Research design and population
The current methodological research was conducted in 2018. A mixture of qualitative and quantitative methods was used to design and develop the desired tool. The qualitative section of the research was performed to identify the preventive behaviors needed by cardiovascular patients when exposed to dust by using an expert panel’s opinions and literature review. The quantitative section of the research was performed to evaluate the psychometric properties of the research tool. The research population included people with heart diseases referred to a heart hospital in Bushehr, Iran. The inclusion criteria of the study were suffering from one type of heart disease and having a history of exposure to dust in the past year. Convenient sampling method was used in this study. A total of 417 heart patients referred to the hospital during the one year of study were enrolled into the research. It should be noted that the sample size was 4-10 times more than the number of items in the questionnaire to analyze the factors structure [10]. Given the number of items in the questionnaire, the number of participants was suitable for evaluating its psychometric properties.
Before distributing the questionnaire, the objectives and methodology of the study were explained to the participants and their written informed consents were obtained. In this way, the participants were ascertained that participation in the study was voluntary and that the data would only be analyzed collectively. Additionally, anonymous questionnaires were collected through face-to-face interviews. The research design was approved by the Scientific and Ethical Committee of Bushehr University of Medical Sciences.
Design of the questionnaire
This research tool was designed in four steps based on Waltz Tool Design Method[11]:
Step 1: Defining the concept of dusts complications for heart patients based on the health belief model using literature review and panel of experts. To review the texts, valid databases including SID, Iranmedex, Scopus, and Pubmed were searched using the following keywords: heart disease, cardiopulmonary problems, dust, and the health belief model. The panel of experts included cardiovascular, health education, and environmental health professionals. The concept of dust was defined based on the health belief model in different dimensions and the preventive behaviors needed by patients were identified in each dimension. The aim of the concept explanation was to present a comprehensive definition of dust problems for heart patients based on the health belief model.
Step (2): Defining the tool design goals: The design of each measurement tool is done for a specific purpose. In this step, it is determined that the designed tool is applicable in a particular setting. The operational definition of each of the desired dimensions is essential in this step [12]. This step aims to define appropriate objective indicators to measure a structure in a tool.
Step (3): Formulating the tool items: In the first step, some of the questionnaire items were formulated by examining the questionnaires designed for this purpose and using the panel of experts. The items formulated in the panel of experts, including 10 health education professors, cardiologists, and environmental health professionals, were reviewed in three sessions. After reaching a consensus on the final dimensions and based on the operational definition of each dimension, the final items were modified and selected. The initial questionnaire consisted of 30 items.
Step 4: Examining the psychometric characteristics of the tool: Tests and other measurement tools must have characteristics in order to be useful for the goal for which they have been designed. One of the most important characteristics that indicate the applicability of a tool is its reliability and validity[12]. In this step, the following measures were done:
Face validity and content validity:
Content validity was examined using both qualitative and quantitative methods in this study. Content validity refers to the extent to which the tool items are related to the studied content or conceptual dimensions[13]. To evaluate the validity of the qualitative content, the initial design of the questionnaire was provided to 10 health education professors, cardiologists, and environmental health professionals to check its grammar, wording, and proper placement of the phrases and to express their opinions. Moreover, the opinions of 10 heart patients were used to examine the ambiguities of the items and to make them comprehensible. After applying their opinions, the content validity of the tool was quantitatively evaluated. Content Validity Ratio (CVR) and Content Validity Index (CVI) were used for this purpose. According to Lawshe, CVR determined the necessity of an item based on a three-point Likert scale (necessary, useful but not necessary, and not necessary). Then, an item with a ratio higher than 0.62 (according to 10 experts) was retained using the following formula and the Lawshe table to determine the minimum value of the CVR index. In other words, the presence of an item with an acceptable level of significance (P<0.05) was essential in the tool[14]. The following formula was used: (see Formula 1 in the Supplementary Files)
