ACUTE CORONARY SYNDROME RESPONSE INDEX: CULTURAL ADAPTATION AND EVALUATION OF MEASUREMENT PROPERTIES - BRAZILIAN VERSIONABSTRACT
Objective: to conduct the cultural adaptation and evaluate the measurement properties of the Brazilian version of the Acute Coronary Syndrome Response Index.
Method: a methodological study was conducted with adults of both sexes diagnosed with coronary artery disease. Individuals with an inability to understand Portuguese, significant hearing deficits, or those disoriented were excluded. Data were collected from November 2020 to August 2022 at a university hospital in the state of São Paulo. The cultural adaptation followed the steps recommended in the literature. The measurement properties were assessed in a sample of 244 individuals, testing the reliability and the face, content, construct, and structural validity.
Results: the Brazilian version of the instrument retained semantic, idiomatic, conceptual, and cultural equivalences similar to the original version. The internal consistency results for the subscales were: KR-20 of 0.399 for the Knowledge subscale, and Cronbach’s alpha of 0.735 and 0.577 for the Attitudes and Beliefs subscales, respectively. Confirmatory factor analysis results indicated adequate fit indices to the original model of three subscales (Knowledge, Attitudes, and Beliefs). Convergent construct validity showed a weak but positive correlation between the Knowledge and Beliefs subscales and between the Knowledge and Attitudes subscales.
Conclusion: the Brazilian version preserved textual equivalences, face, content, and construct validity and showed an adequate fit to the original model when applied to individuals with coronary artery disease treated at a public and teaching hospital. Additional studies in different health institutions and with samples of varying characteristics are recommended.
DESCRIPTORS:
Acute coronary syndrome; Coronary disease; Validation study; Translating; Nursing methodology research; Knowledge; Attitude
RESUMO
Objetivo: realizar a adaptação cultural e avaliar as propriedades de medida da versão brasileira do Acute Coronary Syndrome Response Index.
Método: estudo metodológico desenvolvido com adultos, de ambos os sexos e com doença arterial coronariana. Foram excluídos os indivíduos com incapacidade de compreender o português, com déficit auditivo e desorientados. Os dados foram coletados de novembro de 2020 a agosto de 2022 em um hospital universitário no interior do estado de São Paulo. A adaptação cultural seguiu as etapas preconizadas pela literatura. As propriedades de medida foram avaliadas em uma amostra de 244 indivíduos, sendo testadas a confiabilidade e as validades de face, conteúdo, constructo e estrutural.
Resultados: a versão brasileira do instrumento manteve as equivalências semântica, idiomática, conceitual e cultural, assim como a versão original. Os resultados da consistência interna das subescalas foram: KR-20 de 0,399 para a subescala de Conhecimento e Alfa de Cronbach de 0,735 e 0,577, respectivamente, para as subescalas de Atitudes e Crenças. Resultados da análise fatorial confirmatória evidenciaram índices de ajustes adequados ao modelo original de três subescalas (Conhecimento, Atitudes e Crenças). A validade de constructo convergente mostrou correlação fraca e positiva entre as subescalas de Conhecimento e Crenças e de Conhecimento e Atitudes.
Conclusão: a versão brasileira preservou as equivalências textuais, validades de face, conteúdo e constructo e obteve ajuste adequado ao modelo original quando utilizada em pessoas com doença arterial coronariana atendidas em um hospital público e de ensino. Sugere-se a realização de estudos adicionais em diferentes instituições de saúde e em amostras com características distintas.
DESCRITORES:
Síndrome coronariana aguda; Doença das coronárias; Estudos de validação; Tradução; Pesquisa metodológica em enfermagem; Conhecimento; Atitude
RESUMEN
Objetivo: realizar la adaptación cultural y evaluar las propiedades de medición de la versión brasileña del Acute Coronary Syndrome Response Index.
Método: estudio metodológico desarrollado con adultos de ambos sexos y con enfermedad arterial coronaria. Se excluyeron las personas que no pudían comprender el portugués, tenían deficiencias auditivas y estaban desorientadas. Los datos fueron recolectados entre noviembre de 2020 y agosto de 2022 en un hospital universitario del interior del estado de São Paulo. La adaptación cultural siguió las etapas recomendadas por la literatura. Las propiedades de medición fueron evaluadas en una muestra de 244 individuos, probándose la confiabilidad y validez aparente, de contenido, de constructo y estructural.
Resultados: la versión brasileña del instrumento mantuvo equivalencias semánticas, idiomáticas, conceptuales y culturales, al igual que la versión original. Los resultados de consistencia interna de las subescalas fueron: KR-20 de 0,399 para la subescala de Conocimiento y Alfa de Cronbach de 0,735 y 0,577, respectivamente, para las subescalas de Actitudes y Creencias. Los resultados del análisis factorial confirmatorio mostraron índices de ajuste adecuados al modelo original de tres subescalas (Conocimientos, Actitudes y Creencias). La validez de constructo convergente mostró una correlación débil y positiva entre las subescalas de Conocimiento y Creencias y Conocimiento y Actitudes.
Conclusión: la versión brasileña conservó equivalencias textuales, validez aparente, de contenido y de constructo y logró un adecuado ajuste al modelo original cuando se utilizó en personas con enfermedad arterial coronaria atendidas en un hospital público universitario. Se sugiere realizar estudios adicionales en diferentes instituciones de salud y en muestras con diferentes características.
