Construction and validation of the CADE-Q for patient education in cardiac rehabilitation programs

Ghisi, Gabriela Lima de Melo; Durieux, Adriana; Manfroi, Waldomiro Carlos; Herdy, Artur Haddad; Carvalho, Tales de; Andrade, Alexandre; Benetti, Magnus

doi:10.1590/S0066-782X2010005000045

Abstracts

BACKGROUND: Being aware of the coronary artery disease can be considered the first step to reduce the risk of cardiac complications. OBJECTIVE: Building and validating a tool to assess and describe coronary patients' awareness in cardiac rehabilitation programs, with the purpose of education. METHODS: For the construction, we analyzed articles and field studies to submit items to multidisciplinary team associated to cardiac rehabilitation. After this analysis, we generated the version tested in a pilot study. The tool, named CADE-Q (Questionnaire for Coronary Patient Education) was applied in 155 patients aged 61 ± 9 (min = 36, max = 86) in response to cardiac rehabilitation programs. Out of the 155 patients, 114 were men. Internal consistency was measured by Cronbach's alpha coefficient. Reproducibility was tested by the intraclass correlation coefficient (ICC) and construct validity was performed by exploratory factor analysis. The analysis compared the total scores as a function population characteristics and rehabilitation groups (private and public). RESULTS: The final version has 19 questions with 4 alternatives, with 4 quadrants of awareness. Cronbach's alpha was 0.68 and ICC was 0.783. The factor analysis revealed 6 factors, covering three areas of awareness, which demonstrates the multifactorial nature of the instrument. The population characteristics as a function of the total score showed significant differences depending on the socioeconomic status variables (type of rehabilitation, household income and education level). CONCLUSION: CADE-Q has proper validity and reliability to be used in the Brazilian population in future research.

Questionnaires; education; coronary diseases; rehabilitation

FUNDAMENTO: O conhecimento sobre a doença arterial coronariana pode ser considerado o primeiro passo para reduzir o risco de complicações cardíacas. OBJETIVOS: Construir e validar um instrumento capaz de avaliar e descrever o conhecimento do paciente coronariano em programas de reabilitação cardíaca, com a finalidade de educação. MÉTODOS: Para construção, foi realizada análise de artigos e estudo de campo para a apresentação de itens a uma equipe multidisciplinar associada à reabilitação cardíaca. Após análise, foi gerada a versão testada em um estudo-piloto. O instrumento, nomeado CADE-Q (Questionário para Educação do Paciente Coronariano), foi aplicado em 155 pacientes com idade de 61 ± 9 anos (mín = 36 ; máx = 86), participantes de programas de reabilitação cardíaca. Dos 155 pacientes, 114 eram homens. A consistência interna foi verificada pelo coeficiente Alpha de Cronbach. A reprodutibilidade foi testada através do coeficiente de correlação intraclasse (CCIC) e a validade de construto por análise fatorial exploratória. Foi realizada análise comparando os escores totais em função de características da população e entre os grupos de reabilitação (privado e público). RESULTADOS: A versão final possui 19 questões com 4 alternativas, com 4 quadrantes de conhecimento. O Alpha de Cronbach foi de 0,68 e CCIC foi de 0,783. A análise fatorial revelou 6 fatores, abrangendo três áreas de conhecimento, o que demonstra a multifatoriedade do instrumento. A análise das características da população em função do escore total apresentou diferenças significativas em função das variáveis do nível socioeconômico (tipo de reabilitação, renda familiar e escolaridade). CONCLUSÃO: O instrumento CADE-Q apresenta validade e confiabilidade adequadas para sua utilização na população brasileira em futuras pesquisas.

Questionários; educação; doença das coronárias; reabilitação

ORIGINAL ARTICLE

^ICentro de Ciências da Saúde e do Esporte (CEFID) - Universidade do Estado de Santa Catarina (UDESC), Florianópolis, SC - Brazil

^IIUniversidade Federal do Rio Grande do Sul (UFRGS), Porto Alegre, RS - Brazil

^IIIHospital Regional de São José (HRSJ), São José, SC - Brazil

Mailing address

ABSTRACT

BACKGROUND: Being aware of the coronary artery disease can be considered the first step to reduce the risk of cardiac complications.

OBJECTIVE: Building and validating a tool to assess and describe coronary patients' awareness in cardiac rehabilitation programs, with the purpose of education.

