2026 JSH指南：使用数字技术控制血压

Preamble

1. Purpose of the guideline

The "Guidelines for Blood Pressure Control Using Digital Technologies"

(hereafter referred to as “the Guideline”) has been developed for the following two purposes.

To enable the Guideline users (the general public, hypertensive patients, companies, municipalities, insurers, healthcare providers, etc.) to select the appropriate digital technologies for blood pressure control. To indicate the direction for digital technology developers (companies) in developing new technologies.

In 2022, the Japanese Society of Hypertension established the Committee on the Guidelines for Blood Pressure Control Using Digital Technologies, which published the "Guidelines for Blood Pressure Control Using Digital Technologies" in 2025 (Tokyo: Japanese Society of Hypertension; 2025). This is the English version of that report. Chairperson: Hisatomi Arima (Department of Preventive Medicine and Public Health, Faculty of Medicine, Fukuoka University) www.jpnsh.jp/digitalguide.html. A list of authors and their affiliations appears at the end of "Future research questions".

☑ Hisatomi Arima harima@fukuoka-u.ac.jp Extended author information available on the last page of the article

2. Intended users and sites

The Guideline has been developed to help select appropriate digital technologies in various situations, such as health promotion for the general public and blood pressure control for people with hypertension.

The intended users include the general public, hypertensive patients, companies, local governments, insurers, healthcare providers (physicians, occupational health practitioners, public health nurses, nurses, pharmacists, dieticians, etc.), and digital technology developers (companies).

The intended settings for use include the homes and workplaces of the general public and hypertensive patients, industrial health sites in companies (companies that practice excellent health and productivity management, etc.), community health sites in public health centers, health insurers (health insurance associations, Japan Health Insurance Association, National Health Insurance, etc.), health checkup facilities, nursing care facilities, clinics, hospitals, digital technology development companies, and so on.

The Guideline is intended to serve as a reference and for use for the general public and hypertensive patients in selecting appropriate digital technology for blood pressure control, in health guidance or healthcare and treatment in occupational health, community health, general healthcare, and health checkups.

Furthermore, future research questions are included as information for the direction of technology expected in the

H. Arima et al.

future. Digital technology developers will be expected to use and refer to this information.

3. The guideline's scope

The following describes what is within/outside the scope of the Guideline:

[Within the scope of the Guideline]

1) Adults aged 18 years and older (including hypertensive and non-hypertensive individuals).

2) The following digital technologies:

(1) Self-measuring devices for blood pressure, heart rate, urinary sodium concentration, urinary potassium concentration, calorie intake, energy balance, water balance, number of steps, amount of exercise, calories burned, sleep duration, sleep quality, blood oxygen level, stress level, mental checkup (wearable devices, devices measuring the urinary sodium/potassium, etc.)

(2) Digital technologies to record blood pressure, heart rate, weight, food intake, urinary sodium level, urinary potassium level, alcohol consumption, smoking status, number of steps, amount of exercise, sleep duration, sleep quality, stress level, and mental status

(3) Digital technologies to share monitoring results with healthcare providers and others (cloud, etc.)

(4) Digital technologies to provide knowledge and information (pre-contemplation phase), motivation (contemplation phase), and planning support such as goal setting (preparation phase), motivation enhancement and rewarding (implementation phase), removing barriers for continuation (maintenance phase), and promoting behavior changes [e.g., applications, interactive communication support (interventions using smartphone applications, text messages, web screens on computers, etc.)]

(5) Digital technologies that enable healthcare appointments in telemedicine, healthcare treatment using smartphone applications, text messages, web screens of PCs, payment of healthcare fees, and mailing of prescriptions, as well as digital technologies that support routine healthcare treatment and health guidance.

(6) Other digital technologies (stress reduction programs, mental health applications, etc.)

(7) Combination of multiple digital technologies

[Outside the scope of the Guideline]

1) Individuals aged under 18 years

2) Pregnant women

3. Digital technologies for which the effects on blood pressure have not been clarified

4. Identification of essential healthcare issues

The essential healthcare problem is "blood pressure control using digital technology". Various digital technologies and sites of their application (interventions) used for blood pressure control were hypothesized to solve this problem, and six healthcare questions (HQs) were identified. Each issue is verified: "Can blood pressure control using digital technology for adults be recommended?".

First, as digital technologies that promote the continuation of behavior change through self-monitoring (including information sharing with healthcare providers, etc.), self-measurement of blood pressure (home blood pressure monitoring [HBPM]), using a cuff sphygmomanometer (HQ1), wearable devices (HQ2), and devices measuring the urinary sodium/potassium ratio, or dietary or urinary sodium concentration (HQ3) were set.

Next, as a combination of multiple digital technologies (self-monitoring, information sharing with healthcare providers and others, behavior change support, two-way communication, etc.), smartphone applications (HQ4), and telemedicine and health guidance using digital technology (HQ5) were set.

Finally, artificial Intelligence (AI)-based healthcare and health guidance support (HQ6) is a new technology that is expected to be incorporated into various digital technologies.

Ultimately, the following six HQs were established.

HQ1 Is intervention with self-measurement of blood pressure at home (HBPM) using a cuff sphygmanometer effective in lowering blood pressure in adults?

HQ2 Do interventions with different wearable devices have a beneficial effect on blood pressure in adults?

HQ3 Do interventions using devices that measure the urinary sodium/potassium ratio, or dietary or urinary sodium concentration reduce blood pressure in adults?

HQ4 Does a smartphone application-based intervention for blood pressure control lower blood pressure in adults?

HQ5 Do telemedicine and health guidance using digital technology have a beneficial effect on blood pressure in adults?

HQ6 Does AI-based healthcare and health guidance support lower blood pressure in adults?

The following are applicable service examples and life-style habits expected to be improved for each HQ.

The Japanese Society of Hypertension Guidelines for blood pressure control using digital technologies

|HQ|Applicable service examples|Expected Lifestyle Improvement| |---|---|---| |1|Cuff sphygmomanometer Blood pressure recording application, software Feedbacks using smartphone applications, text messages, computers, etc.|Diet Physical activity Proper weight maintenance Moderate alcohol consumption Quitting smoking Stress Management etc.| |2|24 h wearable devices (altigraph, smartwatch, smart ring, smart glasses, fitness tracker, pedometer, etc.) that measure biometric information such as blood pressure, heart rate, pulse rate, sleep duration, activity level, standing and sitting time, number of steps|Diet Physical activity Proper weight maintenance Stress Management etc.| |3|Devices measuring the urinary sodium/potassium ratio Devices measuring the urinary sodium concentration Devices measuring the dietary sodium concentration|Diet etc.| |4|Smartphone applications for health promotion|Diet Physical activity Proper weight Maintenance Proper weight maintenance Stress Management| |5|Telemedicine and telehealth guidance using smartphone applications, text messages, and web screens such as PCs|Quitting smoking Stress Management| |6|Automatic feedback on self-monitoring results Strengthening motivation through personalized lifestyle improvement guidance and chat functions via smartphone applications|etc.| |8|Personalized lifestyle guidance tools Personalized healthcare support tools that can be implemented in electronic medical records|Diet Physical activity Proper weight maintenance Moderate alcohol consumption Quitting smoking Stress Management|

5. Systematic review

Relevant literature was collected for the six established HQs, and a systematic review was conducted.

The results of the systematic review will be published in Hypertension Research. For more information on the systematic review, please see the paper.

1) Implementation period

The systematic review was conducted between April 2023 and December 2023. Subgroup analyses classified as hypertensive or normotensive were also performed when possible.

2) Evidence searched

(1) Type of Evidence

Individual research papers and systematic review papers were searched in this priority order. In principle, randomized controlled trials (RCTs), non-RCTs, and observational studies were included in the search for individual research papers, although some systematic reviews included only RCTs.

(2) Database

Medline or PubMed, Cochrane Library, and ICHUSHI were used for individual research articles and systematic reviews.

(3) Basic search policy

The PICO format was used for the search, which was based on a combination of P (patient) and I (intervention), with O (outcome) in some cases.

(4) Search period

A literature search was conducted between January and December 2023 for all databases.

3) Criteria for selection and exclusion of literature

As a rule, we selected individual research articles (RCTs, non-RCTs, and observational studies) that examined digital technology interventions' effects on adults' blood pressure. Systematic reviews and meta-analyses of individual research articles and health economic evaluations were also searched to identify articles omitted from the search.

In principle, the systematic review included all RCTs, non-RCTs, and observational studies, although some systematic reviews included only RCTs.

We did not conduct a systematic review when none of the individual studies met the recruitment criteria.

4) Methods of evaluating and integrating evidence

This was based on the "Minds Manual for Guideline Development 2020 ver. 3.0". In evaluating evidence, we evaluated individual studies and the evidence as a whole.

H. Arima et al.

5) Preparation of recommendations

A draft recommendation was prepared based on the results of the systematic review.

6. Method of determining recommendations

The Guideline is based on the "Minds Manual for Guideline Development 2020 ver. 3.0", and the strength of recommendation of the HQ Recommendation Statement was rated on a 5-point scale.

$\bullet$ Strongly recommended action $\bullet$ Action not suggested $\bullet$ Action strongly discouraged $\bullet$ Recommendation/suggestion with held due to insufficient evidence

The Delphi method was used to determine the recommendations, and the respondents' opinions were collected by voting. A recommendation was considered to be decided if it received at least 70% of the votes cast for the item that received the most votes, and voting was conducted up to three times until a decision was made.

In the voting, the following items were considered: the certainty of the evidence confirmed in the systematic review, the balance between the expected effects (benefits) and adverse events (harms) of the intervention, the values of the citizens, and cost-effectiveness.

The voting results are listed on the Recommendations and Explanations page of each HQ.

7. Creation fund

The Guideline was developed with financial support from the Japan Agency for Medical Research and Development (AMED).

8. COI management rules and disclosure

The committee members who developed the Guideline were organized according to the "Guideline for Blood Pressure Control Using Digital Technology: Conflict of Interest (COI) Management Policy".

In addition, the COIs of the committee members for the past 3 years (2022–2024) are available on the Japanese Society of Hypertension website (https://www.jpnsh.jp/digitalguide.html).

9. Notes on the use of the guideline

The Guideline focuses on using digital technology for blood pressure control, reflecting the state of knowledge at the

time of its preparation. It is not intended to restrict continuously advancing digital technology for blood pressure control. Digital technology should be applied flexibly, considering users' specific circumstances and individual needs.

Although the production organization is responsible for the statements in the Guideline, it does not assume responsibility for the results of using digital technologies.

Digital technologies in healthcare continue to advance, and new digital devices are being introduced to help people improve their health. We will update our recommendations as needed, but they may not reflect the latest developments.

When using the digital technologies recommended in the Guideline, it is advisable to use them under the guidance of a healthcare provider such as a certified cardiovascular disease prevention and treatment advisor from the Japanese Society of Hypertension, the Japanese Society for Cardiovascular Disease Prevention, the Japanese Society for Arteriosclerosis Research and the Japanese Society of Cardiology, or a certified hypertension specialist from the Japanese Society of Hypertension. If you are unsure about your judgment, such as when abnormal values are detected, please consult with your healthcare provider and visit a healthcare institution if necessary. When sharing medical information with your healthcare provider, please ensure that measures are taken to protect your personal information.

Guidelines for blood pressure control using digital technologies: conflict of interest (COI) management policy

1. The guideline was financially supported only by the Japan Agency for Medical Research and Development (AMED).

2. The management of conflicts of interest (COI) for the members of the Guideline Development Committee will, in principle, be based on the "Japanese Association of Medical Sciences: Guidance on the Criteria for Participation in the Creation of Clinical Practice Guidelines 2023".

Members of the Guideline Development Committee with the following COI regarding digital technology shall not have voting rights and shall not participate in systematic review (SR).

* Affiliation with a sponsored department * Lecture fees: 2 million yen or more per company * Writing fees: 2 million yen or more per company * Research expenses: 20 million yen or more per company

The Japanese Society of Hypertension Guidelines for blood pressure control using digital technologies

$\bullet$ Scholarship donations: 10 million yen or more per company $\bullet$ Other compensation: 500,000 yen or more per company

However, members of the Formulation Committee who have COI with companies or profit-making organizations that are not thought to impact blood pressure control using digital technologies may have voting rights, only if the relevant conflict of interest is not believed to impact recommendations significantly, etc.

The number of members with Level 2 COI shall not exceed 1/2 of the total.

* Lecture fees: 1 million yen or more per company

Writing fees: 1 million yen or more per company

Research funds: 10 million yen or more per company

Scholarship donations: 5 million yen or more per company

Other compensation: 200,000 yen or more per company

If a committee member in a serious COI situation abstains from voting at the meeting to make recommendations, the person's name in question will be disclosed.

The COI Management Guidelines and the COI of all committee members will be deliberated and approved by the Japanese Society of Hypertension's Conflict of Interest Committee.

When the guideline is made public, the COI of the participants in formulating the guideline (including external evaluation committee members) and the Japanese Society of Hypertension will be disclosed.

Healthcare Questions (HQs) and Recommendations

HQ 1 Is intervention with self-measurement of blood pressure at home (HBPM) using a cuff sphygmomanometer effective in lowering blood pressure in adults?

Recommendation

In adults, intervention with self-measurement of blood pressure at home (HBPM) using an upper arm cuff sphygmomanometer is strongly recommended, especially when combined with telemonitoring or support from a healthcare provider.

HQ 2 Do interventions with different wearable devices have a beneficial effect on blood pressure in adults?

Recommendation

As the evidence on the intervention with various wearable devices in adults is insufficient to date, we withhold any recommendation/suggestion.

HQ 1

Recommendation As the evidence on the intervention with various wearable devices in adults is insufficient to date, we withhold any recommendation/suggestion.

HQ 3 Do interventions using devices that measure the urinary sodium/potassium ratio, or dietary or urinary sodium concentration reduce blood pressure in adults?

Recommendation

In adults, we suggest interventions using devices that measure the urinary sodium/potassium ratio, or dietary or urinary sodium concentration, especially when combined with educational programs.

HQ 4 Does a smartphone application-based intervention for blood pressure control lower blood pressure in adults?

Recommendation

We suggest a smartphone application-based intervention for blood pressure control in adults (normotensive and hypertensive individuals). However, evidence regarding the long-term effects (beyond 6 months) remains insufficient.

HQ 5 Do telemedicine and health guidance using digital technology have a beneficial effect on blood pressure in adults?

Recommendation

We suggest the use of telemedicine and health guidance using digital technology for blood pressure control in adults (normotensive and hypertensive individuals).

HQ 6 Does AI-based healthcare and health guidance support lower blood pressure in adults?

Recommendation

Any recommendation or suggestion is withheld due to insufficient evidence on blood pressure control using AI-based healthcare and health guidance support in adults.

1 Self-measurement of blood pressure at home using a cuff sphygmomanometer

Overview of digital technology

Self-measurement of blood pressure at home (home blood pressure measurement [HBPM]) using an upper arm cuff sphygmomanometer is widely used in Japan and significantly contributes to preventing, diagnosing, and treating hypertension (Table 1). (Fig. 1)

H. Arima et al.

Table 1 Benefits and points to note about HBPM using a cuff sphygmomanometer

$\blacksquare$ Benefits

Measurement with a periodic fixed point under the same conditions

Understanding variability

Self-awareness, self-feedback

Data analysis and visualization

Data sharing with healthcare providers (certified cardiovascular disease prevention educators, public health nurses, nurses, pharmacists, dietitians, hypertension specialists, etc.)

$\blacksquare$Points to note

Validity and reliability of measurements

Compliance with measurement methods for each instrument

Compliance with HBPM conditions

Measuring blood pressure over a fixed period under the same conditions and observing the average value can help us understand the baseline blood pressure.

Long-term measurements allow us to monitor day-to-day blood pressure variability, seasonal blood pressure variations, and increase or decrease in blood pressure over several years.

Understanding the relationship between lifestyle habits, treatments, and blood pressure levels and providing self-feedback can promote better self-management of lifestyle and medication.

By linking a blood pressure monitor with data storage and communication features to a smartphone application, the day-to-day blood pressure variability and long-term trends can be graphed, facilitating a visual grasp of the blood pressure values.

Sharing measurement results with healthcare providers allows for more appropriate advice, diagnosis, and treatment.

A reliable upper arm cuff sphygmomanometer (wraps around the upper arm type) is necessary. It is also essential to use an appropriately sized cuff. Wrist cuff sphygmomanometers tend to provide inaccurate readings and should be used with caution. For a list of reliable sphygmomanometers, please refer to the website of the Scientific Committee, the Japanese Society of Hypertension (https://www.jpnsh.jp/com_ac_wgl.html), which lists the results of accuracy tests of sphygmomanometers available in Japan.

It is essential to read the device's instruction manual carefully before taking a measurement and to use the correct posture and method. Incorrect measurement methods will not provide accurate blood pressure readings.

To determine one's baseline blood pressure values, it is necessary to measure under the correct conditions by following the guidelines of the Japanese Society of Hypertension (Fig. 1).

Healthcare question

$\blacksquare$ HQ 1

Is intervention with self-measurement of blood pressure at home (HBPM) using a cuff sphygmomanometer effective in lowering blood pressure in adults?

$\blacksquare$ Recommendation

In adults, intervention with self-measurement of blood pressure at home (HBPM) using an upper arm cuff sphygmomanometer is strongly recommended, especially when combined with telemonitoring or support from a healthcare provider.

Summary of evidence

A literature search identified 65 randomized controlled trials (RCTs) (21,053 individuals in total; adults aged ≥18 years) that compared HBPM intervention using a cuff sphygmomanometer with a usual care group (group without HBPM intervention), using changes in blood pressure as the outcome.

