Accurate and reliable estimates of hypertension prevalence and management are essential for monitoring CVD risks and informing community-based public health decision-making, particularly in aging societies [8]. This study provides critical insights into the agreement and its associated factors between self-reported and objectively measured data, not only on the diagnosis but also on the control of hypertension among middle-aged and older adults in China, offering a nationwide representative perspective. Our findings revealed moderate agreement on hypertension diagnosis but only low concordance on hypertension control when comparing self-reported data with biometric measurements. This discrepancy, on the one hand, highlights a significant underestimation of hypertension prevalence through self-reporting among middle-aged and older Chinese adults, consistent with prior studies indicating an awareness gap between self-reports and actual prevalence [20, 41]. Despite the excellent specificity in self-reported hypertension diagnosis, which mirrors similar situations in both developed and developing countries [37, 38, 42,43,44], the issue of unawareness or under-diagnosis of hypertension among the older population persists as a critical public health concern. In our study, approximately 30% of middle-aged and older adults were unaware of their hypertensive condition, primarily due to the limited sensitivity of self-reporting. Previous studies have indicated that poor sensitivity of self-reporting often results from inadequate BP screening, which may be linked to insufficient access to or poor quality of healthcare services [17, 22, 43]. Consequently, asymptomatic hypertension in its early phase may remain undetected. This finding demonstrates the urgent need for governments to strengthen public health strategies for improving hypertension detection and screening. It also emphasizes the importance of encouraging older adults to adopt self-management practices for NCDs, such as regularly monitoring of BP and blood glucose levels at home, actively participating in health check-ups or recommended screenings, and fostering healthy behaviours and lifestyles.

On the other hand, the present study demonstrated a significant overestimation of hypertension control through self-reporting. Our results showed a sensitivity of 20.84% (95%CI: 18.05%, 23.71%), indicating that only one in five middle-aged and older hypertensive patients with uncontrolled hypertension in China accurately self-reported their condition, while the remaining 80% mistakenly believed their BP was under control. This low sensitivity, substantially lower than the approximately 80% reported in developed countries such as Canada [29], suggests an important issue with FN self-reporting, where the majority of patients with uncontrolled hypertension are unaware of their condition. A possible contributing factor to this discrepancy could be patients’ limited health literacy, which is often associated with lower educational attainment [45]. This was evident in our sample, where over a quarter of among middle-aged and older respondents were illiterate, and the majority had only completed primary education. This lack of health understanding may lead to misconceptions about hypertension control, such as confusing “controlled” disease with “being without apparent symptoms”, relying on subjective feeling rather than self-monitoring to report chronic conditions, or unknowing of normal BP levels [16]. As a result, hypertension, often asymptomatic, might be perceived as controlled until severe symptoms like headaches or dizziness occur.

Identifying factors associated with agreement on hypertension is crucial for its accurate diagnosis and effective management [41]. Regarding individual-level demographic factors, we initially observed an increasing disparity between self-reported and clinically measured hypertension with advancing age. Older adults, particularly the old-old, were significantly less likely to report accurate diagnoses or control of hypertension compared to middle-aged individuals. The discrepancy between self-reported data and biomedical measurements was mainly attributed to older adults being more prone to FN reports for both hypertension diagnosis and control, suggesting a tendency for unawareness of their hypertensive status. This finding is in line with previous studies indicating that the reliability of self-reported data decreases with age across various chronic conditions [21, 38, 46, 47]. This trend highlights the importance of considering potential memory or recall biases when evaluating self-reported responses from older people [41, 48]. They may not reliably interpret their health symptoms, such as headaches, as indicators of high BP but rather as normal aspects of aging. Additionally, we found that married individuals were more likely to accurately report hypertension diagnoses than their unmarried peers, as evidenced by a significant reduction in the risk of FN reports. This outcome is likely due to better health monitoring and support from their spouses, a trend similarly observed in other developing countries [47], where marital relationships contribute to a greater awareness of one’s health status.

Investigation into the role of socioeconomic factors in the observed discrepancies revealed that, compared to those who were illiterate or from poorer households, respondents with formal education or higher family economic levels were more likely to accurately report their hypertension diagnosis, showing a significantly reduced risk in FN reporting. This aligns with previous findings indicating that individuals with higher levels of education or greater economic status are typically more cognizant of chronic conditions such as hypertension or diabetes, largely due to their enhanced health literacy, more consistent monitoring practices, and earlier access to diagnoses [22]. However, this association of socioeconomic status was not apparent with the agreement on hypertension control, indicating the need for further exploration of other potential factors. Meanwhile, although no significant disparity in the likelihood of the overall agreement on hypertension diagnosis was found between rural and urban respondents, rural residents exhibited a higher risk of FP reporting, suggesting poor validity of self-reported hypertension among this group. This finding is in line with earlier studies, which indicate that rural residents often demonstrate less accuracy in self-reporting due to limited access to continuous monitoring and professional diagnosis, and insufficient knowledge of hypertension symptoms and diagnostic criteria compared to their rural counterparts [17, 20, 40]. Despite improvements in rural primary healthcare services since China’s New Healthcare Reform in 2009 [49], the discrepancy in healthcare resource accessibility and chronic disease detection between urban and rural areas remains a significant concern.

