INTRODUCTION
The recent 2021 data by the Korean Society of Hypertension & Hypertension Epidemiology Research Working Group estimated that approximately 12.07 million adults or approximately 28% of those over the age of 20 in Korea had hypertension (HTN) [1]. Among these adults aged 20 years or older with HTN, about 70% were aware of their condition, and approximately 66% were undergoing treatment [1]. However, the rate of successful management and control of HTN was a mere 48% [1]. This highlights the need for concerted efforts to effectively control HTN in the population. The prevalence of HTN is particularly high among these older adult population and is expected to rise even further as their population continues to grow [2]. In fact, among these patients with HTN, those aged 65 years or older make up 40% [1].
HTN increases the risk of heart disease and stroke, which are the leading causes of death in the adult population [3]. However, a 10-mmHg reduction in systolic pressure can reduce the risk of cardiovascular diseases, such as myocardial infarction, heart failure, and stroke by up to 28% [4,5]. The use of antihypertensive drugs can reduce the incidence of complications in the management of HTN; however, the problem with antihypertensive drugs in old adult people is that the rate of control is lower than the treatment rate [6]. Furthermore, older people complain of difficulty in managing chronic diseases due to aging, mobility difficulties, literacy, or other diseases. It is important for public healthcare providers (PHPs) involved in HTN management of older adults to be aware of the patient's health status and to make appropriate recommendations, including the prescribing of targeted medications that address the causes of HTN and the patient's health concerns [6]. Targeted HTN management can reduce HTN-related physical and psychological distress and the associated financial burden by reducing HTN-related costs [7].
In South Korea, PHPs who are responsible for primary healthcare centers in rural areas comprise nurses or licensed midwives who have completed at least 24 weeks of training provided by the Korean Ministry of Health and Welfare [8]. PHPs work on behalf of doctors and are responsible for undertaking physical examinations; treating trauma patients and providing first aid for patients in an emergency; prevention, care, and management of patients with chronic diseases; patient rehabilitation during healthy deliveries; administering vaccinations; and administering medication in accordance with medical guidelines [9]. In addition to being rurally located, the proportion of senior citizens is often high in areas where PHPs work and access to hospitals is restricted owing to factors such as distance. In Korea, the proportion of older adults in urban areas is 10∼18%, whereas that in rural areas is 20∼25%. Thus, in addition to supplying prescription medicines, PHPs are also responsible for delivering public health services such as health education on diseases, nutrition, and lifestyle management, and HTN management to old adult patients in rural communities [10]. To improve the quality of healthcare services for HTN management by PHPs, rural residents with HTN should assess the quality of care they receive, considering whether it has a positive effect on HTN. However, due to the lack of assessment tools, information on the perspectives of Korean patients with HTN in relation to HTN management by PHPs is insufficient.
Till date, the Korean Primary Care Assessment Tool (KPCAT) stands as the sole tool for evaluating the quality of primary care services provided by local family doctors [11]. Comprising 21 questions across 5 domains-personalized care, coordination, inclusiveness, family/community orientation, and first contact-the KPCAT validity and reliability were assessed among patients utilizing primary healthcare services [11]. However, this tool includes content that is unrelated to the work of PHPs, such as the number of rectal and oral cancer screenings or regular checkups. Therefore, the use of KPCAT in assessing the quality of HTN management by the PHPs has limitations. Moreover, other existing tools primarily concentrate on evaluating patients’ selfcare behaviors when assessing HTN management [7,12,13]. Although some tools do exist for assessing the activities carried out in medical clinics by patients [14], their main focus lies in evaluating prevention and management programs rather than specifically addressing HTN management. Additionally, considering the challenges faced in self-management by older adult patients with HTN, it is crucial to examine how the management behaviors of PHPs effect the self-management practices and BP control rates in these patients. However, there is presently a lack of tools to evaluate the quality of HTN management from the perspective of patients receiving care from healthcare providers. Therefore, this study aimed to develop and assess the validity of a quality assessment tool for hypertension management (QAT-HTN), to evaluate the quality of hypertension management of public healthcare providers working in community healthcare centers.
