Remote Patient Monitoring Insights: Benefits, Technologies, and Best Practices

Remote patient monitoring (RPM) is transforming how clinicians manage chronic diseases, acute conditions and recovery. By deploying digital devices in patients’ homes, providers can track vital signs, medication adherence and lifestyle habits in real time, bridging the gap between clinic visits and everyday life. RPM is not just a telehealth add‑on; it’s a critical component of modern care pathways that helps detect problems early, reduce hospitalizations and empower patients to take charge of their health. Health agencies define RPM as the use of digital devices that collect health information from individuals in one location and electronically transmit that information securely to healthcare providers in a different locationtelehealth.hhs.gov. This data sharing facilitates continuous monitoring, early intervention and patient engagement, delivering benefits for both patients and providers.

Remote patient monitoring scene: clinicians review a tablet while a patient appears on a video call; vitals dashboard on a desktop and wearables, pulse oximeter, and hub device on the desk.

{getToc} $title={Table of Contents} $count={Boolean} $expanded={Boolean}

Why Remote Patient Monitoring Matters

A Paradigm Shift Toward Continuous, Patient‑Centered Care

Traditional healthcare has revolved around episodic encounters: a patient visits the clinic, gets examined and receives a treatment plan. RPM flips this model. Digital devices measure vital signs—weight, heart rate, blood pressure, glucose and more—at home and send the data to clinicians. This shift enables proactive care: providers can spot subtle changes before they become emergencies and adjust treatment plans quickly. Such continuous, patient‑centered monitoring is especially valuable for chronic conditions like heart disease, hypertension, diabetes and pulmonary disorderstelehealth.hhs.gov.

RPM also supports hospital‑at‑home programs, where patients receive hospital‑level care in their homes. Evidence shows these programs improve short‑ and long‑term outcomes and reduce costs compared to traditional hospitalizationspmc.ncbi.nlm.nih.gov. Studies estimate that hospital‑at‑home could serve up to one‑third of U.S. hospitalizations and that frequent monitoring via RPM could reduce mortality by 39 % and lower intensive care transferspmc.ncbi.nlm.nih.gov. These impressive outcomes highlight why RPM is no longer experimental—it’s an integral part of high‑value healthcare.

Rapid Adoption and Market Growth

RPM adoption has accelerated dramatically in recent years. Telehealth utilization jumped from 11 % of U.S. adults before the COVID‑19 pandemic to 46 % during the pandemic, and physician adoption of RPM technologies surged from 14 % to 80 % between 2016 and 2022pmc.ncbi.nlm.nih.gov. Surveys suggest that physicians’ perception of RPM’s value increased from 87 % in 2016 to 95 % in 2022. Analysts predict that about 26 % of the U.S. population will use some form of RPM device by 2025pmc.ncbi.nlm.nih.gov. Market forecasts bear this out: the U.S. RPM market was valued at roughly $14–15 billion in 2024 and is expected to exceed $29 billion by 2030, while the global market—about $39–40 billion in 2023—is projected to reach $77 billion by 2029intuitionlabs.ai. Adoption has been particularly robust in internal medicine, cardiology and family practice, which account for 29 %, 21 % and 19 % of RPM usage, respectivelyintuitionlabs.ai.

Documented Benefits

A 2024 systematic review of 30 randomized controlled trials found that RPM interventions reduce hospital admission or readmission rates, shorten lengths of hospital stay, decrease follow‑up visits and lower non‑hospitalization costspmc.ncbi.nlm.nih.gov. RPM also improves patient mobility and functional status and increases adherence to medications and lifestyle prescriptions. A separate analysis of hospital‑at‑home programs reported that frequent monitoring with wearable and environmental sensors reduces mortality, lowers intensive care unit transfers, decreases delirium and infection rates and improves fall detectionpmc.ncbi.nlm.nih.gov. These clinical benefits translate into financial savings—one hospital program using AI‑enabled RPM reduced 30‑day readmissions by 70 % and cut costs by 38 %intuitionlabs.ai.

Remote Patient Monitoring Devices and Technologies

RPM relies on a range of connected devices and platforms. Selecting the appropriate tool depends on the targeted condition, patient preferences and integration requirements. Below are key categories of devices and technologies.