The changes in blood pressure values during the intervention by HBPM was -3.27 mmHg (95% CI -4.15 to -2.40) for systolic blood pressure (SBP) and -1.61 mmHg (95% CI -2.07 to -1.14) for diastolic blood pressure (DBP) compared to the usual care group, indicating a significant reduction in blood pressure.

The effect was more pronounced when additional interventions were involved, such as telemonitoring or support from a healthcare provider. No significant intervention effects on SBP were observed in studies that used wrist cuff sphygmomanometers rather than upper arm cuff sphygmomanometers.

Explanation

Many studies have reported that HBPM is more reliable and reproducible than office blood pressure (OBP) [1, 2] and such measurements are strongly associated with identification of cardiovascular disease and target organ damage [3–7]. Based on this evidence, the Japan Society of Hypertension (JSH) Guidelines for the Treatment of Hypertension 2014 (JSH2014) clearly states that when there

The Japanese Society of Hypertension Guidelines for blood pressure control using digital technologies

 is a difference between OBP and home blood pressure (HBP), “a home blood pressure-based diagnosis should have priority” [8]. In addition, a systematic review in CQ 1 of the Guidelines for the Management of Hypertension 2019 (JSH2019) showed that antihypertensive treatment using HBP as an outcome is more effective than treatment using OBP as an outcome in lowering mean ambulatory blood pressure [9]. However, there are limited systematic reviews that have quantitatively evaluated the effect of HBPM on blood pressure reduction compared to the control group without HBPM; there are no systematic reviews that include recent studies. In addition, cuffless phygmomanometers, which claim to be sufficiently accurate to measure blood pressure without the use of a cuff, have become more widespread in recent years. However, the validity of their measurements has been reported to be extremely low [10]. Therefore, we established this HQ in the Guideline and conducted a systematic review [11].

As a result, 65 RCTs were identified that compared the HBPM intervention using a cuff sphygmomanometer with a control group; the outcome was a change in blood pressure. Of the 65 trials, 63 trials included hypertensive patients. Twenty-six studies were from the United States, representing the largest number by country; no such study was conducted in Japan.

The intervention significantly changes blood pressure values by 3.27 mmHg (95% CI 2.40–4.15) for SBP and 1.61 mmHg (95% CI 1.14–2.07) for DBP compared to the usual care group (Figs. 2, 3).

A sensitivity analysis was conducted on 42 studies that added telemonitoring and healthcare provider support (such as regular follow-up, lifestyle guidance, and treatment adjustment suggestions) as intervention elements to HBPM. The results showed that the reduction in blood pressure due to the intervention was more significant compared to the usual care group, with SBP changes of -4.27 mmHg (95% CI -5.39 to -3.16) and DBP changes of -1.89 mmHg (95% CI -1.24 to -2.55), indicating that additional intervention may enhance the blood pressure lowering effect of HBPM (Figs. 4, 5).

A sensitivity analysis was conducted on the four studies that used HBPM with a wrist cuff sphygmomanometer. The results showed that the reduction in the blood pressure value

due to HBPM compared to the usual care group was not statistically significant for SBP changes of -0.06 mmHg (95% CI -1.53 to -1.40), but the effect was significant for DBP changes of -1.20 mmHg (95% CI -2.22 to -0.17). Intervention by HBPM using a wrist cuff sphygmomanometer showed no significant effect when SBP was used as the outcome. (Figs. 6, 7).

Summary

HBPM interventions with a cuff sphygmomanometer were shown to help lower blood pressure in adults. The intervention effect was more pronounced when accompanied by telemonitoring/healthcare provider intervention, but the effect was attenuated when a wrist cuff sphygmomanometer was used. Based on the above, this HQ strongly recommends intervention through HBPM using an upper arm cuff sphygmomanometer, especially with telemonitoring and healthcare provider intervention. In addition, verification of appropriate HBPM measurements and their interpretation during health and medication counseling should preferably be performed by a certified cardiovascular disease prevention educator.

In the studies selected for this systematic review, there were no reports of interventions lasting longer than 2 years, and the long-term antihypertensive effect of HBPM and its effect on the long-term risk of death and events is not apparent. In addition, only two studies reported the effects of HBPM in non-hypertensive individuals, and those studies were conducted in patients with chronic kidney disease or previous pregnancy-induced hypertension. Therefore, further research is needed on HBPM in the long-term prevention of hypertension.

Literature search flow

A literature search was conducted using PubMed, ICHUSHI and Cochrane Library, with 4378 studies selected for primary screening (title and abstract). This was followed by a secondary screening (full text) of 364 studies, excluding 4014 studies that met the exclusion criteria. A meta-analysis was conducted using the resulting 65 studies that met the HQ.

H. Arima et al.

![|First author, Year of publication|BP|Intervention period (months)|Difference in SBP change between groups [95%CI] (mmHg)|\n|:---:|:---:|:---:|:---:|\n|Vetter, 2000|OBP|2|-0.50 [-3.76, 2.76]|\n|Mehos, 2000|OBP|2|-10.10 [-24.43, 4.23]|\n|Rogers, 2001|24h-ABP|3|-4.80 [-9.43, -0.17]|\n|Broege, 2001|OBP|3|2.00 [-27.52, 23.52]|\n|Rudd, 2004|OBP|6|-8.50 [-14.52, -2.48]|\n|Halme, 2005|OBP|6|-3.20 [-8.23, 1.83]|\n|Zillich, 2005|OBP|3|-4.50 [-11.70, 2.70]|\n|Marquez-Contreas, 2000|OBP|6|-4.50 [-8.91, 0.95]|\n|Verberk, 2007|OBP|12|0.50 [-4.75, 5.75]|\n|Kauric-Klein, 2007|OBP|3|-7.00 [-20.38, 6.38]|\n|Artinian, 2007|OBP|12|-4.00 [-10.22, 2.22]|\n|Tobe, 2008|OBP|1.5|-2.00 [-6.26, 2.26]|\n|Madsen, 2008|Day ABP|6|-2.30 [-6.10, 1.50]|\n|Green, 2008|OBP|12|-2.70 [-5.14, -0.26]|\n|Parati, 2009|OBP|6|-0.20 [-4.42, 4.02]|\n|da Silva, 2009|24h-ABP|6|-11.00 [-20.34, -1.66]|\n|DeJesus, 2009|OBP|6|-1.56 [-10.27, 7.15]|\n|Bosworth, 2009|OBP|24|-0.60 [-3.55, 2.35]|\n|Rinfret, 2009|24h-ABP|12|-4.80 [-7.57, -2.03]|\n|Godwin, 2010|24h-ABP|12|-0.30 [-4.24, 3.64]|\n|McManus, 2010|OBP|12|-5.50 [-8.79, -2.21]|\n|Varis, 2010|OBP|14|3.40 [-4.37, 11.17]|\n|Bosworth, 2011|OBP|18|-1.20 [-5.65, 3.25]|\n|Magid, 2011|OBP|6|-6.00 [-10.35, -1.65]|\n|Hebert, 2012|OBP|9|1.20 [-5.35, 7.75]|\n|Fuchs, 2012|OBP|3|-1.20 [-7.60, 5.20]|\n|Piette, 2012|OBP|1.5|-4.20 [-9.10, 0.70]|\n|Williams, 2012|OBP|9|-3.90 [-12.67, 4.87]|\n|Kerry, 2013|OBP|12|-2.40 [-6.89, 2.09]|\n|Magid, 2013|OBP|6|-12.40 [-16.25, -8.55]|\n|McKinstry, 2013|Day ABP|6|-4.08 [-6.39, -1.77]|\n|Margarolis, 2013|OBP|12|-9.70 [-13.40, -6.00]|\n|Ogedegbe, 2014|OBP|12|-0.10 [-3.25, 3.05]|\n|Stewart, 2014|OBP|6|-5.30 [-10.60, -0.00]|\n|McManus, 2014|OBP|12|-8.80 [-12.70, -4.90]|\n|Hosseiniinasab, 2014|OBP|6|1.00 [-1.14, 3.14]|\n|Kim, 2014|OBP|6|-7.00 [-12.46, -1.54]|\n|Yi, 2015|OBP|9|-0.60 [-3.65, 2.45]|\n|Hanley, 2015|Day ABP|6|-6.40 [-17.14, 4.34]|\n|Aekplakorn, 2016|OBP|12|-2.50 [-7.25, 2.25]|\n|Kim, 2016|OBP|6|3.00 [-6.67, 12.67]|\n|Tzourio, 2017|OBP|24|-2.10 [-4.05, 0.15]|\n|Klarskov, 2018|OBP|10|-4.60 [-6.93, -2.27]|\n|Martinez, 2018|24h-ABP|24|-0.90 [-7.21, 5.41]|\n|McManus, 2018|OBP|12|-3.20 [-6.43, 0.03]|\n|Pan, 2018|OBP|6|-6.60 [-11.32, -1.88]|\n|Skolarus, 2018|OBP|6|3.10 [-8.42, 14.62]|\n|Cuffee, 2019|24h-ABP|3|-3.50 [-7.78, 0.78]|\n|Gu, 2020|OBP|12|0.10 [-4.90, 5.10]|\n|Ojji, 2020|OBP|1|-6.00 [-13.07, 1.07]|\n|Zha, 2020|OBP|6|-3.60 [-8.83, 1.63]|\n|Ionov, 2021|OBP|3|-13.50 [-18.77, -8.23]|\n|McManus, 2021|OBP|12|-3.50 [-6.84, -0.16]|\n|Zhang, 2021|OBP|3|0.10 [-2.70, 2.90]|\n|Aki, 2021|OBP|1.5|1.90 [-4.78, 8.58]|\n|Margarolis, 2022|OBP|12|-0.76 [-2.84, 1.32]|\n|Okoro, 2022|OBP|12|-2.73 [-12.47, 7.01]|\n|Mujsers, 2022|OBP|12|-5.60 [-10.54, -0.66]|\n|Calderon-Anyosa, 2023|OBP|1|-9.10 [-19.16, 0.96]|\n|Doogue, 2023|OBP|3|-17.20 [-32.44, -1.96]|\n|Hoppe, 2023|OBP|12|0.63 [-3.54, 4.80]|\n|Leupold, 2023|OBP|12|-5.60 [-9.09, -2.11]|\n|Martinez-Ibanez, 2023|OBP|12|-2.90 [-6.58, 0.78]|\n|Nejams, 2023|OBP|3|2.45 [-0.98, 5.88]|\n|Ramos-Zavala, 2023|OBP|3|-5.80 [-10.22, -1.38]|\n|Pooled estimate (I² = 60.8%)||||\n|BP was lowered in HBPM group||||\n|BP was lowered in the usual care group→||||\n|Difference in SBP change between groups (mmHg)|-20|-15|-10|-5|0|5|10|\n](1/parseimages/figure327126e9938e4a56b8b8213a032ac362_0_1.png) SBP: systolic blood pressure, HBPM: home blood pressure monitoring, BP: blood pressure, OBP: office blood pressure, ABP: ambulatory blood pressure

Recommendation decision (voting results)

The results of the systematic review showed that intervention by HBPM using a cuff sphygmomanometer helped lower blood pressure in adults, and the intervention effect was more pronounced when accompanied by telemonitoring/healthcare provider intervention, but not when a wrist

cuff sphygmomanometer was used. The evidence, the balance of benefits and harms, the public's perception of value, and cost-effectiveness led to the decision to "strongly recommend intervention by HBPM using an upper arm cuff sphygmomanometer in adults, especially combined with remote management and intervention by a healthcare provider".

The Japanese Society of Hypertension Guidelines for blood pressure control using digital technologies

![|First author, Year of publication|BP|Intervention period (months)|Difference in DBP change between groups [95%CI] (mmHg)|\n|:---:|:---:|:---:|:---:|\n|Vetter, 2000|OBP|2|1.00 [-32.91, 34.91]|\n|Mehos, 2000|OBP|3|1.00 [-32.91, 34.91]|\n|Rogers, 2001|24h-ABP|3|1.00 [-32.91, 34.91]|\n|Broege, 2001|OBP|3|1.00 [-32.91, 34.91]|\n|Rudd, 2004|OBP|6|1.00 [-32.91, 34.91]|\n|Halme, 2005|OBP|6|1.00 [-32.91, 34.91]|\n|Zillich, 2005|OBP|6|1.00 [-32.91, 34.91]|\n|Marquez-Conteras, 2000|OBP|6|1.00 [-32.91, 34.91]|\n|Verberk, 2007|OBP|12|1.00 [-32.91, 34.91]|\n|Kauric-Klein, 2007|OBP|3|1.00 [-32.91, 34.91]|\n|Artinian, 2007|OBP|12|1.00 [-32.91, 34.91]|\n|Tobe, 2008|OBP|1.5|1.00 [-32.91, 34.91]|\n|Madsen, 2008|Day ABP|6|1.00 [-32.91, 34.91]|\n|Green, 2008|OBP|12|1.00 [-32.91, 34.91]|\n|Parati, 2009|OBP|6|1.00 [-32.91, 34.91]|\n|da Silva, 2009|24h-ABP|6|1.00 [-32.91, 34.91]|\n|De Jesus, 2009|OBP|6|1.00 [-32.91, 34.91]|\n|Bosworth, 2009|OBP|24|1.00 [-32.91, 34.91]|\n|Rinfret, 2009|24h-ABP|12|1.00 [-32.91, 34.91]|\n|Godwin, 2010|24h-ABP|12|1.00 [-32.91, 34.91]|\n|McManus, 2010|OBP|12|1.00 [-32.91, 34.91]|\n|Varis, 2010|OBP|12|1.00 [-32.91, 34.91]|\n|Bosworth, 2011|OBP|18|1.00 [-32.91, 34.91]|\n|Magid, 2011|OBP|6|1.00 [-32.91, 34.91]|\n|Hebert, 2012|OBP|9|1.00 [-32.91, 34.91]|\n|Fuchs, 2012|OBP|3|1.00 [-32.91, 34.91]|\n|Piette, 2012|OBP|1.5|1.00 [-32.91, 34.91]|\n|Williams, 2012|OBP|9|1.00 [-32.91, 34.91]|\n|Kerry, 2013|OBP|12|1.00 [-32.91, 34.91]|\n|Magid, 2013|OBP|6|1.00 [-32.91, 34.91]|\n|McKinstry, 2013|Day ABP|6|1.00 [-32.91, 34.91]|\n|Margolis, 2013|OBP|12|1.00 [-32.91, 34.91]|\n|Ogedegbe, 2014|OBP|12|1.00 [-32.91, 34.91]|\n|Stewart, 2014|OBP|6|1.00 [-32.91, 34.91]|\n|McManus, 2014|OBP|12|1.00 [-32.91, 34.91]|\n|Hosseinnasab, 2014|OBP|6|1.00 [-32.91, 34.91]|\n|Kim, 2014|OBP|12|1.00 [-32.91, 34.91]|\n|Yi, 2015|OBP|9|1.00 [-32.91, 34.91]|\n|Hanley, 2015|Day ABP|6|1.00 [-32.91, 34.91]|\n|Aekplakorn, 2016|OBP|12|1.00 [-32.91, 34.91]|\n|Kim, 2016|OBP|6|1.00 [-32.91, 34.91]|\n|Tzourio, 2017|OBP|24|1.00 [-32.91, 34.91]|\n|Klarskov, 2018|OBP|10|1.00 [-32.91, 34.91]|\n|Martinez, 2018|OBP|24|1.00 [-32.91, 34.91]|\n|McManus, 2018|24h-ABP|12|1.00 [-32.91, 34.91]|\n|Pan, 2018|OBP|6|1.00 [-32.91, 34.91]|\n|Skolarus, 2018|OBP|6|1.00 [-32.91, 34.91]|\n|Cuffee, 2019|24h-ABP|3|1.00 [-32.91, 34.91]|\n|Gu, 2020|OBP|12|1.00 [-32.91, 34.91]|\n|Ojji, 2020|OBP|1|1.00 [-32.91, 34.91]|\n|Zha, 2020|OBP|6|1.00 [-32.91, 34.91]|\n|lonov, 2021|OBP|3|1.00 [-32.91, 34.91]|\n|McManus, 2021|OBP|12|1.00 [-32.91, 34.91]|\n|Zhang, 2021|OBP|3|1.00 [-32.91, 34.91]|\n|Aki, 2021|OBP|1.5|1.00 [-32.91, 34.91]|\n|Margois, 2022|OBP|12|1.00 [-32.91, 34.91]|\n|Okoro, 2022|OBP|12|1.00 [-32.91, 34.91]|\n|Mujsers, 2022|OBP|1|1.00 [-32.91, 34.91]|\n|Calderon-Anyosa, 2023|OBP|3|1.00 [-32.91, 34.91]|\n|Doogue, 2023|OBP|12|1.00 [-32.91, 34.91]|\n|Hoppe, 2023|OBP|12|1.00 [-32.91, 34.91]|\n|Leupold, 2023|OBP|12|1.00 [-32.91, 34.91]|\n|Martinez-Ibanez, 2023|OBP|12|1.00 [-32.91, 34.91]|\n|Nejami, 2023|OBP|3|1.00 [-32.91, 34.91]|\n|Ramos-Zavala, 2023|OBP|3|1.00 [-32.91, 34.91]|\n|Pooled estimate (I² = 46.7%)|||1.61 [-2.07, -1.14]|\n](1/parseimages/figuread78241a48374462895ad3bf02efcae6_0_1.png) The 36 votes cast included 29 voting “strongly recommended action” (85.3% of the valid votes), 5 “suggested action”, zero “action not suggested”, zero “action strongly discouraged”, zero “recommendation/suggestion withheld due to insufficient evidence”, and two abstentions.

Future research questions

(1) Does long-term HBPM intervention prevent death and cardiovascular disease in adults?

