decided to use "minhas mãos" only once in the sentence for conciseness.
The word "reencape" was used because it is more common among professionals.
It was decided to use the word "retiro" instead of "removo". As to location, "local designado" was used.
The expression "sangue e derivados" was used.
The face shield was removed, since nursing professionals in Brazil do not commonly use this PPE in combination with goggles.
The words "avental/capote" were used, since some professionals know this PPE by these names.
The expression "lixo" was considered as "material contaminado". Current legislation requires that materials of this type be discarded in properly identified white plastic bags.
The term "líquidos corporais" was replaced by "fluidos corporais". The presentation of some items raised questions.
In question 16, the word "gown" was added because "apron" may be confused for a white coat when the professional answers this question. Question 20 was extensively discussed in relation to the importance of determining the meaning of "surfaces" and what type of "disinfectant" will be used. However, changes were not made in this item. All scale items were answered. The CSPS-PB version proved to be understandable and easy to complete. The nurses who participated in this phase did not suggest

modifications.
Taking into account all changes that were suggested and made during the stages described above, the CSPS-PB was obtained as presented in Figure 2.
Cleaning spills is not a nursing competence in Brazil, but it is an activity executed by the cleaning team. Thus, the review of item 20 was necessary. Another aspect was related to the product used. Although alcohol is considered a disinfectant, this name can also be associated with other products in Brazil (e.g. washing up liquid, dishwashing liquid). In order to reduce confusion and enhance clarity, the word alcohol was added, enclosed in parentheses.
The scope of CSPS includes the use of PPE, waste and sharp disposal, handling of articles and prevention of cross infection. These measures are an immediate concern because current compliance often falls short of recommendations.
Different studies around the world have evaluated the compliance with SP among nursing professionals (6)(7)(8)(11)(12)(13)(14)(15)(16)27) . A review of literature for the purpose of identifying instruments already used to assess awareness of and compliance with SP revealed that most of the publications were international; 12 instruments were selected, of which eight have been published in English, three in Portuguese, and one in Spanish (28) . Notably, 66.6% of the studies involved nurses and described the development, origin and construction of the questionnaires. Among the instruments, 58.3% were validated. In another review, 18 instruments were found for verifying compliance with infection control practices. Most of the studies stated that content validity and reliability analyses were performed using internal consistency analysis and test-retest (29) .
The previous study of the original version of the CSPS had satisfactory reliability results (Cronbach α = 0.73; ICC = 0.79 for two-week test-retest and 0.74 for threemonth test-retest) (16) .
The quality of the adaptation process determines the validity of the instrument to measure the construct in question. Therefore, it is important that an instrument chosen for performing cultural adaption has been well developed and comprehensively validated with satisfactory psychometric properties. Considering the growing number of instrument translation, adaptation, and validation studies performed by nurses, it is necessary to adopt appropriate techniques and methods, and to evaluate reliability and validity, in order to ensure the quality and methodological rigor of the research (30) . The adaptation process used in this study was developed in accordance with the methodological criteria recommended in the literature. It is important to consider the comment given by the developer of a given instrument at every stage and discuss the conceptual meaning of each item in the adaptation process. The developer of CSPS participated in the referred process and approved all changes that were made in the Portuguese version.

Limitation
Some limitations of this study are noted. The sample for conducting psychometric testing is limited to professional nurses (i.e., nursing staff) at a single hospital institution, which decreases the generalizability of the results. Further study will be valuable if the psychometric properties of CSPS-BP are tested in nursing students or staff working in different types of hospitals. Furthermore, the validity of the CSPS-BP should be reported elsewhere.

Conclusion
The CSPS adaptation consisted of the translation, consensus among judges, back-translation, and semantic validation stages. Execution of these stages enabled the cultural adaptation of the CSPS for Brazilian nurses. The CSPS-PB revealed excellent interpretability and items were considered important and relevant to nurses' clinical practices. The reliability was satisfactory.
Though the internal consistency value not was considered high, it is acceptable for initial validation studies. The stability was already very good. The initial study showed that CSPS-PB is appropriate to assess compliance with standard precautions among nurses in Brazil. Additional study is needed to evaluate psychometric properties.