Given the high prevalence of multimorbidity among middle-aged and older adults in China [50], this study also investigated the relationship between the presence of multiple chronic diseases and the agreement between self-reporting and biomarkers on hypertensive conditions. Our results suggest that respondents with CKD were more likely to achieve agreement on hypertension diagnosis, largely attributed to a lower risk of FN reporting compared to those without the condition. This is possibly due to the increased exposure of comorbid patients to monitoring programs, more frequent interactions with healthcare professionals, and better access to health information [19]. Conversely, having a combination of diabetes or CVD was associated with a lower likelihood of agreement on hypertension control. Specifically, hypertensive patients combined with CVD or diabetes were at a significantly higher risk of FN reporting compared to those without multimorbidity, as they were more likely to incorrectly perceive their BP as being controlled. Two factors may contribute to this discrepancy. First, symptoms of CVD, such as chest pain and dizziness, often overlap with those of hypertension, leading patients to attribute these symptoms primarily to their heart condition, thus overestimating their hypertension control. Second, managing multiple chronic conditions adds complexity to patients’ ability to accurately access and report their health status. In particular, patients with diabetes and cardiovascular conditions are more prone to BP fluctuations, which may make self-reporting of hypertension control less accurate [51]. These findings highlight the necessity for regular follow-up appointments specifically designed for middle-aged and older patients with multiple chronic diseases.

In examining the associations of behavioural risk factors and the accuracy of hypertension self-reporting, we found that heavy drinking was associated with a significantly lower likelihood of agreement in both hypertension diagnosis and control. Nearly 40% of middle-aged and older hypertensive patients who were heavy drinkers were unaware of their condition, indicating a higher risk of FN reporting in hypertension diagnosis. Furthermore, over 91% of hypertensive patients who drank heavily incorrectly reported their BP as being under control, further highlighting an elevated risk of FN reporting in hypertension management among this group. This may be partly due to the adverse effects of excessive alcohol consumption on the cardiovascular system, which leads to fluctuations in BP and complicates accurate self-reporting of hypertensive conditions [52]. Moreover, since alcohol initially lowers BP for up to 12 h after ingestion, followed by a subsequent increase [53], many individuals mistakenly believe that alcohol consumption helps managing high BP, while disregarding the long-term risks of excessive drinking. Based on our routine observations, this misconception seems particularly prevalent among middle-aged and older Chinese males. They may hold cognitive biases about the impact of alcohol on BP, thus compromising the reliability of their self-reported hypertensive condition. Meanwhile, previous research has indicated that heavy drinking is often associated with other unhealthy lifestyle choices, such as irregular dietary habits and lack of physical activity [20, 22], further disrupting BP regulation and reducing the accuracy of self-reported hypertension. This is supported by our findings that being overweight or obese was associated with a lower likelihood of agreement on hypertension control, consistent with previous research linking higher BMI with decreased health awareness and worse chronic conditions [38, 43, 47].

In terms of healthcare-seeking behaviours, we found that respondents who had recently utilized outpatient or inpatient services were more likely to accurately report both their hypertension diagnosis and control status. Specifically, recent engagement of healthcare services was related to a reduced risk of FN reporting for both hypertension diagnosis and control, suggesting that regular healthcare contact enhances patients’ ability to recognize and accurately report uncontrolled high BP. This finding is consistent with previous research, likely due to enhanced detection and monitoring of health conditions resulting from their frequent interactions with healthcare providers, which in turn improves the validity of self-reported hypertension [19, 22, 43]. This phenomenon may also explain the differences in the sensitivity of self-reported hypertension diagnosis between males and females in our study. Previous studies have shown that men are generally less likely to seek medical advice or visit healthcare facilities compared to women, who tend to have more frequent healthcare interactions due to reproductive health needs [54]. Meanwhile, under-diagnosis of chronic diseases often stems from barriers in accessing healthcare services [17]. As a result, women may have better awareness of their health status and are therefore more inclined to provide accurate self-reports of their chronic conditions.

The present study also found a significant reduction in the risk of FN reporting for hypertension diagnosis among middle-aged and older Chinese adults who received preventive care services, particularly routine physical examinations. This finding is in line with prior studies and highlights the benefits of regular health checkups in improving hypertension awareness [24, 55]. However, no association was observed between the overall agreement on hypertension control and the use of routine physical examination or regular BP monitoring, suggesting that these preventive care services may have a limited impact on hypertension management. This may be attributed to the limited quality and uneven distribution of such services in China. Although individuals may undergo regular health check-ups, they may not receive sufficient follow-up care or medical guidance for managing chronic diseases [56], which could reduce the effectiveness of preventive service utilization in enhancing the accuracy of self-reports.

