Efficiency of remote monitoring and guidance in blood pressure management: a randomized controlled trial

Abstract

Background

Regular monitoring of blood pressure (BP) is essential for managing hypertension. The long-term effectiveness of remote monitoring in BP control requires further validation. Although self-monitoring has some benefits, its overall effectiveness remains inconclusive. In this study, we aimed to compare whether remote monitoring and self-monitoring differ from usual care in terms of their effects on BP reduction in patients with hypertension.

Methods

This multicentre randomised controlled trial enrolled participants diagnosed with primary hypertension in Qingdao, China, from December 2020 to December 2021. Patients were randomly assigned to one of the following three groups: remote monitoring, self-monitoring, or usual care. All groups underwent scheduled outpatient follow-up visits every 3 months throughout the 24-month study period. The primary outcome was the adjusted mean difference in systolic BP, which was used to evaluate the effectiveness of remote monitoring and self-monitoring compared to usual care.

Results

A total of 1006 participants were analysed: 332 in the usual care group, 337 in the self-monitoring group, and 337 in the remote group. After 24 months, BP was significantly reduced in all the groups, particularly in the remote group with suboptimal baseline BP, where systolic and diastolic BP decreased by 17.7 mm Hg and 10.2 mm Hg, respectively. Compared to the usual care group, the adjusted mean difference in systolic BP for the remote group was -2.7 (-4.45, -0.96), p = 0.002, while the self-monitoring group showed no significant difference. No significant differences in diastolic BP were observed between either group and the usual care group. Attainment rates of target BP increased rapidly in the first 3 months, and continued to rise only in the remote group, reaching 69.65%. The remote group had a higher proportion of calcium channel blockers, diuretics, and combination therapy than the usual care and self-monitoring groups.

Conclusions

Remote monitoring and guidance enhanced the efficiency of BP management, whereas self-monitoring alone proved to be ineffective.

Trial registration

ClincialTrials.gov Identifier: NCT04690478.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12916-025-04278-6.

Background

Hypertension is a cardiovascular syndrome characterised by elevated systemic arterial pressure, which damages the structure and function of vital organs, including the heart, brain, and kidneys, ultimately leading to their failure [1–3]. Hypertension is a leading risk factor for cardiovascular disease and mortality worldwide [4]. According to a 2019 report, hypertension is the leading cause of death among women and the second leading cause among men, and is responsible for approximately 11 million deaths globally [5]. Despite effective treatments, hypertension management and control remain inadequate, with poor patient compliance and clinical inertia being common problems [6]. Digital health technologies such as remote monitoring, wearables, and mobile health platforms are increasingly becoming important tools in hypertension management [7].

Medical workers often monitor patients' blood pressure (BP) in the following three ways. 1) Usual care primarily involves repeated office BP (OBP) measurements, which remain the most common approach for long-term management of hypertension [8]. However, continuous monitoring of a patient’s BP is not feasible. Additionally, the clinical setting may induce tension and anxiety in patients, potentially leading to ‘white coat hypertension’ (WCH) [9]. This ultimately leads to misdiagnosis and overtreatment. The European Society of Hypertension guidelines recommend that patients with suspected WCH undergo home or ambulatory BP monitoring to confirm the diagnosis [10]. 2) Self-monitoring can reflect the BP level in patients'daily lives better than OBP, but measurement errors can occur in different brands and models of BP monitors. Some patients may be unable to correctly operate the BP monitor or understand the readings [11]. However, studies have indicated that isolated self-monitoring in itself did not improve BP control but was effective when combined with other interventions [12]. 3) Remote monitoring enables patients to track their BP at home in a regular and quantitative manner. After each measurement, BP data is automatically uploaded, and in cases of abnormal BP, an immediate telephone follow-up is conducted to check for hypertension-related symptoms. Based on this consultation, the patient is advised to promptly seek medical care or adjust their medications.

Each of the three methods has advantages and limitations. Current perspectives suggest that while self-monitoring of BP may be used as a diagnostic option for hypertension, it is particularly suitable for long-term management of patients with WCH [13]. Remote BP monitoring avoids the shortcomings of other monitoring methods and has significant advantages in enhancing doctor-patient communication and strengthening patient awareness. The TASMINH4 trial demonstrated that over a 12-month period, both remote and self-monitoring of BP were significantly more effective in reducing systolic BP than usual care treatments, with remote monitoring having the greatest effect. However, these differences were not statistically significant with respect to controlling diastolic BP [14]. Notably, although the effectiveness of self-monitoring in BP management is somewhat recognised, its actual impact remains unconfirmed [15]. Therefore, in this study, we aimed to identify whether remote monitoring and self-monitoring differ from usual care with respect to BP reduction, over an extended follow-up period, to provide robust data for future clinical decisions. The significance of this study is further emphasised by the fact that similar studies focusing on the Chinese population are lacking.

Methods

For this multicentre prospective study, we enrolled participants from December 2020 to December 2021 and followed them up for over 24 months. Patients with primary hypertension who were treated in the cardiology departments of various hospitals in Qingdao, including The Affiliated Hospital of Qingdao University, The Affiliated Cardiovascular Hospital of Qingdao University, Qingdao Beixin Minsheng Hospital, Qingdao Beitai Hospital, and Qingdao Ankang Hospital were included. This was an open-label randomised controlled trial with automated outcome assessments. The effectiveness of remote BP monitoring was compared with usual care and self-monitoring in patients with hypertension. Detailed experimental methods are provided in Additional file 1.

