Designing a national model for assessment of nursing informatics competency

Background Due to the need for informatics competencies in the field of nursing, the present study was conducted to design a psychometric instrument to determine the qualification of informatics competencies of employed nurses in educational care centers. Methods The questionnaire was made by reviewing existing scientific resources and assessment tools. Two hundred nurses were selected using simple random sampling. Structural equation modeling was used using the measurement model technique and the average variance was calculated. Linear structural relations (LISREL) software was used to test the assumptions and correlations of the model. Results Findings showed relatively good estimation in the fit of first-order measurement model. The informatics knowledge subscale with a determining rate of 0.90 had the greatest explanatory effect among the subscales and informatics skill with a determining rate of 0.67 and basic computer skill with a determining rate of 0.60 were observed. The second-order measurement model of fitness indicators showed that the three factors can well explain the multidimensional construct of informatics competency. Conclusions The designed tool can be used to develop educational strategies in relation to nursing students in the field of informatics and prepare them in the rich environment of information technology, which can be helpful in training nursing instructors.

entry (CPOE), and electronic health records (EHR) [5]. There is evidence that awareness of the nursing informatics competencies is necessary to fulfill their professional responsibilities [6].
Competence is defined as the state of being well qualified, and it means having the knowledge, skills, and ability to perform specific tasks, activities, or professions [7]; however, it also includes the concept of values, attitudes, critical thinking, and clinical decision-making [8]. Therefore, nursing informatics competencies can be defined as adequate knowledge, skills, and abilities to perform specific informatics tasks [9]. Today, to maintain public health as one of nursing education's main goals, much emphasis has been placed on evaluating and empowering nursing informatics competencies [10,11]. Without informatics standards and competencies, information technology would be ineffective in the healthcare setting and impose risks on patient safety.
By 1988, nurses in the International Medical Informatics Association and the National League for Nursing developed the first informatics competencies, and other studies soon followed [12,13]. However, these identified competencies only described entry-level competencies such as computer skills for nurses, rather than the more sophisticated informatics skills needed by experienced nurses, especially informatics nurse specialists (INS). Renewed interest in informatics competencies for nurses began in the early 2000s. Staggers et al. [14][15][16] defined the initial standard for determining, ranking, and evaluating nursing informatics competencies. This was the first study to cover four levels of nursing, developed competencies for both entry-level and experienced INS, and examined the categories of computer skills, informatics knowledge, and informatics skills [14].
In Iran, as a developing country, the need to consider the concept of nursing informatics competencies has been emphasized in recent years. Some of the topics related to competency mainly focus on clinical competencies, such as nursing managers, nurse educators, nursing students, and nursing graduates [6]. However, extensive research in this area is limited and so far, only a few studies have been conducted concerning the concept of the nursing informatics competencies in this country. In a study (2019), the effect of a training program was examined on the nursing informatics competencies of critical care nurses in Iran based on the Nursing Informatics Competency Assessment Tool (NICAT) [17]. This tool is developed based on the need for an acute care setting and is not generalizable to all healthcare settings [18]. Therefore, due to the changes in technology and newer technologies required in nursing, more attention should be paid to the informatics competencies for the development and standardization of this concept in various aspects of the nursing profession. To that end, the present study aimed to design a psychometric instrument to qualify the informatics competencies of nurses employed in the educational health care centers.

Study design, setting and participants
This descriptive, cross-sectional study was conducted in Iran in 2018 to investigate the psychometric properties and factor structures of nursing informatics competencies. Eligible nurses with Bachelor's degrees participating in this study needed to have full-time job experience of at least 3 years in eight teaching hospitals in the country. Table 1 shows the research outline in January-October, 2018.