Lastly, our findings indicated a reduced likelihood of concordance between self-reported and objectively measured BP control among middle-aged and older patients who reported adhering to antihypertensive medication, compared to those who did not take the medication regularly. This discrepancy was primarily driven by an increased risk of FN reporting on hypertension control. Importantly, these findings do not diminish the well-established importance of medication adherence in hypertension management, but rather suggest that patients adhering to treatment may exhibit over-confidence in the efficacy of their medication. In China, the BP control rate among hypertensive adults on medication is less than 40% [57]. Despite the availability and widespread use of low-cost generic antihypertensive medications with generally good adherence [58], the high prevalence of uncontrolled hypertension among patients who report adherence remains a significant public health challenge [59]. This discrepancy between self-reports and clinical measurements appears to be multifactorial. First, low health literacy and limited health awareness among middle-aged and older Chinese adults, particularly those with chronic conditions, contribute to inadequate self-management of hypertension [45]. Many patients lack a comprehensive understanding of proper medication use and the broader importance of lifestyle modifications [60]. Some may overestimate the effectiveness of medication, mistakenly believing that simply adhering to their prescription is sufficient for BP control, while neglecting other essential factors such as diet, physical activity, and mood management. This can contribute to biased self-assessments and inaccurate reporting of hypertension control. Additionally, the insufficiency of consistent and high-quality disease monitoring and inadequate medication guidance from community healthcare services in China further hinder effective hypertension management and the accuracy of self-reported data. Limited access to timely follow-ups, personalized treatment adjustments, and patient education exacerbates this issue [58, 59], leaving many patients without the necessary support to manage their condition effectively. As a result, the gap between perceived and actual BP control widens, contributing to both poor health outcomes and unreliable self-reporting. These findings highlight the urgent need to enhance the quality of primary public health service and promote the broader use of person-centred combination therapy [61] to ensure valid self-reporting and foster effective self-management of hypertensive and other chronic conditions.

This study offers several implications for public health researchers and policymakers. Firstly, given the observed discrepancies between self-reported and clinically measured hypertensive conditions, integrating questionnaire surveys with clinical measurements in public health research could help improve the accuracy of disease burden estimates and health technology assessments. Secondly, there is a need for the Chinese government to enhance the accessibility and quality of community-based public health services for older population. Targeting resources towards specific groups, such as the socioeconomically disadvantaged and the old-old adults, may encourage broader participation in regular health screenings and disease monitoring. Furthermore, comprehensive disease management programs focusing on patients with multiple chronic conditions are needed, including regular follow-up appointments with family doctors, encouragement of self-monitoring, and provision of person-centred medical services and medication guidance. Lastly, given the importance of cultivating healthy lifestyle and effective medication adherence in managing chronic conditions, initiatives to promote health behaviours and improve medication literacy through health promotion and medical education should be strengthened. Collectively, these strategies can inform public health policy and enhance community public health practices, potentially improving the management of hypertension and other chronic conditions in aging populations.

This study offers an important examination on the agreement between self-reported and objectively measured hypertension, addressing both diagnosis and control, among middle-aged and older Chinese adults. To the best of our knowledge, this is the first investigation to integrate large-scale, community-based questionnaire survey data with health examination data to evaluate the accuracy of self-reported hypertension control in this demographic. By extending previous studies that primarily focused on the validity of self-reported hypertension diagnosis, we offer a broader evaluation of both diagnosis and control using a nationally representative sample. Incorporating individual, household and community-level variations strengthens the relevance of our findings to the middle-aged and older population in China. Additionally, BP values in our study were measured three times for each respondent to reduce the risk of FP reporting, aligning with clinical best practices for accuracy [38].

Despite the strengths of this study, several limitations should be acknowledged. Firstly, potential misclassifications due to white-coat syndrome or undetected nocturnal hypertension [19] may have introduced bias into our estimates. Secondly, since participation in the physical examinations conducted by the CHARLS team was voluntary, the exclusion of respondents with missing biomarker data might have led to sample selection bias. Of the sample in the CHARLS dataset, approximately 23% of the interviewees had missing anthropometric measurements. While the overall response rate was relatively high (exceeding 80%), the valid response rate for BP measurements was 63%, which could potentially have caused selection bias. Specifically, women were more likely to participate in physical examinations, whereas younger men were underrepresented, possibly due to their work commitments [20, 31]. Additionally, individuals in very poor health status were also less likely to complete the physical examinations [31], potentially introducing estimation bias. However, given that CHARLS respondents are fairly evenly distributed across various demographic and socioeconomic background characteristics and that strict quality control measures were applied throughout the study [20], our findings retain a certain level of generalizability to the middle-aged and older population in China. Future studies may consider matching survey data with official health monitoring data from health departments or healthcare facilities to create a quota sample, thereby addressing potential voluntary selection biases. Thirdly, although we have controlled for factors at the individual, household, and community levels, other relevant factors, such as family history of hypertension and dietary habits, were not included due to data limitations. Future research could incorporate these factors when available to build more robust models of agreement between self-reported and biomedically measured hypertension. Lastly, the cross-sectional design of this study limits our ability to draw causal inferences about the observed relationships. Longitudinal studies with follow-up data would be beneficial in addressing this limitation and allow for more comprehensive analyses of BP trends over time.