Vital‑Sign Monitoring Devices

Digital Scales

For patients with congestive heart failure or obesity, weight fluctuations can signal fluid retention or other complications. Digital scales in RPM programs capture weight daily and automatically transmit data to clinicianstelehealth.hhs.gov. Alerts can trigger when weight changes exceed predetermined thresholds, prompting medication adjustments or nurse check‑ins.

Heart Monitors

Heart monitors include wearable devices like smartwatches, ECG patches and chest straps. Modern smartwatches measure heart rate, rhythm and sometimes oxygen saturation. Adhesive ECG patches provide continuous cardiac rhythm monitoring and can detect arrhythmias and atrial fibrillation. Clinical‑grade patches integrate with remote platforms and use AI algorithms to flag irregular patterns. These devices are particularly beneficial for patients with atrial fibrillation, arrhythmias and post‑cardiac surgerytelehealth.hhs.gov.

Glucometers and Continuous Glucose Monitors (CGMs)

Traditional glucometers require finger‑stick blood samples, but continuous glucose monitors (CGMs) are revolutionizing diabetes management. CGMs use small sensors inserted under the skin to measure glucose levels every few minutes. They transmit readings to mobile apps and cloud dashboards, allowing patients and providers to monitor trends and adjust insulin doses. RPM programs integrate CGM data with telehealth consultations to help diabetics maintain better glycemic control and reduce hypoglycemia risktelehealth.hhs.gov.

Blood Pressure Monitors

Remote blood pressure monitors—wrist or arm cuffs—record blood pressure and heart rate. Devices can be manually activated by patients or programmed to take scheduled readings. Data are sent via Bluetooth or cellular networks to providers. Hypertensive patients benefit from regular monitoring, which can detect early changes and inform medication adjustmentstelehealth.hhs.gov.

Oxygen Monitors and Spirometers

Pulse oximeters measure blood oxygen saturation and pulse rate. They are critical for patients with chronic obstructive pulmonary disease (COPD), asthma and COVID‑19. Some wearable devices track oxygen levels continuously during sleep to detect apnea. Spirometers measure lung function (forced expiratory volume and other parameters) and are used in remote respiratory therapy. Connected spirometers can help identify deterioration in COPD and asthma patients, prompting timely interventionstelehealth.hhs.gov.

Fetal and Maternal Monitors

For high‑risk pregnancies, remote fetal monitoring devices track fetal heart rate and maternal contractions. Data are shared with obstetricians for early detection of complications. These monitors reduce the need for frequent in‑clinic visits and provide peace of mind for expectant motherstelehealth.hhs.gov.

Advanced Diagnostic Tools

Digital Stethoscopes and Remote Auscultation

Digital stethoscopes convert acoustic signals into digital data and transmit them through telehealth platforms. Devices like Littmann CORE amplify sound by up to 40× and allow physicians to switch between analog and digital modes. The Eko DUO combines a stethoscope with a single‑lead ECG and uses AI to detect atrial fibrillation and heart murmursjscimedcentral.com. The Thinklabs One packs high‑fidelity sensors into a pocket‑sized device, while TytoCare offers a multimodal exam kit (stethoscope, otoscope, thermometer) with a digital stethoscope component. In virtual wards and home hospitalization programs, connected stethoscopes enable clinicians to assess cardiac and pulmonary status remotely, extending specialist reach to rural or underserved communitiesjscimedcentral.com.

Ambient Sensors and Fall Detection

Ambient sensors (motion detectors, bed sensors, pressure mats) track movement patterns, falls and behaviors. In hospital‑at‑home programs, these sensors alert caregivers if a patient falls or deviates from normal routines, helping to prevent injuries and improving patient safety. When integrated with AI, ambient sensors can also detect early signs of delirium or infection in hospital wardspmc.ncbi.nlm.nih.gov.

AI‑Enabled Analytics and Predictive Algorithms

The true power of RPM emerges when device data feed into analytics platforms. Machine learning models analyze vital signs and patient‑reported outcomes to identify patterns, predict deterioration and optimize care. For example, Biofourmis’ AI platform analyzes continuous heart rate and respiratory data from wearables; in a clinical trial it reduced 30‑day readmissions by 70 % and lowered costs by 38 %intuitionlabs.ai. Predictive algorithms help care teams prioritize patients requiring intervention and automate triage. However, robust training datasets and careful evaluation are essential to minimize bias and ensure safetypmc.ncbi.nlm.nih.gov.

