Barriers and Determinants to the Underutilized Hypertension Screening in Primary Care Patients in Hong Kong: A Mixed-Method Study

(1) Background: Hypertension (HT) is the most common chronic condition, affecting approximately 1.13 billion people worldwide. Despite freely available blood pressure (BP) devices in primary care (PC) clinics, many patients do not regularly screen for HT and are untreated. (2) Methods: This study investigated the proportion of PC patients who did not screen for HT and the underlying reasons in Hong Kong. An explanatory mixed-method cross-sectional study was conducted in 2020, which included a questionnaire survey, office BP measurements, and subsequent semi-structured interviews. Adult patients who had no diagnosis of HT were recruited in a large PC clinic by convenience sampling. The relationships between not having HT screening and sociodemographic data were investigated by logistic regression. Twenty-four patients were purposefully sampled (based on demographics) and were interviewed until data saturation. (3) Results: Among 428 participants, 190 (44.4%) had not had HT screening in the last two years, but 197 (46.0%) had HT. No HT screening in the last two years or ever was associated with being male, being single, being of younger age, having no family history of HT, having no clinic visits in the last two years, employment status, and self-perceived HT condition. Most participants (77.8%) misinterpreted their BP readings. Individual, social, and healthcare service barriers were identified in patients’ interviews. Many PC patients had no regular HT screening but around half had elevated BP. (4) Conclusion: The study results indicate that the barriers to HT screening were multifactorial. HT screening in PC is urgently needed.


Introduction
Hypertension (HT) is the most common chronic condition, affecting around one-third of the adult population worldwide [1]. Approximately 1.13 billion people are suffering from HT worldwide [2]. HT is the major cause of cardiovascular mortality, accounting for 9.4 million deaths annually [2,3]. Due to the asymptomatic nature of HT, international guidelines and the Hong Kong (HK) guideline recommend regular HT screening, which is cost-effective for preventing cardiovascular complications through early and effective treatments [4,5].
Despite these recommendations, regular screening of HT is often not conducted. Around 30-50% of patients with HT are unaware of the condition and are, therefore, untreated [1,6]. A systematic review identified patients' barriers to HT awareness, which included: knowledge, skills, motivation, social influences on HT screening, availability of screening, and other social determinants [7]. These barriers were identified in both developing and developed countries, the elderly, and multi-ethnic populations [8]. Around half of the people with elevated blood pressure (BP) in HK were unaware of their condition in a population-based survey [4]. 2 of 13 Although the prevalence of people who did not screen for HT in the general population and their corresponding barriers were well described, similar research was yet conducted in primary care (PC). Screening in PC may be more cost-effective by targeting high-risk patients because HT is associated with numerous other diseases [8,9]. Although an official screening program for HT is lacking in HK, regular HT screening is promoted in doctors' offices and hospitals. BP devices are provided freely in social centers and publicly funded general outpatient clinics (GOPCs). Once diagnosed, HT treatment is readily available and affordable in GOPCs. In contrast to the general population, we hypothesize most PC patients received HT screening due to convenience and free service in GOPCs, and the reasons for not having HT screening would be different. This is a sequential explanatory mixed-method study. For primary aim, the proportions of patients who have not been screened for HT in the last two years or ever were estimated. For secondary aims: (i) patients' demographic factors associated with no HT screening were identified. (ii) Barriers to HT screening were identified by qualitative interviews. Patients who had not had HT screening in the past two years, as identified from the survey, were purposefully sampled based on their demographics, which is expected to further explain findings from (i).
The results would facilitate researchers, clinicians, and policymakers to promote HT screening in primary care, which may thereby reduce cardiovascular events and deaths.

Materials and Methods
This study was approved by the Joint CUHK-NTEC Clinical Research Ethics Committee (CREC Ref. No.: 2021.076). Written consent was obtained from all participants.

Participants
Participants (aged ≥ 18) were recruited from a large GOPC (Lek Yuen clinic) in HK from April to June 2020. Participants who fulfilled any of the following criteria were excluded: (i) diagnosed with HT, diabetes mellitus, stroke, heart diseases, or chronic kidney diseases because BP measurements were compulsory for these patients, (ii) unable to communicate in Cantonese, Mandarin, or English, (iii) had a physical or mental illness that prevented the BP measurements, or (iv) unable to consent.

Instruments
As there is no well-validated questionnaire to detect the proportion of patients with no HT screening, a questionnaire was designed by a team of academics including 2 family medicine specialists and academics and 1 public health academic after reviewing the existing literature. The questionnaire was designed to collect (i) demographic data (e.g., age, sex, education level, presence of a regular family doctor), (ii) whether the patient had ever conducted HT screening, and in the last two years, and (iii) whether they interpreted their BP correctly (see BP measurements). The questionnaire was piloted in 20 primary care patients prior to data collection to ensure relevancy and clarity. The questionnaire can be found in (Supplementary File S1).

