Development and Psychometrics Test of Home Blood Pressure Monitoring Perception Scale for Patients With Chronic Kidney Disease

Background: Home blood pressure monitoring (HBPM) has been proved to be benecial to blood pressure control for both patients with hypertension and patients with chronic kidney disease (CKD). However, what are the psychodynamic predictors like perception for the establishment and persistence of HBPM have not been deeply explored, and there is a lack of instruments for assessing psychodynamic characteristics on HBPM from patients’ perspectives, which has limited the in-depth understanding of HBPM behavior. The study aimed to develop an instrument for evaluating HBPM perception in patients with CKD, and to test the reliability and validity of the instrument. Methods: The original item pool of the HBPM perception scale was developed according to the framework of the health promotion model and literature review. The psychometric characteristics of the instrument were examined with a sample of 436 CKD patients in China. Internal consistency reliability, split-half reliability, test-retest reliability and construct reliability were used to verify the reliability of the scale. And content validity, construct validity and criterion-related validity were used to test the validity of the scale. Results: The expert consultation showed satisfactory content validity of the HBPM perception scale and produced a rst draft of the survey with 43 items. By exploratory factor analysis and conrmatory factor analysis (CFA), we found evidence for the construct validity of the following factors: perceived benets of HBPM, perceived barriers of HBPM, perceived self-ecacy of HBPM, situational inuences and commitment to a plan of HBPM, and immediate competing demands and preferences. The nalized ve-component HBPM perception scale has 27 items. CFA suggested the model t the data well ((cid:0) 2 =679.649, df=310, (cid:0) 2 /df=2.192, root mean square error of approximation =0.074, conrmatory t index=0.902). The HBPM perception scale was positively associated with the Self-Ecacy for Managing Chronic Disease 6-item Scale ( r= 0.256, p(cid:0)0.001). The Cronbach’s α coecient of all dimensions was above 0.750, the split-half reliability was above 0.624, and the test-retest reliability was above 0.749. The construct reliability of dimensions ranged from 0.749 to 0.951.