Population

Participants were selected from the outpatient clinics of the five hospitals involved in this study. All potential participants underwent the initial screening process, including a review of general information, medical history, physical examination, and laboratory tests. After assessment by the medical team, patients who met the inclusion criteria were included in the study.

The following inclusion criteria were considered: participants aged 18–75 years and a diagnosis of primary hypertension [8]. Participants were required to own a smartphone, be proficient in using smart devices, and be permanent residents; they had to provide voluntary consent to participate in the study, including agreeing to follow-up visits. Exclusion criteria included the following: systolic BP ≥ 180 mmHg or diastolic BP ≥ 110 mmHg; pregnancy or breastfeeding; arrhythmias (such as atrial fibrillation or atrioventricular block) that could interfere with the operation of electronic BP monitors; severe diabetes with glycosylated haemoglobin levels > 11%; acute cardiovascular events within the past 3 months; severe valvular heart disease, cardiomyopathy, or chronic heart failure; chronic kidney disease stages 4 or 5; terminal diseases, precluding follow-up due to limited life expectancy; and other factors as detailed in supplementary materials.

All participants signed informed consent forms, and the study was approved by the Ethics Committee of the Qingdao University Affiliated Cardiovascular Hospital (approval number: 2019–01-01).

Procedure

This study included three groups: remote care, self-monitoring, and usual care. Neither the participants nor the investigators were blinded to group allocation. The participants were assessed for eligibility at the clinic, provided informed consent, and underwent baseline clinical data collection and laboratory tests. The researchers used an automatic lottery program uploaded to a cloud platform. After entering patients'basic information, the program assigned them to groups in a 1:1:1 ratio using simple randomisation. The system ensured that neither the researchers nor the participants could alter or predict the allocation. All participants received training upon enrolment and formally confirmed the scheduled time and location of the first follow-up visit in writing. The time for subsequent follow-up visits were confirmed at the end of each visit, with a follow-up interval of 3 months, and continued until 24 months. During each outpatient visit, the BP was measured with the assistance of a dedicated research nurse. The measurement method is detailed in Additional file 2. Experienced senior cardiologists performed consultations and BP assessments at outpatient clinics, adjusted medications, and provided health education. The research staff assisted in recording key indicators such as office BP, lifestyle changes, and medication use through structured questionnaires.

Remote group

Each participant in the remote group was provided with an automatic upper-arm oscillometric BP monitor (Maibobo, RBP-9801, Shenzhen, China) that uploaded data to a cloud platform, along with training on proper use of the monitor. Participants in the remote-monitoring group received appropriately sized cuffs and were instructed on their correct use to ensure accurate measurements. Participants were required to measure their blood pressure at home three times each morning (within 1 h of waking, on an empty stomach, before taking medication, and after urinating), with 1-min interval between readings. During the first 3 months, they measured their blood pressure daily for 7 consecutive days (or at least three times per week if unable to complete daily measurements). During follow-up, patients with stable and well-controlled BP were asked to undergo measurements three fixed days per week (or at least once if they were unable to complete the required frequency). Patients with unstable or uncontrolled BP were advised to maintain daily measurements (or at least three times per week if daily measurements were not possible). Well-controlled BP was defined as having BP within the target range in at least three of the last five home-based self-monitored measurements, as determined by a clinic doctor during follow-up [16]. After completing the measurements, the data were automatically uploaded. Patients were instructed to download the public version of the “Medical Home” app on their smartphones, and were guided regarding its usage. The “cloud platform” automatically generated a BP trend graph based on patients'daily morning measurements.

Intermediate-level physicians in the research team reviewed BP data of the participants in the remote-monitoring group daily between 8:00 AM and 12:00 PM. If a patient's systolic BP was ≥ 135 mm Hg and/or diastolic BP was ≥ 85 mm Hg for two consecutive readings, an orange indicator was triggered. If a patient's systolic BP reached ≥ 175 mm Hg and/or diastolic BP ≥ 105 mm Hg in a single reading, a red indicator was displayed. In both cases, the physician immediately contacted the patient to enquire regarding the symptoms and assess their condition, with higher priority given to those with the red indicator. The physician provided remote guidance, which included lifestyle modifications, medication adjustments, recommendations for follow-up BP monitoring, or advice for clinic visits. Specific interventions were determined based on the condition of each patient. In addition, participants could reach out to the research team at any time for assistance with device use, follow-up scheduling, or health consultations, particularly during poor BP control or when they felt unwell. However, contacting a physician was at the discretion of the patient.

Researchers provided health education to patients via the"Medical Home" app, which included: (1) publishing two articles weekly on hypertension management covering diet, lifestyle, drug side effects, and coping strategies; (2) hosting biweekly hypertension lectures and promptly notifying the patients. After the patients watched, their participation was automatically recorded, helping researchers set reminders for those who missed sessions; (3) for participants whose BP remained uncontrolled after 1 month of management, research staff initiated remote expert support via the app’s expert studio function, which was conducted by senior internal medicine physicians to provide timely and individualised guidance. These senior physicians conducted personalised assessments through video calls or phone consultations, asked patients regarding their symptoms, evaluated their BP, identified the cause of poor control, and provided tailored strategies. These strategies included psychological counselling, medication adjustments, dietary advice, physical activity recommendations, and sleep management. The physician also determined whether further in-person consultations were required (Additional file 2: Fig. S1).