BP measurements and Body Mass Index
These are collected because no HT screening may associate with elevated BP because of under-diagnosis. Furthermore, obesity is a known risk factor for HT and may prompt regular HT screening.
After the questionnaire survey, BP was measured by a trained master student using an office BP monitor TM-2657P (A&D Company, Limited, Tokyo, Japan), which was validated according to the British and Irish Hypertension Society standards. BP was first measured on each arm. Two further BP readings were obtained for analysis and averaged using the arm with higher BP. Elevated office BP was defined as systolic BP of ≥140 mmHg or diastolic BP ≥ 90 mmHg [10]. Body weight and height were measured using the validated MUW 300 L ultrasonic Health and fitness scale (Adam Equipment Company, Milton, UK).

Sample Size Calculation for Questionnaire Survey
Using a margin of error of 5%, a confidence level of 95%, and the presumed proportion of patients who had not received screening for HT to be 50% (which required the largest sample size), 384 participants were needed. To allow a dropout rate of 10%, at least 426 participants were required.
z is the z score. is the margin of error. n is the population size. p is the population proportion.

Qualitative Interview
The inclusion criteria of the qualitative interview were patients (1) who had not screened for HT in the previous two years; and (2) who had been found to have elevated BP during the BP measurements as part of the quantitative survey. Patients meeting the inclusion criteria were further sampled based on age, gender, marital status, employment status, clinical visit in the past two years, and self-perceived HT condition.
A semi-structured interview guide was developed based on the questionnaire results and literature review (Supplementary File S2). Interviews were conducted by the trained master student via telephone due to the COVID-19 pandemic. Each interview lasted from 15 to 30 min. Preliminary qualitative data analysis was conducted immediately after each interview. Interviews were continued until data saturation (no new information can be further identified) [11].

Statistical Analysis
Quantitative data was performed using the Statistical Package for Social Science version 25.0 (SPSS, IBM Corp., Armonk, NY, USA). Demographic data were presented as mean with standard deviation and numbers with percentages for continuous and categorical data, respectively. Univariate analysis and Mann-Whitney U test were performed to examine the association between no HT screening and sociodemographic data. Significant factors from the univariate analysis were further analyzed by simple logistic regression. Statistical significance was defined as p-value < 0.05. Interviews were audio-recorded and transcribed verbatim. Two investigators (RYKC, DD) repeatedly read and familiarized themselves with the transcripts. Thematic analysis was conducted, in which concepts were coded, and grouped into patterns/themes using excel (Microsoft 365 Excel Version 2021) [12]. Quantitative and qualitative results were compared and presented in a side-by-side table.

Participant Characteristics
Among 918 patients who were approached, 432 were excluded, mostly due to diagnosed HT. (428 were known HT patients, and 4 were aged under 18). A total of 428 participants were included ( Figure 1).

Factors Associated with no HT Screening in the Previous Two Years or Ever
Similar factors were associated with no HT screening in the past two years and never having had HT screening. In univariate analysis, these factors included gender, marital status, employment status, family history of HT, having a clinic visit in the last two years, and age. Regarding their BP values as normal was associated with no BP measurements in the last two years only. In logistic regression, no HT screening in the last two years or ever was associated with being male, being single, being of younger age, having no family history of HT, and having clinic visits in the last two years. Patients who were not able to interpret their BP values had higher odds of not having had HT screening in the last two years ( Table 2).

Qualitative Results
Patients who participated in the survey and found themselves to have HT only through this study were interviewed via telephone. They were sampled based on their age, gender, marital status, employment status, and clinical visit in two years, and they considered their BP as normal. Interviews continued until data saturation was reached (Supplementary Table S1 shows details of interviewees). Eventually 24 interviews were completed. Since the purpose of the follow-up qualitative interviews was to identify the barriers for these patients not to have BP measurement, three overarching themes concerning their barriers were identified: (1) Individual barriers; (2) Social barriers, and (3) Healthcare service barriers (Table 3).    Barriers to individual patients not having HT screening vary; yet they can broadly be summarized at three levels: individual, social, and institutional. The barriers at the individual level can be further divided into two sub-themes: (1) insufficient knowledge about HT and (2) feeling suspicious, careless, or even in denial of HT.