H. Arima et al.

a) Addition of telemonitoring and support from healthcare providers as intervention elements

![|First author, Year of publication|Intervention period (months)|Difference in SBP change between groups [95%CI] (mmHg)|\n|:---:|:---:|:---:|\n|Mehos, 2000|6|-10.10 [-24.43, 4.23]|\n|Rogers, 2001|3|-4.80 [-9.43, -0.17]|\n|Broege, 2001|3|-2.00 [-27.52, 23.52]|\n|Rudd, 2004|6|-8.50 [-14.52, -2.48]|\n|Zillich, 2005|3|-4.50 [-11.70, 2.70]|\n|Artinian, 2007|12|-4.00 [-10.22, 2.22]|\n|Madsen, 2008|6|-2.30 [-6.10, 1.50]|\n|Green, 2008|12|-2.70 [-5.14, -0.26]|\n|Parati, 2009|6|-0.20 [-4.42, 4.02]|\n|Bosworth, 2009|24|-0.60 [-3.55, 2.35]|\n|Rinfret, 2009|12|-4.80 [-7.57, -2.03]|\n|McManus, 2010|12|-5.50 [-8.79, -2.21]|\n|Varis, 2010|14|3.40 [-4.37, 11.17]|\n|Bosworth, 2011|18|-1.20 [-5.65, 3.25]|\n|Magid, 2011|6|-6.00 [-10.35, -1.65]|\n|Hebert, 2012|9|-8.20 [-14.15, -2.25]|\n|Piette, 2012|1.5|-4.20 [-9.10, 0.70]|\n|Williams, 2012|9|-3.90 [-12.67, 4.87]|\n|Kerry, 2013|12|-2.40 [-6.89, 2.09]|\n|Magid, 2013|6|-12.40 [-16.25, -8.55]|\n|McKinstry, 2013|6|-4.08 [-6.39, -1.77]|\n|Margarolis, 2013|12|-9.70 [-13.40, -6.00]|\n|Stewart, 2014|6|-5.30 [-10.60, -0.00]|\n|Kim, 2013|12|-7.00 [-12.46, -1.54]|\n|Yi, 2015|9|-0.60 [-3.65, 2.45]|\n|Hanley, 2015|6|-6.40 [-17.14, 4.34]|\n|Kim, 2016|6|3.00 [-6.67, 12.67]|\n|McManus, 2018|12|-4.50 [-7.74, -1.26]|\n|Pan, 2018|6|-6.60 [-11.32, -1.88]|\n|Skolarus, 2018|6|3.10 [-8.42, 14.62]|\n|Gu, 2020|12|0.10 [-4.90, 5.10]|\n|Zha, 2020|6|-3.60 [-8.83, 1.63]|\n|Ionov, 2021|3|-13.50 [-18.77, -8.23]|\n|McManus, 2021|12|-3.50 [-6.84, -0.16]|\n|Margarolis, 2022|12|-0.76 [-2.84, 1.32]|\n|Okoro, 2022|12|-2.73 [-12.47, 7.01]|\n|Mujisers, 2022|12|-5.60 [-10.54, -0.66]|\n|Calderon-Anyosa, 2023|1|-9.10 [-19.16, 0.96]|\n|Doogue, 2023|3|-17.20 [-32.44, -1.96]|\n|Hoppe, 2023|12|0.63 [-3.54, 4.80]|\n|Leupold, 2023|12|-5.60 [-9.09, -2.11]|\n|Ramos-Zavala, 2023|3|-5.80 [-10.22, -1.38]|\n](1/parseimages/figureea915ccadd3d4dd2b8315454d2fcf743_0_2.png) SBP: systolic blood pressure, BP: blood pressure, OBP: office blood pressure, ABP: ambulatory blood pressure, HBPM: home blood pressure monitoring

Fig. 4 Effect of additional intervention with telemonitoring or healthcare provider support on the SBP change. Results of stratified analysis of: (a) studies with the addition of telemonitoring and support from healthcare providers as intervention elements. (Adapted from ref. [12]). (b) Effect of additional intervention with telemonitoring or healthcare provider support on the SBP change. Results of stratified analysis of: (b) studies without the addition of telemonitoring or support from healthcare providers. (Adapted from ref. [12])

The Japanese Society of Hypertension Guidelines for blood pressure control using digital technologies

![|First author, Year of publication|Intervention period (months)|Difference in SBP change between groups [95%CI] (mmHg)|\n|:---:|:---:|:---:|\n|Vetter, 2000|2|-0.50 [-3.76, 2.76]|\n|Halme, 2005|6|-3.20 [-8.23, 1.83]|\n|Marquez-Conterras, 2006|6|-4.50 [-8.91, -0.09]|\n|Verberk, 2007|12|0.50 [-4.75, 5.75]|\n|Kauric-Klein, 2007|3|-7.00 [-20.38, 6.38]|\n|Tobe, 2008|1.5|-2.00 [-6.26, 2.26]|\n|da Silva, 2009|6|-11.00 [-20.34, -1.66]|\n|DeJesus, 2009|6|-1.56 [-10.27, 7.15]|\n|Godwin, 2010|12|-0.30 [-4.24, 3.64]|\n|Hebert, 2012|9|1.20 [-5.35, 7.75]|\n|Fuchs, 2012|3|-1.20 [-7.60, 5.20]|\n|Ogedegbe, 2014|12|-0.10 [-3.25, 3.05]|\n|McManus, 2014|12|-8.80 [-12.70, -4.90]|\n|Hosseinasab, 2014|6|1.00 [-1.14, 3.14]|\n|Aekplakorn, 2016|12|-2.50 [-7.25, 2.25]|\n|Tzourio, 2017|24|-2.10 [-4.05, -0.15]|\n|Klarskov, 2018|10|-4.60 [-6.93, -2.27]|\n|Martinez, 2018|24|-0.90 [-7.21, 5.41]|\n|McManus, 2018|12|-3.20 [-6.43, 0.03]|\n|Cuffee, 2019|3|-3.50 [-7.78, 0.78]|\n|Ojji, 2020|1|-6.00 [-13.07, 1.07]|\n|Zhang, 2021|3|0.10 [-2.70, 2.90]|\n|Akl, 2021|1.5|1.90 [-4.78, 8.58]|\n|Martinez-Ibanez, 2023|12|-2.90 [-6.58, 0.78]|\n|Nejamis, 2023|3|2.45 [-0.98, 5.88]|\n](1/parseimages/figure0090bebeb1224aa4a76ed9a1c59a6f16_0_1.png) Fig. 4 Continued

(2) Does HBPM alter risk factors for cardiovascular disease (blood pressure, BMI, etc.) in non-hypertensive individuals?

2 Wearable devices

Overview of digital technology

Recently, there has been an increase in health management devices called wearable devices, worn like a wristwatch.

Wearable devices include devices that can measure blood pressure (e.g., wearable sphygmomanometers) and devices that do not have blood pressure measurement functions. A wearable sphygmomanometer is a blood pressure monitor that can be worn at all times and comes in various forms, including a wristwatch, wristband, ring, and upper arm types. There are cuffless and cuff-based sphygmomanometers, but their accuracy needs to be verified in future studies [12]. Using these devices, we can quickly check daily health data such as blood pressure and body condition, and the accumulation of such data enables us to evaluate blood pressure and health condition based on changes over

H. Arima et al.

a) Addition of telemonitoring and support from healthcare providers as intervention elements

![|First author, Year of publication|BP|Intervention period (months)|Difference in DBP change between groups [95%CI] (mmHg)|\n|:---:|:---:|:---:|:---:|\n|Mehos, 2000|OBP|6|-6.70 [-19.22, 5.82]|\n|Rogers, 2001|24h-ABP|3|-4.10 [-7.20, -1.00]|\n|Broege, 2001|OBP|3|1.00 [32.91, 34.91]|\n|Rudd, 2004|OBP|6|-3.10 [-6.12, -0.08]|\n|Zillich, 2005|OBP|3|-3.20 [-9.81, 3.41]|\n|Artinian, 2007|OBP|12|-0.80 [-4.79, 3.19]|\n|Madsen, 2008|Day ABP|6|-0.80 [-3.05, 1.45]|\n|Green, 2008|OBP|12|-0.80 [-2.14, 0.54]|\n|Parati, 2009|OBP|6|0.40 [-2.23, 3.03]|\n|Bosworth, 2009|OBP|24|-1.20 [-2.85, 0.45]|\n|Rinfret, 2009|24h-ABP|12|-2.10 [-5.15, 0.95]|\n|McManus, 2010|OBP|12|-2.60 [-4.23, -0.97]|\n|Varis, 2010|OBP|14|2.40 [-1.54, 6.34]|\n|Bosworth, 2011|OBP|18|-0.50 [-3.15, 2.15]|\n|Magid, 2011|OBP|6|-2.30 [-4.65, 0.05]|\n|Hebert, 2012|OBP|9|-1.50 [-5.05, 2.05]|\n|Piette, 2012|OBP|1.5|-3.20 [-6.80, 0.40]|\n|Williams, 2012|OBP|9|0.85 [-3.22, 4.92]|\n|Kerry, 2013|OBP|12|-1.40 [-4.83, 2.03]|\n|Magid, 2013|OBP|6|-5.70 [-7.80, -3.60]|\n|McKinstry, 2013|Day ABP|6|-2.17 [-3.55, -0.79]|\n|Margarolis, 2013|OBP|12|-5.10 [-7.40, -2.80]|\n|Stewart, 2014|OBP|6|-0.30 [-3.40, 2.80]|\n|Kim, 2014|OBP|12|-4.00 [-7.17, -0.83]|\n|Yi, 2015|OBP|9|-1.00 [-2.66, 0.66]|\n|Hanley, 2015|Day ABP|6|-3.20 [-10.44, 4.04]|\n|Kim, 2016|OBP|6|4.30 [-2.93, 11.53]|\n|McManus, 2018|OBP|12|-0.70 [-2.86, 1.46]|\n|Pan, 2018|OBP|6|-3.00 [-7.92, 1.92]|\n|Skolarus, 2018|OBP|6|0.90 [-5.71, 7.51]|\n|Gu, 2020|OBP|12|-0.90 [-3.30, 1.50]|\n|Zha, 2020|OBP|6|-0.56 [-9.20, 8.08]|\n|Ionov, 2021|OBP|3|-5.90 [-8.89, -2.91]|\n|McManus, 2021|OBP|12|-0.70 [-3.04, 1.64]|\n|Margarolis, 2022|OBP|12|0.30 [-0.95, 1.55]|\n|Okoro, 2022|OBP|12|-0.54 [-6.11, 5.03]|\n|Mujisers, 2022|OBP|12|-5.00 [-8.39, -1.61]|\n|Calderon-Anyosa, 2023|OBP|1|-6.00 [-11.26, -0.74]|\n|Doogue, 2023|OBP|3|-4.40 [-25.34, 16.54]|\n|Hoppe, 2023|OBP|12|1.27 [-2.06, 4.60]|\n|Leupold, 2023|OBP|12|-3.10 [-5.50, -0.70]|\n|Ramos-Zavala, 2023|OBP|3|-4.50 [-7.50, -1.50]|\n](1/parseimages/figurea4d893f4f78c42edb78dcecce9abc3be_0_2.png) DBP: diastolic blood pressure, HBPM: home blood pressure monitoring, BP: blood pressure, OBP: office blood pressure, ABP: ambulatory blood pressure

Fig. 5 Effect of additional intervention with telemonitoring or healthcare provider support on the DBP change. Results of stratified analysis of: (a) studies with the addition of telemonitoring or support from healthcare providers as intervention elements. (Adapted from ref. [12]). Effect of additional intervention with telemonitoring or healthcare provider support on the DBP change. Results of stratified analysis of: (b) studies without the addition of telemonitoring or support from healthcare providers. (Adapted from ref. [12])

The Japanese Society of Hypertension Guidelines for blood pressure control using digital technologies

![|First author, Year of publication|Intervention period (months)|Difference in DBP change between groups [95%CI] (mmHg)|\n|:---:|:---:|:---:|\n|Vetter, 2000|2|-1.30 [-2.88, 0.28]|\n|Halme, 2005|6|-1.50 [-3.95, 0.95]|\n|Marquez-Contreras, 2006|6|-3.10 [-5.83, -0.37]|\n|Verberk, 2007|12|1.80 [-0.92, 4.52]|\n|Kauric-Klein, 2007|3|-1.00 [-8.87, 6.87]|\n|Tobe, 2008|1.5|-1.00 [-4.52, 2.52]|\n|da Silva, 2009|6|-6.00 [-10.84, -1.16]|\n|DeJesus, 2009|6|1.58 [-3.68, 6.84]|\n|Godwin, 2010|12|-1.20 [-3.37, 0.97]|\n|Hebert, 2012|9|-0.70 [-4.60, 3.20]|\n|Fuchs, 2012|3|-2.30 [-5.70, 1.10]|\n|Ogedegbe, 2014|12|-0.70 [-2.09, 0.69]|\n|McManus, 2014|12|-3.10 [-5.50, -0.70]|\n|Hosseinasab, 2014|6|-1.00 [-3.39, 1.39]|\n|Aekplakorn, 2016|12|-1.20 [-4.45, 2.05]|\n|Tzourio, 2017|24|-0.20 [-1.20, 0.80]|\n|Klarskov, 2018|10|-2.90 [-4.34, -1.46]|\n|Martinez, 2018|24|-2.80 [-6.90, 1.30]|\n|McManus, 2018|12|-1.20 [-3.40, 1.00]|\n|Cuffee, 2019|3|-2.00 [-8.42, 4.42]|\n|Ojji, 2020|1|-1.00 [-5.74, 3.74]|\n|Zhang, 2021|3|-0.20 [-1.95, 1.55]|\n|Akl, 2021|1.5|1.30 [-2.95, 5.55]|\n|Martinez-Ibanez, 2023|12|-1.10 [-3.19, 0.99]|\n|Nejamis, 2023|3|0.38 [-2.02, 2.78]|\n](1/parseimages/figure699612cc90e64dbba97aec8a82937b55_0_1.png) Fig. 5 Continued

time, which was previously not possible. This section describes the benefits and points to note about these devices (Table 2).

Healthcare question

$\blacksquare$HQ 2 Do interventions with different wearable devices have a beneficial effect on blood pressure in adults?

$\blacksquare$ Recommendation

As the evidence on the intervention with various wearable devices in adults is insufficient to date, we withhold any recommendation/suggestion.

Summary of evidence

A literature search identified 21 RCTs (2474 individuals in total; adults aged ≥18 years) that compared changes in SBP

H. Arima et al.

a) Studies using upper arm cuff sphygmomanometers

![|First author, Year of publication|BP|Intervention period (months)|Difference in SBP change between groups [95%CI] (mmHg)|\n|:---:|:---:|:---:|:---:|\n|Mehos, 2000|24h-ABP|6|-10.10 [-24.43, 4.23]|\n|Rogers, 2001|24h-ABP|3|-4.80 [-9.43, 0.17]|\n|Broege, 2001|OBP|3|-2.00 [-27.52, 23.52]|\n|Rudd, 2004|OBP|6|-8.50 [-14.52, -2.48]|\n|Halme, 2005|OBP|6|-3.20 [-8.23, 1.83]|\n|Zillich, 2005|OBP|6|-4.50 [-11.70, 2.70]|\n|Marquez-Conteras, 2006|OBP|6|-4.50 [-8.91, 0.09]|\n|Verberk, 2007|OBP|12|-0.50 [-4.75, 5.75]|\n|Kauric-Klein, 2007|OBP|3|-7.00 [-20.38, 6.38]|\n|Artinian, 2007|OBP|12|-4.00 [-10.22, 2.22]|\n|Tobe, 2008|OBP|1.5|-2.00 [-6.26, 2.26]|\n|Madsen, 2008|Day ABP|6|-2.30 [-6.10, 1.50]|\n|Green, 2008|OBP|12|-2.70 [-5.14, -0.26]|\n|Parati, 2009|OBP|6|-0.20 [-4.42, 4.02]|\n|da Silva, 2009|24h-ABP|6|-11.00 [-20.34, -1.66]|\n|DeJesus, 2009|OBP|6|-1.56 [-10.27, 7.15]|\n|Rinfret, 2009|24h-ABP|12|-4.80 [-7.57, -0.31]|\n|Godwin, 2010|24h-ABP|12|-0.30 [-4.24, 3.64]|\n|McManus, 2010|OBP|12|-5.50 [-8.79, -2.21]|\n|Varis, 2010|OBP|14|3.40 [-4.37, 11.17]|\n|Bosworth, 2011|OBP|18|-1.20 [-5.65, 3.25]|\n|Magid, 2011|OBP|6|-6.00 [-10.35, -1.65]|\n|Hebert, 2012|OBP|9|1.20 [-5.35, 7.75]|\n|Fuchs, 2012|OBP|3|-1.20 [-7.60, 5.20]|\n|Piette, 2012|OBP|1.5|-4.20 [-9.10, 0.70]|\n|Williams, 2012|OBP|5|3.90 [-12.67, 4.87]|\n|Kerry, 2013|OBP|9|-2.40 [-6.89, 2.09]|\n|Magid, 2013|OBP|6|-12.40 [-16.25, -8.55]|\n|McKinstry, 2013|Day ABP|6|-4.08 [-6.39, -1.77]|\n|Margolis, 2013|OBP|12|-9.70 [-13.40, -6.00]|\n|Ogedegbe, 2014|OBP|6|-0.10 [-3.25, 3.05]|\n|Stewart, 2014|OBP|6|-5.30 [-10.60, -0.00]|\n|McManus, 2014|OBP|12|-8.80 [-12.70, -4.90]|\n|Kim, 2014|OBP|12|-7.00 [-12.46, -1.54]|\n|Yi, 2015|OBP|9|-0.60 [-3.65, 2.45]|\n|Hanley, 2015|Day ABP|6|-6.40 [-17.14, 4.34]|\n|Aekplakorn, 2016|OBP|12|-2.50 [-7.25, 2.25]|\n|Kim, 2016|OBP|6|3.00 [-6.67, 12.67]|\n|Tzourio, 2017|OBP|24|-2.10 [-4.05, -0.15]|\n|Klarskov, 2018|OBP|10|-4.60 [-6.93, -2.27]|\n|Martinez, 2018|24h-ABP|24|-0.90 [-7.21, 5.41]|\n|McManus, 2018|OBP|12|-3.20 [-6.43, 0.03]|\n|Pan, 2018|OBP|6|-6.60 [-11.32, -1.88]|\n|Skolarus, 2018|OBP|6|3.10 [-8.42, 14.62]|\n|Cuffee, 2019|24h-ABP|6|-3.50 [-7.78, 0.78]|\n|Gu, 2020|OBP|12|0.10 [-4.90, 5.10]|\n|Oiji, 2020|OBP|1|-6.00 [-13.07, 1.07]|\n|Ionov, 2021|OBP|12|-13.50 [-18.77, -8.23]|\n|McManus, 2021|OBP|3|-3.50 [-6.84, 2.16]|\n|Zhang, 2021|OBP|1.5|0.10 [-2.70, 3.90]|\n|Aki, 2021|OBP|12|1.90 [-4.78, 8.58]|\n|Margolis, 2022|OBP|12|-0.76 [-2.84, 1.32]|\n|Okoro, 2022|OBP|12|-2.73 [-12.47, 7.01]|\n|Mujisers, 2022|OBP|12|-5.60 [-10.54, -0.66]|\n|Calderon-Anyosa, 2023|OBP|1|-9.10 [-19.16, 0.96]|\n|Doogue, 2023|OBP|3|-17.20 [-32.44, -1.96]|\n|Hoppe, 2023|OBP|12|0.63 [-3.54, 4.80]|\n|Leupold, 2023|OBP|12|-5.60 [-9.09, -2.11]|\n|Martinez-Ibanez, 2023|OBP|12|2.90 [-6.58, 0.78]|\n|Nejamis, 2023|OBP|3|2.45 [-0.98, 5.88]|\n|Ramos-Zavala, 2023|OBP|3|-5.80 [-10.22, -1.38]|\n|Vetter, 2000|OBP|2|-0.50 [-3.76, 2.76]|\n|Bosworth, 2009|OBP|24|-0.60 [-3.55, 2.35]|\n|Hosseininasab, 2014|OBP|6|1.00 [-1.14, 3.14]|\n|Zha, 2020|OBP|6|-3.60 [-8.83, 1.63]|\n|Pooled estimate (I² = 58.8%)|BP was lowered in HBPMM group|BP was lowered in the usual care group|\n|Pooled estimate (I² = 0.0%)|BP was lowered in HBPMM group|BP was lowered in the usual care group|\n](1/parseimages/figuredbe1b0b3de1440719c457823ae185ebe_0_2.png) SBP: systolic blood pressure, BP: blood pressure, OBP: office blood pressure, ABP: ambulatory blood pressure, HBPM: home blood pressure monitoring