Self-monitoring group

The participants in the self-monitoring group used the same BP monitor as those in the remote group and followed the same measurement methods and frequencies. The results were recorded in a logbook for assessment and treatment during follow-up visits; however, the monitor lacked remote functionality. The follow-up frequency and methods were consistent with those of the remote-monitoring group, with participants receiving necessary medical guidance during outpatient visits.

Usual care group

Patients in the usual care group were instructed to manage their BP according to their habits or local community practices with no additional interventions implemented. In addition to the absence of home BP blood pressure monitoring data for physician evaluation, the follow-up procedures were identical to those in the other two groups.

BP targets and measurements

Blood pressure targets were defined as < 140/80 mmHg for individuals with diabetes and < 140/90 mmHg for others based on office measurements, and < 135/75 mmHg and < 135/85 mmHg respectively for home self-monitoring [14, 17]. Failure to meet these criteria was classified as uncontrolled BP. The method for OBP measurement is detailed in Supplementary Materials.

Outcome

The primary outcome was office-measured systolic BP at 24 months, which served as the key indicator for assessing treatment effects and BP control. Secondary outcomes included BP levels, BP reduction, and changes in target BP attainment rate at 6, 12, and 24 months. Additionally, changes in parameters such as smoking, alcohol consumption, sleep duration, anxiety, salt intake, body mass index (BMI), and medication use at 24 months were assessed.

Statistical analysis

To detect a 5 mm Hg systolic BP difference between groups with 90% power and an adjusted alpha of 0.017 for a two-by-two comparison of the three groups, 298 cases per group were needed, as indicated by the results of the TASMINH-4 study [14]. This calculation was based on a common standard deviation of 17 mm Hg and a three-factor pairwise comparison. Considering a 15% dropout rate, 350 participants were required in each group, resulting in a total sample size of 1050 participants.

Statistical analysis was performed using SPSS software (version 22.0). The analyses were performed on an intention-to-treat basis. Continuous variables are expressed as mean ± standard deviation (x ± s), and one-way analysis of variance (ANOVA) was used for comparisons between groups. Count data are expressed as cases (%), and chi-square or Fisher's exact test was used for comparisons among groups. For non-continuous missing data, nearest neighbor interpolation was used. To better compare the differences in the antihypertensive effects among the three groups, a linear mixed-effects model was used. The model was adjusted for baseline BP, healthcare institutions, and target BP levels to accurately assess treatment effects and control for individual differences. Subgroup analyses were conducted to assess whether treatment effects differed according to predefined baseline characteristics, including sex, age, baseline systolic blood pressure (SBP), body mass index (BMI), diabetes status, anxiety, and sleep duration. There was a statistically significant difference (p < 0.05) between the two groups. In this study, we controlled for the risk of Type I error from multiple comparisons using Bonferroni correction, applying p < 0.0167 as the significance level for between-group comparisons.

Results

Description of study participants

As shown in Fig. 1, 1,258 individuals met the inclusion criteria and underwent further screening. Of these, 32 withdrew their consent and 220 were excluded for various reasons. The main reasons for exclusion (based on 1,258 participants) were as follows: 78 individuals (6.2%) had systolic BP ≥ 180 mmHg or diastolic BP ≥ 110 mmHg, 42 (3.3%) had atrial fibrillation, 33 (2.6%) had chronic heart failure, 25 (2.0%) had chronic kidney disease stages 4–5, and 15 (1.2%) had spouses or relatives already enrolled in the study.

Fig. 1.

Flowchart of participant recruitment

Ultimately, 1,006 participants were randomised: 332 to the usual care group, 337 to the self-monitoring group, and 337 to the remote group. Among all the participants, 668 (66.4%) had uncontrolled baseline BP with the following distribution across the groups: 217 (65.4%) in the usual care group, 229 (68%) in the self-monitoring group, and 222 (65.9%) in the remote care group. During the follow-up period, 35, 29, and 22 participants were lost to follow-up in the usual care, self-monitoring, and remote groups, respectively. Additionally, although some individuals completed the follow-up visits, several participants missed visits due to transportation issues, health status, and lack of telephone contact; the number of visits were 57 in the usual care group, 87 in the self-monitoring group, and 165 in the remote group. The groups were well matched, with a mean age of 61.02 ± 9.02 years, a slightly higher proportion of males (464, 51.21%), and a mean BP of 144.56/86.07 ± 15.86/10.41 mm Hg (Table 1).

Table 1.