Individual Barriers: Insufficient Knowledge about HT
We were surprised to find out that, regardless of their age or gender, many patients had many misconceptions about HT. For example, the majority of participants expressed their HT knowledge insufficiency. When they were asked about the BP index classification, most of them did not know or failed to answer correctly. A number of participants reported that a BP over 140 is normal (participant 5, 7, 20, and 22). Some explained they had experienced difficulties in reading the index and were unable to interpret it which become a major barrier in measuring BP.
In addition to BP index interpretation, the knowledge of HT's nature is unclear regardless of age. Some patients thought that elevated BP was just a temporary thing, and it would not cause any harm to their health. An 18-year-old and a 50-year-old participant both believed HT is temporary and that BP will return to normal shortly (participant 16 and 17). Over half of the participants perceived they were in good health and that they never expected to have HT. A middle-aged female participant further explained that it is unexpected to have HT as she is still young, and her body is in a good shape (participant 4). Such misconceptions might also happen to those who perceive themselves as having quite high health literacy. A 62-year-old man who claimed to read medical books stated that HT and diabetes could be felt by oneself (Participant 7), but, apparently, his knowledge about HT was also off base. With the asymptomatic nature of HT, it could be difficult for patients to realize that they were sick by themselves. In comparison, patients who admitted having low health literacy also felt relaxed about being found to have HT, but the main reason was that they did not know the meanings of the BP numbers (participant 3, 4, and 10).

Individual Barriers: Feeling Suspicious, Careless, or Even Denial
At the attitudinal level, men or women, most patients were optimistic about their current and future health. Two interviewees commented that they were healthy without any health problems and living normally. They further emphasized that measuring BP is unnecessary for them as they will never get HT (Participant 4, 6, and 22). After our BP measurements, some interviewees denied having HT and claimed that "medical consultation is meaningless" (Participant 19). A 63-year-old female later explained that the reason why she refused to measure BP was that she understood her lifestyle was unhealthy and was reluctant to face that reality (Participant 12).
The denial attitude was further demonstrated through their behaviors. Some patients showed distrust towards office BP devices and believed that measuring BP was a waste of time; some suspected the BP machines were broken, especially when they were found to have a high BP (Participant 24). One even accused the measurements of being high and inaccurate for everyone (Participant 19).

Social Barriers
At the social level, work constraints and insufficient social support were the most salient social barriers that emerged from the interviews. The working patients could not optimally take care of their health due to insufficient time and occupation by their work schedule. This phenomenon appeared in employed young to middle adults. A 24-year-old employed young adult reported that the packed work schedule disrupted his daily health routine. He further supplemented that he had insufficient time to sleep, and, thus, it was barely possible for him to have time to measure his BP (Participant 2). Participant 2 also mentioned that his working environment prohibited him from adopting a healthy lifestyle including doing physical activities and choosing healthy food. Another working adult (Participant 19) complained that he had no spare time to consider healthy food choices as he was fully occupied by work. A similar situation was experienced by Participant 4 as she expressed that there were limited healthy food choices around her workplace.
Lacking social resources and support could preclude HT screening. In particular, the elderly lacked medical knowledge, and skills, thereby requiring extra support for HT screening. They had difficulties in interpreting BP values and required assistance with other healthcare needs. An 80-year-old lady expressed her concern about measuring BP because she was not able to take care of her health when her children were out for work, plus there was no extra social support provided by the government to assist her after she left the clinic (Participant 3). Another 83-year-old man commented that district counsellors would measure BP for them in the past, but there had not been such services for a long time (Participant 20).

Healthcare Service Barriers
At the institutional level, despite free and available BP devices in GOPCs, patients often perceived a shortage and reduced accessibility of BP machines in the communities. A participant reported the decreased number of BP devices in the GOPC and difficulty finding BP devices in his own community (Participant 22). With the limited availability of BP devices, it became more difficult for people to monitor or screen their HT. In addition, some interviewees who lived in poverty disclosed their hardship to afford a BP device at home. The decreased number of public BP devices indirectly discouraged their BP measurement (Participant 14).

Mixed-Method Analysis Results
To juxtapose and compare the findings from quantitative surveys and qualitative interviews, further insights on the reasons for patients not having HT screening emerge. Simply put, qualitative results echo and provide further evidence to support our quantitative findings. The comparison of quantitative and qualitative findings is listed in Table 4. In our quantitative findings, numbers of factors were found to have an association with no HT screening in the past two years: male, single, younger age, unemployed, without a family history of HT, and without clinic visits in the last two years. Aligning these findings with the qualitative results, our further explanation on such associations are: (1) the younger age was associated with no HT screening as they regard HT will only affect older people; (2) being single means that no family or spouse would have experienced BP-related diseases, paid more attention to their BP, or helped them maintain their health; (3) for individuals without a family history, they would consider themselves as low-risk population; and (4) absence of clinic visit in 2 years, and in combination with the insufficient BP devices provided in the community, both demotivated individuals in BP measurement. The qualitative results, however, cannot further explain the gender differences.
In addition, from the survey, some patients were found to be unable to correctly interpret their BP values, which can be further explained by poor HT knowledge among the patients who might also consider HT screening as irrelevant, as the qualitative results show. Subsequently, these patients had a wrong belief in their BP conditions and, therefore, avoided HT screening.