DESCRIPTORES:
Síndrome coronario agudo; Enfermedad coronaria; Estudio de validación; Traducción; Investigación metodológica en enfermería; Conocimiento; Actitud
INTRODUCTION
Cardiovascular diseases are the leading cause of mortality in Brazil and worldwide, with coronary artery disease (CAD) being one of its most prevalent forms1. CAD results from atherosclerosis, a condition caused by the abnormal accumulation of lipids and fibrous tissue in the walls of coronary arteries, obstructing blood flow, which may reduce or stop myocardial perfusion, leading to an acute coronary syndrome (ACS)1-2.
It is known that the quicker the reperfusion of the heart muscle in ACS cases, the better the chances of survival without damage. However, lack of knowledge and delays in recognizing ischemic symptoms are the primary factors influencing delays in pre-hospital care, thereby determining the extent of myocardial ischemia and pre- and in-hospital mortality3-5.
Educational programs focused on recognizing ACS signs and symptoms, responding to these symptoms, and managing risk factors are crucial strategies, as they improve survival rates and reduce recurrent events6. The implementation of educational programs or interventions requires monitoring and evaluation. In this context, using instruments to assess individuals' knowledge and responses to a cardiac event is essential for measuring the effectiveness of these programs and preventing deaths from ACS.
Measurement instruments developed/adapted and validated for application in individuals with CAD in Brazil are written in complex language and focus primarily on the risk factors associated with this condition7-9. Furthermore, no validated instruments for assessing knowledge about ACS symptoms were found.
The Acute Coronary Syndrome (ACS) Response Index is an instrument that evaluates an individual's response to ACS by assessing their knowledge, attitudes, and beliefs10. It was developed and validated in English by Dr. Barbara Riegel, a nurse and professor at the University of Pennsylvania, along with a group of nurses from the United States, Australia, and New Zealand10. It has been culturally adapted and validated for use in China11, Iran12, and Lebanon13. The instrument is easy to apply and can be useful for healthcare professionals, especially nurses, to assess individuals’ responses to ACS, identifying knowledge about ischemic event symptoms as well as appropriate information and responses in such situations. This instrument can also contribute to measuring the effectiveness of educational programs aimed at increasing knowledge about ACS in public and private institutions14-15.
Given the importance of having an instrument available for nurses to evaluate individual responses to ischemic events in both hospital and primary care settings, this study aimed to culturally adapt and evaluate the measurement properties of the ACS Response Index in Brazilian individuals with CAD.
METHOD
This is a methodological study for the cultural adaptation and evaluation of the measurement properties of the ACS Response Index in the Brazilian context.
A non-probabilistic, consecutive sample was used, including adult individuals (18 years or older) of both sexes with CAD. Individuals unable to understand Portuguese, those with severe and uncorrected hearing deficits, and those disoriented regarding time, space, and/or person were excluded.
Data were collected from November 2020 to August 2022 at a university hospital in the interior of São Paulo state by the lead researcher and two well-trained assistant researchers. Data collection was conducted through individual interviews in a private setting during a single meeting. Two instruments were used: 1) A sociodemographic and clinical history questionnaire (developed by the authors); and 2) The Brazilian version of the ACS Response Index.
The primary author of the ACS Response Index, Dr. Barbara Riegel, granted authorization in August 2019 for the cultural adaptation, measurement properties evaluation, and publication of the research results. This research followed all ethical principles related to studies involving human subjects and was submitted to and approved by the Research Ethics Committee (Approval No. 4.307.783). Consent from all individuals who agreed to participate in the study was obtained through a signed informed consent form.
Acute coronary syndrome response index (ACS response index)
The ACS Response Index consists of 33 items divided into three subscales: Knowledge, Attitudes, and Beliefs. The Knowledge subscale contains 21 items addressing true (15 items) and false (six items) ACS symptoms with a dichotomous response scale (no/yes), scoring zero for incorrect responses and one for correct responses. Six of these items are reverse-scored as they are considered incorrect ACS symptoms. The total score for the Knowledge subscale can range from zero to 21 points, with higher scores indicating a better level of knowledge about ACS symptoms10.
The Attitudes subscale has five items that evaluate perceptions of decision-making during an ACS emergency. Responses are given on a four-point ordinal scale: 1) not at all sure; 2) somewhat sure; 3) very sure; 4) completely sure. The total score can range from five to 20 points, with higher scores indicating better perceptions of correct attitudes when recognizing symptoms and seeking help10.
The Beliefs subscale comprises seven items, also rated on a four-point ordinal scale: 1) strongly agree; 2) agree; 3) disagree; 4) strongly disagree. The total score ranges from seven to 28 points, with higher scores indicating better expectations and potential actions taken in the event of a cardiac incident. Three items on this subscale are reverse-scored, meaning they score higher when the individual agrees with the statement, in contrast to other beliefs that score higher when the individual disagrees with the statement. The scores for the three subscales are computed independently, though they are assumed to be related concepts influencing the response to ACS10.
Cultural adaptation
The cultural adaptation process was carried out according to the following steps16-17: 1) initial translation into Portuguese; 2) synthesis of Portuguese versions; 3) expert committee review; 4) back-translation; 5) synthesis of English versions and comparison with the original version; 6) review by the original version’s author; and 7) pre-test (Figure 1).
Flowchart of the steps in the Cultural Adaptation Process of the ACS Response Index for the Brazilian Population. Ribeirão Preto,SP, Brazil, 2023.