Integrating RPM with Other Digital Health Tools

RPM doesn’t operate in isolation. Many programs combine it with telemedicine visits, electronic health records (EHRs), mobile health apps and digital therapeutics. Modern RPM platforms integrate with EHRs so that clinicians can view remote data alongside lab results and clinic notes. Telehealth visits provide opportunities to discuss data trends and adjust care plans in real time. Mobile apps allow patients to log symptoms, medications and lifestyle factors, feeding additional information into analytics platforms. Some systems incorporate digital coaching or behavioral interventions, providing personalized guidance based on data trends.

Adoption Statistics and Market Trends

The rapid uptake of RPM is reshaping the healthcare landscape. Key data points illustrate its momentum and potential.

Telehealth Adoption – Telehealth use among U.S. adults jumped from 11 % pre‑pandemic to 46 % during the pandemic, reflecting a paradigm shift in healthcare deliverypmc.ncbi.nlm.nih.gov. Usage continues to remain high even as in‑person visits resume.

Physician Acceptance – Physicians’ perception of RPM initiatives increased from 87 % in 2016 to 95 % in 2022, while actual adoption of RPM by physicians rose from 14 % to 80 % during the same period.

Projected Usage – Analysts estimate that roughly 26 % of the U.S. population will use an RPM device by 2025pmc.ncbi.nlm.nih.gov.

Market Size – The U.S. RPM market is projected to grow from $14–15 billion in 2024 to over $29 billion by 2030. Globally, the RPM market is expected to expand from $39–40 billion in 2023 to $77 billion by 2029intuitionlabs.ai.

Specialty Distribution – Internal medicine (29 %), cardiology (21 %) and family practice (19 %) account for the largest share of RPM usage, reflecting its importance in chronic disease management.

Program Success – AI‑enabled RPM programs have demonstrated dramatic improvements in outcomes: Biofourmis’ program cut 30‑day readmissions by 70 % and reduced costs by 38 %intuitionlabs.ai. Hospital‑at‑home programs using RPM have expanded to over 350 hospitals in 39 states, delivering lower mortality and spending in the 30 days after discharge.

These figures underscore RPM’s significance. Health systems that adopt RPM gain a competitive advantage by offering more convenient, data‑driven care.

Developing a Remote Patient Monitoring Program

Implementing RPM requires careful planning and alignment with clinical workflows. The following step‑by‑step framework summarizes best practices from U.S. federal guidance and industry leaders.

Step 1: Assess Patient Needs and Program Goals

Start by identifying target patient populations and conditions. Consider which clinical problems can benefit most from continuous monitoring—heart failure, hypertension, diabetes, COPD, perinatal care, post‑surgical recovery, obesity, vertigo, cancer and neurological disorders are common use casestelehealth.hhs.gov. Determine the goals of the program: reducing readmissions, improving medication adherence, enhancing quality of life or supporting early discharge. Assess patient readiness and technology comfort; some individuals may need training or caregiver support.

Step 2: Choose Appropriate Devices and Technologies

Select devices that align with clinical goals and patient preferences. Decide whether data will be transmitted via cellular, Wi‑Fi or manual upload and whether devices should integrate with smartphone apps. Evaluate devices for usability, reliability, battery life, HIPAA compliance and potential for reuse. Multi‑condition devices (e.g., a device that measures both heart rate and oxygen saturation) may reduce complexity, while single‑use devices can be cost effective for short‑term monitoringtelehealth.hhs.gov.

Consider partnering with technology vendors that offer integration with EHRs and analytics platforms. Ensure devices are FDA‑approved or appropriately validated. Evaluate whether AI‑enabled algorithms are necessary and ensure they are trained on diverse datasets to avoid biaspmc.ncbi.nlm.nih.gov.

Step 3: Plan Staffing and Workflow

Assign roles for patient enrollment, device training, data monitoring and technical support. Determine whether nursing staff, medical assistants or specialized RPM coordinators will monitor incoming data and respond to alerts. Develop protocols for handling abnormal readings—what constitutes an alert, who gets notified, and what follow‑up actions are required. Coordinate with physicians to determine thresholds for intervention and escalation. Plan how the program will integrate with existing care pathways (e.g., heart failure clinics or primary care visits). Ensure administrative staff understand reimbursement processes and documentation requirementstelehealth.hhs.gov.