Baseline characteristics

Characteristic	Total (n = 1006)	Usual care group (n = 332)	Self-monitoring group (n = 337)	Remote group (n = 337)	P value
Age, years	61.02 ± 9.02	61.01 ± 8.97	60.92 ± 9.18	61.11 ± 8.95	0.966
Systolic BP, mm Hg	144.56 ± 15.86	143.93 ± 15.56	145.45 ± 15.45	144.28 ± 16.53	0.47
Diastolic BP, mm Hg	86.07 ± 10.41	85.96 ± 10.48	86.63 ± 10.24	85.63 ± 10.52	0.482
High, cm	167.11 ± 8.10	167.25 ± 8.25	167.07 ± 8.12	167.01 ± 7.90	0.933
Weight, kg	73.53 ± 11.98	73.88 ± 11.64	73.70 ± 11.51	73.03 ± 12.73	0.661
BMI, kg/m2	26.24 ± 3.28	26.32 ± 3.01	26.32 ± 3.12	26.09 ± 3.65	0.622
Waist circumference, cm	92.42 ± 10.06	92.77 ± 9.44	93.12 ± 10.11	91.42 ± 20.52	0.09
White blood cell, × 109/L	6.22 ± 1.40	6.22 ± 1.40	6.47 ± 1.57	6.28 ± 1.65	0.23
Red blood cell, × 109/L	4.69 ± 0.51	4.73 ± 0.43	4.67 ± 0.59	4.67 ± 0.47	0.429
Hemoglobin, g/L	142.67 ± 14.51	143.58 ± 15.83	141.74 ± 14.09	142.79 ± 13.69	0.426
Platelet, × 109/L	227.20 ± 56.67	225.69 ± 49.04	229.59 ± 54.13	226.22 ± 55.36	0.741
ALT, U/L	23.19 ± 16.60	21.78 ± 11.36	23.84 ± 14.85	23.86 ± 21.41	0.313
Creatinine, mmol/L	82.41 ± 21.18	83.02 ± 20.14	82.55 ± 19.91	81.73 ± 23.21	0.799
Total cholesterol, mmol/L	5.09 ± 1.15	5.06 ± 1.16	5.04 ± 1.15	5.16 ± 1.13	0.455
Triglyceride, mmol/L	1.88 ± 1.70	1.88 ± 1.81	1.80 ± 1.31	1.94 ± 1.91	0.679
HDL-C, mmol/L	1.41 ± 0.46	1.39 ± 0.52	1.40 ± 0.44	1.45 ± 0.41	0.289
LDL-C, mmol/L	3.09 ± 0.92	2.99 ± 0.90	3.14 ± 0.90	3.13 ± 0.94	0.162
Glucose, mmol/L	6.33 ± 2.37	6.52 ± 3.15	6.31 ± 2.18	6.19 ± 1.60	0.329
Urine acid, μmol/L	349.81 ± 110.73	340.92 ± 89.68	353.35 ± 126.92	355.12 ± 112.90	0.367
BP control rate, %	338(33.6)	115(34.6)	108(32%)	115(34.1)	0.753
Male, %	521(51.8)	175(52.7)	172(51)	174(51.6)	0.908
Diabetes mellitus, %	198(19.7)	72(21.7)	66(19.6)	60(17.8)	0.45
Smoking, %	166(16.5)	53(16)	57(16.9)	56(14.6)	0.944
Drinking, %					0.622
Never	722(71.8)	238(71.7)	247(73.3)	237(70.3)
Intermittent	203(20.2)	72(21.7)	61(18.1)	70(20.8)
Daily	81(8.1)	22(6.6)	29(8.6)	30(8.9)
Salt intake, %					0.757
< 6 g	415(41.3)	139(41.9)	132(39.2)	144(42.7)
6-12 g	464(46.1)	154(46.4)	163(48.4)	147(43.6)
> 12 g	127(12.6)	39(11.7)	42(12.5)	46(13.6)
Hours of sleep, %					0.24
< 4 h	170(16.9)	48(14.5)	60(17.8)	62(18.4)
4–8 h	744(74)	257(77.4)	248(73.6)	236(70)
> 8 h	95(9.4)	27(8.1)	29(8.6)	39(11.6)
Anxiety, %					0.914
Never	212(62.9)	618(61.4)	200(60.2)	206(6.11)
Sometimes	290(28.8)	99(29.8)	96(28.5)	95(28.2)
Often	98(9.7)	33(9.9)	35(10.4)	30(8.9)

BP Blood pressure, BMI Body mass index, ALT Alanine aminotransferase, HDL-C High-density lipoprotein cholesterol, LDL-C Low-density lipoprotein cholesterol