Step 1 involved the initial translation of the instrument from English to Portuguese by two independent Brazilian translators fluent in English (Translator 1 and Translator 2). This step resulted in two Portuguese versions of the instrument. Next, Translator 1 and Translator 2 met with the study researchers to synthesize the Portuguese versions (Step 2), leading to the Consensual Portuguese Version 1.
Step 3, the expert committee review, aimed to verify the semantic, idiomatic, cultural, and conceptual equivalences between the Original Version and Consensual Portuguese Version 1. Semantic equivalence evaluates the meaning of words while maintaining the sense across languages; idiomatic equivalence considers expressions unique to a specific location/language, which are difficult to translate and are replaced by equivalent expressions in the target culture; cultural equivalence refers to the context in which the instrument will be used, checking if the terms and situations in the original version are coherent with the target population's experiences; and conceptual equivalence highlights the importance of the concept within the cultural context, as some items may be semantically equivalent but lack conceptual equivalence18. This step was conducted before the back-translation to identify and correct potential errors and inappropriate items that might not be easily noticed during back-translation16. Five healthcare professionals were chosen to join the expert committee. Among them, four were nurses with post-doctorates and one was a physical therapist with a doctorate. All met at least two of the following criteria: expertise in cardiology, fluency in Portuguese and English, and knowledge in instrument adaptation and validation methodology. The discussion took place in a single meeting with all committee members19. The outcome of this step was the Consensual Portuguese Version 2.
To develop Step 4, which involves back-translation, the Consensual Portuguese Version 2 was translated back into English. Two different translators from the initial stage, both native English speakers fluent in Portuguese, were selected (Translator 3 and Translator 4). This step resulted in two English versions, which were evaluated and synthesized into a single version during Step 5, generating the Final Consensual English Version, which was then sent to the original authors for evaluation and approval (Step 6).
Subsequently, the pre-test (Step 7) was conducted to verify the understanding of the pre-final version of the instrument within the target population and to identify any wording issues. Potential participants were invited during hospital care, using the same inclusion and exclusion criteria applied for the measurement properties evaluation. Thirty individuals from the target population17 were individually interviewed and responded to Consensual Portuguese Version 2. The items were read up to three times upon the participant’s request for additional explanation. If, after these attempts, the participant still could not understand the item, no response was recorded (considered a missing data point), and the issue was documented for later discussion. The pre-test participants were not included in the evaluation of measurement properties. At the end of this stage, the final version, titled ACS Response Index-BR, was obtained.
Measurement properties evaluation
To evaluate the measurement properties, the ACS Response Index-BR was applied to a sample of 244 individuals to verify the reliability and the face, content, construct, and structural validity. Descriptive statistical analyses were conducted using the International Business Machines Corporation Statistical Package for the Social Sciences (IBM-SPSS) software, version 25.0.
The sample size was defined based on the protocol developed by researchers from the Consensus-based Standards for the Selection of Health Measurement INstruments (COSMIN) initiative, which suggests a sample of five to seven times the number of items in the instrument being tested, and ≥ 100 individuals for Confirmatory Factor Analysis (CFA)20.
Face and content validity were assessed in conjunction with the adaptation process by an expert committee, through consensus among professionals regarding clarity and perception of what is being measured (face validity) and the relevance of each item to the construct being studied (content validity). A consensus of 80% or higher among committee members was required for the modification of any item19.
Reliability, measured through internal consistency, was assessed using KR-20 coefficients (Knowledge subscale) and Cronbach’s alpha (Attitudes and Beliefs subscales). Internal consistency values greater than 0.65 and 0.70 for KR-20 and Cronbach’s alpha, respectively, were considered acceptable21.
Construct validity was estimated through hypothesis testing for known groups and by evaluating convergent validity. To assess the instrument’s ability to distinguish known groups, subscale results were compared based on age group (up to 59 years/60 years or older), whether participants received professional guidance about their heart condition or treatment (yes/no), and education level (years of schooling), as verified in previous studies using the ACS Response Index3,10,13,22-23.
Independent sample t-tests and ANOVA were used for group mean comparisons. Convergent construct validity was assessed by analyzing correlations between the three subscales of the ACS Response Index-BR, based on the original instrument study10. Correlation strength was categorized as follows: values below 0.30 as weak correlation, between 0.30 and 0.50 as moderate correlation, and values above 0.50 as strong correlation24. The level of significance adopted for the analyses was 0.05.
Structural validity was evaluated using CFA, where the original structure with four components in the Knowledge subscale (1. Stereotypical symptoms; 2. Other common symptoms; 3. Incorrect symptoms; 4. Symptoms consistent with stroke), and with two components in the Attitudes subscale (1. Symptom recognition; 2. Help-seeking) and in the Beliefs subscale (1. Expectations; 2. Action) was tested. The goal was to confirm whether the measured variables represent the analyzed construct and if the original factor structure fits the results obtained with the translated and adapted version of the instrument.
CFA was conducted using R software via its RStudio interface with the aid of the sem and lavaan packages. To test model fit, the following indices were analyzed: chi-square (X2), degrees of freedom, Comparative Fit Index (CFI), Tucker-Lewis Index (TLI), Root Mean Square Error of Approximation (RMSEA), and Standardized Root Mean Square Residual (SRMR). The following criteria were considered indicative of adequate model fit: CFI close to or above 0.90, TLI close to or above 0.90, RMSEA below 0.06, and SRMR below 0.1025.
RESULTS
The translation stage resulted in two Portuguese versions, which presented some language discrepancies related to words or expressions with similar meanings (Examples: "tensão no peito" and "pressão no peito"; "frequência cardíaca acelerada" and "frequência cardíaca rápida"). These discrepancies were discussed during the translation synthesis, aiming to maintain the meaning of each item while using terms and expressions consistent with Brazilian culture.