Questionnaire development
After a review of previous studies and evaluation of existing tools and scientific resources associated with nursing informatics competencies, the basic items of the questionnaire were extracted and then classified into three domains (computer skills, informatics knowledge, and informatics skills) based on the study conducted by Staggers et al. (2002) [14]. The questionnaire consisted of 74 items at three subscales: Basic Computer Skills (22 items), Informatics Knowledge (25 items), and Informatics Skills (27 items). It should be noted that computer skills in this study is defined as the proficiency in the use of computer hardware and software and should not be confused with computer science.
The questionnaire was reviewed in a focus group of twelve experts, including eight nursing experts holding a Master's degree or faculty members with at least five years of work experience with hospital information systems (HIS) and four information technology (IT) professionals holding a Bachelor's degree with at least five years of work experience in the IT department of the hospital. Experts scored the importance of each item on a 5-point Likert scale ranging from 'strongly disagree' to 'strongly agree' . The average importance of informatics competencies items was estimated separately for each item, and then the final questionnaire was designed. The content validity of the questionnaire was assessed by the same experts using the content validity ratio (CVR) [19] and content validity index (CVI) [20]. In CVR, experts' responses for each item were measured based on the three scales of (1) 'necessary' , (2) 'helpful but not necessary' , and (3) 'not necessary' . Then, according to the Lawshe table [21], items with a content validity ratio (CVR) higher than 0.56 were considered acceptable. To calculate CVI, relevance, clarity, and simplicity of all items were checked using a 4-point Likert scale, and items with CVI > 0.79 were considered appropriate. Finally, the items were confirmed in three basic computer skills (17 items), informatics knowledge (15 items), and informatics skills (16 items) subscales. Following validity assessment, content reliability of the questionnaire was assessed using the test-retest method. Thus, a valid version of the questionnaire was administered to 20 nurses. After two weeks, the participating nurses were asked to fill out the questionnaire again. Cronbach's alpha was used to assess data reliability. Data reliability was calculated as 0.97.

Data collection
Before conducting the research, verbal consent was obtained from all participants. They were informed that they could withdraw from the study at any time during filling out the questionnaires if they do not want to continue. All questionnaires were anonymous, and the participants were assured of the confidentiality of their information. This study was approved by the ethics committee of Kashan University of Medical Sciences (KAUMS).

Study size
The confirmatory factor analysis (CFA) designed to validate an appropriate and meaningful instrument was used to assess nursing informatics competencies. In this study, the sample comprised 200 nurses eligible for the study. According to Kline [20], in the factor analysis, the minimum sample size of 200 is defensible. Thus, the sample size of the present study was larger because, in the confirmatory factor analysis, the minimum sample size is determined based on the subscales, not the items. All samples were selected using a simple random sampling technique according to a list of nurses with a Bachelor's or higher degree and three years of work experience with Hospital Information System (HIS) in teaching and medical hospitals.

Data sources and measurement
The questionnaire comprising two parts was distributed among nurses. The first part included demographic characteristics including age, sex, marital status, employment, level of education, work experience, familiarity with computers, frequency of computer usage, and interactions with the HIS. The second part included questions related to the subscale of informatics competencies on a scale of 48 questions.
The validity of the constructed instrument was analyzed through confirmatory factor analysis [22] using LISREL software (version 8.5). LISREL has also been used for the structural equation modeling in two parts: confirmatory factor analysis and path analysis. In order to assess the amount of fitness in the developed measurement model, confirmatory factor analysis and experimental data were used to get the chi-square index, chi-square goodness of fit index, adjusted goodness of fit index, comparative fit index, and the root mean square error of approximation (RMSEA) [23]. In this study, structural equation modeling was used with the help of measurement model, composite reliability, and construct validity (convergent and discriminant) using the maximum likelihood approach through LISREL 8.5 software to test the assumptions and correlations of the model. Convergent validity shows the compatibility of items measuring the same construct. Three methods were evaluated to achieve convergent validity and solidarity: (a) factor loading suggested by Haire et al. [24] as the accreditation criterion for the factor loading of the numeric amount of 6.0 and above. It is noteworthy that if the amount of the factor loading of an item in conjunction with a construct is higher, the item plays a greater role in explaining the constructs. If the factor loading is less than 3.0, the role of the item is not significant, so that it will be ignored. The factor loading between 3.0 and 6.0 is considered acceptable. If it is greater than 6.0, it will be desirable. (b) Composite reliability shows the number of item reflections on the desired construct. The amount of the proposed criterion of composite reliability is 7.0, confirming that the reliability is acceptable [25]. (c) The average extracted variance is equal to or more than 0.5 [26,27]. In discriminant validity, the amount of difference between the items of the construct and those of other constructs was evaluated. The most common way to assess discriminant validity needs to be formed in the correlation matrix (Fornell and Larker) [28]. Therefore, the original diagonal values are the root of Average Variance Extracted (AVE) coefficients of each construct, and lower values of original diagonal are the correlation