Primary outcome: differences in systolic and diastolic BP across interventions

BP decreased significantly during the follow-up period in all groups, particularly in the remote group. Specifically, mean BP in the remote group was 134.4/79.05 ± 11.25/8.15 mm Hg at 24 months of follow-up, with mean reductions of 9.88 ± 16.95 mm Hg and 6.58 ± 9.66 mm Hg in systolic and diastolic BP, respectively. For patients with substandard baseline BP, the mean reductions in systolic and diastolic BPs in the remote group were 17.70 ± 3.79 mm Hg and 10.20 ± 8.41 mm Hg, respectively. Adjusted mean differences were used to compare between-group reductions among the three groups. In all enrolled patients, the mean systolic BP in the remote group was significantly lower than that in the usual care group at 6 months (135.10 ± 13.53 vs. 137.34 ± 13.70 mm Hg), with an adjusted mean difference of −2.38 mm Hg (95% confidence interval (CI): −4.31, −0.45), p = 0.016. By 24 months, the mean systolic BP in the remote group further decreased to 134.4 ± 11.25 mm Hg, widening the adjusted difference to −2.7 mm Hg (95% CI: −4.45, −0.96), p = 0.002, compared to 137.01 ± 12.19 mm Hg in the usual care group. However, mean systolic BP at 12 months and mean diastolic BP at 6, 12, and 24 months decreased more in the remote care group than in the usual care group; however, the difference was not statistically significant. The mean systolic and diastolic BPs were not significantly lower in the self-monitoring group than in the usual care group at any time point. At 24 months, the adjusted mean difference in diastolic BP was slightly higher in the self-monitoring group (1.39 mm Hg, 95% CI: 0.21, 2.59, p = 0.021) but was not statistically significant (Table 2). Moreover, compared to the self-monitoring group, the remote-monitoring group showed significantly greater reductions in both systolic and diastolic BPs (Additional file 2: Table S1).

Table 2.

Differences in systolic and diastolic BP across interventions in all participants

	Baseline	6 month	12 month	24 month	Difference in BP from baseline to 24-month	6-month adjusted mean difference vs usual care	12-month adjusted mean difference vs usual care	24-month adjusted mean difference vs usual care
All participants
Systolic BP (mm Hg)
Usual care (n = 332)	143.93 ± 15.56	137.34 ± 13.70	136.28 ± 12.24	137.01 ± 12.19	−6.29 ± 14.74	-	-	-
Self-monitoring (n = 337)	145.45 ± 15.45	138.13 ± 14.0	138.16 ± 12.64	138.46 ± 12.92	−6.98 ± 16.37	0.13(−1.83,2.08), p = 0.9	1.33(−0.46,3.11), p = 0.146	0.97(0.90,2.84), p = 0.309
Remote (n = 337)	144.28 ± 16.53	135.10 ± 13.53	134.72 ± 11.79	134.4 ± 11.25	−9.88 ± 16.95	−2.38(−4.31,−0.45), p = 0.016*	−1.67(−3.42,0.83), p = 0.062	−2.7(−4.45,−0.96), p = 0.002*
Diastolic BP (mm Hg)
Usual care (n = 332)	86.07 ± 10.41	81.68 ± 8.93	80.28 ± 8.71	80.02 ± 7.72	−5.94 ± 10.38	-	-	-
Self-monitoring (n = 337)	85.96 ± 10.48	82.31 ± 9.02	81.6 ± 8.23	81.63 ± 8.45	−4.99 ± 9.89	0.33(−0.91,1.56), p = 0.602	1.06(−0.15,2.27), p = 0.085	1.39(0.21,2.59), p = 0.021
Remote (n = 337)	86.63 ± 10.24	80.99 ± 9.32	79.43 ± 8.17	79.05 ± 8.15	−6.58 ± 9.66	−0.55(−1.80,0.70), p = 0.391	−0.74(−1.92,0.45), p = 0.221	−0.86(−2.01,0.29), p = 0.143

BP Blood pressure

After Bonferroni adjustment, significant at p < 0.0167

An analysis of patients with substandard blood pressure at enrolment showed that the remote group had a notably greater decrease in systolic BP compared to the usual care group at 6, 12, and 24 months of follow-up, with a difference of −4.79 mm Hg (95% CI: −7.01, −2.57) at 24 months (p < 0.001). The difference in mean diastolic BP between the groups increased progressively at 6, 12, and 24 months, with a difference of −1.66 mm Hg (95% CI: −3.08, −0.24) at 24 months (p = 0.022), which was not statistically significant. In patients with well-controlled BP at the time of selection, there were no significant differences between the groups (Table 3).

Table 3.

Differences in systolic and diastolic BP across interventions in participants with uncontrolled and controlled baseline BP