In the expert committee review, the suggested modifications generally involved adding or removing words or substituting them with synonyms. In some items, the committee suggested adding a brief explanation at the end of the sentence, placed in parentheses, to enhance the clarity of the information. All changes achieved over 80% agreement among committee members.
The two English versions obtained through back-translation were approximately 61% identical, with differences mostly involving synonyms. The Final Consensual English Version, obtained after synthesizing the two back-translated English versions, was sent to the original authors for approval of the modifications made to meet semantic, idiomatic, and cultural equivalences. The authors did not request any additional changes.
In the pre-test, Consensual Portuguese Version 2 was completed by 30 subjects, most of whom were male (n=21; 70%), with an average age of 64.9 years (SD=9.9), and married or in a consensual union (n=17; 56.7%). It was identified that some participants had difficulty with items 11 ([…] jaw pain), 29 ([…] if I thought I was having a heart attack, I would wait until I was very sure before going to the hospital), 30 ([…] if I thought I was having a heart attack, I would prefer someone take me to the hospital rather than an ambulance coming to my house), and 31 ([…] due to difficulties accessing healthcare services and financial/economic challenges, I would want to be absolutely sure I was having a heart attack before going to the hospital). As a result, the Consensual Portuguese Version 2 and the Original Version were sent to an expert with extensive experience in adaptation and validation methodology and proficient in English. Based on discussions and suggestions from the expert, additional adjustments were made to achieve cultural and conceptual equivalence for the items. This stage culminated in the final version, named the ACS Response Index-BR.
The sample for the measurement properties evaluation consisted of 244 participants, predominantly male (n=150; 61.5%), married or in a consensual union (n=151; 61.8%), self-identified as white (n=150; 61.5%), with an average age of 62.19 years (SD=10.2). Among the participants, 61.5% reported having received guidance from a healthcare professional regarding heart disease and/or treatment at some point in their lives (Table 1). The mean scores for the Knowledge, Attitudes, and Beliefs subscales were 12.75 (SD=2.6), 13.48 (SD=3.39), and 22.40 (SD=3.16), respectively. In the Knowledge subscale, 26.2% achieved more than 70% correct answers, what is considered a high level of knowledge about the disease.
The internal consistency results for the ACS Response Index-BR subscales were unsatisfactory for the Knowledge and Beliefs subscales. The KR-20 coefficient value was 0.399 for the Knowledge subscale (21 items), while the Cronbach’s alpha coefficient values were 0.577 for the Beliefs subscale (5 items) and 0.735 for the Attitudes subscale (7 items) (Table 2).
The results of the construct validity for known groups are presented according to the statistical tests for the three pre-established hypotheses. The first hypothesis was that younger individuals (up to 59 years old) would present higher scores compared to those aged 60 years or older across the three subscales of the ACS Response Index-BR. The results confirmed this hypothesis only for the Knowledge subscale (t=3.007; p<0.05).
The second hypothesis was that participants who had received guidance from a healthcare professional regarding their heart disease or treatment would present higher scores across the three subscales compared to those who had not received such guidance. This hypothesis was only confirmed for the Beliefs subscale (t=2.687; p<0.05). The third and final hypothesis was that participants with higher educational levels (13 years or more of formal education) would present higher scores than those with lower educational levels (up to nine years and between 10 and 12 years), across the three subscales of the ACS Response Index-BR. The results confirmed this hypothesis only for the Beliefs subscale (F=3.878; p<0.05) (Table 3).
In the evaluation of convergent construct validity, positive and statistically significant correlations of weak magnitude were found between the Knowledge and Beliefs subscale scores (r=0.217; p=0.001) and between the Knowledge and Attitudes subscale scores (r=0.128; p=0.046). The correlation between the Attitudes and Beliefs subscale scores was positive and of weak magnitude, but not statistically significant (r=0.052; p=0.42).
Structural validity was assessed through CFA, where the results demonstrated adequate fit indices for the four-component model of the Knowledge subscale: CFI=0.792, TLI=0.761, RMSEA=0.058, and SRMR=0.065. For the two-component models of the Attitudes and Beliefs subscales, the index results were also satisfactory, showing a good model fit (Table 4).
For the Knowledge subscale, the CFA indicated that most relationships between latent and observed variables had moderate factor loadings, ranging from 0.30 to 0.69, except for items 1 [Lower abdominal pain (pain in the lower belly)] and 5 (Chest pain/pressure/tightness), which had factor loadings of 0.25 and 0.10, respectively. The correlations between latent variables were strong and positive for "stereotypical symptoms” and “other common symptoms" (r=0.72) and for "incorrect symptoms” and “symptoms consistent with stroke" (r=0.58). The other correlations were strong and negative (Figure 2).
Graphical representation of the 21 items in the Knowledge subscale of the ACS Response Index-BR. Ribeirão Preto,SP, Brazil, 2023. (n=244)
For the Attitudes subscale, the CFA revealed moderate to strong relationships between latent and observed variables, with a minimum factor loading of 0.56 and a maximum of 0.75. The correlation between the latent variables "Symptom recognition” and “Help-seeking" was strong and positive (r=0.79) (Figure 3).