METHODS
1. Study Design
A cross-sectional descriptive study was designed to develop and assess the validity and reliability of QAT-HTN.
2. Participants
The study included adult participants who were diagnosed with HTN by a doctor and were being managed by a PHP. Recruitment for the study took place in rural areas of C-do, Korea, and involved 149 primary healthcare centers. The criteria for selecting specific candidates were as follows: those who were diagnosed with HTN by a doctor; those who received HTN management from PHP or received HTN-related education at primary healthcare centers; and voluntarily agreed to participate in the study. Those who did not agree to participate in the study or who had difficulty answering the research questions due to cognitive decline or dementia were excluded from the study. The sample size for this study was calculated as the ratio of study participants to the number of items on a scale (5∼10 per item) to enable exploratory factor analysis (EFA) [15], and >200 samples were needed to enable confirmatory factor analysis (CFA) [16]. To enable EFA and CFA, the required sample size was a minimum of 460 participants. The questionnaire was distributed to 550 participants, 22 of whom did not respond. In total, 528 questionnaires were used in the data analysis.
3. Procedure and Data Analysis
1) Instrument development
The QAT-HTN was developed based on the Evidence-based Recommendations for Hypertension in Primary Care [17] and public community healthcare guidelines for patient care from which we extracted essential items for managing HTN in primary care. We then compared and adjusted these items to align with the responsibilities of the PHPs. Items outside the PHP's scope of work were removed, and we included items related to nursing. The developed tool consisted of 27 items divided into six sub-domains: information provision (e.g., “ I was educated on the definition, symptoms, and complications of high blood pressure by my PHP”[disease knowledge]), BP measurement (e.g., “ When I visit the primary healthcare center, the PHP directly measures my blood pressure [including an automatic blood pressure monitor]”), physical examination before medication (e.g., “ My PHP listens to my heart or bowel sound with a stethoscope when my PHP re-prescribes my high blood pressure medications”), medication prescription and guidance (e.g., “ My PHP prescribes me a dose of antihypertensive drugs within a month”), lifestyle improvement (e.g., “ My PHP explained that I should not eat salty food”), and preventive action (e.g., “ My PHP recommends me to have a test for early detection and prevention of HTN complications at least once a year”). The tool was evaluated for the appropriateness of its contents by seven PHPs and three nursing professors with research experience on patients with HTN. This tool was evaluated using a content validity index (CVI) involving a 4-point Likert scale. We checked to determine whether the question content was within the scope of the PHPs’ work and whether the general public would find any phrases or words difficult to understand. Through measuring the concordance ratio for relevance, items with a score of ≥.80 were adopted for the scale. Meanwhile, items with a score ranging from .50 to .80 were modified in consultation with experts. For example, in item 1, “ explanation” for hypertension was modified to “ education.” In item 9, “ drug prescription” was modified to “ drug re-prescribing.” The content in the evaluated tool was then reduced to 26 items. Each item was measured using a 5-point Likert scale: “ every time,” 5 points; “ frequently,” 4 points; “ sometimes,” 3 points; “ rarely,” 2 points; “ never,” 1 point.