METHODS: For the construction, we analyzed articles and field studies to submit items to multidisciplinary team associated to cardiac rehabilitation. After this analysis, we generated the version tested in a pilot study. The tool, named CADE-Q (Questionnaire for Coronary Patient Education) was applied in 155 patients aged 61 ± 9 (min = 36, max = 86) in response to cardiac rehabilitation programs. Out of the 155 patients, 114 were men. Internal consistency was measured by Cronbach's alpha coefficient. Reproducibility was tested by the intraclass correlation coefficient (ICC) and construct validity was performed by exploratory factor analysis. The analysis compared the total scores as a function population characteristics and rehabilitation groups (private and public).

RESULTS: The final version has 19 questions with 4 alternatives, with 4 quadrants of awareness. Cronbach's alpha was 0.68 and ICC was 0.783. The factor analysis revealed 6 factors, covering three areas of awareness, which demonstrates the multifactorial nature of the instrument. The population characteristics as a function of the total score showed significant differences depending on the socioeconomic status variables (type of rehabilitation, household income and education level).

CONCLUSION: CADE-Q has proper validity and reliability to be used in the Brazilian population in future research.

Key words: Questionnaires/utilization; education; coronary diseases; rehabilitation.

Introduction

In recent years, the concept of health has evolved and changed from 'state of absence of disease' to 'a positive concept focusing on personal and social aspects as well as physical, mental and spiritual aspects' related to many factors, including the opportunity to education and awareness^1,2.

Every year, cardiovascular diseases (CVD) are responsible for 16.7 million deaths around the world, and according to the World Health Statistics (2008), these deaths will increase to 23.4 million worldwide in 2030³.

Cardiac rehabilitation programs aim getting coronary patients back to an optimal level of recovery, enabling this individual to acquire and maintain better health conditions and reduce the risk of death and acute events related to their disease. To achieve these goals, interventions are combined, and correctly and conveniently applied. These interventions include educational programs^4,5.

The raising of awareness associated with multidisciplinary programs to control coronary artery disease (CAD) helps patients achieve an adequate perception of their health status, changing beliefs, behaviors and bad habits^5-9.

However, for coronary patients to be truly aware of their disease, symptoms, lifestyle, risk factors and treatments, it is necessary to evaluate how much patients know about their disease. Therefore, the development and/or improvement of instruments capable of undertaking such an evaluation become critical^10,11.

The use of questionnaires or instruments of assessment is an important resource in educational health programs, because they measure the effects of teaching and learning, and the potential changes in attitudes on CAD. It is also a way to learn individual needs and the conditions for the implementation of an educational process^7,9-12.

Therefore, the purpose of this study was to construct and validate a tool to assess and describe coronary patients' awareness in cardiac rehabilitation programs, for the purposes of education.

Methods

Preparation of the questionnaire

Based on Pasquali¹³, the process of developing and validating the tool named CADE-Q - Questionnaire for Education on Coronary Artery Disease - included three steps with three different procedures: theoretical, empirical and analytical. The theoretical procedure referred to the theoretical foundation on the construct for which we intended to develop a measurement tool. The empirical procedure is also called experimental and consisted of steps and techniques for applying the pilot instrument, as well as collecting information that could evaluate the properties of the instrument. The analytical procedure determined the statistical analyses of data in order to validate the instrument.

Therefore, the development of the CADE-Q instrument consisted of the following steps:

1) Searching texts and articles to build up an inventory. Such inventory was supposed to carry key information that patients need to know about CAD, as well as, for the selection of items that should be known by coronary patients, a qualitative field research was performed with patients in cardiac rehabilitation programs with 50 items to be part of the instrument;

2) Presenting items to a multidisciplinary team of professionals associated with the Cardiac Rehabilitation Program - composed of physicians, nurses, physical educators, physiotherapists, nutritionists and psychologists - to assess the content, relevance and conceptual and cultural analysis of the questions, generating the First Version with 20 multiple choice questions, each question containing 4 choices;

3) The First Version was submitted to a pilot study in 30 patients to evaluate the understanding of the issues (clarity) and dispel doubts about the structure of questions and answers. Using a scale from zero to 10, each patient should assign a score to the clarity of the questions, giving rise to the level of clarity, according to Pasquali's¹³ clarity criterion.