Graphical representation of the 5 items in the Attitudes subscale of the ACS Response Index-BR. Ribeirão Preto,SP, Brazil, 2023. (n=244)
Lastly, the CFA of the Beliefs subscale revealed moderate to strong relationships between latent and observed variables (factor loadings ranging from 0.48 to 0.80), except for item 30 (If I thought I was having a heart attack, I would prefer someone take me to the hospital rather than an ambulance coming to my house), which had a factor loading of 0.14. The correlation between the latent variables “Expectations” and “Action” was moderate and positive (r=0.46) (Figure 4).
Graphical representation of the seven items in the Beliefs subscale of the ACS Response Index-BR. Ribeirão Preto,SP, Brazil, 2023. (n=244)
DISCUSSION
The high mortality rate from coronary artery disease (CAD), both nationally and globally, is a well-known reality, especially in low- and middle-income countries1,26. The lack of knowledge and delayed recognition of acute coronary syndrome (ACS) symptoms by individuals, particularly atypical symptoms, is one of the main factors that delays seeking healthcare3-5. Educational programs can be strong allies of the public health system and the population itself by teaching and preparing individuals to take appropriate actions and respond to an event like ACS. In this context, measurement instruments are important tools to assess individuals’ knowledge and responses to a cardiac event, thereby measuring the effectiveness of these programs and preventing deaths from ACS.
In this study, the ACS Response Index was adapted to Brazilian Portuguese, and its measurement properties were tested in Brazilians with CAD. During the review, it was noted that none of the instruments adapted and validated within the national context assessed individuals’ knowledge about ACS. Moreover, it was found that the ACS Response Index has been used in various languages and countries10-13 but had not yet been adapted and validated for use in Brazil, which was the motivation for developing this study.
The process of cultural adaptation and measurement properties evaluation was conducted according to the recommended literature16-17,20-21,25. It is important to highlight that the literature offers a variety of methodological approaches for the cultural adaptation of instruments; however, there is no consensus on the most appropriate methodology or strategy. A measurement instrument developed in one country can be used in other countries as long as it undergoes a proper cultural adaptation process that considers the context and culture of the target population. This practice promotes knowledge sharing among researchers and contributes to the comparison of results in studies using the same tool27.
The initial translation stage should be conducted by at least two qualified and independent translators whose native language is the one into which the instrument is being translated. These recommendations also apply to the back-translation stage. The translation, back-translation, and translation synthesis stages allow for the detection of errors and divergent interpretations16. In this study, these stages were successfully completed, following the literature’s guidelines.
During the expert committee evaluation, members thoroughly discussed the instrument’s content to verify and preserve the equivalencies between the original and translated versions. This stage is crucial to ensure understanding by the target population. In the original validation study, the expert committee consisted of five nurses with doctorates who assessed the completeness, wording, and format10. This evaluation method was also used in other studies describing the validation process of the ACS Response Index11-13. The face and content validity of the ACS Response Index-BR were found to be adequate in this study.
It is noteworthy that 30 participants were included in the pre-test stage, in accordance with the recommendations17. During this stage, it was possible to understand the patient’s perspective regarding the instrument, allowing for the necessary adjustments to make it clearer and achieve a satisfactory level of equivalence.
The reliability of the instrument was assessed through internal consistency, but satisfactory values were only obtained for the Attitudes subscale. The internal consistency results for the Knowledge and Beliefs subscales differed significantly from those obtained by the authors who assessed the reliability of the English10, Chinese11, and Arabic23 versions. On the other hand, the Beliefs subscale also presented an inadequate Cronbach’s alpha in the Iranian12 and Lebanese13 versions. A possible explanation for the result obtained, particularly with the Beliefs subscale, is that the items may not be fully applicable to the Brazilian population, leading to inconsistent responses from participants.
The reliability of an instrument can be influenced by several factors, one of which is sample heterogeneity; the more homogeneous the sample, meaning the more similar the participants' responses, the lower the reliability indices28. The purpose of instruments is to detect variations in what is being measured. Therefore, if the sample members are too similar, the instrument will have more difficulty reliably distinguishing different levels of the attribute. In light of this, it was decided not to modify the Knowledge and Beliefs subscales and to encourage the application of the instrument in diverse samples.
Construct validity, assessed through analysis of known groups, showed that individuals aged 59 or younger had higher levels in the Knowledge subscale, and those who received guidance from a healthcare professional and those with 13 or more years of education had higher scores in the Beliefs subscale. The literature shows that higher education levels are associated with higher scores in all three subscales of the ACS Response Index13,23. Similarly, some studies have pointed out that older individuals tend to have lower knowledge and attitude scores3,13.
In the original study10, construct validity was tested by comparing scores between individuals who had received prior guidance from healthcare professionals and those who had not. The test showed that the instrument was able to identify differences between the groups, as those with higher scores were part of the group that had access to guidance. A study applying the ACS Response Index22 also demonstrated significant differences in the knowledge scores of subjects who had access to guidance from trained professionals. In the sample of this study, most participants reported receiving guidance from a healthcare professional regarding their heart disease and/or treatment, which significantly influenced their responses on the Beliefs subscale. However, this guidance did not lead to an increase in knowledge levels or more positive attitudes, unlike what was observed by the authors of the original version10.
Convergent construct validity revealed a weak yet positive correlation between the Knowledge and Beliefs subscales and between the Knowledge and Attitudes subscales. In the original study, this test also showed significant correlations across all subscales10. In the adaptation and validation study conducted in China11, the correlation was moderate and positive between the Attitudes and Beliefs subscales (r=0.49, p<0.01), while the correlations were weak and positive between the Knowledge and Attitudes subscales (r=0.22, p<0.01) and the Knowledge and Beliefs subscales (r=0.21, p<0.01). A Jordanian study23 pointed to moderate and positive correlations between the Knowledge subscale and the Attitudes (r=0.55, p<0.01) and Beliefs subscales (r=0.56, p<0.01), as well as between the Attitudes and Beliefs subscales (r=0.58, p<0.01).