2) Construct validity
For construct validity, item analysis and EFA were verified using SPSS version 24.0, and CFA was verified using AMOS version 22.0 (both by IBM Corp., Armonk, NY, USA). The item analysis evaluated the mean, standard deviation, skewness, and kurtosis to determine the degree of bias in each item. Additionally, item-total correlation coefficients were calculated to determine whether each item reflected the concept of a measure for the satisfactory HTN management in primary community healthcare [18]. The validity of this tool was assessed through EFA and CFA. EFA was conducted using the SPSS 24.0 program, and the suitability of the factor analysis was evaluated using the Kaiser-Meyer-Olkin (KMO) index and Bartlett's sphericity test. The principal factor extraction method was used to extract the factors, while the varimax method was used for rotation. Factor extraction was based on eigenvalues of 1 or higher, while the cumulative variance was required to be 50.0% or more [19]. Loading values below .40 and items with low commonality were excluded, in line with the approach described by Song [20]. CFA was performed using AMOS 22.0 program. Metrics that describe the fit of a model included the χ2 to the degrees of freedom ratio (CMIN/df), root mean square residual (RMR), root mean square error of approximation (RMSEA), comparative fit index (CFI), goodness of fit index (GFI), and Tucker-Lewis index (TLI). CMIN/df should be <2, RMR and RMSEA are judged to be good if<.05 [21], and CFI is considered acceptable if ≥.80 [22]. Discriminant validity was assessed by employing the mean variance extraction to determine distinctions in measured values among various variables. The criterion dictates that discriminant validity exists when the square of the correlation coefficient between each variable is less than the average variance extracted estimate (AVE). To evaluate convergent validity, which asserts that different measurement methods should exhibit a strong correlation when measuring the same concept, concept reliability was examined. The criterion for judging convergent validity is a composite reliability (CR) value of≥.70.
3) Reliability
Reliability of the QAT-HTN scale was evaluated by calculating the Cronbach's ⍺ value, which indicates internal consistency.
4) Concurrent validity
Concurrent validity is a method of testing validity by measuring the developed tool and external criterion con-currently. A correlation coefficient ranging from .30 to .70 is recommended [23]. The concurrent validity of the QAT-HTN scale was evaluated by calculating the Pearson's correlation coefficient using SPSS version 24.0. In a previous study [24], the quality of primary care service was found to be related to HTN selfcare behavior. The concurrent validity was evaluated by a HTN selfcare behavior scale developed by Han et al. [25]. This tool was translated into Korean by An et al. [7] and its reliability and validity were confirmed for Korean older adult patients with HTN. This tool consists of 20 items divided into two sub-domains: 11 HTN selfcare diet behaviors and 9 HTN selfcare health behaviors (excluding diet). This tool was measured using a 4-point Likert scale. During the tool development, Cronbach's ⍺ was .92 (HTN selfcare diet behavior, .91; HTN selfcare health behavior [excluding diet], .85) [7]. In this study, Cronbach's ⍺ was .91 (HTN selfcare diet behavior, .90; HTN selfcare health behavior [excluding diet], .87).
4. Data Collection
The study period was from August to September 2019. Data were collected through the PHPs working at 149 primary healthcare centers located in South Chungcheong Province. Prior to the survey, the purpose and study procedures were explained to 149 PHPs. The PHPs in turn explained these to the patients who met the inclusion criteria, asked them about their willingness to participate in the study, and conducted a survey targeting only patients with HTN who consented to participate in the study. A self-reported questionnaire, which included general characteristics, developed QAT-HTN tool items, and HTN selfcare behavior, was used. If requested, the PHPs assisted the study participants in completing the questionnaire. The average time to completion of the questionnaire was 15 to 20 minutes. After completing the survey, a small gift was provided. The questionnaire was distributed according to the number of patients with HTN visiting the public health center, with 550 distributed overall. Of these, 22 were not returned; therefore, 528 were analyzed. The 528 participants who completed the questionnaire were divided randomly into dataset A (n=260) for EFA and dataset B (n=268) for CFA.
5. Ethical Considerations
This study was conducted following approval of the research proposal (JIRB-2019061001-01-190704) by the institutional review board of the Joongbu University. A written consent form was added on the first page of the written survey, which stated the background, purpose, anonymity, and confidentiality of the survey; that the results of the survey will not be used for purposes other than research; and that the survey form will be destroyed at the end of the study. In addition, it explained that there was no disadvantage in refusing to participate in the study or withdrawing participation during the study. Completed questionnaires were stored as an anonymous computer-coded Excel file, stored in a password-protected file in a locked place, and stored in a place accessible only to the researcher to prevent leakage. According to Article 15 of the Enforcement Rules of the Bioethics Act, research data will be stored for 3 years from the time the research is completed and then shredded. All participants provided written consent.