4) After the analysis, some items were developed and one question was removed, creating the Final Version of CADE-Q with 19 questions.

Validation

Data collection took place in Florianópolis, state of Santa Catarina, from August to November 2008, 155 coronary patients selected at random, provided that they had been participating in cardiac rehabilitation programs for at least a month. The application was done by researchers in a standardized, (before or after the sessions), collective and supervised manner (to ensure privacy and anonymity of the results) without any interaction between evaluators and respondents. The data were analyzed using the program SPSS 13.0 - Statistical Package for Social Sciences, and the significance level adopted was 0.05.

Psychometric analyses were performed to evaluate the validity and reliability of the CADE-Q.

The construct validity was assessed through the factor analysis. To check the suitability of the data to such analysis, we assessed Keiser-Meyer Olkin's index (KMO), to assess the factorability of intercorrelation matrices on which the factor analysis is based. Then, Bartlett' sphericity test was performed to determine whether the data met the sphericity requirement^14,15.

We used the principal component analysis to extract factors, considering only those with eigenvalues greater than one, combined with scree plot. After the factors have been selected, a factorial matrix was generated to observe the relationships between items and factors by means of factor loadings. To interpret the matrix, we applied the method of Principal Component Extraction by orthogonal rotation, using the method Equamax^14,15.

The reliability assessment includes two main ways: internal consistency and reproducibility. Internal consistency was assessed by Cronbach's alpha in all individuals subject to the application of the instrument, based on the minimum value of 0.60¹⁴. Reproducibility was assessed using the intraclass correlation coefficient, using test and retest situations. The time interval between tests was two weeks, testing 25 individuals of the pilot study, also selected in a simple and random manner.

To check CADE-Q's outcomes for some population characteristics (age, sex, comorbidities, surgical procedures related to CAD, type and time of cardiac rehabilitation, education level and household income), we compared the total scores according to such variables. We also checked the outcomes as a function of the 4 quadrants of awareness. By testing the distribution of data through the Kolmogorov-Smirnov test, it was found that the variables of the scale resultants were normally distributed (p> 0.1). Considering this, we used ANOVA and Bonferroni's post hoc test.

There are other types of validity of instruments found in literature¹², such as the Concurrent Validation, which consists in the correlation of an instrument under validation with a scientifically validated one¹². However, as there are few instruments in literature that assess coronary patients' awareness and most of them consist of yes/no and true/false questions - such as MICRO-Q⁷ - which may not reflect patient's real awareness of the disease, this method was discarded.

The research was conducted within the standards required by Helsinki Declaration and approved by the Ethics and Human Research Committee of the University of Santa Catarina (UDESC) according to resolution CNS 196/96. All individuals were informed about the goals of the research, data confidentiality, and signed a Consent Form.

Results

Preparation of the questionnaire

To select important items for the composition of the instrument, we performed a field research with patients in cardiac rehabilitation programs. Patients were mainly interested in learning about the following: physical symptoms, such as shortness of breath, chest pain (48.0%), exercising and sport (39.0%), recovery of the heart (29.0%) and questions about how life goes with the disease (22.0%). These results are similar to other studies evaluating the quantity and quality of information gained about the disease in everyday life of coronary patients^16,17.

After this research, were prepared and presented 50 items to the multidisciplinary team of professionals associated to the Cardiac Rehabilitation Program, which generated the First Version of CADE-Q with 20 questions.

In the pilot test performed in 30 patients, on a zero to 10 scale, the index of clarity¹³ of questions was 9.4 ± 1.3, indicating that the questionnaire is easily understood by the target population. However, examining the questions separately, question 17 (Angiotensin-converting Enzyme Inhibitors (ACE) are?) showed a level of clarity of 1.5 ± 1.4, and was later excluded from the questionnaire¹³. Still in the pilot test, we observed the average time required for patients to fill out the questionnaire: 13 ± 4 minutes.

CADE-Q's final version is a self-applied questionnaire comprising 19 multiple choice questions with 4 alternatives each: a correct statement representing full knowledge, a correct statement representing incomplete knowledge, an incorrect statement representing wrong knowledge, and an 'I do not know' statement representing no knowledge (Annex 1). Both alternatives, 'full knowledge' and 'incomplete knowledge', are correct. The difference lies in the point of view of the level of knowledge: the complete alternative brings a more scientific and thorough knowledge, which is probably transmitted by health professionals to the patient; the incomplete alternative brings a correct knowledge of the question, but a popular one, which may be acquired, for example, through non-scientific media.