Fig. 6 Differences in the SBP change by the type of sphygmomanometer used. Results of stratified analysis of: (a) studies using upper arm cuff sphygmomanometers, (b) studies using wrist cuff sphygmomanometer. (Adapted from ref. [12])

The Japanese Society of Hypertension Guidelines for blood pressure control using digital technologies

a) Studies using upper arm cuff sphygmomanometers

![|First author, Year of publication|Intervention period (months)|Difference in DBP change between groups [95%CI] (mmHg)|\n|:---:|:---:|:---:|\n|Mehos, 2000|6 months|-6.70 [-19.22, 5.82]|\n|Rogers, 2001|3 months|-4.10 [-7.20, -1.00]|\n|Broege, 2001|3 months|1.00 [-32.91, 34.91]|\n|Rudd, 2004|6 months|3.10 [-6.12, -0.08]|\n|Halme, 2005|6 months|-1.50 [-3.95, 0.95]|\n|Zillich, 2005|3 months|-3.20 [-9.81, 3.41]|\n|Marquez-Contreras, 2006|6 months|-3.10 [-5.83, -0.37]|\n|Verberk, 2007|12 months|1.80 [-0.92, 4.52]|\n|Kauric-Klein, 2007|3 months|-1.87 [-8.87, 6.87]|\n|Artiawan, 2007|12 months|-0.80 [-4.79, 3.19]|\n|Tobe, 2008|1.5 months|-1.00 [-4.52, 2.52]|\n|Madsen, 2008|6 months|-0.80 [-3.05, 1.45]|\n|Green, 2008|6 months|-0.80 [-2.14, 0.54]|\n|Parati, 2009|6 months|0.40 [-2.23, 3.03]|\n|da Silva, 2009|6 months|-6.00 [-10.84, -1.16]|\n|DeJesus, 2009|6 months|1.58 [-3.68, 6.84]|\n|Rinnefret, 2009|12 months|-2.10 [-5.15, 0.95]|\n|Godwin, 2010|12 months|-1.20 [-3.37, 0.97]|\n|McManus, 2010|12 months|-2.60 [-4.23, -0.97]|\n|Varis, 2010|14 months|2.40 [-1.54, 6.34]|\n|Bosworth, 2011|18 months|-0.50 [-3.15, 2.15]|\n|Magid, 2011|6 months|-2.30 [-4.65, 0.05]|\n|Hebert, 2012|6 months|-0.70 [-4.60, 3.20]|\n|Fuchs, 2012|3 months|-2.30 [-5.70, 1.10]|\n|Piette, 2012|1.5 months|-3.20 [-6.80, 0.40]|\n|Williams, 2012|9 months|0.85 [-3.22, 4.92]|\n|Kerry, 2013|12 months|-1.40 [-4.83, 2.03]|\n|Magid, 2013|6 months|-5.70 [-7.80, -3.60]|\n|McKinstry, 2013|6 months|-2.17 [-3.55, -0.79]|\n|Margarolis, 2013|12 months|-5.10 [-7.40, -2.80]|\n|Ogedegbe, 2014|12 months|-0.70 [-2.09, 0.69]|\n|Stewart, 2014|6 months|-0.30 [-3.40, 2.80]|\n|McManus, 2014|12 months|-3.10 [-5.50, -0.70]|\n|Kim, 2014|12 months|-4.00 [-7.17, -0.83]|\n|Yi, 2015|9 months|-1.00 [-2.66, 0.66]|\n|Hanley, 2015|6 months|-3.20 [-10.44, 4.04]|\n|Aekplakorn, 2016|12 months|-1.20 [-4.45, 2.05]|\n|Kim, 2016|6 months|4.30 [-2.93, 11.53]|\n|Tzourio, 2017|24 months|-0.20 [-1.20, 0.80]|\n|Klarskov, 2018|10 months|-2.90 [-4.34, -1.46]|\n|Martinez, 2018|24 months|-2.80 [-6.90, 1.30]|\n|McManus, 2018|12 months|-1.20 [-3.40, 1.00]|\n|Pan, 2018|6 months|-3.00 [-7.92, 1.92]|\n|Skolarus, 2018|6 months|0.90 [-5.71, 7.51]|\n|Cuffee, 2019|3 months|-2.00 [-8.42, 4.42]|\n|Gu, 2020|12 months|-0.90 [-3.30, 1.50]|\n|Ojiji, 2020|1 months|-1.00 [-5.74, 3.74]|\n|lonov, 2021|3 months|-5.90 [-8.89, -2.91]|\n|McManus, 2021|12 months|-0.70 [-3.04, 1.64]|\n|Zhang, 2021|3 months|-0.20 [-1.95, 1.55]|\n|Aki, 2021|1.5 months|1.30 [-2.95, 5.55]|\n|Margarolis, 2022|12 months|0.30 [-0.95, 1.55]|\n|Okoro, 2022|12 months|-0.54 [-6.11, 5.03]|\n|Mujisers, 2022|12 months|-5.00 [-8.39, -1.61]|\n|Calderon-Anyosa, 2023|1 months|-6.00 [-11.26, -0.74]|\n|Doogue, 2023|3 months|-4.40 [-25.34, 16.54]|\n|Hoppe, 2023|12 months|1.27 [-2.06, 4.60]|\n|Leupold, 2023|12 months|-3.10 [-5.50, -0.70]|\n|Martinez-Ibanez, 2023|12 months|-1.10 [-3.19, 0.99]|\n|Nejami, 2023|3 months|0.38 [-2.02, 2.78]|\n|Ramos-Zavala, 2023|3 months|-4.50 [-7.50, -1.50]|\n|Pooled estimate (I² = 49.9%)| | |\n|Vetter, 2000|2 months|-1.30 [-2.88, 0.28]|\n|Bosworth, 2009|24 months|-1.20 [-2.85, 0.45]|\n|Hosseininasab, 2014|6 months|-1.00 [-3.39, 1.39]|\n|Zha, 2020|6 months|-0.56 [-9.20, 8.08]|\n|Pooled estimate (I² = 0.0%)| | -1.20 [-2.22, -0.17]|\n](1/parseimages/figure501ca45f99214b4cb84ca3a40e1c3723_0_2.png) Fig. 7 Differences in the DBP change by the type of sphygmomanometer used. Results of stratified analysis of: (a) studies using upper arm cuff sphygmomanometers, (b) studies using wrist cuff blood pressure monitors. (Adapted from ref. [12])

H. Arima et al.

Table 2 Benefits and points to note about blood pressure control with wearable devices

Benefits

|Convenience|Wearable devices can be monitored for extended periods, allowing the collection of biometric data such as blood pressure and heart rate throughout the day. This provides a detailed understanding of daily blood pressure fluctuations and patterns.| |---|---| |Usability and wearability|Many wearable devices are comfortable and designed for everyday use, making them easy for users to continuous use.| |Automatic data collection and tracking|Wearable devices can be linked to smartphones and tablets to automatically record, manage, and analyze data through dedicated applications and cloud services. This allows users to easily access their health data and share it with doctors or other healthcare providers as needed.| |Early detection|Some wearable devices can detect unusual changes and abnormal values and an increased likelihood of illness at an early stage by performing various analyses using artificial intelligence and other means. This enables the early prediction of health problems and the implementation of countermeasures.| |Points to note|| |Validity and reliability|The validity of a wearable device's blood pressure measurement function is highly dependent on its quality and accuracy. A product certified as a medical device is recommended for blood pressure measurement. Compliance with the operating conditions and wearing instructions is also essential.| |Collaboration with healthcare providers|Blood pressure data from wearable devices should be shared with physicians and other healthcare providers and used as reference information for the appropriate diagnosis and treatment planning.| |Data security and privacy|Health data is crucial personal information. When using wearable devices, you must ensure that appropriate safeguards are in place for your data.|

and DBP as the outcome between a group that wore a wearable device and a control group (no intervention). Pedometers accounted for 28.6%, and devices measuring activity accounted for 23.8%, about half the total. There was no study on devices that measure blood pressure. None of the studies involved interventions with healthcare provider feedback. All of these studies used OBP. There were no significant differences in SBP and DBP values at 12, 24, or 48 weeks of intervention.

Explanation

Wearable devices not only directly measure biometric information using embedded sensors but also monitor and track heart rate, pulse rate, sleep time, activity level, time spent standing and sitting, and number of steps taken, and provide feedback to encourage behavioral change. Through continuous data management and visualization, user's awareness of blood pressure control will be raised, promoting sustained lifestyle improvements and behavioral change. In particular, wearable blood pressure monitors can potentially increase the number of blood pressure measurements in various situations while minimizing the burden on the user and detecting blood pressure abnormalities that hurt the prognosis of cardiovascular diseases, such as masked hypertension and abnormal blood pressure fluctuations [13]. In addition, intervention with wearable devices may also lead to increased physical activity and stress management [14]. In a systematic review in 2022, 8 out of 15 wearable cuffless phymomanometers showed that the mean blood pressure bias was within 5 mmHg of the reference device. Although the validity of the measurement

accuracy has been demonstrated to some extent, its long-term sustainability has not been clarified, and issues such as cost remain [15].

(1) SBP

RCTs compared the use of a wearable device as an intervention with a control group (no intervention) in adults aged 18 years and older, with a change in SBP as the outcome. The lengths of the intervention periods were as follows: 10 trials were 12 weeks; 11 trials, 24 weeks; and 6 trials, 48 weeks. The changes in SBP values with the intervention compared to controls were -1.97 mmHg (95% CI -4.69 to 0.75) at 12 weeks, -1.57 mmHg at 24 weeks (95% CI -4.23 to 1.09), and -0.58 mmHg (95% CI -2.98 to 1.81) at 48 weeks, with no significant reduction in blood pressure observed during any intervention period (Fig. 8).

(2) DBP

RCTs were conducted in adults aged 18 years and older, comparing the use of a wearable device as an intervention with a control group, with changes in DBP as the outcome. The lengths of the intervention periods were as follows: 10 trials were 12 weeks; 9 trials, 24 weeks; and 4 trials, 48 weeks. The changes in DBP values with the intervention compared to controls were -0.03 mmHg (95% CI -1.44 to 1.38) at 12 weeks, -0.25 mmHg (95% CI -2.56 to 2.05) at 24 weeks, and -1.39 mmHg (95% CI -2.84 to 0.07) at 48 weeks. No significant reduction in blood pressure was observed during any intervention period (Fig. 9).

The Japanese Society of Hypertension Guidelines for blood pressure control using digital technologies


(3) Weight

RCTs compared the use of a wearable device as an intervention with a control group in adults aged 18 years and older, with a change in body weight as the outcome. The lengths of the intervention periods were as follows: 6 trials

were 12 weeks; 9 trials, 24 weeks; and 4 trials, 48 weeks. The changes in weight due to the intervention compared to controls were -0.89 kg (95% CI -1.82 to 0.04) at 12 weeks, -0.86 kg (95% CI -1.99 to 0.28) at 24 weeks, -0.07 kg (95% CI -0.47 to 0.60) at 48 weeks, with no significant reduction in any of the intervention period.

H. Arima et al.

![|First author, Year of publication|Difference in DBP change between groups (mmHg)|\n|:---:|:---:|\n|Engel, 2006|0.40 [ -4.34, 5.14]|\n|Kolt, 2012|-0.50 [ -5.94, 4.94]|\n|Petrella, 2014|2.40 [ -2.52, 7.32]|\n|Frias, 2017|-3.10 [ -9.76, 3.56]|\n|Kempf, 2018|1.00 [ -2.92, 4.92]|\n|Maxwell-Smith, 2019|-4.21 [ -12.33, 3.91]|\n|Romero, 2019|-7.75 [ -17.89, 2.39]|\n|Gu, 2020|1.30 [ -2.73, 5.33]|\n|Pelletier, 2021|1.00 [ -8.51, 10.51]|\n|Gómez-Sánchez, 2022|-0.31 [ -2.55, 1.93]|\n|Pooled estimate(² = 0%)|-0.03 [ -1.44, 1.38]|\n](1/parseimages/figure66b5de47c74e4758887fd843ebdf9496_0_1.png)

(4) Fasting blood glucose

RCTs were conducted in adults aged 18 years and older, comparing the use of a wearable device as an intervention with a control group, with changes in fasting blood glucose as the outcome. The lengths of the intervention periods were as follows: 2 trials were 12 weeks; and 3 trials, 24 weeks (there was no applicable 48-week study). The changes in fasting blood glucose with the intervention compared to control were

-1.24 mg/dL (95% CI -8.40 to 5.92) at 12 weeks and -3.19 mg/dL (95% CI -7.36 to 0.97) at 24 weeks, with no significant reduction in either intervention period.

(5) HbA1c

RCTs were conducted in adults aged 18 years and older, comparing the use of a wearable device as an intervention with a control group, with changes in HbA1c as the

The Japanese Society of Hypertension Guidelines for blood pressure control using digital technologies

outcome. The lengths of the intervention periods were as follows: 5 trials were 12 weeks; and 3 trials, 24 weeks (there was no applicable 48-week study). The changes in HbA1c levels with the intervention were -0.00 % (95% CI -0.07 to 0.08) at 12 weeks and -0.08 % (95% CI -0.20 to 0.04) at 24 weeks, compared to controls. There was no significant reduction in either of the intervention periods.

Summary

Using the wearable device did not significantly affect SBP or DBP for the 12, 24, or 48-week intervention periods compared to the control group.

Literature search flow

A literature search was conducted using PubMed, ICHUSHI, and Cochrane Library, and 3940 studies were selected for the primary screening (title and abstract), followed by the secondary screening (full text) of 159 studies, excluding 3781 that met exclusion criteria. A meta-analysis was performed using 21 studies applicable to this HQ [16].

Recommendation decision (voting results)

The systematic review showed that the wearable device did not significantly differ from the control group in either SBP or DBP, with the primary outcomes at 12, 24, or 48 weeks of intervention. There were also no significant differences in other outcomes, such as changes in body weight, in fasting blood glucose, and in HbA1c, between the wearable device group and the control group at 12 and 24 weeks of intervention, and no significant difference in weight change at 48 weeks of intervention. Based on the certainty of the evidence, the balance of benefits and harms, public values, and cost-effectiveness, we decided that “evidence for various wearable device interventions in adults is insufficient, and we withhold our recommendation”.

The 37 votes cast included 3 "strongly recommended action", 8 "suggested action", 1 "action not suggested", zero "action strongly discouraged", 25 "recommendation/sugestion withheld due to insufficient evidence" (67.6% of valid votes), and zero abstentions.

Future research question

(1) Does an intervention using various wearable devices and feedback from healthcare providers positively affect blood pressure in adults?