Step 4: Onboard Patients and Caregivers

Preparing patients is critical to success. Use telehealth appointments, phone calls or email handouts to explain the program’s purpose and benefits. Demonstrate how devices work—show patients how to turn them on, attach sensors, take readings and charge batteries. Provide written instructions and video tutorials. During onboarding, verify the device’s connectivity and ensure data transmissions are working. Encourage patients to practice using the device while you watch. Engage family members or caregivers to support the patient, especially if the patient is elderly or has cognitive limitationstelehealth.hhs.gov. Caregivers can help set up devices, remind patients to measure, monitor readings and contact providers if there are issues.telehealth.hhs.gov

Step 5: Launch and Monitor the Program

Once patients are onboarded, begin collecting data. Establish a schedule for routine readings and instruct patients to alert the care team if they experience symptoms outside of typical patterns. Clinicians should review incoming data daily or according to program protocols. Use dashboards to visualize trends and highlight out‑of‑range values. When alarms trigger, follow established protocols: call the patient, adjust medications, schedule a telehealth visit or arrange an in‑person evaluation. Document all interventions in the EHR. Track program metrics such as hospital readmissions, emergency visits, patient satisfaction and cost savingstelehealth.hhs.gov.

Step 6: Evaluate and Refine

A successful RPM program is iterative. Collect feedback from patients, caregivers and clinicians. Assess whether devices are easy to use and whether data transmission is reliable. Measure progress toward program goals (e.g., readmission rate reduction, improved blood pressure control). Identify barriers or workflow bottlenecks and adjust protocols accordingly. Consider adding new device types or expanding eligibility if the program meets its objectives. Document lessons learned and share insights with stakeholders.

Best Practices and Considerations

Integrating Human Touch

While digital devices and algorithms are powerful, they cannot replace human connection. The systematic review emphasizes that interventions combining RPM with phone calls, coaching and educational support yield better adherence and satisfactionpmc.ncbi.nlm.nih.gov. Regular check‑ins build trust, encourage patients to stay engaged and allow clinicians to assess subtle changes not captured by sensors. Many programs assign a dedicated nurse or care coordinator to each RPM patient to answer questions and provide encouragement.

Ensuring Accessibility and Equity

Equity is an important consideration. Older adults, low‑income patients and those with limited digital literacy may struggle with technology. The digital health innovations literature notes that inadequate internet access and low technology adoption can create disparitiespmc.ncbi.nlm.nih.gov. To mitigate this, programs can offer devices with cellular connectivity (which don’t require home broadband), provide training sessions, simplify user interfaces and involve caregivers. Health systems should also consider offering devices in multiple languages and ensuring accessibility for individuals with disabilities.

Maintaining Data Privacy and Security

RPM generates sensitive health information that must be protected. Devices and platforms should comply with HIPAA and other privacy laws. Use encryption for data in transit and at rest, and require multi‑factor authentication for platform access. Adopt vendor contracts with clear responsibilities for data security. Conduct regular risk assessments and establish incident response plans. As AI becomes more embedded in RPM, pay attention to algorithm transparency and explainability; patients and clinicians should understand how decisions are madepmc.ncbi.nlm.nih.gov.

Addressing Reimbursement and Regulatory Requirements

RPM reimbursement policies vary by country and payer. In the United States, Medicare and many private insurers reimburse for RPM services under specific CPT codes. However, billing requires detailed documentation of time spent reviewing data and interacting with patients. It’s crucial to stay informed about evolving regulations—temporary pandemic waivers may expire or be modified. Some states still have barriers to cross‑state telehealth practice; check licensure rules and ensure your clinicians are covered.

Evaluating Device Usability and User Experience

Human factors matter. Research on RPM usability stresses that devices must be intuitive, comfortable to wear and easy to maintainpmc.ncbi.nlm.nih.gov. Poor usability can lead to non‑adherence, data gaps and frustration. When selecting devices, consider battery life, charging requirements, display readability and connectivity reliability. Solicit feedback from patients early and frequently. Don’t assume a technically advanced device is better if patients find it cumbersome.

Challenges and Barriers

Despite its promise, RPM faces obstacles that stakeholders must address:

Data Overload and Alarm Fatigue – Continuous monitoring produces large volumes of data. Without intelligent filtering and prioritization, clinicians may become overwhelmed, leading to missed alerts or unnecessary interventions. AI and configurable thresholds can reduce noise, but providers must calibrate systems carefully.

Integration Complexity – Integrating data from multiple devices into EHRs and workflows can be technically challenging. Disparate standards and proprietary formats hinder interoperability. Vendors and health systems need to collaborate on open standards and APIs.