Demographic characteristics
Out of 200 questionnaires distributed, 197 were completed. The response rate was 98.5%. About 77% of the respondents were employed in the position of a ward nurse, and 11.5% in the position of a head nurse. The majority of participants were females (75.5%) and married (68.5%). Regarding the level of education, 95.6% had a Bachelor's degree with 3 to 11 years of work experience (56.3%). In terms of familiarity with computers, 66.5% were relatively good, 59% of nurses used computers more than once a day, and spent less than 1 h working with the HIS in each shift (59%).

Correlation between Demographic characteristics and nursing informatics competencies
The correlation analysis results revealed no significant relationship between sex, marital status, level of education, work experience, frequency of computer usage, and interaction with the HIS and nurses' basic computer skills (P > 0.05). However, there was a significant relationship between age, familiarity with computers, and nurses' basic computer skills (P < 0.02). The results did not show any significant relationship between sex, age, level of education, work experience, and frequency of computer usage, and nursing informatics knowledge (P > 0.05). Nevertheless, there was a significant relationship between marital status, familiarity with computers, and interaction with the HIS and nursing informatics knowledge (P < 0.02). In the subscale of informatics skills, there was no significant relationship between sex, age, marital status, level of education, work experience, and frequency of computer usage and nursing informatics skills (P < 0.05), while a significant relationship was observed between familiarity with computers, interaction with the HIS and nursing informatics skills (P < 0.02). Table 2 shows CVI and CVR for each item. Among the questionnaire items, five items from the basic computer skills, 10 items from informatics knowledge, and 11 items from the informatics skills subscales were removed. Therefore, the number of items in the questionnaire was reduced to 48 (Additional file 1).

Confirmatory factor analysis of the Questionnaire
To determine which item belongs to each factor, the confirmatory factor analysis was used. The results showed a relatively good estimation in the fit of the first-order measurement model (χ 2 /df = 1.59, GFI = 0.79, CFI = 0.99, NFI = 0.97, RMSEA = 0.055). It is worth mentioning that each subscale was separately analyzed through the confirmatory factor analysis using the LISREL, such that some items were removed to obtain a relatively appropriate measurement model (Fig. 1). Two items of basic computer skills and informatics knowledge and six items of informatics skills subscales were removed because the cut-off point was lower than 0.5 (R 2 < 0.5). Table 3 shows the discriminant validity and reliability for each subscale. Furthermore, to examine the fit of subscales, the subscales were analyzed along with all items in the first-order measurement model. Therefore, none of the items were removed (Fig. 1). Finally, the secondorder factor analysis was used to evaluate the effect of basic computer skills, informatics knowledge, and informatics skills on nursing informatics competencies and assess whether the entire subclass was put into the form as a concept (Fig. 2). It was observed that the subscale of informatics knowledge with the determining rate of 0.90 has the greatest explanatory effect among the subscales and informatics skills with the determining rate of 0.67 and basic computer skills with the determining rate of 0.60 were ranked second and third.
The second-order measurement model of fit indices showed that the three factors of basic computer skills, informatics knowledge, and informatics skills could explain the multidimensional construct of informatics competency and represent the dimensional accuracy in introducing a three-dimensional framework of informatics competency in the best way (χ 2 /df = 1.51, GFI = 0.80, CFI = 0.99, NFI = 0.97, RMSEA = 0.051). Finally, the reliability assessment results from the test-retest method, using Spearman's rank correlation-coefficient (ρ), showed that the nursing informatics competencies tool has the highest correlation in each subscale (more than 0.9).