Introduction
Chronic kidney disease (CKD), a group of diseases with chronic progressive deterioration of renal function and structure, with high morbidity, high mortality and high medical expenditure, has become a chronic non-communicable disease that seriously affects global public health. CKD has a high global prevalence of between 11-13%, with the majority stage 3 1 . The prevalence of CKD among Chinese adults is 10.8%, with an estimated 119.5 million CKD patients nationwide 2 , and the prevalence of CKD in people over 45 years of age was 11.5% 3 . The results of the global burden of disease research program 4 showed that in 2013, 956,200 patients died of CKD worldwide, with a 1.34-fold increase since 1990. Cardiovascular events are the main cause of death of patients with CKD in the world, accounting for more than 60% 5 . Hypertension is the most common cardiovascular complication in patients with CKD. The prevalence of hypertension in patients with CKD in the United States was > 60%, and the prevalence of hypertension in patients with stage 3 and above was as high as 85.7% in the U.S. 6 . In China, the prevalence of hypertension in non-dialysis CKD patients is nearly 70% 7 . With the decline of renal function, the prevalence of hypertension in patients with CKD gradually increased, and was as high as 91% in CKD stage 5 7 .
Hypertension is one of the major risk factors for the progression of CKD and cardiovascular disease. Blood pressure level and cardiovascular risk showed a continuous, independent and direct positive correlation 8 . Controlling blood pressure can reduce the risk of cardiovascular diseases such as atherosclerosis, heart failure, as well as cardiovascular events and vascular death 9 . Currently, patients' blood pressure is mainly controlled by medication treatment 10 and non-drug therapy 11 (such as no smoking, weight control, reduce mental stress). But blood pressure control in patients with CKD remains unsatisfactory. Literature reports that the control rate of hypertension in patients with CKD was 67.1% for the standard of blood pressure < 140/90mmHg, and 46.1% for the standard of blood pressure < 130/80mmHg in the U.S. 6 . In China, the control rate of blood pressure < 140/90mmHg in non-dialysis CKD patients was 33.1%, and the control rate of blood pressure < 130/80mmHg was only 14.1% 7 . In addition to antihypertensive drugs, how to help patients effectively establish the scienti c lifestyle and related health promotion behaviors has become a key and urgent need to be addressed for non-drug therapy.
Blood pressure monitoring plays an essential role in the diagnosis, condition evaluation, curative effect evaluation and prognosis judgment of hypertension. Among them, home blood pressure monitoring (HBPM) has the advantages of strong practicability and maneuverability (simple operation, low cost, strong repeatability) 12,13 , and its role in hypertension monitoring has been increasingly valued. At present, clinical practice guidelines in China 8 , the United States 14 , Japan 15 , Europe 16 and other countries and regions recommend HBPM for hypertensive patients for e cacy evaluation of antihypertensive drugs. HBPM can effectively predict the risk of cardiovascular events, stroke and death in patients with hypertension 17 . Meanwhile, HBPM behavior can also be regarded as a health promotion behavior. Hypertensive patients with HBPM behavior have higher level of understanding of antihypertensive drugs, higher medication compliance, and more satisfactory blood pressure control 18, 19 .
Although HBPM has important value for CKD therapy, HBPM in patients with CKD has not been studied in-depth, and the subjects of the HBPM study are mainly hypertensive patients. In the hypertensive population, only about 50% had their blood pressure monitored at home 20 . Demographic factors, such as age, gender and education level have been identi ed associating with HBPM behavior of hypertensive patients in multiple studies [20][21][22][23][24] . From the patient's experiences, the main reasons for not measuring blood pressure at home include failing to recognize the bene ts of HBPM 25 , lacking skills to operate the device 25 , having di culty adhering to time limits for HBPM 26 , and holding the belief that their blood pressure measured by themselves is not as accurate as the doctor's 27,28 , which hinder patients to perform HBPM regularly. However, studies on the promotion and adherence of HBPM in CKD patients are not as rich as that in patients with hypertension. Therefore, it is very necessary to explore the psychodynamic mechanism and action process of HBPM in patients with CKD, so as to improve the adherence of HBPM for these patients. Moreover, studies have investigated patients' psychodynamic predictors to monitor home blood pressure mainly through single item questions 25,26,29 , or interview patients for information 18,27,28,30 . The speci c instruments to assess the psychodynamic predictors of patients' HBPM behavior is still lacking until now.
The health promotion model (HPM), as a comprehensive theoretical model, be able to identify complex biological, psychological and social processes associated with behavior change, and it can be used to explain various health promotion behaviors 31,32 . The model is composed of three groups of ten categories of factors that in uence health behaviors, which includes personal characteristics and experiences (i.e., pre-related behaviors, personal factors), behavior-speci c cognitions and affect (i.e., perceived bene ts of action, perceived barriers of action, perceived self-e cacy, situational in uences, interpersonal in uences and activity-related affect), and desirable health promotion behavior (i.e., commitment to a plan of action and immediate competing demands and preferences) 32 . The health promotion model had been widely used to predict the establishment of behavioral predictors such as hearing protection 33 , regular breakfast eating 34 , and physical activity in adolescents 31 and patients with chronic disease like chronic kidney disease, diabetes and stroke 29,31,35 . And in previous studies, the health promotion model had demonstrated high levels of predictive power. Wu and Pender 36 studied the determinants of physical activity based on health promotion models, and the results showed that variables accounted for 30% of the variance in physical activity, and the perceived self-e cacy was the strongest predictor of physical activity among adolescents in Taiwan.
Since monitoring home blood pressure is a health promotion behavior, the intervention program based on the theory of health promotion behavior will be more targeted and systematic. This study assumes that the scale based on the health promotion model can effectively evaluate and predict HBPM behavior. The purpose of this study was to develop the home blood pressure monitoring perception scale and conduct reliability and validity tests so as to provide a reliable instrument for medical staff to assess the psychodynamic predictors of HBPM.

Participants
The convenience sample was selected from the department of nephrology of two tertiary hospitals in Guangdong province and  Of the 436 patients surveyed in this study, 89.91% (392) of patients had a blood pressure measuring device at home. Among these patients, electronic sphygmomanometer was the main type of blood pressure measuring device (97.45%, 382). Blood pressure was mainly measured in the upper arm (92.35%, 362). 65.83% (287) of patients had their blood pressure measured at least once a week.
The mean o ce systolic blood pressure was 147.50 ± 21.19 mmHg, and the mean o ce diastolic blood pressure was 87.30 ± 14.22mmHg. There were 358 (82.1%) patients using at least one type of antihypertensive medication, and more than half of the patients (56.9%) took a combination of two or more antihypertensive drugs to control blood pressure. Speci c participant characteristics were reported in the previous article 37 . Instruments The questionnaire consisted of three parts, which were the home blood pressure monitoring perception scale, the Self-E cacy for Managing Chronic Disease 6-item Scale and the characteristics questionnaires.