The aim of the questionnaire is educating patients with coronary disease, by assessing and describing the level of knowledge of such patients. Patients with other heart diseases cannot be assessed with such instrument.

The questions were divided into 4 areas of knowledge: A1, related to coronary artery disease; A2, diagnosis and drug therapy; A3, on risk factors and lifestyle; and A4, related to exercising. The questions were randomly arranged and could be part in more than one quadrant.

For each alternative, we established scores which, together, lead to a total score, which indicate the degree of knowledge about CAD, as well as knowledge of each specific area. The scores established for each alternative are: correct statement representing full knowledge = three; correct statement representing incomplete knowledge = one; incorrect statement representing wrong knowledge = zero; and 'do not know' statement, representing no knowledge = zero. Based on Sommaruga et al⁷, in spite of representing different situations, the 'incorrect' and 'do not know' statement received equal scores, because in the case of instruments of knowledge, they have similar implications. Whereas the purpose of this instrument - patient education - having a wrong understanding or not having any understanding about a given subject determines the need for education, which justifies the scores given.

The sum of the scores establishes the patient's total level of knowledge, and the maximum score is 57 points. Table 1 shows the classification of knowledge level based on other studies^7,13.

Thumbnail

The participants are characterized by a personal data sheet attached to the instrument.

Validation

The sample consisted of 155 patients, of which 114 were men. The 155 patients participated in cardiac rehabilitation programs (40.0% private, 60.0% public), with participation time of 33 ± 46 months (min = 1, max = 360, Md = 18) and age 61 ± 9 (min = 36, max = 86). All patients had comorbidities associated with CAD and hypertension (HBP) was the most frequently found (72.0%), followed by dyslipidemia (64.0%) and congestive heart failure (CHF) (24.0%). Regarding the surgical procedures related to the disease, 105 individuals have undergone some type of cardiac surgery.

Socioeconomic status was characterized by household income and educational level. In the analysis of the types of rehabilitation - public and private - we observe that patients in public programs have lower income and educational level, and 66.0% earn 1 to 5 minimum wages per month and 39.0% have completed primary education. Patients in private programs have greater income and higher education levels: 45.0% earn more than 20 monthly wages and 61.0% have university degree.

The characteristics of the population under study are described in Table 2.

Thumbnail

Validity

Regarding construct validity, we performed a preliminary analysis to ensure normality of data and suitability for factor analysis. The Kolmogorov-Smirnov test revealed normal distributions for all items. After the data were found normal, we analyzed Keiser-Meyer Olkin's index (KMO) and Bartlett's sphericity test to check suitability of data for factor analysis^14,15. The KMO index was 0.608 and Bartlett's sphericity indicated that the correlation matrix is not an identity matrix (X ² = 348.8, p < 0.001). Together, these indicators ensured that the data set has the necessary requirements for factor analysis¹⁵.

The factor analysis of data was done by "Principal Component Extraction" by "orthogonal" rotation, by the method Equamax. Considering the load factors above 0.3¹⁴ with the aid of the method of "Principal Axis Analysis", several extractions were made and the solution found to be most appropriate¹³ was the 6-factor one, which explain 56.1% of total variance of individuals' responses. Table 3 presents the empirical structure of the instrument with the factor loadings of each question in the 6 factors extracted.

Thumbnail

The first factor "General Factor" refers to the 12 questions that assess all fields of knowledge, with greater predominance in the area of pathophysiology, signs and symptoms. Factor two included 8 items called "Causal Factor", with greater predominance in the field of diagnosis, treatment and medicines (4 items), risk factors and lifestyles (4 items) and exercising (4 items). The third factor included 4 items, with greater predominance in the field of risk factors and lifestyles. Therefore, it was called "Risk Factors Factor". As factors 4, 5 and 6 did not prevail in areas, they have been termed "Other Factors".

Internal consistency - tested via Cronbach's alpha coefficient - was 0.68. The test-retest reliability was assessed by intraclass correlation coefficient (ICC), whose value was 0.783.