Regarding structural validity, the ACS Response Index-BR demonstrated adequate fit to the original model. However, comparison with other validation studies was not possible since the authors opted to conduct new exploratory factor analyses instead of confirming the original proposed model11-12. Items one, five, and 30 had factor loadings below the acceptable threshold (0.30)20. The decision was made to present the results related to the original theoretical model. Subsequently, after excluding the three mentioned items, the new version of the instrument should be administered to other samples of people with CAD. This would allow the new theoretical model to be verified using exploratory factor analysis21,29.
Regarding the Knowledge subscale score, some authors opted to convert the result into a percentage for analysis purposes. In this context, the cutoff point to determine a high level of knowledge would be a score of 70% or higher13,22-23. In this study, 26.2% of the sample achieved a score classified as a high level of knowledge. A similar result was observed in a study conducted in Lebanon, where 26% of the sample showed a high level of knowledge13. In a study conducted in Guyana, the sample also demonstrated a low level of knowledge, with only 3.3% of the participants answering more than 70% of the items on the Knowledge subscale correctly30.
Other studies reported higher rates of high knowledge levels, such as a study conducted in Jordan, where approximately 35% of participants achieved more than 70% correct answers on the Knowledge subscale23. In a study conducted in Australia, 61% of the sample achieved a high level of knowledge22.
The disparity observed in the Knowledge subscale scores can be explained by the differing socioeconomic conditions that characterize developed and developing countries. This occurs because social inequalities tend to be more pronounced in underdeveloped countries, which can, in turn, impact access to healthcare and health education.
A limitation of this study was the inclusion of individuals with CAD treated at a single tertiary-level hospital, which predominantly serves patients from the public unified healthcare system (SUS). These individuals mostly have less favorable socioeconomic conditions, low educational levels, and, since it is a tertiary hospital, many already present more severe clinical conditions than those treated in private or complementary healthcare services. Another limitation is the discussion of results given the lack of published articles describing the cultural adaptation process of the ACS Response Index, as well as the results of exploratory or confirmatory factor analyses.
CONCLUSION
This study presents the Brazilian version of the ACS Response Index, titled ACS Response Index-BR, which retained the 33 items and preserved the semantic, idiomatic, cultural, and conceptual equivalences in line with the original version. Construct validity showed satisfactory results. Items one, five, and 30 had factor loadings below the acceptable threshold, but this did not impact the structural validity, which demonstrated adequate fit to the original proposed model. The internal consistency results demonstrated satisfactory reliability only in the Attitudes subscale.
Despite the identified limitations, this study contributes to the advancement of knowledge and evidence-based practice by providing a tool that helps assess individuals during an ACS episode.
Given the particularities of the sample, it is recommended that additional methodological studies be conducted with the adapted version to test the instrument’s measurement properties in different healthcare settings and in samples with varied sociodemographic characteristics.
ACKNOWLEDGMENT
To the Hospital das Clínicas, Faculdade de Mecidina de Ribeirão Preto, for providing the opportunity for data collection and the development of this research.
REFERENCES
-
1. World Health Organization (WHO). Cardiovascular diseases. Geneva: World Health Organization [Internet]. 2021 [cited 2023 Nov 27]. Available from: http://www.who.int/en/news-room/fact-sheets/detail/cardiovascular-diseases-(cvds)
» http://www.who.int/en/news-room/fact-sheets/detail/cardiovascular-diseases-(cvds) -
2. Nicolau JC, Feitosa Filho GS, Petriz JL, Furtado RHM, Précoma DB, Lemke W, et al. Brazilian Society of Cardiology Guidelines on Unstable Angina and Acute Myocardial Infarction without ST-Segment Elevation - 2021. Arq Bras Cardiol [Internet]. 2021 [cited 2023 Nov 27];117(1):181-264. Available from: https://doi.org/10.36660/abc.20210180
» https://doi.org/10.36660/abc.20210180 -
3. Albarqouni L, Smenes K, Meinertz T, Schunkert H, Fang X, Ronel J, et al. Patients' knowledge about symptoms and adequate behaviour during acute myocardial infarction and its impact on delay time: Findings from the multicentre MEDEA Study. Patient Educ Couns [Internet]. 2016 [cited 2023 Nov 27];99(11):1845-51. Available from: https://doi.org/10.1016/j.pec.2016.06.007
» https://doi.org/10.1016/j.pec.2016.06.007 -
4. Garrido D, Petrova D, Catena A, Ramírez-Hernández JA, Garcia-Retamero R. Recognizing a heart attack: Patients' knowledge of cardiovascular risk factors and its relation to prehospital decision delay in acute coronary syndrome. Front Psychol [Internet]. 2020 [cited 2023 Nov 27];11:2056. Available from: https://doi.org/10.3389/fpsyg.2020.02056