We hope that future research will answer the question, "Does an intervention using various wearable devices and

feedback from healthcare providers have a beneficial effect on blood pressure in adults?" The results of this systematic review indicate that the intervention effect of using a wearable device alone does not result in significant changes in blood pressure levels. In addition to the direct intervention effects of using the wearable device, indirect effects of interventions, such as feedback from the healthcare provider on the evaluation of the collected data, behavioral change promotion, and applying the gamification (applying design elements and principles of computer games to non-game activities) also require further investigation.

3 Devices measuring urinary sodium/potassium ratio, or dietary or urinary sodium concentration

Overview of digital technology

Reducing salt (sodium) intake and increasing potassium intake are very important in managing and preventing hypertension. However, it has been challenging to objectively determine an individual's salt intake, which has been a major obstacle to salt reduction guidance. Since most dietary sodium and potassium are excreted in the urine, devices that measure the urinary sodium/potassium (Na/K) ratio or urinary sodium concentration have recently been developed. Their effectiveness in blood pressure control has been studied.

Devices that measure the sodium concentration of meals (dishes) have also been used for salt reduction guidance. One can estimate an individual's saltiness preferences and salt intake by measuring the sodium concentration of miso soup, cooking soup, broths, etc.

These digital technologies are intended to assess the quality of an individual's diet, not to diagnose disease. In addition, urine-based tests are noninvasive and can be used for general home, community, and workplace health management.

In contrast, devices that estimate salt intake based on the number of seasonings used, devices that estimate vegetable intake based on skin carotenoid levels, and devices that use electricity to make tableware taste salty have also been developed. Still, those effects on blood pressure control have not yet been verified, and further evidence is needed (Table 3).

Healthcare question

$\blacksquare$ HQ 3

Do interventions using devices that measure the urinary sodium/potassium ratio, or dietary or urinary sodium concentration reduce blood pressure in adults?

H. Arima et al.

Table 3 Benefits and points to note about blood pressure control with devices measuring urinary sodium/potassium ratio, or dietary or urinary sodium concentration

$\blacksquare$ Benefits

Objective evaluation of dietary habits by urinary level analysis

Objective evaluation of meals (dishes)

Real-time presentation of measurement results

Non-invasive measurements

$\blacksquare$ Points to note

Handling of urine specimens

Interpretation of data and healthcare providers

Measuring the urinary sodium/potassium ratio or sodium concentration objectively assesses an individual's diet and salt and potassium intake.

You can evaluate individual saltiness preferences by measuring the sodium concentration of miso soup, broth, etc.

These devices measure urinary sodium/potassium ratios and urinary sodium concentrations in real time and provide results on the spot.

Urine and dietary (food) measurements are non-invasive and can be used in general home, community, and workplace health management. Since these are not used to diagnose diseases, manufacturers are not limited to healthcare providers (such as cardiologists, public health nurses, nurses, pharmacists, dietitians, hypertension specialists, etc.).

Individuals can handle urine specimens at home, but community and work settings require the involvement of health care providers. Proper device calibration and training are also needed to ensure accurate measurements.

Although the tests are not intended to diagnose disease, advice should be sought from a healthcare provider when interpreting the data.

$\blacksquare$ Recommendation

In adults, we suggest interventions using devices that measure the urinary sodium/potassium ratio, or dietary or urinary sodium concentration, especially when combined with educational programs.

Summary of evidence

A literature search identified eight RCTs (1442 individuals in total; adults aged ≥18 years), which compared the differences in blood pressure, urinary sodium/potassium ratio, salt intake (including differences in urinary sodium excretion), potassium intake (including differences in urinary potassium excretion) with the use of health management, nutritional and dietary guidance, and clinical practice, to the control group (no intervention). The meta-analysis showed significant reductions in SBP, the urinary sodium/potassium ratio, and salt intake (including urinary sodium excretion).

Explanation

The relationship between salt (sodium) intake and blood pressure has been reported in the past, and the blood pressure-lowering effect of salt reduction (sodium intake restriction) has been confirmed in many RCTs [17, 18]. It has also been confirmed that high salt intake increases the risk of future cardiovascular disease [19], and many domestic and international guidelines indicate the target amount of salt intake and the need for salt reduction [20–22]. On the other hand, there is also much evidence that potassium, which is abundant in fruits and vegetables, has a blood pressure-lowering effect [17, 23, 24], and increasing

potassium intake is recommended to prevent hypertension (however, some patients with chronic kidney disease may require potassium restriction, so caution is advised) [20, 22]. In addition, it has recently been reported that the dietary sodium/potassium ratio or urinary sodium/potassium ratio is often associated with hypertension and future risk of cerebrovascular disease [19, 25, 26].

Since it is challenging to determine salt (sodium) intake from dietary surveys, the most reasonable estimate has been based on 24-h urinary sodium excretion. Potassium intake is also reflected mainly in urinary potassium excretion. However, since it is difficult to obtain the value with 24-h urine storage, spot or nocturnal urine sodium and potassium concentrations have also been used. In addition, in recent years, utilization of the sodium/potassium ratio of spot urine has been promoted, and the Japanese Society of Hyper-tension has announced a target value for the urinary sodium/potassium ratio in Japanese people in 2024 [27]. In addition, guidance on salt reduction has also been provided by measuring sodium concentrations in foods such as soups and juices.

This has led to the development of devices that measure urinary sodium/potassium ratios, or dietary or urinary sodium concentrations, and research has been conducted to clarify the effects of health guidance using these devices on reducing salt intake, increasing potassium intake, and even lowering blood pressure.

A literature search identified 8 RCTs (6 from Japan, 2 from other countries). [28–35] Of the 8 studies, one used a device to measure urinary sodium/potassium ratio [28], 5 used a device to measure urinary sodium concentration to estimate excretion [29–31, 33, 34], and 2 used a device to measure sodium concentration in food (soup, broth, etc.)

The Japanese Society of Hypertension Guidelines for blood pressure control using digital technologies

 [32, 35]. The results of the meta-analysis [36] of these eight studies are shown below.

(1) Blood pressure

All eight RCTs analyzed the effect on blood pressure reduction. The meta-analysis showed that the changes in SBP values with the intervention compared to controls were -2.45 mmHg (95% CI -4.86 to -0.04), with a statistically significant reduction. Still, the variability between studies (heterogeneity) was substantial (Fig. 10) [36]. In addition, a possible publication bias was suggested for the effect on SBP.

There was no statistically significant reduction in DBP, and there was greater variability (heterogeneity) between studies (Fig. 11) [36].

The result of the sub-analysis showed a significant reduction in SBP in three studies with a longer intervention period of 2–6 months than in the five studies of 4 weeks. In addition, three studies that added an educational program to the device measurement showed a significant reduction in SBP compared to five studies with only device measurement.

(2) Urinary sodium/potassium ratio

Three studies examined the effect of lowering the urinary sodium/potassium ratio (Fig. 12) [36]. The changes in urinary sodium/potassium ratio with the intervention were -0.60 (95% CI -0.97 to -0.23) compared to the control group, a statistically significant reduction.

(3) Urinary sodium excretion

Five studies examined the effect of lowering urinary sodium excretion (Fig. 13) [36]. The changes in urinary sodium excretion with the intervention were -525.50 mg/day (95% CI -785.35 to -265.66) compared to the control group, a statistically significant reduction.

(4) Urinary potassium excretion

Three studies examined the effect of the intervention on increasing urinary potassium excretion (Fig. 14) [36]. The increase in urinary potassium excretion with the intervention was 144.03 mg/day (95% CI -17.61 to 305.67)

H. Arima et al.

Fig. 12 Differences in urinary sodium/potassium ratio change with devices measuring urinary sodium/potassium ratio, or dietary or urinary sodium concentration. (Adapted from ref. [20])

 Fig. 13 Differences in urinary sodium excretion change with devices measuring urinary sodium/potassium ratio, or dietary or urinary sodium concentration. (Adapted from ref. [20])

 Fig. 14 Differences in urinary potassium excretion change with devices measuring urinary sodium/potassium ratio or dietary urinary sodium concentration. (Adapted from ref. [20])

 compared to the control group, which tended to increase but was not statistically significant.

Summary

Health management, nutritional dietary guidance, and clinical practice utilizing devices that measure urinary sodium/potassium ratio, or dietary or urinary sodium concentrations were shown to be more helpful in lowering SBP, urinary sodium/potassium ratio, and salt intake (including urinary sodium excretion) than controls who did not use such devices. The effect on SBP was more

substantial in long-term intervention studies of ≥2 months, and in interventions that added an educational program to using the device. However, the number of studies is small, and the variability (heterogeneity) between studies tends to be large; more evidence needs to be accumulated in the future.

Literature search flow

A literature search was conducted in PubMed, ICHUSHI, and Cochrane Library, and 1525 studies were selected for the primary screening (title and abstract). This was followed

The Japanese Society of Hypertension Guidelines for blood pressure control using digital technologies

by the secondary screening (full text) of 80 studies, excluding 1445 that met the exclusion criteria. A meta-analysis was conducted using the 8 studies that met this HQ.

Recommendation decision (voting results)

The systematic review showed that the devices helped lower SBP, urinary sodium/potassium ratio, and salt intake (including urinary sodium excretion). Still, the overall number of studies was small, and there was no evidence on the long-term effects. Based on the certainty of the evidence, the balance of benefits and harms, the public's perception of value, and cost-effectiveness, the committee suggested "interventions using devices measuring urinary sodium/potassium ratio, or dietary or urinary sodium concentration in adults, mainly when used in conjunction with educational programs".

The 38 votes cast included one "strongly recommended action", 33 "suggested action" (86.8% of the valid votes), 4 "action not suggested", zero "action strongly discouraged", zero "recommendation/suggestion withheld due to insufficient evidence", and zero abstentions.

Future research questions

(1) To confirm these findings, more studies are necessary due to the small number of studies and subjects, and the sizeable inter-study variability (heterogeneity). (2) Since there has been only one study conducted using a device measuring urine sodium/potassium ratio, more intervention studies using this device are needed.

(3) Most of the studies cited here were conducted in Japan; studies on diverse populations outside Japan are needed to generalize the evidence.

(3) Most of the studies cited here were conducted in Japan; studies on diverse populations outside Japan are needed to generalize the evidence.

(4) Since more than half of the patients received the intervention for only 4 weeks, and the most extended intervention period was 6 months, more studies with longer intervention periods are needed to examine the effects.

(5) In addition to the sodium measurement, the implementation of educational programs tended to improve the effectiveness of the intervention. Further research is needed to optimize the contents to achieve more effective educational programs.

(6) Numerous devices have been developed related to salt in the diet. Some estimate salt intake based on the amount of seasonings used, or the vegetable intake based on skin carotenoid levels, and also, some use electricity to make dishes taste salty. Still, the effectiveness of these devices must be verified in the future.

4 Smartphone applications for blood pressure control

1. Overview of digital technology

The widespread adoption of smartphones has led to the development of numerous applications designed to support blood pressure control. These applications enable users to record their blood pressure anytime and anywhere, analyze trends over time, and visualize changes through graphical displays. Additionally, some applications facilitate data sharing with healthcare providers, including certified cardiovascular disease prevention educators—such as public health nurses, nurses, pharmacists, registered dietitians, and hypertension specialists—allowing for professional feedback and enhanced communication. Furthermore, certain applications incorporate features aimed at promoting behavioral change through education and motivational support.

While smartphone applications provide several advantages for blood pressure control, it is important to be aware of their points to note (Table 4). Selecting a suitable application and using it appropriately are essential for ensuring safe and effective blood pressure control.

Healthcare question

$\blacksquare$ HQ 4

Does a smartphone application-based intervention for blood pressure control lower blood pressure in adults?

$\blacksquare$ Recommendation

We suggest a smartphone application-based intervention for blood pressure control in adults (normotensive and hypertensive individuals).

However, evidence regarding the long-term effects (beyond 6 months) remains insufficient.

Summary of evidence

A literature search identified 69 RCTs and 30 observational studies assessing the efficacy of smartphone application-based interventions on blood pressure involving a total of 78,084 participants (adults ≥ 18 years). Of these, 76 trials (46,459 participants) were included in the meta-analysis, which showed a significant reduction in office SBP (-2.76 mmHg) after 6 months of intervention. In healthy individuals, a reduction in SBP was observed at 3 months; however, no effect was found beyond 6 months. In hypertensive patients, office SBP was decreased at both 3 and 6 months, but no significant effect was shown at 12 months. When examining the functionality of smartphone applications, those that used a wireless connection with a sphygmomanometer to obtain accurate blood pressure readings showed a greater blood pressure lowerlingeffect.

H. Arima et al.

Table 4 Benefits and points to note about blood pressure control with smartphone applications

$\blacksquare$ Benefits

Convenience and accessibility

Automatic data collection and tracking

Data analysis and visualization

Data sharing with healthcare providers

Facilitating behavioral change

$\blacksquare$ Points to note

Data validity and reliability

Privacy and security

Collaboration with healthcare providers

Evolution of technology

Users can record their blood pressure and review collected data anytime, anywhere.

Some smartphone applications can sync with blood pressure monitors and wearable devices via Bluetooth or other wireless connections, enabling automatic data collection. This feature improves the accuracy of daily tracking.

Some smartphone applications allow users to analyze their blood pressure data, displaying average values and seasonal variations. These visual representations make it easier to identify trends and patterns over time.

Certain smartphone applications enable users to share their blood pressure data with healthcare providers, facilitating timely feedback, diagnosis, and treatment.

Some smartphone applications incorporate educational resources and motivational tools to encourage healthy lifestyle changes.

It is essential to verify the validity of both the smartphone application and any devices it links to. Collecting inaccurate data may lead to inappropriate recommendations or guidance.

It is crucial to select a smartphone application with strong security measures to protect sensitive personal information, including medical history and test results.

Users are encouraged to consult a healthcare provider with expertise in blood pressure control, if the application allows for data sharing to receive informed advice.

As digital technologies continue to evolve, it is essential to select smartphone applications based on the most current and reliable information available.

Explanation

This summary presents the findings of a systematic review on smartphone applications for blood pressure control [37]. It is recommended that smartphone application-based interventions for blood pressure control be conducted under the supervision of a healthcare provider with expertise in blood pressure control, for both healthy individuals and patients with hypertension.

(1) Blood pressure

A total of 24 RCTs and 3 observational studies reported OBP outcomes at 6 months post-intervention. The result of meta-analysis (Fig. 15) including observational studies showed significant changes in office SBP of -2.76 mmHg (95% CI -3.94 to -1.58) at 6 months in the smartphone application group compared to the control group. However, there was substantial heterogeneity between trials ($l^2 = 79.2\%$). Similarly, changes in office DBP were -1.23 mmHg (95% CI -1.80 to -0.67), with substantial heterogeneity ($l^2 = 61.3\%$). The high degree of heterogeneity may be attributed to variations in study populations and the diverse functionalities incorporated into the smartphone applications.

(2) Stratified analysis by intervention duration in normotensive individuals and hypertensive patients

A stratified analysis was conducted to assess the effect of smartphone application-based interventions on blood pressure in normotensive individuals and hypertensive patients,

categorized by intervention duration (Fig. 16). In healthy individuals, changes in office SBP were -4.44 mmHg (95% CI -6.96 to -1.92) with significant reduction after 3 months of intervention. However, no significant effect was observed beyond 6 months. Among hypertensive patients, changes in office SBP were -7.71 mmHg (95% CI -10.63 to -4.79) at 3 months and -1.88 mmHg (95% CI -3.41 to -0.35) at 6 months with significant reductions. However, no significant effect was observed at 12 months or beyond.

(3) Stratified analysis of smartphone applications by function

A stratified analysis was conducted to assess the effect of smartphone application-based interventions on blood pressure based on their functionality (Fig. 17). No clear difference in blood pressure reduction was observed among applications based on features such as blood pressure recording, educational content, feedback, or communication with healthcare providers or other users. Meanwhile, smartphone applications that obtained accurate blood pressure readings via wireless connection to an upper arm cuff sphygmomanometer demonstrated a greater reduction compared to those without wireless connectivity (−5.42 mmHg vs. −1.85 mmHg, interaction P = 0.047).

(4) Cost-effectiveness

A comprehensive cost-effectiveness analysis incorporating all the articles extracted from the systematic review was not

The Japanese Society of Hypertension Guidelines for blood pressure control using digital technologies

Fig. 15 Differences in office SBP change after 6 months of smartphone application-based intervention. (Adapted from ref. [1])

![Fig. 15 Differences in office SBP change after 6 months of smartphone application-based intervention. (Adapted from ref.[1])\n#First author,\nYear of publication\nDifference in SBP change between groups\n(95%CI) (mmHg)\nWang, 2023\nWong, 2023\nPletcher, 2022\nSakboonyarat, 2022\nXia, 2022\nZhang, 2022\nKario, 2021a\nKario, 2021b\nLi, 2021\nLim, 2021a\nLim, 2021b\nCho, 2020\nLee, 2020\nLim, 2020\nTakeyama, 2020\nWang, 2020\nYu, 2020\nApiñaniz, 2019\nChoi, 2019\nContreras, 2019\nZha, 2019\nMaddison, 2019\nPan, 2018\nRosario, 2018\nSenecal, 2018\nKim, 2016\nRossi, 2010\nPooled estimate (P=79.2%)\n](1/parseimages/figure2f65d5165c1f4b2bbddb695332d10901_0_2.png) feasible due to the limited number of studies reporting cost effectiveness. However, one selected RCT conducted a cost-effectiveness evaluation, estimating that the cost per quality-adjusted life year (QALY) gained through the use of the smartphone application in the study was ~1.2 million yen [38]. This suggests that an intervention costing 1.2 million yen can provide one additional year of life in perfect health for an individual. However, it remains uncertain whether this finding is applicable to all smartphone applications for blood pressure control.