Home Blood Pressure Monitoring Perception Scale
This scale was developed through item pool formation, expert consultation and item selection. Based on the health promotion model, we generated the item pool for the HBPM perception scale after reviewing a large pool of literature and conducting research group discussions, which consisted of seven aspects and 58 items. The HBPM perception scale is a self-report scale.
A panel of eight experts was employed to conduct two rounds of reviews of the item pool of the HBPM perception scale. Experts had various academic backgrounds, including nephrology, cardiology, nursing, and education. They aged 24 ~ 54 (37.3 ± 9.1) years old and had clinical or academic experiences for 13.0 ± 11.2 years. There were two males and six females; three had doctoral degrees, one had master's degree; three of them hold associate senior titles and three hold intermediate titles. Experts were asked to evaluate the importance, relevance, and linguistic expression of each item, and were encouraged to add necessary items to the item pool. The importance and relevance were evaluated with a Likert 5-rating scale and a Likert 4-rating scale, respectively. The active coe cient of experts in the two rounds was 100%. The expert authority coe cient was 0.84, and the coordination coe cients of the two rounds of experts were 0.206 and 0.746, respectively. The full-score ratio ≥ 50%, with the average score of importance assignment ≥ 4.00, itemlevel content validity index (I-CVI) ≥ 0.78 was considered as the item selection standard [10] . After expert consultation and discussion by the research group, the draft of the HBPM perception scale was formed, including 43 items. Likert 5 rating was used in the scale, including "strongly disagree = 1, disagree = 2, neutrality = 3, agree = 4, strongly agree = 5".

Self-E cacy for Managing Chronic Disease 6-item Scale
This scale has been used in chronic diseases to measure patients' self-e cacy 38 . The scale consists of 6 items with each item is scored on a scale of 1 to 10. The average score of the items represents the level of self-e cacy, and the higher the score, the better the self-e cacy. In this study, the scale was used to as a criterion, and criterion-related validity was calculated by analyzing its correlation with this developed scale.

Characteristics Questionnaire
The questionnaire was composed of demographic and sociological information of patients (gender, age, level of education), diseaserelated information such as o ce blood pressure and the use of antihypertensive drugs, etc.

Data Collection
Participants were asked to complete the questionnaires on the spot. The investigator explained the purpose, content and ethical issues of the study to patients. Patients participating in the study all signed informed consent forms. The patients completed the questionnaire by themselves. A total of 436 questionnaires were issued, and 436 were effectively collected, with an effective collected rate of 100%. Thus, 436 valid respondents were included in the data analyses. Twenty participants completed the questionnaires again two weeks after the rst survey.

Data Analysis
Data analyses were performed on the basis of three samples. First, the data collected from 436 participants were randomly split into two halves. The rst half of the participants (sample 1) was used for item analysis and selection and exploratory factor analysis (EFA).
Item frequency analysis 39 , coe cient of variation (C.V.), critical ratio (C.R.), item-total correlation, corrected item-total correlation (CITC), internal consistency reliability coe cient, commonalities value and factor loading 40 were adopted to analyze and screen items, items that meet three or more standards were considered to be retained 41 . The standards for the above coe cients are shown in table 1.
Sampling suitable for factor analysis was evaluated with the Kaiser-Meyer-Olkin (KMO) measure and Bartlett's Test of Sphericity.
Principal component analysis (PCA) and varimax rotation method were performed to decide the optimal number of components, and the number of extraction factors was determined by the eigenvalue (greater than 1) and scree plot. Items with factor loading values greater than 0.40 and no repeat load (the same item has a factor loading greater than 0.4 in 2 or more components) were kept. Sample 2 included the other half of the participants, was used for validity and reliability test. Con rmatory factor analysis (CFA) was adopted to test the construct validity. Content validity and criterion-related validity were also analyzed to test the scale validity 40,42,43 .
At the same time, the discriminability of the scale was veri ed in patients with different monitoring frequencies. Internal consistency reliability, split-half reliability, test-retest reliability and construct reliability were used to test the reliability of the scale 40,44 . Sample 3, which included 20 participants, was used to gauge test-retest reliability. SPSS 20.0 and Amos 23.0 were used for data analysis.