» https://doi.org/10.3389/fpsyg.2020.02056 -
5. Rodrigues JA, Melleu K, Schmidt MM, Gottschall CAM, Moraes MAP, Quadros AS. Independent Predictors of Late Presentation in Patients with ST-Segment Elevation Myocardial Infarction. Arq Bras Cardiol [Internet]. 2018 [cited 2023 Nov 27];111(4):587-93. Available from: https://doi.org/10.5935/abc.20180178
» https://doi.org/10.5935/abc.20180178 -
6. Smith SC Jr, Benjamin EJ, Bonow RO, Braun LT, Creager MA, Franklin BA, et al. AHA/ACCF Secondary Prevention and Risk Reduction Therapy for Patients with Coronary and other Atherosclerotic Vascular Disease: 2011 update: A guideline from the American Heart Association and American College of Cardiology Foundation. Circulation [Internet]. 2011 [cited 2023 Nov 19];124(22):2458-73. Available from: https://doi.org/10.1161/CIR.0b013e318235eb4d
» https://doi.org/10.1161/CIR.0b013e318235eb4d -
7. Ghisi GL, Durieux A, Manfroi WC, Herdy AH, Carvalho TD, Andrade A, et al. Construction and validation of the CADE-Q for patient education in cardiac rehabilitation programs. Arq Bras Cardiol [Internet]. 2010 [cited 2023 Nov 27];94(6):813-22. Available from: https://doi.org/10.1590/s0066-782x2010005000045
» https://doi.org/10.1590/s0066-782x2010005000045 -
8. Ghisi GL, Leite CM, Durieux A, Schenkel IC, Assumpção MS, Barros MM, et al. Validation into portuguese of the Maugerl CaRdiac preventiOn-Questionnaire (MICRO-Q). Arq Bras Cardiol [Internet]. 2010 [cited 2023 Nov 27];94(3):372-8:394-400. Available from: https://doi.org/10.1590/s0066-782x2010000300018
» https://doi.org/10.1590/s0066-782x2010000300018 -
9. Saffi MAL, Macedo Junior LJJ, Trojahn MM, Polanczyk CA, Rabelo-Silva ER. Validity and reliability of a questionnaire on knowledge of cardiovascular risk factors for use in Brazil. Rev Esc Enferm USP [Internet]. 2013 [cited 2023 Nov 27];47(5):1083-9. Available from: https://doi.org/10.1590/S0080-623420130000500011
» https://doi.org/10.1590/S0080-623420130000500011 -
10. Riegel B, McKinley S, Moser DK, Meischke H, Doering L, Dracup K. Psychometric evaluation of the Acute Coronary Syndrome (ACS) Response Index. Res Nurs Health [Internet]. 2007 [cited 2023 Nov 19];30(6):584-94. Available from: https://doi.org/10.1002/nur.20213
» https://doi.org/10.1002/nur.20213 -
11. Cao X, Cao Y, Salamonson Y, Digiacomo M, Chen Y, Chang S, et al. Translation and validation of the Chinese version of the Acute Coronary Syndrome Response Index (C-ACSRI). Int J Nurs Stud [Internet]. 2012 [cited 2023 Nov 19];49(10):1277-90. Available from: https://doi.org/10.1016/j.ijnurstu.2012.04.008
» https://doi.org/10.1016/j.ijnurstu.2012.04.008 -
12. Rezaei H, Ismil Jamil D, Ebadi A, Ghanei Gheshlagh R. Psychometric properties of the Persian version of the instrument for assessing cardiac patients’ knowledge, attitude, and beliefs regarding heart attack. Int Cardiovasc Res J [Internet]. 2019 [cited 2023 Nov 19];13(2):62-68. Available from: https://brieflands.com/articles/ircrj-84462
» https://brieflands.com/articles/ircrj-84462 -
13. Noureddine S, Dumit NY, Maatouk H. Patients' knowledge and attitudes about myocardial infarction. Nurs Health Sci [Internet]. 2020 [cited 2023 Nov 19];22(1):49-56 Available from: https://doi.org/10.1111/nhs.12642
» https://doi.org/10.1111/nhs.12642 -
14. Gallagher R, Roach K, Belshaw J, Kirkness A, Sadler L, Warrington D. A pre-test post-test study of a brief educational intervention demonstrates improved knowledge of potential acute myocardial infarction symptoms and appropriate responses in cardiac rehabilitation patients. Aust Crit Care [Internet]. 2013 [cited 2023 Nov 19];26(2):49-54. Available from: https://doi.org/10.1016/j.aucc.2012.01.002
» https://doi.org/10.1016/j.aucc.2012.01.002 -
15. O'Brien F, McKee G, Mooney M, O'Donnell S, Moser D. Improving knowledge, attitudes and beliefs about acute coronary syndrome through an individualized educational intervention: A randomized controlled trial. Patient Educ Couns [Internet]. 2014 [cited 2023 Nov 19];96(2):179-87. Available from: https://doi.org/10.1016/j.pec.2014.05.022
» https://doi.org/10.1016/j.pec.2014.05.022 -
16. Ferrer M, Alonso J, Prieto L, Plaza V, Monsó E, Marrades R, et al. Validity and reliability of the St George's Respiratory Questionnaire after adaptation to a different language and culture: The Spanish example. Eur Respir J [Internet]. 1996 [cited 2023 Nov 27];9(6):1160-6. Available from: https://doi.org/10.1183/09031936.96.09061160
» https://doi.org/10.1183/09031936.96.09061160 -
17. Beaton DE, Bombardier C, Guillemin F, Ferraz MB. Guidelines for the process of cross-cultural adaptation of self-report measures. Spine (Phila Pa 1976) [Internet]. 2000 [cited 2023 Nov 27];25(24):3186-91. Available from: https://doi.org/10.1097/00007632-200012150-00014
» https://doi.org/10.1097/00007632-200012150-00014 -
18. Schmidt S, Bullinger M. Current issues in cross-cultural quality of life instrument development. Arch Phys Med Rehabil [Internet]. 2003 [cited 2023 Nov 27];84(4 ) Suppl 2:S29-34. Available from: https://doi.org/10.1053/apmr.2003.50244
» https://doi.org/10.1053/apmr.2003.50244 - 19. Pasquali L. Instrumentos psicológicos: manual prático de elaboração. Brasília: LabPAM/IBAPP; 1999.