	Baseline	6 month	12 month	24 month	Difference in BP from baseline to 24-month	6-month adjusted mean difference vs usual care	12-month adjusted mean difference vs usual care	24-month adjusted mean difference vs usual care
Participants with baseline BP uncontrolled
Systolic BP (mm Hg)
Usual care (n = 217)	152.8 ± 10.08	141.61 ± 12.69	139.83 ± 11.8	140.36 ± 12.15	−12.44 ± 2.22	-	-	-
Self-monitoring (n = 229)	153.26 ± 11.27	141.41 ± 11.85	141.09 ± 12.21	140.66 ± 12.68	−12.59 ± 4.32	−0.39(−2.88,2.09), p = 0.756	1.01(−1.19,3.38), p = 0.345	0.14(−2.21,2.49), p = 0.906
Remote (n = 222)	153.48 ± 11.43	137.76 ± 13.02	136.81 ± 11.42	135.78 ± 11.13	−17.70 ± 3.79	−4.12(−6.54,−1.69), p = 0.001*	−3.22(−5.44,−1),p = 0.005*	−4.79(−7.01,−2.57), p < 0.001*
Diastolic BP (mm Hg)
Usual care (n = 217)	89.89 ± 9.85	83.51 ± 9.51	81.9 ± 8.87	81.35 ± 7.85	−8.54 ± 10.44	-	-	-
Self-monitoring (n = 229)	90.21 ± 9.50	83.70 ± 9.16	83.09 ± 8.33	82.89 ± 8.69	−7.32 ± 9.97	0.046(−1.57,1.66), p = 0.956	1.06(−0.47,2.59), p = 0.175	1.44(−0.10,2.97), p = 0.66
Remote (n = 222)	89.89 ± 9.49	82.31 ± 9.24	80.41 ± 8.03	79.69 ± 8.10	−10.20 ± 8.41	−1.20(−2.82,0.43), p = 0.149	−1.49(−2.96,−0.02), p = 0.047	−1.66(−3.08,−0.24), p = 0.022
Participants with baseline BP controlled
Systolic BP (mm Hg)
Usual care (n = 115)	127.17 ± 8.81	129.26 ± 11.8	129.56 ± 10.06	130.67 ± 9.5	3.5 ± 11.5	-	-	-
Self-monitoring (n = 108)	128.89 ± 8.29	131.12 ± 11.58	131.93 ± 11.02	133.78 ± 12.23	4.99 ± 13.91	1.20(−1.96,4.36), p = 0.456	1.84(−1.04,4.72), p = 0.209	2.78(−5.87,0.277), p = 0.074
Remote (n = 115)	126.57 ± 8.35	129.98 ± 12.96	130.68 ± 11.47	131.76 ± 11.06	5.20 ± 11.4	1.062(−2.09,4.21), p = 0.507	1.36(−1.48,4.19), p = 0.347	1.30(−1.44,4.03), p = 0.351
Diastolic BP (mm Hg)
Usual care (n = 115)	78.53 ± 7.05	78.23 ± 6.49	77.23 ± 7.53	77.49 ± 6.8	−1.041 ± 8.33	-	-	-
Self-monitoring (n = 108)	78.99 ± 7.13	79.36 ± 7.98	78.44 ± 7.09	78.96 ± 7.25	−0.31 ± 7.69	0.91(−0.92,2.74), p = 0.328	1.06(−0.89,3.01), p = 0.286	1.306(−0.546,3.16), p = 0.166
emote (n = 115)	77.44 ± 6.96	78.44 ± 6.98	77.51 ± 8.12	77.81 ± 8.15	0.37 ± 7.8	0.82(−1.07,2.71), p = 0.395	0.759(−1.25,2.76), p = 0.456	0.756(−1.18,2.69), p = 0.442

BP Blood pressure

After Bonferroni adjustment, significant at p < 0.0167

The trend in BP changes

Trend analysis showed that among all enrolled patients, the target BP attainment rate in the self-monitoring group was slightly lower than that in the usual care group at baseline (31.92% vs. 34.62%), whereas the remote monitoring group had a similar rate (34.19%). During the follow-up period, the attainment rate gradually increased in all groups, with the fastest increase observed in the first 3 months. While the rates in the usual care and self-monitoring groups stabilised after 3 months, the rate in the remote group stabilised after 6 months and then significantly increased again after 12 months, reaching 69.65% at 24 months compared to 60.14% in the usual care group and 53.09% in the self-monitoring group (Fig. 2A). In patients with substandard BP at baseline, the achievement rate increased in all three groups during the follow-up period, with the fastest rate of increase in the first 6 months, after which it levelled off. The remote group showed a significant upward trend from month 12, reaching an attainment rate of 67.48% at 24 months (Fig. 2B).

Fig. 2.

Trends in target BP attainment rates and values among the overall groups (A, C) and uncontrolled baseline BP groups (B, D). BP targets: < 140/80 mm Hg for diabetics, < 140/90 mm Hg for all others. BP, blood pressure

The overall trend in BP value changes mirrored the attainment rate, with a rapid decrease during the first 6 months, and the remote group showed the greatest reduction (Fig. 2C). For participants with suboptimal baseline BP, the remote group continued to exhibit a slow decline after 6 months; at 24 months, the BP was approximately 135 mmHg, aligning with the group's average systolic BP, whereas the other two groups had average systolic BPs slightly > 140 mm Hg, with no further significant decline (Fig. 2D).

Subgroup analyses for differences in mean systolic BP

Subgroup analyses revealed that among participants with uncontrolled baseline BP, the remote monitoring group showed significantly greater reductions in mean systolic BP at 24 months than the usual care group, across subgroups defined by sex, age, diabetes status, baseline systolic BP, BMI, anxiety, and sleep duration. The reduction was higher in females, individuals with diabetes, and those with baseline systolic BP ≥ 160 mmHg, but interaction effects were not statistically significant. No significant differences in BP reduction were observed between the self-monitoring and usual care groups in any subgroup (Fig. 3).

Fig. 3.

Forest plot of subgroup analyses for differences in mean systolic BP from baseline to 24 months among participants with uncontrolled BP at baseline. BP targets: < 140/80 mm Hg for diabetics, < 140/90 mm Hg for all others. BP, blood pressure

Treatment-induced changes in participants with uncontrolled baseline BP

We further analysed participants with uncontrolled BP at baseline and identified the changes within each group and differences between the groups, from baseline to the end of follow-up. At the end of follow-up, all three groups showed significant lifestyle improvements: the proportion of smokers decreased and the average sleep duration increased. Both the usual care and remote monitoring groups exhibited significant improvements in salt intake, with a notable reduction in the proportion of participants consuming > 12 g/day and an increase in those consuming less than 6 g/day. Although the self-monitoring group exhibited a similar trend, the changes were not statistically significant (Additional file 2: Table S2).