Summary
This was the first mixed-method study in PC exploring the prevalence and reasons for no HT screening. Around 44% of PC patients had not received HT screening in the previous two years despite free screening. Population-based studies indicated around onethird of the Chinese population were unaware of their HT [6]. Aligning with a systematic review, knowledge, attitude, and social factors played important roles in not receiving HT screening; however, this was the first study to elicit prevalence and barriers specific to PC patients [13,14]. Our study suggests that regular universal HT screening in PC would identify a large number of undiagnosed patients with HT because approximately one-third of our participants had elevated BP whilst 80% were unable to interpret their BP values.

Implication for Practice
Our results suggest that despite free HT screening in the PC setting, many patients did not receive HT screening due to various reasons. At individual levels, patients should be educated about the details of screening (including the threshold of HT diagnosis, and proper BP measurement techniques), complications, prognosis, and treatments of HT. It is commonly found that there is a lack of medical knowledge among the elderly, and extra social support is encouraged [15]. Our result shows that those of younger age perceived themselves as a low-risk population whereas further public health education should be provided to clear the misconception. A denial attitude is typical when patients are first told that they have some unexpected disease [16]; a deeper medical knowledge may assist patients with disease acceptance. At clinical practice levels, if universal screening for all patients during every clinical visit is infeasible, computerized systems can be designed to alert healthcare professionals when the patient has had no screening in the previous two years. The threshold for HT diagnosis should be posted in the clinics. Free BP measurements can be made available in clinics and in the community to facilitate regular screening [17]. Finally, governments can set up universal screening programs and reimburse clinicians to make definite HT diagnoses by out-of-office BP methods, which have superior reproducibility and predictability to cardiovascular events than office BP measurements. Out-of-office BP is recommended for HT diagnosis [18,19]. Our results provided important patient characteristics and their reasons for not receiving HT screening which can guide HT screening strategies.

Implications for Research
Our results showed that implementation studies integrating evidence-based HT screening, which is shown to reduce hospitalizations and deaths, into PC are urgently needed [20]. This is hardly unique to HT screening because it took 17 years on average for evidencebased interventions to be successfully implemented into daily practice [21]. Our results suggest that employed participants had difficulties receiving HT screening, and it is currently unclear whether implementing BP devices in the workplace is feasible, accurate, or can improve HT detection. Although clinic BP was recommended to be the yearly screening tool for HT in all adults, it only has fair accuracy to diagnose HT (sensitivity of 0.54 [95% CI, 0.37-0.70] and specificity of 0.90 [95% CI, 0.84-0.95]) [10,22]. More studies may determine whether home BP measurement that is more sensitive, specific to diagnose HT than office BP, and well-accepted by HT patients can be a valid and accepted regular screening tool for HT in PC patients [23,24].

Strength and Limitations
The current mixed-method study was one of the first to explore the proportion and the patients' reasons for not receiving HT screening. The sample size was pre-determined and adequate. The qualitative interview methodology was guided by, subsequently explained, and provided a deep understanding of the quantitative results.
A few limitations should be discussed. Although recruitment was carried out in a single large publicly funded center, the applicability of our results to other clinics, especially private clinics, was unknown. In HK, relatively wealthy and educated patients may prefer private general practitioners. However, >80% of patients with chronic diseases received care from public PC clinics [19]. As a cross-sectional study, causal relationships cannot be established. Although we found that patients without HT screening did not have an increased risk of elevated BP, patients at risk of having elevated BP may measure more frequently and were excluded from our study once they were diagnosed with HT. BP was measured only in one clinic visit. To definitely diagnose HT, BP measurements over a few clinic visits or using out-of-office BP measurements are recommended, but these were infeasible for this entirely self-funded study [10].

Conclusions
Around 44% of PC patients had no regular HT screening and many unscreened patients had elevated BP. Barriers to HT screening were multifactorial and consisted of individual, social, and healthcare factors. Gender, marital status, employment status, family history, clinic visits in two years, and age were associated with no BP screening from the quantitative result, whereas qualitative results supported the identified three barriers that hindered individuals in measuring BP. The individual barrier included knowledge and attitude towards HT; the social barrier included work constraints and lack of care; and the health service barrier reflected the insufficiency of resources in the community. PC doctors should encourage regular screening for HT. On top of that, health education is necessary to clear the misconception of HT and deepen public knowledge of BP. Extra care is particularly needed for the elderly in BP management. Studies investigating the implementation of HT screening in PC are urgently needed.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/ijerph20020985/s1, File S1: Questionnaire; File S2: Interview guide; Table S1: Characteristics of the interviewees. Informed Consent Statement: Written informed consent has been obtained from the patient(s) to publish this paper.

Data Availability Statement:
The data used and analyzed during the current study are available from the corresponding author upon reasonable request.