-
20. Prinsen CAC, Mokkink LB, Bouter LM, Alonso J, Patrick DL, Vet HCW, et al. COSMIN guideline for systematic reviews of patient-reported outcome measures. Qual Life Res [Internet]. 2018 [cited 2024 Apr 14];27(5):1147-57. Available from: https://doi.org/10.1007/s11136-018-1798-3
» https://doi.org/10.1007/s11136-018-1798-3 - 21. Fayers PM, Machin D. Quality of Life: The assessment, analysis and reporting of patient-reported outcomes. 3rd ed. Oxford: John Wiley & Sons; 2016.
-
22. Boyde M, Grenfell K, Brown R, Bannear S, Lollback N, Witt J, et al. What have our patients learnt after being hospitalised for an acute myocardial infarction? Aust Crit Care [Internet]. 2015 [cited 2024 Apr 14];28(3):134-9. Available from: https://doi.org/10.1016/j.aucc.2014.05.003
» https://doi.org/10.1016/j.aucc.2014.05.003 -
23. Alfasfos N, Darawad MW, Nofal B, Samarkandi OA, Abdulqader B. Knowledge, attitudes, beliefs and perceived risk of Acute Coronary Syndrome among Jordanian patients. Health [Internet]. 2016 [cited 2024 Apr 14];8:1830-44. Available from: https://www.scirp.org/journal/paperinformation.aspx?paperid=73135
» https://www.scirp.org/journal/paperinformation.aspx?paperid=73135 - 24. Ajzen J, Fishben M. Understanding attitudes and predicting social behavior. New Jersey: Prentice-Hall; 1980.
- 25. Brown TA. Confirmatory fator analysis for applied research. 2nd ed. New York: Guilford Publications; 2015.
-
26. Demisse L, Alemayehu B, Addissie A, Azazh A, Gary R. Knowledge, attitudes and beliefs about acute coronary syndrome among patients diagnosed with acute coronary syndrome, Addis Ababa, Ethiopia. BMC Cardiovasc Disord [Internet]. 2022 [cited 2023 Nov 27];22(444):1-9. Available from: https://doi.org/10.1186/s12872-022-02893-2
» https://doi.org/10.1186/s12872-022-02893-2 -
27. Vocci MC, Fontes CMB, Abbade LPF. Cultural adaptation of the Glamorgan Scale to Brazilian Portuguese: Pressure Injury in Pediatrics. Rev Latino-Am Enfermagem [Internet]. 2021 [cited 2023 Nov 27];29:e3424. Available from: https://doi.org/10.1590/1518-8345.4083.3424
» https://doi.org/10.1590/1518-8345.4083.3424 - 28. Polit DF, Beck CT. Fundamentos de pesquisa em enfermagem: avaliação de evidências para a prática de enfermagem. 7th ed. Porto Alegre: Artmed; 2011.
-
29. Keszei AP, Novak M, Streiner DL. Introduction to health measurement scales. J Psychosom Res [Internet]. 2010 [cited 2024 Jul 23];68:319-23. Available from: https://doi.org/10.1016/j.jpsychores.2010.01.006
» https://doi.org/10.1016/j.jpsychores.2010.01.006 -
30. Johnson CAH, Pitta NC, Dessotte CAM, Dantas RAS, Rossi LA. Knowledge, attitudes, and beliefs about acute coronary syndrome among patients with type 2 diabetes. Rev Latino-Am Enfermagem [Internet]. 2021 [cited 2024 Apr 14];29:e3503. Available from: https://doi.org/10.1590/1518-8345.5435.3503
» https://doi.org/10.1590/1518-8345.5435.3503
NOTES
-
ORIGIN OF THE ARTICLE
Extracted from the thesis - Cultural Adaptation and Validation of the Acute Coronary Syndrome Response Index for Use in Brazil, presented to the Graduate Program in Fundamental Nursing, Escola de Enfermagem de Ribeirão Preto, Universidade de São Paulo, in 2023.
-
FUNDING INFORMATION
This paper was carried out with the support of Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brazil (CAPES) - Funding Code 001.
-
APPROVAL OF ETHICS COMMITTEE IN RESEARCH
Approved by the Ethics Committee in Research of the Escola de Enfermagem de Ribeirão Preto, Universidade de São Paulo, opinion No. 4,307,783 and Certificate of Presentation for Ethical Appraisal No. 30391820.9.0000.5393.
-
TRANSLATED BY
Leonardo Parachú.
*Translator 1: Brazilian and fluent in English; †Translator 2: Brazilian and fluent in English; ‡Translator 3: American and fluent in Portuguese; §Translator 4: American, healthcare professional, and fluent in Portuguese.
*F1: factor "Stereotypical symptoms"; †F2: factor "Other common symptoms"; ‡F3: factor "Incorrect symptoms"; §F4: factor "Symptoms consistent with stroke".
*F1: factor “Symptom Recognition”; †F2: factor “Help-seeking”.
*F1: factor “Expectations”; †F2: factor “Action”.