(5) Safety

The analysis did not find an increase in adverse events associated with the use of a smartphone application compared to the control group.

Summary

Smartphone application-based interventions were associated with significantly larger reduction in office SBP (−2.76 mmHg) after 6 months compared to the control group. In healthy individuals, a reduction in office SBP was

observed at 3 months, but no sustained effect was found beyond 6 months. Among hypertensive patients, office SBP was reduced at both 3 and 6 months, but no effect was shown at 12 months. Stratified analysis by application functionality suggested that smartphone applications with wireless blood pressure cuff connectivity, which provide accurate blood pressure data, had a greater effect on blood pressure reduction.

Literature search flow

A literature search was conducted in PubMed, ICHUSHI, and Cochrane Library, identifying 7385 studies for primary screening (title and abstract review). After excluding 6244 based on exclusion criteria, 1141 studies underwent secondary screening (full-text screening). Ultimately, 99 studies met the inclusion criteria, and 76 were eligible for meta-analysis.

Recommendation decision (voting results)

The results of the systematic review suggest that smart-phone applications can help lower blood pressure, though evidence of long-term benefits remains limited. Considering

H. Arima et al.

Fig. 16 Differences in office SBP change after smartphone application-based intervention (per participant). (Adapted from ref. [1])

 Fig. 17 Differences in office SBP change after 6 months of smartphone application-based intervention (by application function). (Adapted from ref. [1])

 the certainty of the evidence, the balance of benefits and harms, public values, and cost-effectiveness, we suggest a smartphone application-based intervention for blood pressure control in adults (both normotensive and hypertensive individuals). However, evidence supporting the long-term effectiveness (beyond 6 months) remains insufficient.

Among the 36 votes cast, 4 voted "strongly recommended action", 31 voted "suggested action" (86.1% of valid votes), one voted "action not suggested", and none voted "action strongly discouraged" or "recommendation/suggestion withheld due to insufficient evidence", and zero abstentions.

Future research questions

(1) Does long-term intervention using smartphone applications lead to sustained reductions in blood pressure? In this study, the blood pressure-lowering effect of the smartphone application-based intervention was demonstrated for up to 3 months in healthy individuals and up to 6 months in hypertensive patients (Fig. 16). Further clinical studies are needed to

In this study, the blood pressure-lowering effect of the smartphone application-based intervention was demonstrated for up to 3 months in healthy individuals and up to 6 months in hypertensive patients (Fig. 16). Further clinical studies are needed to

examine the long-term effects, beyond 6 months in healthy individuals and 12 months in hypertensive patients.

(2) Does the function of smartphone applications affect the antihypertensive effect?

Smartphone applications that provide accurate blood pressure values through a wireless connection with an upper arm cuff sphygmomanometer tended to show a more significant antihypertensive effect compared to those without this functionality (Fig. 17). Therefore, incorporating the ability to obtain accurate blood pressure measurements via a wireless connection with a brachial cuff sphygmomanometer is desirable to enhance the effectiveness of these applications.

(3) Will there be a digital literacy divide?

As digital technologies continue to spread, the “digital divide” — the gap between those who can use digital technology and those who cannot — has become an important issue. When developing smartphone applications for blood pressure control, it is essential to design them to be user-friendly and accessible, especially for individuals with low digital literacy.

The Japanese Society of Hypertension Guidelines for blood pressure control using digital technologies

Table 5 Benefits and points to note about blood pressure control through telemedicine and health guidance using digital technology

$\blacksquare$ Benefits

|Improved access to healthcare|Patients can communicate interactively with health care providers (e.g., cardiologists, health professionals, nurses, pharmacists, dietitians, hypertension specialists, etc.) from home or remote areas, even with geographical, mobility, or time constraints. This can improve access to healthcare and regular blood pressure control.| |---|---| |Real-time data collection|Using digital technology, telemedicine and health guidance systems can collect real-time data through automated blood pressure monitoring devices and other healthcare devices. Healthcare providers will be able to view the data online. Healthcare providers can provide health guidance based on the collected data, and physicians can diagnose and treat patients.| |Cost reduction|The system is expected to reduce the frequency of regular visits to healthcare facilities of patients with chronic diseases, thereby reducing health expenditures and patient transportation expenses. This can be a great advantage, especially for patients in remote areas, such as remote islands or mountainous regions where access to healthcare is geographically difficult.| |Points to note|| |Diffusion and availability of technology|The prevalence of telemedicine and health guidance systems based on digital technology depends on the Internet environment, geographical location, and economic factors. Not all patients have access to digital technology, and challenges remain in introducing telemedicine and health guidance using digital technology, especially for the elderly and those unfamiliar with technology. However, as time goes on, it is expected that more and more older adults will be able to use digital technology.| |Data security and privacy|When conducting telemedicine and health guidance using digital technology and handling medical information, sufficient measures must be taken to ensure data security and privacy protection. To date, the degree of confidentiality of personal information for each individual is not yet organized in our society; these data security issues will have to be addressed.| |The importance of face-to-face healthcare|While telemedicine/health guidance using digital technology is highly convenient, face-to-face consultation with a physician is essential, especially in cases where the diagnosis and treatment plan are complicated or in conditions that may require acute changes. Therefore, it is essential that not only users but also their families and others around them, as well as digital technology providers and society, correctly understand the scope of telemedicine and healthcare services that function as necessary support through digital technology.|

(4) Does the function of providing personalized interventions and personalized feedback using artificial intelligence (AI) have any benefits?

Currently, there are not enough smartphone applications that incorporate personalized interventions using AI and other technologies. In the future, it would be beneficial to integrate AI-driven features that offer tailored interventions based on individual characteristics and blood pressure data. These features could prioritize lifestyle habits that need improvements, utilizing education and incentives, and provide personalized feedback on daily blood pressure monitoring results.

5 Digital health technologies for telemedicine and health guidance for blood pressure control

Overview of digital technology

Telemedicine and health guidance using digital technologies, such as smartphone applications and short messages for blood pressure control, have several advantages, and with recent technological advances, more effective methods are

being developed [39–43]. In particular, it can be a convenient and helpful method for patients in areas with limited access to healthcare facilities, patients who have difficulty traveling, and older adults. At the same time, however, some points must be kept in mind: appropriate selection and use of the technology must be ensured, and collaboration with the healthcare setting is necessary (Table 5).

Healthcare question

$\blacksquare$ HQ 5

Do telemedicine and health guidance using digital technology have a beneficial effect on blood pressure in adults?

$\blacksquare$ Recommendation

We suggest the use of telemedicine and health guidance using digital technology for blood pressure control in adults (normotensive and hypertensive individuals).

Summary of evidence

We conducted a systematic review of RCTs and observational studies of the usefulness of telemedicine and health guidance utilizing digital means, such as smartphone applications and short message services (hereafter referred

H. Arima et al.

to as “digital technology”), on blood pressure in adults aged 18 years and older [6]. A total of 117 studies, including 68,677 individuals (from 116 RCTs and one quasi-RCT) were extracted, which compared the change in blood pressure from baseline between the telemedicine/healthcare guidance group using digital technology and the control group (no intervention) and followed for more than 3 months. Many telemedicine and health guidance groups that utilized digital technology were hybrids of the face-to-face response. A meta-analysis of 38 of these studies, for which data were available after 3 months of intervention, showed a significant changes in office SBP of -3.21 mmHg. The change in blood pressure by intervention duration [3 months, 6 months (n = 54), 12 months (n = 43), and more than 12 months (n = 9)] was similar. By intervention modality [smartphone application (n = 38), text message intervention (n = 35), and PC or other web-based intervention (n = 34)] and study site type [healthcare facility (n = 74), non-healthcare facility (n = 33)], all groups showed a significant increase in telemedicine/healthcare using digital technology. The effect of telemedicine and health guidance intervention using digital technology was substantial in both groups, and there were no differences in the amount of change between intervention modalities or types of study sites. On the other hand, the change in blood pressure was more significant in hypertensive individuals (n = 41) than in non-hypertensive individuals (n = 66) [44].

Explanation

The results of a systematic review and meta-analysis of telemedicine and health guidance using digital technology for blood pressure control are listed below. It is desirable that these interventions for blood pressure control in hypertensive patients and healthy individuals be conducted under the supervision of a healthcare provider with accurate blood pressure knowledge and experience in treatment and health guidance.

(1) By intervention period

Although there was a statistically significant reduction in office SBP (-3.21 mmHg, 95% CI -4.51 to -1.90) after 3 months of intervention, there was significant heterogeneity with an $I^2$ of 75%. The same was valid for office DBP. Office DBP similarly decreased significantly (-1.47 mmHg, 95% CI -2.18 to -0.77), $I^2$ was 48%. The high levels of heterogeneity across the studies may be due to heterogeneous subject background, various intervention modalities, etc. Stratified analysis by intervention duration (3 months, 6 months, 12 months, and more than 12 months) revealed a significant decrease in office SBP regardless of intervention duration, with no difference in the amount of change (Fig. 18) [44].


(2) By intervention modality

In the telemedicine and health guidance group using the smartphone application, compared to the control group, changes in office SBP were -3.01 mmHg (95% CI -4.40 to -1.61) with significant reduction, but the I$^{2}$ was 83%. Office DBP also decreased significantly (-1.32 mmHg, 95% CI -2.02 to -0.62), with an I$^{2}$ of 70%. In the telemedicine and health guidance group using text messages, office SBP decreased significantly (-2.78 mmHg, 95% CI -3.87 to -1.69) compared to the usual healthcare group. Still, there was large heterogeneity (I$^{2}$ = 67%). Office DBP decreased significantly (-1.69 mmHg, 95% CI -2.73 to -0.67), I$^{2}$ was 78%. Furthermore, the office SBP was significantly decreased (-3.87 mmHg, 95% CI -5.18 to -2.56) in the telemedicine and health guidance group using a web screen (such as a personal computer), compared to the control group, but this was also with large heterogeneity (I$^{2}$ = 83%). Office DBP was significantly decreased (-1.95 mmHg, 95% CI -3.15 to -0.76), with I$^{2}$ of 86%. The large heterogeneous is thought to be due to subject characteristics and the diversity of combinations of functions added to the intervention modality. The results of the stratified analysis examined by intervention modality are shown below (Fig. 19) [44].

(3) Hypertensive and non-hypertensive individuals

The results of the stratified analysis of intervention effects in hypertensive ($n=41$) and non-hypertensive individuals ($n=66$) are shown in Fig. 20. In hypertensive individuals, the intervention significantly reduced office SBP ($-4.81$ mmHg, 95% CI -6.33 to -3.29). In non-hypertensive individuals, office SBP was also significantly reduced ($-2.17$ mmHg, 95% CI -3.15 to -1.19), but the change was significantly greater in the hypertensive individuals. The results for office DBP were similar [44].

The Japanese Society of Hypertension Guidelines for blood pressure control using digital technologies

Fig. 19 Differences in office SBP change by intervention modality for telemonitoring and health guidance using digital technology. (Adapted from ref. [6])

  Fig. 20 Differences in office SBP change in hypertensive and non-hypertensive cohorts in telemonitoring and health guidance using digital technology. (Adapted from ref. [6])

(4) Healthcare and non-healthcare Facilities

The results of the stratified analysis of intervention effects in healthcare (n = 33) and non-healthcare facilities (n = 74) are shown in Fig. 21. Healthcare facilities included hospitals and clinics, while non-healthcare facilities included community facilities such as community centers and health centers, workplace facilities, and research facilities, such as universities and research institutes. In the studies conducted at healthcare facilities, the intervention significantly reduced office SBP (-3.64 mmHg, 95% CI -4.63 to -2.65). In studies conducted at non-healthcare facilities, the intervention also reduced office SBP (-2.51 mmHg, 95% CI -3.86 to -1.15). There was no difference in the change between centers. The results for office DBP were similar [44].

(5) HBP

Of the trials included in the meta-analysis, 12 of them analyzed HBP. After the intervention, home SBP was significantly reduced (-4.27 mmHg, 95% CI -5.64 to -2.90) and heterogeneity was acceptable ($I^2$ = 23%). Similarly, home DBP was significantly reduced (-1.47 mmHg, 95% CI -2.33 to -0.62), and the variability (heterogeneity) was acceptable $I^2$ of 21% [44].

(6) Ambulatory blood pressure

In seven studies, 24-h blood pressure was analyzed by ambulatory blood pressure monitoring. After the

Fig. 21 Differences in office SBP change in healthcare and non-healthcare facilities in telemonitoring and health guidance using digital technology. (Adapted from ref. [6]) intervention, 24-h SBP was significantly reduced (−3.64 mmHg, 95% CI −6.99 to −0.29), but with considerable heterogeneity ($I^2$ = 90%). The 24-h DBP was also significantly reduced (-2.06 mmHg, 95% CI −3.33 to −0.78), with $I^2$ of 68% [44].

(7) Safety

We found no evidence that telemedicine and health guidance using digital technology increased adverse events compared to usual healthcare [44].

Summary

Telemedicine and health guidance interventions utilizing digital technology have been shown to significantly reduce blood pressure in adults without increasing adverse events. The results do not indicate that telemedicine and health guidance interventions using digital technology are more beneficial for blood pressure control than usual care. Results of the stratified analysis showed that the change in blood pressure with the intervention was significant regardless of intervention duration (3, 6, 12, or more than 12 months), intervention modality (smartphone application, text message, or use of a web screen such as PC), and study site type (healthcare or non-healthcare facility). On the other hand, the intervention produced more considerable changes in blood pressure in hypertensive individuals than in non-hypertensive individuals.

H. Arima et al.

Literature search flow

A literature search was conducted in PubMed, Cochrane Library, and ICHUSHI to identify 8,090 studies for the primary screening (title and abstract). The secondary screening (full text) of 352 studies was conducted, excluding 7,738 studies that met the exclusion criteria, and a meta-analysis of 117 studies applicable to this HQ was conducted [44].

Recommendation decision (voting results)

The results of systematic review and meta-analysis showed that telemedicine and health guidance using digital technology effectively lowered blood pressure in the office, at home, and over 24 h. However, the study also showed that many of the intervention arms were hybrids of face-to-face care and the use of digital technology, that there was high heterogeneity, that there was little evidence of long-term effects, and that the various intervention modalities could not be categorized. Based on the certainty of the evidence, the balance of benefits and harms, the values of citizens, cost-effectiveness, and the current status of telemedicine and health guidance in Japan, we suggest "telemedicine and health guidance using digital technology for blood pressure control in adults (nor-motensive and hypertensive individuals)".

The 36 votes cast included 8 for "strongly recommend action" and 25 for "suggested action" (71% of the valid votes), for a total of 94%. Two respondents "action not suggested", zero "action strongly discouraged", zero "recommendation/suggestion withheld due to insufficient evidence", and one abstention.

Future research questions

(1) Does long-term intervention through telemedicine and health guidance using digital technology sustainably lower blood pressure?

At present, the antihypertensive effects of telemedicine and health guidance interventions using digital technology are mainly demonstrated for up to 12 months. It is expected to accumulate evidence from future interventions and observational studies examining the effects of interventions over more extended periods beyond 12 months.

(2) Does long-term intervention through telemedicine and health guidance using digital technology improve medication adherence and the patient's quality of life (QOL)? It is necessary to examine whether long-term intervention through telemedicine and health guidance using digital technology has merits in blood pressure

It is necessary to examine whether long-term intervention through telemedicine and health guidance using digital technology has merits in blood pressure

reduction, medication adherence, and the patient's quality of life.

(3) Is long-term intervention through telemedicine and health guidance using digital technology cost-effective? It is necessary to study whether long-term intervention through telemedicine and health guidance using digital technology has benefits in cost-effectiveness and antihypertension.

It is necessary to study whether long-term intervention through telemedicine and health guidance using digital technology has benefits in cost-effectiveness and antihypertension. (4) Support for healthcare providers in telemedicine and health guidance using digital technology.

Since the antihypertensive effect of telemedicine and health guidance interventions using digital technology has been observed regardless of the intervention modality (smartphone app, text message, or use of a web screen such as a PC) (Fig. 19), healthcare providers need to address the convenience of patients and counselors in using digital technology. Therefore, healthcare providers need to provide educational opportunities for healthcare providers on telemedicine and health guidance using digital technologies.

6 Al-based healthcare and health guidance support

Overview of digital technology

Artificial intelligence (AI) is a generic term for technologies and theories that use machines to mimic human intellectual activities [45]. AI can be broadly classified into two categories: (1) machine learning, in which patterns are learned from data to make predictions and classifications, and (2) deep learning, in which predictions are made by building a multilayer network that mimics human behavior of learning from examples. In recent years, the IoT (Internet of Things) has been expanded due to the improvement of computational power through the development of computer technology, the development of algorithms, data storage technology, and sensor technology, and big data has been widely utilized. AI is expected to be useful in the healthcare setting. Table 6 lists the expected benefits of AI blood pressure control and points to note, since this is a relatively new technology.

Although AI is considered a promising technology for blood pressure control, many issues remain for its implementation in society. These include expanding clinical trials to ensure the accuracy and reliability of the technology, enhancing personal privacy protection, and developing educational programs to promote understanding and acceptance of AI technology among healthcare providers and patients. Addressing these challenges will require cooperation among healthcare organizations, technology developers, and regulators to develop practical guidelines and policies.