In terms of medication use, all three groups showed significantly increased usage of calcium channel blockers (CCBs) and a higher proportion of patients receiving combination therapy at 24 months. The remote care group showed significantly increased use of angiotensin-converting enzyme inhibitors (ACEIs)/angiotensin receptor blockers (ARBs), and both the usual care and remote care groups had substantially higher use of diuretics (Table 4).

Table 4.

Changes in medication use before and after the intervention in participants with uncontrolled BP

Antihypertensive medication, %		Conventional group, n = 217	Self-monitoring group, n = 229	Remote group, n = 222	P Value
ACEI/ARB/ARNI	Baseline	156(71.9)	162(70.7)	159(71.6)	0.961
	24 months	144(75.4)	151(72.6)	160(77.7) a	0.44
Calcium channel blockers	Baseline	108(49.8)	124(54.1)	121(54.5)	0.542
	24 months	121(63.4) a	142(68.3) a	155(75.2) a, b	0.036
Beta-blockers	Baseline	51(23.5)	52(22.7)	44(19.8)	0.617
	24 months	45(23.6)	52(25)	48(23.3)	0.922
Diuretics	Baseline	35(16.1)	48(21)	44(19.8)	0.41
	24 months	47(24.6) a	51(24.5)	77(37.4) a, b	0.004
Drug combination	Baseline	116(53.5)	126(55)	122(55)	0.933
	24 months	120(62.8) a	131(63)a	151(73.3) a, b	0.034

BP Blood pressure

^aSignificant difference between baseline and 24 months within the group, P < 0.05

^bSignificant difference in the remote monitoring group compared with both the usual care and self-monitoring groups, P < 0.0167

Furthermore, at 24 months, no significant differences in lifestyle changes were observed among the three groups. However, the remote group had a significantly higher proportion of participants using CCBs, diuretics, and combination therapy than the usual care and self-monitoring groups, indicating the advantage of the remote group in medication adjustment (Table 4; Additional file 2: Table S2).

Discussion

Main findings

The remote BP monitoring and management model is a hospital-community-individual management model, with the objective to include guidance from higher-level hospitals to community management. This includes remote monitoring of BP, timely detection of abnormal BP, lifestyle adjustment, and health education to achieve optimal management. After 24 months, the patients using remote monitoring had significantly lower systolic and diastolic BPs than those receiving clinical adjustment therapy. This reduction was achieved through increased titration or a combination of antihypertensive medications, demonstrating that remote monitoring reduces clinical inertia and optimises treatment.

Notably, the self-monitoring group had the lowest target BP achievement rate at 24 months (53.09%), compared to the remote monitoring (69.65%) and usual care groups (60.14%), during the same period. A higher proportion of participants in the self-monitoring group (41.8%) had a BP close to the target (systolic BP 140–145 mmHg), compared to the usual care (27.9%) and remote monitoring groups (30.2%). This suggests that despite improved awareness of BP changes, the self-monitoring group probably did not adopt appropriate lifestyle or treatment adjustments owing to a lack of necessary health guidance and professional intervention. Additionally, when BP is close to the target, patients may develop a sense of self-satisfaction, leading to treatment inertia, reduced medication adherence, and lifestyle changes. Finally, there was no significant difference in BP reduction between the self-monitoring and usual care groups, which is consistent with previous studies showing that self-monitoring without additional medical interventions has limited effects on BP control [18]. This highlights the importance of health education regarding remote monitoring.

This study first compared BP changes among the three patient groups at different time points. Significant reductions in BP levels were found after specific interventions, with the greatest improvement occurring within the first 6 months. This emphasises the importance of prompt action when BP control is inadequate, as timely measures can yield rapid health benefits. Further observation showed that the attainment rates in the remote group continued to decline significantly 1 year after intervention, highlighting the long-term effectiveness of remote intervention in BP management. A thorough assessment showed that among all enrolled patients, the systolic BP decreased by an additional 2.7 mm Hg in the remote group compared to the usual care group. Patients with suboptimal baseline BP had a significantly higher reduction in systolic BP (additional decrease of 4.79 mm Hg). The difference in reduction of diastolic BP between the remote and usual care groups gradually increased, reaching −1.66 mm Hg at the 24 months. In particular, the Bonferroni correction used in the study design to adjust for significance levels is a relatively conservative approach that may miss meaningful differences. However, it is very likely that such differences would become statistically significant if the treatment duration were extended. This result shows that remote intervention is more effective in reducing BP, and surpassed other strategies with superior timeliness, continuity, and personalised treatment plans, emphasising its benefits in BP management.

Strengths and weaknesses

The distinguishing feature of our study is that remote monitoring of hypertension was performed through a 24-month follow-up period and quarterly assessments, offering robust evidence of BP trends and intergroup differences. Notably, the remote monitoring group exhibited sustained improvement beyond 12 months, demonstrating the effectiveness of consistent monitoring and intervention in achieving long-term BP control. Additionally, the self-monitoring group did not show greater benefits than the usual care group, likely because of the rigorous follow-up in the usual care group, which produced similar effects. This is the first large-scale randomised controlled trial on remote monitoring of hypertension in China, which further enhances the generalisability and representativeness of our findings.