Descriptive analysis

Through the Personal Data Sheet attached to the instrument, we analyzed the characteristics of the population according to the total score obtained: age, sex, comorbidities associated (hypertension, CHF, diabetes mellitus type I - DMI, diabetes mellitus type II - DMII, peripheral arterial disease - PAD, dyslipidemia, and chronic obstructive pulmonary disease - COPD), surgical procedures related to CAD, type of cardiac rehabilitation (public or private), time of cardiac rehabilitation, education level and monthly household income^16,17.

It was observed that there are no significant differences in total scores due to age (p = 0.240), comorbidities - having hypertension or not (p = 0.487), having CHF or not (p = 0.310), having DMI or not (p = 0.612), having DMII or not (p = 0.335), having PAD or not (p = 0.781), having dyslipidemia or not (p = 0.625), having COPD or not (p = 0.999), duration of cardiac rehabilitation (p = 0.599).

However, total scores revealed significant differences according to gender (p = 0.04), surgical procedures (p = 0.02), type of cardiac rehabilitation (p = 0.02), monthly household income (p < 0.001) and education level (p = 0, 02). These data demonstrate the influence of socioeconomic status in patient's level of awareness, as reported in studies^7,18-25.

By analyzing cardiac rehabilitation groups (public and private) according to the characteristics, we found no significant differences regarding age, gender, comorbidities, surgical procedures related to CAD and cardiac rehabilitation time. However, significant differences (p <0.001) were found according to educational level, monthly household income and total score.

At this stage, the average time of application of the instrument was 14 ± 4 minutes. This time was higher in individuals with lower levels of knowledge (17 minutes).

The implementation of CADE-Q revealed a total score of 43.01 ± 6.5, which corresponds to a knowledge of 75.0% of the total scoring presented in the questionnaire. As for classification, we found a prevalence of 'good knowledge' type in 66.0% of patients (n = 102). As for the alternatives identified, it is observed that 69.63% of the questions had the full knowledge option checked, 17.9% of incomplete knowledge, 5.7% of wrong knowledge and 6.8% of "do not know".

Examining the questions, we observe that such behavior does not occur in questions 9, 10 and 11, considered critical items. The interval and the average of correct answers per question are shown in Table 4.

Thumbnail

Regarding the 4 areas of knowledge into which the questionnaire is divided, the comparison between the averages pointed out significant differences between the areas studied (p = 0.03). The best scores were observed in A2 (risk factors and lifestyles), concentrating 90.0% of responses between great and good. The worst levels of knowledge were found in the area of diagnosis, treatment and medicines (A3), with 34.0% of responses ranging from acceptable to poor.

Discussion

The process of constructing and validating an instrument in the field of health requires a great effort¹⁰. Because of this, in Brazil we note a lack of instruments to measure knowledge about CAD. This study contributes towards equipping health professionals with an instrument to establish educational strategies focused on patients' real needs.

In the construct validity, although the KMO indicator and Bartlett's sphericity test have pointed out that the data sets had the prerequisites for factor analysis, the factor solution was not appropriate because the instrument items have multidimensional characteristics, mostly comprising more than one field of knowledge. However, the factor analysis alone cannot determine that an instrument is invalid¹³. In this case, CAD is a multifactorial disease. Because of that, the instrument was adapted to cover the largest number of factors related to the disease, which the factor analysis eventually corroborated, since each of the 6 factors includes at least three fields of knowledge.

Also concerning validity, the results show that there is consistency of measures in successive applications in the same group, that is, accurate and stable data¹⁴. This is because, in terms of internal consistency, Cronbach's alpha revealed no homogeneity among the questions, since it was above the minimum value of 0.60 reported by some authors¹⁴, despite the multifactorial nature of the disease previously reported.

Looking at the test-retest reliability, the ICC value was very close to what literature suggests¹⁴ (0.8), demonstrating stability of the instrument, i.e., successive applications of the same instrument produced the same or similar outcomes.

The final version of CADE-Q (Questionnaire for Education on Coronary Artery Disease) is self-applied and consists of 19 multiple-choice questions with 4 alternatives each, divided into 4 quadrants of knowledge: coronary artery disease, diagnosis and medication, risk factors and lifestyle, and exercising.

Concerning the descriptive analysis, total scores revealed significant differences according to the variables representing socioeconomic status (type of cardiac rehabilitation, monthly household income and educational level). This shows that socioeconomic status is influencing coronary patients' knowledge of their disease, as reported in studies^7,18-25.