The Japanese Society of Hypertension Guidelines for blood pressure control using digital technologies

Table 6 Benefits and points to note about AI-based healthcare and health guidance support

$\blacksquare$ Benefits

|Personalized healthcare|AI can be used to create a personalized blood pressure control plan using individual patient data (genetic information, lifestyle, past history, etc.). This is expected to maximize the effectiveness of treatment while minimizing the risk of side effects. Furthermore, a prophylactic approach can be taken by predicting future changes in blood pressure.| |---|---| |Early detection of abnormalities|AI may be beneficial for early detection of abnormal values and trends by monitoring continuous patient data (e.g., blood pressure data collected from devices over time). Consequently, it may be also useful to detect emergencies and for the early detection of hypertension.| |Support patient self-management|Using AI-powered applications and devices to advise patients on appropriate lifestyle habits, such as diet and exercise, can support and motivate patients to manage their blood pressure actively. Furthermore, introducing AI chatbots and agents can help patients resolve daily questions and concerns in real time, promoting ongoing self-management of their health.| |Improving the healthcare process|The introduction of generative AI may allow visual representation of medical data, making the information quicker and easier to understand for patients and healthcare providers. This would not only increase patient motivation and adherence to treatment but also improve physicians' efficiency in diagnosis and treatment planning. This will support physicians' decision-making, reduce redundant practice, and contribute to improving physicians' work environments.| |Points to note|| |Data security and privacy|A highly secure environment is required to protect the privacy of the data used.| |Transparency and interpretability of algorithms|Many AI algorithms are black boxes, and transparency and interpretability of algorithms (explainable AI) are desirable so that healthcare providers and patients can understand and trust AI suggestions.| |Integration with expertise|It is essential that personnel be familiar with both the mechanics of AI and healthcare because these disciplines work together; ongoing maintenance and updates are necessary to ensure that AI systems continue to work accurately, and knowledge of AI and healthcare is indispensable. Those who use AI with insufficient healthcare knowledge may draw erroneous conclusions from biased data, leading to errors in recommendations. AI must evolve into a more reliable tool through reinforcement learning, in which humans do not overconfidently trust AI's judgment but intervene and adjust the learning process as necessary.| |Cost-effectiveness|Developing and implementing advanced AI systems can be a high-cost investment, as well as time-consuming; one must always be aware of cost-effectiveness.|

Healthcare question

$\blacksquare$ HQ 6

Does AI-based healthcare and health guidance support lower blood pressure in adults?

$\blacksquare$ Recommendation

Any recommendation or suggestion is withheld due to insufficient evidence on blood pressure control using AI-based healthcare and health guidance support in adults.

Summary of evidence

A literature search was conducted on the effectiveness of AI-based healthcare and health guidance support for adults aged 18 years and older, and a total of five studies (n = 1421) were extracted: two RCTs and three observational studies. In two RCTs, Chiang et al. conducted a study in which lifestyle factors that contribute most to an individual's blood pressure were selected by random forest, a method of machine learning, to encourage lifestyle improvement [intervention group (n = 6); control group (no intervention)(n = 19)] [46], and Persell et al. employed a coaching application with conversational AI to manage

hypertension [intervention group (n = 144); control group (n = 152)] [47]. All of the observational studies had a before-and-after comparative design. Branch et al. used a coaching application with conversational AI to coach healthy lifestyle habits (n = 864) [48]. Guthrie et al. used a mobile application to provide feedback and support that motivate lifestyle improvement (n = 172) [49] and Shamanna et al. used AI to create a virtual twin using various biometric data to recommend better diet, exercise habits, etc. (n = 64) [50].

Explanation

The results of a systematic review [51] on blood pressure control using AI are given in the following sections.

(1) Meta-analysis of RCTs

In the meta-analysis pooling RCTs, there was a trend toward a decrease in SBP (-2.13 mmHg, 95% CI -4.72 to 0.46) and DBP (-1.03 mmHg, 95% CI -2.52 to 0.46) in the AI-based group compared with the non-user control group, although no statistically significant reduction was observed (Fig. 22).

H. Arima et al.

Fig. 22 Meta-analysis of RCTs examining the effectiveness of AI-based healthcare and health guidance support. (Adapted from ref. [7])

![a) SBP (RCT only)\nb) DBP (RCT only)\nFirst author,\nYear of publication\nDifference in SBP change between groups\n[95%CI] (mmHg)\nPersell, 2020\nPoesell, 2020\nChiang, 2021\n-3.50 [-8,11, 1.11]\nPooled estimate ($I^2$=0%) -2.13 [-4.72, 0.46]\nDoverline in SBP change between groups\n(mmHg)\nRCT: randomized controlled trial, SBP: systolic blood pressure, DBP: diastolic blood pressure\n](1/parseimages/figurec812449b78ce4000944634f7b695d42e_0_2.png) RCT: randomized controlled trial, SBP: systolic blood pressure, DBP: diastolic blood pressure

Fig. 23 Meta-analysis of RCTs (only the intervention groups) and observational studies examining the effectiveness of AI-based healthcare and health guidance support. (Adapted from ref. [7])

![a) SBP (RCT plus Observational studies)\nb) DBP (RCT plus Observational studies)\nFirst author,\nYear of publication\n[95%CI] (mmHg)\nObservational studies\n[95%CI] (mmHg)\nGurthrie, 2019\nShamanna, 2021\nBranch, 2022\nPooled estimate ($I^2=89\%$)\n-87.1 [-12.52, -4.90]\nIntervention groups of RCT\nPersell, 2020\nChiang, 2021\nPooled estimate ($I^2=73\%$)\n-6.37 [-10.73, -2.00]\nPooled estimate ($I^2=82\%$)\n-7.79 [-10.46, -5.13]\nPooled estimate ($I^2=84\%$)\nDifference in DBP change (mmHg)\nRCT: randomized controlled trial, SBP: systolic blood pressure, DBP: diastolic blood pressure\n](1/parseimages/figure288ae0cccc51480bbbf4b4834311101d_0_5.png) RCT: randomized controlled trial, SBP: systolic blood pressure, DBP: diastolic blood pressure

(2) Meta-analysis of only the intervention groups from RCTs and observational studies

The observational studies did not include a control group because they were designed as a before-and-after comparison study. Thus, to pool the results of observational studies with those of RCTs, we first performed a meta-analysis incorporating the results of the intervention group in the RCTs with those of the observational studies. The results showed a statistically significant reduction in SBP (-7.79 mmHg, 95% CI -10.46 to -5.13), and in DBP (-3.40 mmHg, 95% CI -5.04 to -1.76) (Fig. 23). However, it is noteworthy that there was no comparison with a control group, unlike a typical meta-analysis; therefore, we conducted a further analysis in the next section.

(3) Pooled meta-analysis of RCTs and observational studies

To pool the RCTs and observational studies, we performed a meta-analysis using an arm-based model to calculate the results of the intervention and control groups of all studies using Bayesian methods. The results showed that the AI-based group tended to reduce SBP (-3.03 mmHg, 95% CI -7.08 to 1.15) and DBP (-1.21 mmHg, 95% CI -3.55 to

1.13) compared to the control group, with the difference not statistically significant (Fig. 24).

Summary

Although blood pressure tended to decrease with the use of AI, it was difficult to conclude whether healthcare and health guidance support for hypertension management using AI is effective in this HQ. This is due to the relatively small sample size as well as several other limitations, such as the possible lack of uniformity in the selection criteria and various backgrounds of the trial participants, the lack of consistency in the way the intervention was implemented, and the short duration of the intervention. In addition, the lack of clarity about the details of the AI algorithms used in the trials and the lack of comparative criteria for assessing their effectiveness also contribute to the poor quality of the evidence. Due to these limitations, we found it difficult to draw robust conclusions.

Literature search flow

A literature search was conducted in PubMed, ICHUSHI and Cochrane Library to identify 2677 studies for primary

The Japanese Society of Hypertension Guidelines for blood pressure control using digital technologies

 screening (title and abstract) and a secondary screening (full text) of 39 studies, excluding 2433 that did not meet the inclusion criteria. A meta-analysis was performed using five studies eligible for this HQ.

Recommendation decision (voting results)

As a result of the systematic review, we withhold any recommendation/suggestion on blood pressure control using AI-based healthcare and health guidance support in adults due to the lack of intervention studies, the variability (heterogeneity) of interventions, the certainty of evidence, the balance of benefits and harms, citizens' values, and cost-effectiveness.

The 35 votes cast included 3 "strongly recommended action," 4 "suggested action," 3 "action not suggested," 0 "action strongly discouraged," 25 "recommendation/sugestion withheld due to insufficient evidence" (71.4% of valid votes), and zero abstentions.

Future research questions

The number of studies in this systematic review was small, and no significant findings were obtained. However, several studies are already in progress, and more evidence will be accumulated regarding the effectiveness of AI in supporting healthcare and health guidance. To date, no standardized AI-based method has been established, and each method's algorithm is like a black box. Future research will be necessary to determine which areas AI can be most effective, sustainable, and cost-effective using established standard methods. Below is an example of a Future Research Question. Exploring the most effective intervention is desirable, considering cost-effectiveness, feasibility, and sustainability.

(1) Is AI effective in identifying high-beneficiaries?

In conventional healthcare, the target population for treatment is “high-risk patients” for disease (“high-risk approach”). On the other hand, the application of AI has made it possible to predict the effectiveness of treatment at the individual level, making it possible to target a group of patients who are estimated to be highly effective (high-benefit patients) (“high-benefit approach”). The high-benefit approach is expected to be validated in the future, as it can potentially increase the overall therapeutic effect enjoyed by the patient population as a whole, compared to the conventional high-risk approach, by selecting a treatment group and intervening with them.

(2) Is AI effective in improving lifestyle?

AI-based chat functions can encourage lifestyle improvements more suited to the user's characteristics and habits, which may lead to lifestyle improvement rather than simply presenting pre-registered answers in the conventional system.

(3) Is AI effective in supporting healthcare and health guidance?

AI is expected to improve the quality of healthcare and health guidance by providing the appropriate differential diagnosis, predicting the risk of disease onset in the future, providing appropriate lifestyle improvement and treatment methods according to patient characteristics, and providing a reminder function to prevent careless mistakes. Implementing these features is expected to improve the quality of healthcare and health guidance.

the Japanese Society of Hypertension Committee on the Guidelines for Blood Pressure Control Using Digital Technologies CHAIRPERSON Hisatomi Arima$^{1}$, Executive Committee Hisatomi Arima$^{1}$, Hiroshi Itoh$^{2}$, Takayoshi Ohkubo$^{3}$, Yusuke Ohya$^{4}$, Mai

H. Arima et al.

Kabayama$^{5}$, Takuya Kishi$^{6}$, Nobuhito Hirawa$^{7}$, Katsuyuki Miura$^{8}$, Shin-ichiro Miura$^{1}$, Koichi Yamamoto$^{5}$, Chikara Yoshimura$^{1}$, Hiromi Rakugi$^{9}$, Guideline Development Committee Kei Asayama$^{3}$, Taku Inoue$^{1}$, Mitsuru Ohishi$^{11}$, Miho Kusaka$^{1}$, Kenichi Goto$^{1}$, Tatuo Shimosawa$^{14}$, Akira Nishiyama$^{15}$, Satoshi Hoshide$^{16}$, Yuichiro Yano$^{17}$, Keiko Kaneko$^{18}$, Fujiko Uesugi$^{19}$, Mutsumi Shigyo$^{20}$, Erika Suenaga$^{21}$, Takeshi Masujima$^{22}$, Souichiro Kawachi$^{23}$, Akira Okayama$^{24}$, Masaaki Miyakawa$^{25}$, Naomi Miyamatsu$^{26}$, Koshi Nakamura$^{14}$, Isao Saito$^{77}$, Yoshitaka Muramaku$^{23}$, Mari Ishida$^{24}$, Kazuo Kitagawa$^{30}$, Hirofumi Ohnishi$^{31}$, Hirokazu Okada$^{32}$, Yoshio Fujioka$^{33}$, Yoshiro Ogawa$^{13}$, Kenichiro Shide$^{34}$, Shigeru Hozeu$^{1}$, Systematic Review Group, Healthcare Question 1 Michihiro Satoh$^{36}$, Miki Kawazoe$^{1}$, Ayako Kunimura$^{37}$, Yukiko Shinohara$^{1}$, Yukako Tatsum$^{3}$, Maya Toyama$^{36}$, Shingo Nakayama$^{24}$, Yoichi Nozato$^{4}$, Takahisa Murakami$^{36}$, Tomoko Muroya$^{38}$, Takahito Yagihashi$^{36}$, Healthcare Question 2 Masayuki Koyama$^{31}$, Keisuke Oyatani$^{31}$, Kei Nakata$^{31}$, Nobuaki Himuro$^{31}$, Healthcare Question 3 Takashi Hisamatsu$^{40}$, Kimika Arakawa$^{14}$, Maho Ishihara$^{3}$, Kohei Ueda$^{14}$, Yukiko Okami$^{42}$, Minako Kinuta$^{40}$, Keiko Kondo$^{1}$, Hiroshige Jinnouchi$^{43}$, Shintaro Minegishi$^{44}$, Healthcare Question 4 Makino Abe$^{1}$, Shintaro Ishida$^{1}$, Yoshifumi Kato$^{1}$, Megumi Kawashima, Shin Kawasoe$^{1}$, Takeshi Shibukawa$^{8}$, Hiroyoshi Seguā$^{45}$, Koji Takahashi$^{1}$, Yoichi Takami$^{5}$, Kazuhiro Tada$^{1}$, Tetsuo Hirata$^{1}$, Taku Fujimoto$^{5}$, Natsumi Morito$^{1}$, Toshitaka Yamanokuch$^{7}$, Sumiko Yoshida$^{46}$, Healthcare Question 5 Atsushi Sakima$^{4}$, Yuya Akagi$^{5}$, Yuichi Akasaki$^{11}$, Ken Kitajima$^{1}$, Yusuke Kobayashi$^{1}$, Takashi Sakaguchi$^{1}$, Tetsuara Niijina$^{14}$, Tatsuya Haze$^{48}$, Takako Fujii$^{1}$, Fumiko Mori$^{49}$, Masanobu Yamazato$^{50}$, Healthcare Question 6, Toshiki Maeda$^{1}$, Rie Koyoshi$^{1}$, Yuki Sakamoto$^{43}$, Atsushi Satoh$^{15}$, Satoshi Hosoki$^{37}$, Sumiyo Yamashita$^{35}$, Systematic Review Support Group, Hisatomi Arma$^{1}$, Atsushi Sakma$^{15}$, Makiko Abe$^{1}$, Assessment Members Atsuhiro Ichihana$^{4}$, Yoshinah Uehara$^{1}$, Tomonori Okamura$^{2}$, Sayoko Ogura$^{55}$, Hisashi Kai$^{56}$, Tomohiro Katsuya$^{57}$, Kazuomi Kario$^{16}$, Hiroshi Koriyama$^{5}$, Shigyeru Shibata$^{1}$, Hirotaka Shibata$^{27}$, Akira Sugawara$^{85}$, Yukari Takemi$^{59}$, Kouichi Tamura$^{7}$, Takuya Tsuchihashi$^{60}$, Kazunori Toyoda$^{61}$, Koichi Node$^{62}$, Yukihito Higashi$^{63}$, Akira Fujiyoshi$^{64}$, Masato Furuhashi$^{31}$, Atsushi Mizuno$^{65}$, Masashi Mukoyama$^{66}$, Hirohito Metoki$^{36}$, Masaki Mogi$^{67}$, Hisashi Eguchi$^{68}$, Atsushi Hozawa$^{58}$, Kunihiko Takahashi$^{69}$, Masanari Kuwabara$^{14}$, Yoshiki Yagita$^{70}$, Naoyuki Takashima$^{71}$, Masayuki Yoshida$^{69}$, Tetsuya Kakuma$^{27}$, Takayoshi Tsukahara$^{71}$, Yoshihiki Nato$^{34}$

$^{36}$ToHoku Medical and Pharmaceutical University, Miyagi, Japan; $^{37}$Aichi Medical University, Aichi, Japan; $^{38}$Izumi Hospital, Miyagi, Japan; $^{39}$Okayama University, Okayama, Japan; $^{40}$NHO Kyushu Medical Center, Fukuoka, Japan; $^{41}$Gunma University, Gunma, Japan; $^{42}$Nippon Medical School, Tokyo, Japan; $^{43}$Yokohama City University School of Medicine, Kanagawa, Japan; $^{44}$Hirose Mirai Clinic, Miyazaki, Japan; $^{45}$NHO Shikoku Medical Center for Children and Adults, Kagawa, Japan; $^{46}$Okinawa University, Okinawa, Japan; $^{47}$Yokohama City University Hospital, Kanagawa, Japan; $^{48}$Mitsui Memorial Hospital, Tokyo, Japan; $^{49}$Okinawa Health Promotion Foundation, Okinawa, Japan; "Kobe Pharmaceutical University, Hyogo, Japan; $^{51}$Kobe City Medical Center General Hospital, Hyogo, Japan; $^{52}$Nagoya City University Rehabilitation Hospital, Aichi, Japan; $^{53}$Tokyo Women's Medical University, Tokyo, Japan; $^{54}$Nihon University, Tokyo, Japan; $^{55}$Kurume University, Fukuoka, Japan; $^{56}$Katsa�ula, Clinica, Hyogo, Japan; $^{57}$Tokohu University, Miyagi, Japan; $^{58}$Kagawa Nutrition University, Kagawa, Japan; $^{59}$Steel Memorial Yawata Hospital, Fukuoka, Japan; $^{60}$National Cerebral and Cardiovascular Center, Osaka, Japan; $^{61}$Saga University, Saga, Japan; $^{62}$Hiroshima University, Hiroshima, Japan; $^{63}$Wakayama Medical University, Wakayama, Japan; $^{64}$St. Luke's International University, Tokyo, Japan; $^{65}$Social Insurance Omta Tenryo Hospital, Fukuoka, Japan; $^{66}$Ehime University, Ehime, Japan;

$^{67}$University of Occupational and Environmental Health, Fukuoka, Japan; $^{68}$Institute of Science Tokyo, Tokyo, Japan; $^{69}$Kawasaki Medical School, Okayama, Japan; $^{70}$Kyoto Prefectural University of Medicine, Kyoto, Japan; $^{71}$Nagoya University of Arts and Sciences, Aichi, Japan

Compliance with ethical standards

Conflict of interest The authors declare no competing interests.

Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

1. Imai Y, Obara T, Asamaya K, Ohkubo T. The reason why home blood pressure measurements are preferred over clinic or ambulatory blood pressure in Japan. Hypertens Res. 2013;36:661-72. 2. Imai Y, Kario K, Shimada K, Kawano Y, Hasebe N, Matsuura H, et al. The Japanese Society of Hypertension Guidelines for Self-monitoring of Blood Pressure at Home (Second Edition). Hypertens Res. 2012;35:77-95. 3. Asayama K, Ohkubo T, Emoe R, Kikuya M, et al. Hypertension objective treatment based on measurement by electrical devices of blood pressure (HOMED-BP). Cardiovascular outcomes in the first trial of antihypertensive therapy guided by self-measured home blood pressure. Hypertens Res. 2012;35:1102-10. 4. Ohkubo T, Asayama K, Kikuya M, Metoki H, Hoshi H, Hashimoto J, et al. How many times should blood pressure be measured at home for better prediction of stroke risk? Ten-year follow-up results from the Ohasama study. Hypertens. 2004;22:1099-104. 5. Ohkubo T, Imai Y, Tsuji I, Nagai K, Kato J, Kikuchi N, et al. Home blood pressure measurement has a stronger predictive power for mortality than does screening blood pressure measurement: a population-based observation in Ohasama, Japan. J Hypertens. 1998;16:971-5. 6. Mishra A, Malik R, Hachiya T, Jürgenson T, Namba S, Posner DC, et al. International Database of Home blood pressure in relation to cardiovascular Outcome Investigators. Outcome-driven thresholds for home blood pressure measurement: international database of home blood pressure in relation to cardiovascular outcome. Hypertension. 2013;61:27-34. 7. Satoh M, Asayama K, Kikuya M, Inoue R, Metoki H, Hosaka M, et al. Long-term stroke risk due to partial white-coat or masked hypertension based on home and ambulatory blood pressure measurements: the Ohasama study. Hypertension. 2016;67:48-55. 8. Shimamoto K, Ando K, Fujita T, Hasebe N, Higaki J, Horiuchi M, et al. Japanese Society of Hypertension Committee for Guidelines for the Management of Hypertension. The Japanese Society of Hypertension Guidelines for the Management of Hypertension (JSH 2014). Hypertens Res. 2014;37:253. 9. Satoh M, Maeda T, Hoshide S, Ohkubo T. Is antihypertensive treatment based on home blood pressure recommended rather than that based on office blood pressure in adults with essential hypertension? (meta-analysis). Hypertens Res. 2019;42:807-16. 10. NCD Risk Factor Collaboration; European Society of Hypertension Working Group on Blood Pressure Monitoring and Cardiovascular Variability. Cuffless blood pressure measuring devices: review and statement by the European Society of Hypertension Working Group on Blood Pressure Monitoring and Cardiovascular Variability. J Hypertens. 2021;10:1449-60.

The Japanese Society of Hypertension Guidelines for blood pressure control using digital technologies

11. Satoh M, Tatsumi Y, Nakayama S, Shinohara Y, Kawazoe M, Nozato Y, et al. Self-measurement of blood pressure at home using a fcf device for change in blood pressure levels: systematic review and meta-analysis. Hypertens Res. 2025;48:574-91. 12. Kario K, Williams B, Tomitani N, McManus RJ, Schutte AE, Avolio A, et al. Innovations in blood pressure measurement and reporting technology: International Society of Hypertension position paper endorsed by the World Hypertension League, European Society of Hypertension, Asian Pacific Society of Hypertension, and Latin American Society of Hypertension. J Hypertens. 2024;42:1874-88. 13. Kario K. Management of hypertension in the digital era: small wearable monitoring devices for remote blood pressure monitoring. Hypertension. 2020;76:640-50. 14. Amagasa S, Kojin H, Kamada M, Fukuoka Y, Inoue S. Evaluation of physical activity using smartphones and wearable devices in healthcare: current situation and future perspectives. Nihon Koshu Eisei Zasshi. 2021;68:585-96. 15. Islam SMS, Chow CK, Daryabeygikhotbehsara R, Subedi N, Rawstorn J, Teegne T, et al. Wearable cuffless blood pressure monitoring devices: a systematic review and meta-analysis. Eur Heart J Digit Health. 2022;3:323-37. 16. Koyama M, Himuro N, Nakata K, Oyatani K, Shinohara Y, Ohnishi H, et al. Effects of physical activity-promoting wearable devices on blood pressure in adults: a systematic review and meta-analysis. Hypertens Res. 2025. https://doi.org/10.1038/s41440-025-02260-6. Online ahead of print. 17. Elliott P, Stamler J, Nichols R, Dyer AR, Stamler R, Kesteloot H, et al. Intersalt revisited: further an alyses of 24 hour sodium excretion and blood pressure within and across populations. BMJ, 1996;312:1249-53. 18. Huang L, Trieu K, Yoshimura S, Neal B, Woodward M, Campbell N, et al. Effect of dose and duration of reduction in dietary sodium on blood pressure levels: systematic review and meta-analysis of randomised trials. BMJ, 2020;368:m15 PMID: 32094151. 19. Ma Y, He FJ, Sun Q, Yuan C, Kieneker LM, Curhan GC, et al. 24-hour urinary sodium and potassium excretion and cardiovascular risk. N Engl J Med J. Med. 2022;386:252-63. 20. Umemura S, Arima H, Arima S, Asayama K, Dohi Y, Hirooka Y, et al. The Japanese Society of Hypertension Guidelines for the Management of Hypertension (JSH2019). Hypertens Res. 2019;42:1235-481. 21. World Health Organization. Guideline: Sodium intake for adults and children. Geneva: World Health Organization; 2012. 22. Ministry of Health. Labour and Welfare. Dietary Reference Intakes for Japanese 2020. Tokyo: Ministry of Health, Labour and Welfare; 2020. 23. Appel LJ, Moore TJ, Obarzanek E, Vollmer WM, Svetkey LP, Sacks FM, et al. DASH Collaborative Research Group. A clinical trial of the effects of dietary patterns on blood pressure. N Engl J Med. 1997;336:1117-24. 24. Poorolajal J, Zeraati F, Soltanian AR, Sheikh V, Hooshmand E, Maleki A. Oral potassium supplementation for management of essential hypertension: a meta-analysis of randomized controlled trials. PLoS ONE. 2017;12:0174967. 25. Kogure M, Hirata T, Nakaya N, Tsuchiya N, Nakamura T, Narita A, et al. Multiple measurements of the urinary sodium to potassium ratio strongly related home hypertension: TMM Cohort Study. Hypertens Res. 2020;43:62-71. 26. Okayama A, Okuda N, Miura K, Okamura T, Hayakawa T, Akasaka H, et al. NIPPON DATA80 Research Group. Dietary sodium-to-potassium ratio as a risk factor for stroke, cardiovascular disease and all-cause mortality in Japan: the NIPPON DATA80 cohort study. BMJ Open. 2016;6:e011632.

27. Hisamatsu T, Kogure M, Tabara Y, Hozawa A, Sakima A, Tsu-chihashi T, et al. Practical use and target value of urine sodium-to-potassium ratio in assessment of hypertension risk for Japanese: Consensus Statement by the Japanese Society of Hypertension Working Group on Urine Sodium-to-Potassium Ratio. Hypertens Res. 2024;47:3288–302. 28. Iwahori T, Ueshima H, Ohgami N, Yamashita H, Miyagawa N, Kondo K, et al. Effectiveness of a self-monitoring device for urinary sodium-to-potassium ratio on dietary improvement in free-living adults: a randomized controlled trial. J Epidemiol. 2018;28:41–7. 29. Meuleman Y, Hoekstra T, Dekker FW, Navis G, Vogt L, van der Boog P, et al. ESMO Study Group. Sodium restriction in patients with CKD: a randomized controlled trial of selfmanagement support. Am J Kidney Dis. 2017;69:576–86. 30. Morikawa N, Yamasue K, Tochikubo O, Mizushima S. Effect of salt reduction intervention program using an electronic salt sensor and cellular phone on blood pressure among hypertensive workers. Clin Exp Hypertens. 2011;33:216–22. 31. Takada T, Imamoto M, Sasaki S, Azuma T, Miyashita J, Hayashi M, et al. Effects of self-monitoring of daily salt intake estimated by a simple electrical device for salt reduction: a cluster randomized trial. Hypertens Res. 2018;41:524–30. 32. Wiriyatamakorn S, Mukdadilok A, Kantachuvesiri S, Mekhora C, Yinghoncharoen T. Impact of self-monitoring of salt intake by salt meter in hypertensive patients: a randomized controlled trial (SMAL-SALT). J Clin Hypertens. 2021;23:1852–61. 33. Yasutake K, Miyoshi E, Misumi Y, Kajiyama T, Fukuda T, Ishii T, et al. Self-monitoring of urinary salt excretion as a method of salt-reduction education: a parallel, randomized trial involving two groups. Public Health Nutr. 2018;21:2164–73. 34. Yasutake K, Umeki Y, Horita N, Morita R, Murata Y, Ohe K, et al. A self-monitoring urinary salt excretion level measurement device for educating young women about salt reduction: a parallel randomized trial involving two groups. J Clin Hypertens. 2019;21:730–8. 35. Yokokawa H, Yuasa M, Nedsuwan S, Moolphate S, Fukuda H, Kitajima T, et al. An impact of dietary intervention on blood pressures among diabetic and/or hypertensive patients with high cardiovascular disorders risk in northern Thailand by cluster randomized trial. J Gen Fam Med. 2021;22:28–37. 36. Hisamatsu T, Ueda K, Arakawa K, Minegishi S, Okami Y, Kinuta M, et al. Effectiveness of self-monitoring devices measuring the urinary sodium-to-potassium ratio, urinary salt (sodium) excretion, or salt concentration in foods for blood pressure management: a systematic review and meta-analysis. Hypertens Res. 2025;48:1891–8. 37. Abe M, Hirata T, Morito N, Kawashima M, Yoshida S, Takami Y, et al. Smartphone application-based intervention to lower blood pressure: a systematic review and meta-analysis. Hypertens Res. 2025;48:492–505. 38. Nomura A, Tanigawa T, Kario K, Igarashi A. Cost-effectiveness of digital therapeutics for essential hypertension. Hypertens Res. 2022;45:1538–48. 39. Omboni S, McManus RJ, Bosworth HB, Chappell LC, Green BB, Kario K, et al. Evidence and recommendations on the use of telemedicine for the management of arterial hypertension: an international experiment position paper. Hypertension. 2020;76:1368–83. 40. Siopis G, Moschonis G, Eweka E, Jung J, Kwasnicka D, Asare BY, et al. Effectiveness, reach, uptake, and feasibility of digital health interventions for adults with hypertension: a systematic review and meta-analysis of randomised controlled trials. Lancet Digit Health. 2023;5:e144–59.

H. Arima et al.

41. Katz ME, Mszar R, Grimshaw AA, Gunderson CG, Onuma OK, Lu Y, et al. Digital health interventions for hypertension management in US populations experiencing health disparities: a systematic review and meta-analysis. JAMA Net Open. 2024;7:e2356070. 42. Kario K, Harada N, Okura A. Digital therapeutics in hypertension: evidence and perspectives. Hypertension. 2022;79:2148–58. 43. Nomura A. Digital health, digital medicine, and digital therapeutics in cardiology: current evidence and future perspective in Japan. Hypertens Res. 2023;46:2126–34. 44. Sakima A, Akagi Y, Akasaki Y, Fujii T, Haze T, Kawakami-Mori F, et al. Effectiveness of digital health interventions for tele-medicine/telehealth for managing blood pressure in adults: a systematic review and meta-analysis. Hypertens Res. 2025;48:478–91. 45. Krittanawong C, Bomback AS, Baber U, Bangalore S, Messerli FH, Wilson Tang WH. Future direction for using artificial intel-ligence to predict and manage hypertension. Curr Hypertens Rep. 2018;20:75. 46. Chiang PH, Wong M, Dey S. Using wearables and machine learning to enable personalized lifestyle recommendations to improve blood pressure. IEEE J Transl Eng Health Med. 2021;9:1–13. PMID: 34765324.

47. Persell SD, Peprah YA, Lipiszko D, Lee JY, Li JJ, Ciolino JD, et al. Effect of home blood pressure monitoring via a smartphone hypertension coaching application or tracking application on adults with uncontrolled hypertension: a randomized clinical trial. JAMA Net Open. 2020;3:e200255. 48. Branch OH, Rikhy M, Auster-Gussman LA, Lockwood KG, Graham SA. Relationships between blood pressure reduction, weight loss, and engagement in a digital app-based hypertension care program: observational study. JMIR Form Res. 2022;6:e38215. 49. Guthrie NL, Berman MA, Edwards KL, Appelbaum KJ, Dey S, Carpenter J, et al. Achieving rapid blood pressure control with digital therapeutics: retrospective cohort and machine learning study. JMIR Cardio. 2019;3:e13030. 50. Shamanna P, Dharmalingam M, Sahay R, Mohammed J, Mohamed M, Poon T, et al. Retrospective study of glycemic variability, BMI, and blood pressure in diabetes patients in the Digital Twin Precision Treatment Program. Sci Rep. 2021;11:14892. 51. Maeda T, Sakamoto Y, Hosoki S, Satoh A, Koyoshi R, Yamashita S, et al. Does clinical practice supported by artificial intelligence improve hypertension care management? A pilot systematic review. Hypertens Res. 2024;47:2312–6.

Affiliations

Hisatomi Arima$^{1}$ · Hiroshi Itoh$^{2}$ · Takayoshi Ohkubo$^{3}$ · Yusuke Ohya$^{4}$ · Mai Kabayama$^{5}$ · Takuya Kishi$^{6}$ · Nobuhito Hirawa$^{7}$ · Katsuyuki Miura$^{8}$ · Shin-ichiro Miura$^{1}$ · Koichi Yamamoto$^{5}$ · Chikara Yoshimura$^{1}$ · Hiromi Rakugi$^{9}$ · Kei Asayama$^{3}$ · Taku Inoue$^{10}$ · Mitsuru Ohishi$^{11}$ · Miho Kusaka$^{12}$ · Kenichi Goto$^{13}$ · Tatsuo Shimosawa$^{14}$ · Akira Nishiyama$^{15}$ · Satoshi Hoshide$^{16}$ · Yuichiro Yano$^{17}$ · Keiko Kaneko$^{18}$ · Fujiko Uesugi$^{19}$ · Mutsumi Shigyão$^{20}$ · Erika Suenaga$^{21}$ · Takeshi Masujuma$^{22}$ · Souichiro Kawachi$^{23}$ · Akira Okayama$^{24}$ · Masaaki Miyakawa$^{25}$ · Naomi Miyamatsu$^{26}$ · Koshi Nakamura$^{4}$ · Isao Saito$^{27}$ · Yoshitaka Murakami$^{28}$ · Mari Ishida$^{29}$ · Kazuo Kitagawa$^{30}$ · Hirofumi Ohnishi$^{31}$ · Hirokazu Okada$^{32}$ · Yoshio Fujioka$^{33}$ · Yoshihiro Ogawa$^{13}$ · Kenichiro Shide$^{34}$ · Shigeru Inoue$^{35}$ · on behalf of the Japanese Society of Hypertension Committee on the Guidelines for Blood Pressure Control Using Digital Technologies

1. Fukuoka University, Fukuoka, Japan 2. Keio University, Tokyo, Japan 3. Teikyo University, Tokyo, Japan 4. University of the Ryukyus, Okinawa, Japan 5. The University of Osaka, Osaka, Japan 6. International University of Health and Welfare, Fukuoka, Japan 7. Yokohama City University, Kanagawa, Japan 8. Shiga University of Medical Science, Shiga, Japan 9. Osaka Rosai Hospital, Osaka, Japan

10. Omoromachi Medical Center, Okinawa, Japan 11. Kagoshima University, Kagoshima, Japan 12. Internal Medicine Dr.Kusaka's Office, Hiroshima, Japan 13. Kyushu University, Fukuoka, Japan 14. International University of Health and Welfare, Chiba, Japan

15 Kagawa University, Kagawa, Japan 16 Jichi Medical University, Tochigi, Japan 17 Juntendo University, Tokyo, Japan 18 Japan Industrial Counselors Associationy, Tokyo, Japan 19 Nishitetsu Human Resources Service Co., Ltd, Fukuoka, Japan 20 Fukuoka City Public Health Center, Fukuoka, Japan 21 Iki City, Nagasaki, Japan 22 Tokyo Freight Forwarding Health Insurance Association, Tokyo, Japan 23 Bosch Health Insurance Society, Saitama, Japan 24 Research Institute of Strategy for Prevention, Tokyo, Japan 25 Japan Medical Association, Tokyo, Japan 26 Japanese Association of Cardiovascular Nursing, Tokyo, Japan 27 Oita University, Oita, Japan

The Japanese Society of Hypertension Guidelines for blood pressure control using digital technologies

28 Toho University, Tokyo, Japan29 Hiroshima Shudo University, Hiroshima, Japan30 Suita Municipal Hospital, Osaka, Japan31 Sapporo Medical University, Hokkaido, Japan32 Saitama Medical University, Saitama, Japan

33 Kobe Gakuin University, Hyogo, Japan 34 Mukogawa Women's University, Hyogo, Japan 35 Tokyo Medical University, Tokyo, Japan