RESULTS
1. Participant Characteristics
The participants’ general characteristics are presented in Table 1. In total, 62.7% of the participants were female, and the mean age was 71.5±7.69 years. Most participants had graduated from elementary school (40.0%) and were married (64.4%). In total, 48.3% had jobs and 51.7% were unemployed. Most respondents (75.0%) considered their health to be above average. The mean duration of HTN was 9.7±6.38 years, and 78.2% of the participants received HTN-related education. HTN-related complications occurred in 43.8% of the participants. There was no statistically significant difference between datasets A (used to run the EFA) and B (used to run the CFA).
Table 1.
General and Disease-related Characteristics of Participants (N=528)
2. Construct Validity
Following item analysis, one item (T12) with an item-total correlation <.30 was deleted. This item included the ‘explaining the condition after the physical examination’ content. After deleting this item (T12), the reliability of the final model was.92, and the total-item correlation ranged from .41∼.67.
The KMO index for the 25 questions in the EFA was .89, indicating its suitability for conducting EFA. Additionally, Bartlett's spherical test yielded a significance level <.001, further confirming its appropriateness. The communality values for all items exceeded .40, and the percentage of cumulative variance was 56.5%. EFA was initially performed without rotation to examine the number of sub-domains within the scale. By analyzing the elbow point of the scree plot and eigen values ≥1.0, dividing it into six components was considered the most appropriate. Subsequently, varimax rotation was applied, resulting in factor loadings >.30 for most of the rotated component matrices. This process ultimately led to the classification of the final six factors and their corresponding 25 items, as outlined in Table 2.
Table 2.
Factor Loadings Results of Exploratory Factor Analysis (N=260)
The CFA was performed using dataset B (n=268) to confirm the goodness of fit of the model between the six sub-domains derived from the EFA results (Table 3). The chi-squared degree of freedom (χ2/df) for the CFA analysis of the 25 items for the six factors was 2.90. In the model's goodness-of-fit, the SRMR (.49), RMSEA (.08), CFI (.93), GFI (.90), and TLI (.81) showed that that the model was suitable. Standardized factor loadings ranged from .45 to .95 (p<.001) (Table 3). The results presented in Table 4 demonstrate both discriminant and convergent validity. Discriminant validity was established by comparing the squared values of the correlation coefficients between each variable to the AVE, which confirmed that the squared correlation values were smaller than the AVE. This indicates that there is a clear distinction between each sub-domain. Furthermore, convergent validity was established by the CR values for all sub-domains of ≥.70, providing evidence on convergent validity.
Table 3.
Factor Estimates and Goodness-of-fit Indicators in Confirmatory Factor Analysis (N=268)
Table 4.
Discriminant Validity and Convergent Validity among Subdomains (N=268)
3. Reliability
Cronbach's ⍺ was used to measure the reliability of the tool, which was .92. The reliability for each factor was as follows: HTN management, .82; regular checkup and education, .84; complication prevention, .76; assessment, .62; blood pressure measurement, .60; lifestyle management, .90.
4. Concurrent Validity
There was a significant positive correlation between the total QAT-HTN score and HTN selfcare behavior score (r=.35, p<.001). There was also a significant positive correlation between the QAT-HTN sub-factors and HTN selfcare behavior score (r=.19∼.32, p<.001)(Table 5).
Table 5.
Concurrent Validity (N=528)
DISCUSSION
This study confirmed the validity and reliability of the QAT-HTN. Our findings showed that the 26-item tool adequately measured the quality of HTN management by PHPs in six dimensions: HTN management, regular checkup and education, complication prevention, assessment, BP measurement, and lifestyle management (Appendix 1). Our tool differed from the KPCAT [11], which involves personalized care, coordination function, comprehensiveness, family/community orientation, and first contact. This is because, compared with primary care physicians, the PHPs perform the work of nurses, community managers, and health educators in addition to some of the work of doctors [10]. However, these results are similar to the sub-domains of initial questions classified as information provision, BP measurement, physical examination, prescription and medication guidance, lifestyle improvement, and prevention. This is because the initial questions were developed based on the HTN management guidelines for primary care physicians and the work of the PHPs, and the contents were validated by the working group.