Because 'learning' is at the heart of human adaptation skills, and because patient's education is not only "practice", but "movement", much of patient's knowledge is related to acquired information, values, attitudes, moral judgments, patterns of behavior through observing and socializing. Individuals can educate themselves and acquire knowledge through rules of behavior observed in certain media or social levels, which we define as the influence of socioeconomic status, also reported in this study^{7-9, 19}.

Although only the characteristics related to socioeconomic status have shown significant differences, other factors (age, sex, associated comorbidities, time of cardiac rehabilitation, surgical interventions related to the disease) may influence the acquisition of knowledge of coronary patients and should be described in studies using this instrument^4,7,10,26.

The inclusion of patients with different times of participation in cardiac rehabilitation programs showed no influence on the knowledge neither on the reliability of the questions, perhaps because the samples have shown large asymmetry (SD = 46.21). It was found that patients who have been 1-6 months in the programs had scores of 43 ± 7. Patients with over 120 months in programs had lower knowledge scores (42 ± 8). Despite the insignificant results, the time of cardiac rehabilitation is an important point in studies evaluating patients in treatment programs, which can be explored in future studies, addressing, for example, knowledge before and after a certain period.

The average of total scores, corresponding to 75.0% of knowledge of the total presented in the questionnaire, differs from studies where it is estimated that 30 to 78.0% of coronary patients do not fully understand educational information transmitted to them²⁷. This reveals that education in clinical practice is often inappropriate, inconsistent and inaccurate, but when patients are enrolled in cardiac rehabilitation programs, they have greater contact with knowledge, once the team of professionals involved in these programs reviews concepts, sets aside misleading ideas and beliefs and encourages healthy habits, educating patients⁴.

The critical items (questions 9, 10 and 11) reveal that, although the patients had good total scores, some important and fundamental information to coronary patients are missing¹⁷, such as appropriate diet, optimal blood lipid values and absolute contraindication for physical exercising, respectively.

Although it is not the goal of this study, it is known that using such tools provides groups and individuals with an opportunity of increasing their levels of understanding of CAD, that is, it provides education targeted at cardiac rehabilitation programs^7,27. Moreover, such studies and can be quite useful in developing strategies to encourage patients' adherence to cardiac rehabilitation programs, as well as interfering with the success of this intervention, since it is not known how patient's knowledge is translated into concrete actions towards optimal health^28,29.

We also suggest that further studies associating knowledge with therapy acceptance, follow-up and success, which represent the clinical validation of the tool scientifically designed and validated in this study.

Conclusion

CADE-Q questionnaire produced appropriate levels of reliability and validity and can be used to assess the knowledge of coronary patients in cardiac rehabilitation and, where necessary, establish educational programs focusing on patients' education about their disease.

Potential Conflict of Interest

No potential conflict of interest relevant to this article was reported.

Sources of Funding

There were no external funding sources for this study.

Study Association

This article is part of the thesis of master submitted by Adriana Durieux, from Universidade do Estado de Santa Catarina (UDESC).