Discussion
In the first part of this study, three areas of informatics competency were used as stated by Staggers et al. [14]. Content validity (CVI and CVR) of informatics competency items was assessed by experts with a 48-item validated questionnaire. Staggers et al. [14] designed the classification of 3-factor nursing informatics competencies with 37 items for beginner nurses and 32 items for experienced nurses including basic computer skills, informatics knowledge, and informatics skills. However, evaluation of the need for education at each specific level of nursing informatics competence was not possible [29].
Hunter et al. [9] made the list of nursing informatics competencies in three subscales including basic computer skills (51 items), information literacy (25 items), and clinical data management (9 items). In their study,   while basic computer skills were listed in detail, items such as PowerPoint familiarity, Excel familiarity, working with accessories, and typing skills were not listed.
In conclusion, it seems that the designed tools in the present study are more complete and more comprehensive in the field of basic computer skills and all the necessary dimensions in this context have been taken into consideration. Hart [30] performed a three-round Delphi study to determine core informatics competencies for generic nurse managers resulting in a list of 49 core informatics competencies that were categorized into three groups including computer skills, informatics knowledge, and informatics skills. Nonetheless, the list does not mention the practical, research, and educational competencies [29]. Chang et al. [31] conducted a Delphi technique to identify informatics competencies required for nurses in Taiwan. The results showed that 318 informatics competencies for nurses had 97.8% consensus with the results of the study conducted by Staggers et al. [14].
Hwang et al. [32] designed a specific classification for nursing informatics competencies in their study. They classified parts of basic computer skills in the form of informatics knowledge and its original classification; they introduced computer trends structure as one of the three factors affecting nursing informatics competencies in Taiwan whose significance was more than basic computer skills. The difference between the items of these two studies is due to the variety in the classification of nursing informatics competencies structures. Also, the knowledge of using the HIS as an important variable is considered in the informatics knowledge structure in the study conducted by Hwang et al. [32]. This variable was not considered in the present study since it is only the skill of using it that has been classified in the informatics skill structure. This research is the only study assessing  nursing informatics competence confirming the threefactor structure including basic computer skills, informatics knowledge, and informatics skills with regard to the suitability of various indicators of structural equation modeling. Westra et al. [33] introduced 92 items related to nursing informatics competencies based on the work of Staggers et al. [14], including 24 items related to computer skills, 40 items related to informatics knowledge, and 28 items related to informatics skills. This list is the same as the one provided by Hart [30] in providing specific qualifications of access to data, communication, systems, data dissemination, and training. This list is useful for determining individuals' specific competence while it does not allow the evaluation of the level of competence at a particular level of practice. Westra's list emphasizes the knowledge and skills of informatics in the role of nurses' leadership for planning, supporting, directing, and evaluating information initiatives beyond the practical skills of nurses for action [29].
Lenburg [34] provided an 8-core practice list of essential competencies for nursing including assessment and intervention skills, communication, critical thinking, human caring, teaching, management, leadership, and knowledge integration skills. In competence of communication skills, the sub-skill of computing, it is discussed in which only customer communication, search for resources, and specialized responsibility were provided. In this list, informatics was not presented as a core competence of nursing whereas it is of great importance because of the integration of technology and information communication in contemporary nursing. The difference between Lenburg's study and the present one is that, in the present study, informatics competencies emphasize computer skills and informatics skills when discussing nursing practices. In another study administered by Cronenwett et al. [35], a list of recommended qualifications was required for all nurses to provide safe care to patients. In the classification provided, at the level of informatics competencies, only 3 subjects including knowledge, skill, and attitude were mentioned. The list is generally a useful tool for identifying current nurses' competencies; however, the informatics section is inadequate in identifying vital concepts of nursing competencies. This study lists the qualifications while it does not show the depth of the skills and the level of training required at each particular level [29].
TIGER research [36] presented a list of essential informatics competencies for experienced nurses including levels of basic computer competence, information literacy, and information management. At the level of basic computer competence, varied items concerning hardware and software intelligence, various forms of electronic communication, recognition of safety and operating systems, using the Internet, and many other items were listed. There are also several competencies in the information literacy and information management area which cannot be evaluated using this list [29]. Integrated informatics was considered to improve education, provide care, and practical measures for nurses [37]. TIGER's tool was developed for all nurses without specification for level of practice and included 281 competencies. Nevertheless, the use of information technology in the field of health in this country is at an early stage and nursing experts have not presently considered many of the competencies as important and in practice, there is no need to teach them to all nurses. Therefore, nursing informatics can be created in countries that do not have this field, and higher-level activities in the field of clinical health information systems can be expected from them.