The final 25 items identified were used to validate the CFA for the tool. The CFA assessment confirmed the suitability of the structural model, as evidenced by the evaluation of CMIN/df and model fit. Moreover, the tool exhibited high explanatory power and reliability. Its development was based on HTN management protocol stand-ards and primary care requirements. Furthermore, the content validity was verified by PHPs after its development. However, it is important to note that unlike the number of items included during the development phase, the two sub-domains included only two items. This can be attributed to a mismatch between the intended actions by PHPs and the actions accepted by the target audience. Previous studies have highlighted a communication gap between patients and medical staff [26].
Concurrent validity was confirmed because there was a significant correlation between QAT-HTN and HTN selfcare behavior. There was also a significant correlation between the sub-domains of QAT-HTN and HTN selfcare behavior. However, the “ assessment,” “ blood pressure measurement,” and “ complication prevention” domains had a lower correlation with HTN selfcare behavior when compared with other domains. The limited correlation between HTN selfcare behavior and the sub-domains of the developed tool could be attributed to the tool's content being unrelated to HTN selfcare. Furthermore, the section on ‘lifestyle management’ includes material that is closely associated with selfcare; however, the observed correlation between the two was minimal. This may be because weight management education, smoking cessation, and alcohol abstinence were items in the lifestyle domain. Notably, a previous study examining smoking cessation intentions in patients with HTN found significantly lower intentions among those aged 60 years or older [27]. This implies the need for a program that reinforces abstinence or smoking cessation, rather than focusing solely on educating about a normal lifestyle. HTN management is a major issue in community health, and ensuring the quality of HTN management is important in rural areas where community healthcare centers are located [28]. In HTN management, along with medication therapy, selfcare behavior is important for improving the adherence to non-pharmacological therapy [29]. In addition, selfcare behavior creates economic benefits, prevent complications, and improve health [30]. However, HTN management in primary healthcare centers focused on pharmacotherapy. Furthermore, management plans for HTN selfcare are often not actively pursued in relation to the characteristics of patients with HTN in healthcare centers. Considering that the population composition of the area where community healthcare centers are located is particularly high in older adults, more meticulous care services for selfcare and health promotion should be provided for the management of patients with HTN visiting community healthcare center. The results of our study demonstrated that the quality of PHP's HTN care service was related to the selfcare behavior of patients with HTN. The QAT-HTN developed through our research can be used to assess the quality-of-care service provided by PHPs to patients with HTN. In addition, this tool can be used to conduct a further study to determine the effect of the HTN care service provided by PHPs on the selfcare of patients with HTN.
This study was developed based on the work of PHPs in community healthcare centers in Korea. However, the limitations of this study need to be addressed. Since this study was conducted only at some community healthcare centers in Korea, there may be bias in the study results. Moreover, a stability test using a test-retest or a group comparison validity test was not performed. In addition, although the correlation between this tool and HTN selfcare was proven, the sub-domain of hypertension selfcare showed a lower correlation than the standard. Nevertheless, we developed an objective measure to identify the quality of services for patients with HTN by developing a QAT-HTN. In the future, it is crucial to reassess the reliability and consistency of this tool by including a broader sample. Furthermore, given that this tool was originally designed with specific Korean PHP rules, it is essential to conduct a separate validity test when translating it into different languages or utilizing it to assess the effectiveness of HTN management of other professionals. We believe this tool will be useful for PHPs in primary healthcare centers for managing patients with HTN residing in the community. Furthermore, our study findings may be used to better determine the quality of HTN management in primary healthcare centers and to develop strategies for enhancing self-management behaviors among community-dwelling patients with HTN.
CONCLUSION
This study developed a QAT-HTN that comprise 25 questions with six factors for evaluating the quality of HTN management of PHPs in primary healthcare centers; this tool showed adequate reliability and validity. Further studies are needed to evaluate the quality of HTN management in community healthcare centers using this management scale.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation 
Print 