References

1. Bethel H, Lewin R, Dalal H. Cardiac rehabilitation in the United Kingdom. Heart. 2009; 95 (4): 271-5.
2. Cohen JD. ABCs of secondary prevention of CHD: easier said than done. Lancet. 2001; 357 (9261): 972-3.
3. World Health Organization. World Health Statistics. Genebra; 2008.
4. Sociedade Brasileira de Cardiologia. Diretriz de reabilitação cardiopulmonar e metabólica: aspectos práticos e responsabilidades. Arq Bras Cardiol. 2006; 86 (1): 74-82.
5. Hirani SP, Patterson DLH, Newman SP. What do coronary artery disease patients think about their treatments? An assessment of patients' treatment representations. J Health Psychol. 2008; 13 (3): 311-22.
6. Calwell MA, Peters KJ, Dracup KA. A simplified education program improves knowledge, self-care behaviour, and disease severity in heart failure patients in rural settings. Am Heart J. 2005; 150 (5): 983.
7. Sommaruga M, Vidotto G, Bertolotti G, Pedretti RF, Tramarin R. A self administered tool for the evaluation of the efficacy of health education interventions in cardiac patients. Monaldi Arch Chest Dis. 2003; 60 (1): 7-15.
8. Kayaniyil S, Winstanley J, Parsons C, Brister S, Oh P, Stewart DG, et al.Degree and correlates of cardiac knowledge and awareness among cardiac impatients. Patient Educ Couns. 2009; 75 (1): 99-107.
9. Yehle KS, Sands LP, Rhynders PA, Newton GD. The effect of shared medical visits on knowledge and self-care in patients with heart failure: a pilot study. Heart Lung. 2009; 38 (1): 25-33.
10. Osborne RH, Elsworth GR, Whitfield K. The Health Impact Questionnaire (heiQ): An outcomes and evaluation measure for patient education and self-management interventions for people with chronic conditions. Patient Educ Couns. 2007; 66 (2): 192-201.
11. Gazmararian JA, Williams MV, Peel J, Baker DW. Health literacy and knowledge of chronic disease. Patient Educ Couns. 2003; 51 (3): 267-75.
12. Guillemin F. Cross-cultural adaptation and validation of health status measures. Scand J Rheumatol. 1995; 24 (2): 61-3.
13. Pasquali L. Psicometria: teoria dos testes na psicologia e na educação. Rio de Janeiro: Vozes; 2003.
14. Dancey CP, Reidy J. Statistics without maths for Psychology: using SPSS for Windows. 3rd ed. London: Prentice Hall; 2005.
15. Hair JF, Anderson RE. Multivariate data analysis. 5th ed. New Jersey: Prentice Hall; 1998.
16. Bolman C, Brug J, Bar F, Martinali J, van den Borne B. Long-term efficacy of a checklist to improve patient education in cardiology. Patient Educ Couns. 2005; 56 (2): 240-8.
17. Martinali J, Bolman C, Brug J, van den Borne B, Bar F. A checklist to improve patient education in a cardiology outpatient setting. Patient Educ and Couns. 2001; 42 (3): 231-8.
18. Albert MA, Glynn RJ, Buring J, Ridker PM. Impact of traditional and novel risk on the relationship between socioeconomic status and incident cardiovascular events. Circulation. 2006; 114 (24): 2619-26.
19. Muennig P, Sohler N, Mahato B. Socioeconomic status as an independent predictor of physiological biomarkers of cardiovascular disease: evidence from NHANES. Am J Prev Med. 2007; 45 (1): 35-40.
20. Cohen B, Vitinghoff E, Whooley M. Association of socioeconomic status and exercise capacity in adults with coronary heart disease (from the Heart and Soul Study). Am J Cardiol. 2008; 101 (4): 462-6.
21. Ranjit N, Diez-Roux AV, Shea S, Cushman M, Ni H, Seeman T. Socioeconomic position, race/ethnicity, and inflammation in the multi-ethnic study of atherosclerosis. Circulation. 2007; 116 (21): 2383-90.
22. Alter DA, Naylor CD, Austin P, Tu JV. Effects of socioeconomic status on access to invasive cardiac procedures and on mortality after acute myocardial infarction. N Engl J Med. 2008; 341 (18): 1359-67.
23. Fiscella K, Tandredi D. Socioeconomic status and coronary heart disease risk prediction. JAMA. 2008; 300 (22): 2666-8.
24. Fiscella K, Tandredi D, Franks P. Adding socioeconomic status to Framingham scoring to reduce disparities in coronary risk assessment. Am Heart J. 2009; 157 (6): 988-94.
25. Loucks EB, Lynch W, Pilote L, Fuhrer R, Almeida ND, Richard H, et al. Life-course socioeconomic position and incidence of coronary heart disease - The Framingham Offspring Study. Am J Epidemiol. 2009; 169 (7): 819-26.
26. Piegas LS, Avezum A, Pereira JCR, Neto JM, Hoepfner C, Farran JA, et al. Risk factors for myocardial infarction in Brazil. Am Heart J. 2003; 146 (2): 331-8.
27. Scott JT, Thompson DR. Assessing the information needs of post-myocardial infarction patients: a systematic review. Patient Educ Couns. 2003; 50 (2): 167-77.
28. Santos RD, Sposito AC, Santos JE, Fonseca FH, Moriguchi EH, Martinez TLR, et al. Programa de avaliação nacional do conhecimento sobre a prevenção da aterosclerose (PANDORA). Arq Bras Cardiol. 2000; 75 (6): 289-95.
29. Redfern J, Ellis E, Briffa T, Freedman SB. Development and testing of innovate patient resources for the management of coronary heart disease (CHD): a descriptive study. BMC Health Serv Res. 2006; 6: 95.