Item Analysis and Selection
There was no item with more than 80% distribution on an option. The other analysis results of the items are shown in table 1. After analysis, item 23, 24, 34, 41, and 42 failed to meet the standard of 3 or more, therefore they were considered to delete. Table 1 The analysis results of the items of the rst draft with 43 items n=218) indicating that it was suitable for factor analysis. There were 9 factors whose eigenvalue of initial load matrix was greater than 1, and the cumulative contribution rate of variance reached 72.669%.
According to the results, 5 items with factor loading less than 0.4 (item 25,27,30,35,43) were deleted. Items 22,26,29,37 with loads greater than 0.4 above two factors were deleted. Item 8,11 with loads more than 0.4 on at least two factors, but these two items played an important role in the HBPM behavior of patients during the investigation through communication with patients. Meanwhile, the study in patients with hypertension suggested that the price of the sphygmomanometer and the degree of mastery of HBPM methods are important factors affecting patients' HBPM behavior. Therefore, these two items were decided to keep after expert discussion. At the same time, item 28 has highly similar meaning with item 39 in the a liation factor, so item 28 was deleted. Finally, there were 7 factors whose eigenvalues of the structural matrix are greater than 1, and the cumulative contribution rate of variance reaches 74.373% ( Table   2). The factor loadings and structure were showed in Table 3.  The rst common factor, named perceived bene t of HBPM, included 7 items with the extracted sum of squared loadings was 5.679, and the variance was explained by 20.283%. It is used to investigate the perceived bene t of HBPM. The second common factor contained 4 items, the extracted sum of squared loadings was 3.406, which explained 12.163% of the variation. It was named perceived self-e cacy of HBPM and is used to evaluate the self-e cacy of patients for HBPM. The third common factor was named as situational in uences. It contained 3 items, the extracted sum of squared loadings was 2.605, which accounted for 9.304% of the variation. It is mainly used to investigate the situational in uences on HBPM. The fourth common factor extracted contained 4 items, with the extracted sum of squared loadings of 2.550, which explained for 9.108% of the variation. It was named as commitment to a plan of HBPM and immediate competing demands and preferences. The fth, sixth, seventh common factor extracted contained 4, 4, and 2 items, with a load of 2.541, 2.442, 1.600, which explained 9.076%, 8.723%, and 5.715% of the variation, respectively. Factors ve, six and seven were related to barriers of HBPM behavior. Therefore, these 10 items were combined as one factor after expert consultation.

Con rmatory Factor Analysis
The construct validity of the scale was further veri ed by CFA. The tting indexes of the model were as follows: χ 2 = 963.793,df = 340,  Table 4. The I-CVI of 27 items in this study were 1.00, S-CVI /UA was 1.00, and S-CVI /AVE was 1.00. The correlation coe cient between the score of HBPM perception scale and Self-E cacy for Managing Chronic Disease 6-item Scale was 0.256 (p<0.001). The square root of average variance extracted of dimensions between 0.571 and 0.858. The correlation coe cient of each dimension and the total score were shown in Table 5. Those who had their blood pressure measured at home at least once a week had higher scores on the perception scale than those who didn't (Table 6).  Table 6 The difference of perception scale scores in patients with different frequency of HBPM

Discussion
Home blood pressure monitoring has been shown to improve the accuracy of hypertension burden assessment in patients with CKD, improve the blood pressure management of these patients 18, 19 , and provide reliable data on the relationship between hypertension and prognosis 17 , which has great clinical value in patients with CKD 8,14−16 . However, the cognitive and other psychological factors of HBPM are not well studied and there is a lack of speci c instruments to explore the psychodynamic factors of home blood pressure monitoring behavior. In this study, we have developed a new valid and reliable scale to measure the perception of home blood pressure monitoring based on the health promotion model. The HBPM perception scale consists of 27 items covering ve dimensions. It is evaluated on a Likert 5-point scale ranging from 1 (strongly disagree) to 5 (strongly agree), resulting in a nal score in the range of 27 ~