Our goal was to include individuals of all ages for broad representation and unbiased results. However, ensuring the participation of groups affected by economic hardships, limited technical skills, inadequate education, poor health, and other social factors remains a challenge. These groups often struggle to access and use remote monitoring technology, which can hinder healthcare access and worsen health inequalities. Previous studies have shown that socioeconomic status and age are significant barriers in adapting digital technology, as is the case in most impoverished areas of China [19, 20]. According to the 47th Statistical Report on Internet Development in China (2020), 29.6% of the general population did not use mobile internet, and 46% of this group comprised individuals aged ≥ 60 years [21]. Considering significant regional and economic disparities in the use of smart devices among older adults, we conducted a preliminary survey in Qingdao, China, prior to the initiation of the study. The results showed that the proportion of individuals aged > 75 years who were proficient in using smart devices was significantly lower than that of individuals aged 60–75 years. Therefore, individuals aged > 75 years were excluded from the study due to cognitive and operational limitations that made it difficult for them to meet the technical requirements of remote monitoring. We acknowledge this limitation and aim to refine the remote monitoring program to better support these groups and enhance healthcare equity and accessibility.

Unfortunately, we did not collect precise and complete information on changes in medication dosages, which represents a limitation in the design of our questionnaire. Additionally, we did not use electronic systems to monitor medication use, which is a reliable method for tracking medication usage [22]. Baseline data collection in this study was not comprehensive. It did not categorise hypertension stages or classifications, lacked long-term outcomes, such as cardiovascular and cerebrovascular events, as observational indicators, and did not include a comprehensive cost analysis. Further research is required to address these gaps.

Changes in medication use

It is generally believed that combination therapy can activate different pathophysiological pathways to achieve coordinated BP reduction more effectively, while also resulting in fewer adverse reactions [23]. Empirical analysis revealed that a significantly higher proportion of the remote treatment group received combination therapy compared to the usual care group, highlighting the positive impact of remote monitoring in facilitating optimal BP management and reduction. A 2015 randomised controlled trial published in The Lancet found that sodium retention is likely the main cause of resistant hypertension, with spironolactone showing significant BP-lowering effects [24]. In this study, a significant increase in diuretic use was observed in the remote treatment group, which likely contributed to better BP control in patients with hypertension. However, a detailed analyses of the specific types of diuretics involved was not performed. Additionally, the remote group had a higher rate of CCB use than the control group, which is consistent with the choice of combination therapy [25].

Clinical implications

Investigations indicate that general practitioners rely on self-monitoring practices for managing hypertension and that at least one-third of patients with hypertension are self-monitoring [26, 27]. Digital technologies, including smartphone applications and wearable devices, have demonstrated significant clinical benefits in BP monitoring and management [14]. Telemedicine can alleviate the shortages in medical resources, and offers a viable solution for monitoring and managing BP in large patient populations, without geographical limitations. It enables timely medical advice and reduces the burden on hospitals and patients [28]. Implementing a cost-effective digital intervention to lower BP appears to be an appropriate strategy, although its implementation in real-world settings can be time consuming. For example, in the UK, key barriers in implementing home BP monitoring systems were overcome by incorporating extensive user feedback and optimised design, ensuring clear objectives, and enabling ease of use [29–31]. The development of remote systems requires ongoing technological research to improve patient acceptance, enhance data security, and ensure system compatibility and stability. Its effectiveness relies heavily on a multidisciplinary team of healthcare professionals including clinicians, pharmacists, and nurses. Therefore, enhancing team work and continuously advancing the skills and knowledge of its members are key challenges in optimising the efficiency and quality of remote monitoring services.

Conclusions

Overall, a hypertension management model using remote monitoring enables better BP control than traditional models and has the potential to serve as a powerful tool for reducing BP in patients. The next step is to devise ways to implement these strategies and attain the benefits.

Abbreviations

BP

Blood pressure

OBP

Office blood pressure

WCH

White coat hypertension

CCB

Calcium channel blocker

ANOVA

One-way analysis of variance

CI

Confidence interval

BMI

Body mass index

ACEIs

Angiotensin-converting enzyme inhibitors

ARBs

Angiotensin receptor blockers

Authors’ contributions

All authors read and approved the final manuscript. TT: Conceptualization, methodology, formal analysis, drafting the original, review and editing, project management. GS: Conceptualization, investigation, data organization, drafting the original, review and editing, supervision. ZY, ZL, TW, QW, RQ, MW: Experiment implementation, data collection. RC, JX, NZ, BS, XL, YZ: Follow-up, data organization, and data analysis. HY: Conceptualization, supervision, project management, fund acquisition.

Funding

This study is funded by the Qingdao Municipal Science and Technology Bureau under its Science and Technology Beneficial Demonstration Guidance Special Fund project plan. The Project Leader is Hai-Chu Yu. Award Number: 20–3-4–54-nsh.

Data availability

The datasets analysed during the current study are not publicly available due to protecting participants'privacy, but are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee of Qingdao University Affiliated Cardiovascular Hospital (approval number: 2019–01-01). Written informed consent was obtained from all participants.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.