In the study by Hubner et al. [38], a broad list of core competencies for nurses was identified based on five nursing domains including clinical nursing, quality management, inter-professional coordination, nursing management, and information technology management. They provided six core informatics competencies in each of the five nursing areas. However, this tool was not carefully developed as the tool demonstrates core competencies, such as nurses' documentation and information knowledge management, without identifying the competency items under each core competency [39].
Kaminski [40] also introduced a framework determining a specific classification of nursing skills. This list includes a set of competencies required for different levels of nursing. The tool is capable of assessing the presence or absence of nursing skills while it does not show the degree of nursing mastery to the details of the aforementioned skills [29]. In the instrument presented in this study, nursing competence was generally evaluated since nurses were not assessed in terms of their levels in the working field.
The differences between the present study and other studies are due to the differences in concept definitions, the type and level of technology used and the kind of competence that is required. Competencies such as the use of decision support systems, database management systems, database software, remote patient monitoring systems, effective care design tools and appropriate technology to collect patient-related data were considered essential for nurses in Stagger's study. Since in many developing countries, hospital information systems and applications in the field of public health, especially in nursing care, are in the early and experimental stages and have not reached the implementation stage, these competencies were not agreed upon and approved by the experts in this study. The competencies agreed in this area are mostly related to the basic and primary principles of working with computers.
However, the results of this study in terms of construct validity and confirmatory factor analysis as assessed through the 3-factor structure of the Persian version of the nursing informatics competencies questionnaire were different from the results reported in other studies [11,41,42]. The reasons behind the difference in the number of factors can be linked to different statistical methods such as exploratory functional analysis concerned with the derivation and determination of the relationships between the factors.
One of the findings of the present study is the relationship between the 3-factor model of nursing informatics competencies and their subscales (Figs. 1, 2). The strongest relationship was observed between informatics knowledge (0.95) and informatics competencies while the weakest relationship was observed between basic computer skills (0.78) and informatics qualifications. In analyzing these results, it is noteworthy to mention that informatics knowledge had a stronger impact on informatics competencies than the other two subscales. Moreover, enhancing nursing informatics knowledge plays a greater role in increasing informatics competencies. In the study conducted by Yang et al. [11], sufficient knowledge regarding technology use significantly increased nurses informatics competencies.
Nevertheless, in a study conducted by Arzu Akman et al. [41], the studied participants had no knowledge of nursing informatics. However, 2.86% of them stated that nursing informatics courses must be included in the nursing curriculum. The computer-based nursing curricula should be dynamically based on the regional needs and adaptations to new professional attitudes to assure the possibility of training the nurses in order to guarantee an adequate level of expertise concerning informatics. The result of this study showed that the questionnaire was reliable, with the minimum reliability coefficient being related to test-retest of informatics skills subscale (0.95) and the highest reliability coefficient being related to basic computer skills subscales (0.99). In the study done by Akman et al. [41], the reliability of the designed tools was proper. The results of the present study showed that participants' level of informatics capabilities was medium. In other studies, nursing informatics skills are have been shown to be at an intermediate level [11,42,43].

Implications for practice and future research
The presented tool can be used to develop educational strategies for nursing students in the field of informatics and to prepare them in the rich environment of information technology in the healthcare environment of developing countries. Therefore, the Faculty members of Nursing and Midwifery and the Education Deputy of Ministry of Health, Treatment, and Medical Education can benefit from the results of this study. Future nurses cannot be well-prepared to confront challenges in the various professional fields without the acquisition of technical skills. Therefore, nurses should be able to use data, information, knowledge, and technologies properly to improve nursing care. In this regard, nursing informatics can play a significant role in the areas of education, healthcare provision, research, and management and be used for preservation and storage of clinical care and the extension of research studies that are directly related to the quality of patient healthcare.

Conclusion
Since nurses are regarded as the largest part of the healthcare workforce and the major users of clinical information systems, nursing informatics competencies are important for the successful use of clinical information systems and improving patient safety in computerized environments. Therefore, the nursing profession needs to promote the acceptance and usage of information technology to empower nursing informatics competencies. For this purpose, the informatics competencies approved in the present questionnaire can be helpful for educational administrators in order to train nurses and include a special course for nursing informatics in the educational curriculum. This questionnaire can also be useful in periodically evaluating the level of informatics competencies of working nurses and nursing students in different communities.