135.
Out of the 58 items in the item pool, 27 were retained in the nalized version. The justi cations for reducing the item were as follows.
First, there were 43 items retained after evaluated by experts for the relevance, importance and linguistic expression of the item.
Second, based on ndings from the item analysis, there were ve items with 5 or more substandard indicators, which indicated that the sensitive, differentiated and representative of these items were poor, so these items were deleted. Third, based on ndings from the EFA, ten items with factor loading less than 0.40 or had a factor loading greater than 0.4 in 2 or more components were omitted. Finally, the model tting index suggested and research group discussion agreed that the semantic of item 13 overlaps with the perceived selfe cacy of HBPM dimension, so deleted this item. Based on the ndings and the health promotion model, ve domains were nally identi ed.
HBPM perception scale is a reliable and valid instrument to assess the perception of HBPM. Firstly, the data analysis results demonstrated it had reasonable reliability 40,44 . The Cronbach's α coe cient for total scale was 0.911 and ranged from 0.750 to 0.952.
The split-half reliability was 0.896 for total scale and 0.624 ~ 0.963 for ve dimensions. And the test-retest reliability for total scale was 0.908 ranged 0.829 ~ 0.909. The construct reliability of dimensions ranged from 0.749 to 0.951. Secondly, the results showed that the HBPM perception scale had reasonable validity 40,42,43 . The HBPM perception scale total score was positively correlated with the Self-E cacy for Managing Chronic Disease 6-item Scale, suggesting that criterion-related validity was reasonable. Con rmatory factor analysis showed the model t well: χ 2 = 679.649, df = 310, χ 2 /df = 2.192, RMSEA = 0.074, CFI = 0.902. The I-CVI of 27 items in this study was 1.00, the scale-level content validity index (universal agreement) was 1.00, and the scale-level content validity index(average) was 1.00. The square root of average variance extracted of dimensions between 0.571 and 0.858. For patients who measured their blood pressure at home at least once a week or not, the scale could detect differences in scores, indicating that the scale can predict HBPM behavior well.
According to the analysis results, ve domains were nally identi ed, and this factor structure derived from the health promotion model. The health promotion model puts forward ten categories of factors that in uence health promotion behaviors. Our study demonstrates that the following ve factors were predictors of HBPM behavior among CKD patients.
The dimension of perceived bene ts of HBPM consists of 7 items. The perceived bene ts of a certain behavior can directly promote an individual to adopt this behavior according to the health promotion model 34 . Data from 436 patients demonstrated this structure. And some results obtained in previous studies were retained. For example, the item "Home blood pressure is a good indicator of how my blood pressure uctuates" was developed based on a study in Singapore 25 . These items are representative and can be used to assess the perceived bene t of HBPM behavior.
The perceived barriers of HBPM dimension contains 9 items. Individuals often avoid behaviors that they feel are more di cult to implement 34 . In terms of HBPM behavior, patients are more likely to avoid it because it involves barriers such as measurement equipment 25 , time 26 , and method 45 . Our study found that these barriers, which have been demonstrated in patients with hypertension, are also present in patients with CKD. For instance, items "I don't have time to monitor my blood pressure at home", "I don't know how to measure blood pressure" remained after data analysis. Attention should be paid in clinical practice to identify and help patients overcome the barriers of HBPM behavior.
The dimension of perceived self-e cacy of HBPM includes 4 items. Individuals with high self-e cacy are more likely to maintain healthy behaviors and engage in them frequently 36 . Two qualitative studies embedded within the randomized controlled trials in the U.K. have shown that patients undergoing HBPM are con dent in self-monitoring blood pressure 27,28 . However, for HBPM, regularly, long-term, and accurately measure blood pressure is needed if its clinical value is to be brought into play. Therefore, this study developed items based on these three aspects, and data analysis also proved that perceived self-e cacy of HBPM plays an important role in HBPM behavior in patients with CKD. On the basis of previous studies and combined with the characteristics of HBPM, our study further re ned the perceived self-e cacy of HBPM, so as to comprehensively re ect the self-e cacy of HBPM perceived by patients, and provide reference for clinical practice.
The situational in uences dimension consists of 3 items. Individuals often want their behavior to be consistent with the behavior of others in a given situation 46 . One study showed that putting up posters about HBPM in health care facilities increased the use of HBPM for hypertension patients 47 . In this study, the items were developed mainly based on this result, and analysis of 436 cases of data also con rmed that situational in uences are also important for HBPM behavior in patients with chronic kidney disease. The use of HBPM needs to be encouraged by setting up situations, such as health talks and more publicity.
The dimension of commitment to a plan of HBPM and immediate competing demands and preferences contains 4 items. A commitment to a plan of action is one's own plan to perform a certain behavior, and immediate competing demands and preferences are the various contingencies of a healthy behavior plan 46 . Making a plan is an important step before start acting. Individuals may have immediate competing demands and preferences before performing a planned health behavior, and the attraction or stress brought by such emergencies can affect individuals' health behavior 34 . Our study con rmed that this factor could affect the HBPM behavior in patients with CKD. It is necessary to help patients develop HBPM plan and adjust the plan exibly for patients with special conditions. HBPM perception scale is a reliable and sensitive quantitative instrument, which adopts the health promotion model as the cause pattern that can assist medical personnel in evaluating the determinants of HBPM behavior from the patient's perspective. For the patients who reluctant to perform HBPM according to recommendation, the scale can help clinical medical staff to nd out the potential problems that patients will encounter during HBPM, so as to take preventive and targeted measures to help patients overcome the problems. In the case of patients with irregular monitoring, it can help to nd the reasons for non-standard monitoring and then develop personalized strategies, so as to improve the clinical work e ciency and reduce the workload of clinical medical staff. It can also be used in clinical studies to help to identify intervention targets and establish intervention strategies to improve the monitoring rate of patients' home blood pressure, and effectively predict the HBPM behavior and health outcome of HBPM intervention.
There are some limitations of this study. Although the patients with stage 1-5 CKD were investigated, the subjects were mainly hemodialysis patients. Future studies need to further explore the validity and reliability of the HBPM perception scale in non-dialysis patients and peritoneal dialysis patients. Meanwhile, the sample of this study was only recruited in two tertiary hospitals from Guangdong, China. This might limit the generalizability of the ndings. Therefore, future studies need to assess the psychometrics features of the HBPM perception scale in different samples with different cultural and socio-economic status.