CVI aimed to determine the appropriateness, clarity, ambiguity, and relevance of the items to the research objective from the experts’ points of view. For this purpose, the opinions of the relevant professors (health education, cardiology, and environmental health) on the relevance, clarity, and simplicity of each item were evaluated using a four-point Likert scale. Then, based on the following formula, the CVIs of each item and the whole questionnaire were calculated. (see Formula 2 in the Supplementary Files)
After evaluating the content validity by the experts and excluding the items that did not have appropriate index scores, the questionnaire was submitted to 10 heart patients to determine the importance of each item based on a five-point Likert scale. Considering the face validity, the validity of the questions was assessed in terms of appearance and form and the question whether the appearance of the questionnaire items was appropriate for evaluating the desired goal was answered. The impact score method was used to determine the quantitative face validity of the tool. In doing so, 10 heart patients were asked to indicate the importance of each item using a five-point Likert scale (completely important, important, moderately important, slightly important, and not important) and scores 1 to 5 were respectively assigned to the mentioned options. Then, the impact score of each item was calculated using the following formula: “Impact Score = (% Frequency) importance” and the quantitative face validity of the questionnaire was determined. If the impact score of each item was higher than 1.5, the item was identified as appropriate and was retained for the subsequent analyses[15].
Construct validity
Construct validity examines the relationship between a measurement tool and the theoretical backgrounds or theorems. In other words, construct validity raises the question of to what extent a measurement tool reflects theorems and the higher this reflection, it will have a higher construct validity. In order to use factor analysis, there must be a correlation between the desired variables and when the matrix is significant, all correlation coefficients will be zero [12]. Confirmatory Factor Analysis (CFA) with likelihood maximal method at the level of covariance matrix was used to evaluate the construct validity using the health belief model and to identify the tool domains. CFA is a statistical method that tests hypothetical models. CFA allows models to be used both statistically and theoretically, which is not possible by conventional multivariate methods such as Exploratory Factor Analysis (EFA)[16]. It should be noted that the appropriate sample size for factor analysis is 4 to 10 times more than the number of variables and at least 100[10]. To determine the construct validity in the current investigation, the questionnaire was distributed among 417 heart patients with a history of exposure to dust. To evaluate the fitness of the CFA model, chi-square index was used first. The values smaller than the mentioned index indicate the better fit of the model. However, as this index is sensitive to the large sample size, the researchers did not rely on this index and calculated the chi-square to degree of freedom ratio, which is more statistically significant. Some studies have recommended that the chi-square to degree of freedom ratio should be less than three for the model to be accepted (14). Other used indices were Comparative Fit Index (CFI), Incremental Fit Index (IFI), Root Mean Square Error of Approximation (RMSEA), Goodness of Fit Index (GFI), and Adjusted Goodness of Fit Index (AGFI). The CFI, IFI, GFI, and AGFI take a value between zero and one. The closer the values are to one, the more appropriate the model will be [17]. Additionally, RMSEA values less than 0.08 are considered to be appropriate and those less than 0.05 are regarded as good fit [10, 15]. Finally, the GFI and AGFI values higher than 0.8 and 0.9 and the CFI values higher than 0.9 are considered to be appropriate[10, 13].
Reliability
To determine the internal stability of the questionnaire, the internal consistency method was used for each of the domains as well as for the whole questionnaire. In this regard, using the Cronbach’s alpha coefficient is one of the most common methods based on the internal consistency of the scales within a questionnaire[18].
Statistical analysis:
The data were analyzed using the SPSS software, version 23 and AMOS software, version 23.
Ethical considerations:
After obtaining the necessary legal permissions from the Research Committee and obtaining the ethics code from Bushehr University of Medical Sciences (IR.BPUMS.REC.1395.62), the data were collected by direct referral of the researcher to Bushehr Heart Hospital. After expressing the research goals and reassuring the participants about the voluntary nature of the research and confidentiality of the information, the researcher invited the heart patients who were interested in participating in the study.