Conclusions
The HBPM perception scale provides a valid and reliable instrument for assessing the perception of HBPM. Besides, the items are straightforward and easy for users to understand and use. In addition to enriching our understanding of home blood pressure monitoring among patients, the HBPM perception scale is also useful in establishing a possible link between assessment and health education and helpful in developing appropriate training programs and evaluating the effectiveness of HBPM intervention program to improve blood pressure management in these patients and nally reduce the risk of cardiovascular events. This study was performed in accordance with the Declaration of Helsinki and was approved by the ethics and academic committee of Sun Yat-sen University.

Consent for publication
Not applicable.

Availability of data
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. 9.I don't know how to measure blood pressure.
10.The requirement of blood pressure measurement is too high for me (such as getting enough rest, measure posture, etc.).
12.I don't have time to monitor my blood pressure at home. 13.I think the blood pressure I measured at home was inaccuracy.
14.I don't know how to explain the blood pressure.
15.HBPM is too frequently measured, I feel it troublesome.
16..I can't get my blood pressure measured by myself without the help of others.
Perceived self-e cacy of HBPM 17.I am con dent that I can monitor my blood pressure at home in the right way.
18.I am con dent that I can monitor my blood pressure at home regularly every day. 19.I am con dent that I will be able to monitor my blood pressure at home for the long term. 20.No matter what the situation, I can carefully monitor my blood pressure according to the HBPM plan.
Situational in uences 21.The media (Internet, books, magazines, T.V.) made me want to monitor my blood pressure at home. 22.The publicity from hospitals and community health services made me want to monitor my blood pressure at home. 23.The fellow patient's condition was controlled after monitoring blood pressure at home made me want to monitor my blood pressure at home.
Commitment to a plan of HBPM and immediate competing demands and preferences 24. In order to monitor blood pressure at home, I ask the medical staff for advice on how to measure blood pressure. 25. I have made speci c arrangements for the time and recording method of HBPM.
26. I plan to have family or friends (or fellow patients or colleagues) help remind me to monitor my blood pressure at home. 27.I still monitor my blood pressure regularly when I'm away. Figure 1 Con rmatory Factor Analysis Model of HBPM perception scale. Abbreviations: HBPM, home blood pressure monitoring.