Digital Divide – Lack of internet access, digital literacy or device availability can exclude vulnerable populationspmc.ncbi.nlm.nih.gov. Expanding broadband infrastructure and providing loaner devices or subsidies can help.

Regulatory Uncertainty – Telehealth regulations and reimbursement policies continue to evolve. Temporary waivers may expire or change. Keeping abreast of policy changes and advocating for supportive legislation is essential.

Privacy and Security Risks – Cyberattacks and data breaches threaten patient trust. Strong encryption, secure device design and constant vigilance are needed.

User Resistance – Some patients may be reluctant to adopt technology due to privacy concerns, fear of being monitored or cultural preferences. Clinicians must address these concerns empathetically and explain the benefits clearly.

Real‑World Use Cases and Case Studies

Hospital‑at‑Home Programs

Hospital‑at‑home programs exemplify RPM’s potential to replace traditional inpatient care. Such programs provide hospital‑level treatments—intravenous medications, oxygen therapy, frequent vital sign monitoring and rapid escalation—in patients’ homes. One study notes that hospital‑at‑home could serve up to one‑third of U.S. hospitalizationspmc.ncbi.nlm.nih.gov. Programs use wearable sensors, smart beds and ambient devices to monitor vital signs and activity. Clinicians review data remotely and visit patients daily or more frequently if needed. Evidence shows hospital‑at‑home results in lower mortality, fewer complications and greater patient satisfaction, while decreasing costs and freeing up hospital bedspmc.ncbi.nlm.nih.gov.

AI‑Enabled Heart Failure Monitoring

Heart failure patients often experience fluid retention that leads to shortness of breath and hospitalization. Advanced RPM programs equip patients with digital scales, blood pressure monitors and wearable sensors that track heart rate, respiration and activity. AI algorithms analyze trends and alert clinicians to impending decompensation. In one such program, AI‑guided RPM reduced 30‑day readmissions by 70 % and slashed costs by 38 %intuitionlabs.ai. The program also improved quality of life by enabling patients to remain at home.

Diabetes Management with Continuous Glucose Monitors

Continuous glucose monitors have transformed diabetes care. Instead of intermittent finger‑stick readings, CGMs deliver near‑continuous glucose data. When integrated with RPM platforms, clinicians can adjust insulin regimens, dietary recommendations and physical activity plans in real time. Patients receive alerts when their glucose is trending high or low. Studies show that CGM‑enabled RPM helps maintain tighter glycemic control and reduces hypoglycemia episodes. Many insurance plans now reimburse CGMs for Type 1 diabetes, and their use in Type 2 diabetes is expanding.

Remote Blood Pressure Control

Uncontrolled hypertension is a major risk factor for stroke and heart attack. RPM programs supply patients with connected blood pressure cuffs and coach them via telehealth visits. Data are stored in an online dashboard, where physicians can see long‑term trends. When readings are consistently high, clinicians adjust medication doses or add drugs. A systematic review found that such monitoring leads to lower blood pressure and fewer hospitalizations compared to usual carepmc.ncbi.nlm.nih.gov. Combining blood pressure monitoring with lifestyle coaching further enhances results.

Future Directions and Emerging Innovations

The RPM landscape continues to evolve. Several emerging trends will shape its future:

Artificial Intelligence and Predictive Analytics – AI will become even more integral, enabling automatic triage, predictive modeling and decision support. Digital twins—virtual patient models built from physiological data—may help simulate interventions and optimize treatments.pmc.ncbi.nlm.nih.gov

Smart Textiles and Implantable Sensors – Wearable technology is advancing beyond wristbands. Smart fabrics and adhesive patches can monitor heart rate, respiration and hydration unobtrusively. Miniaturized implantable sensors may provide long‑term monitoring of cardiac function or intracranial pressure.

Interoperable Data Ecosystems – Efforts to adopt open standards (like HL7 FHIR) will improve data exchange among devices, EHRs and analytics platforms. Blockchain and other secure data‑sharing technologies may enhance interoperability and patient control over datapmc.ncbi.nlm.nih.gov.

Integration with Digital Therapeutics – RPM data will power personalized digital therapeutics—software that delivers cognitive behavioral therapy, medication reminders or lifestyle interventions. Combining monitoring with digital treatments could improve adherence and outcomes.

Regulatory Evolution – Regulators are increasingly recognizing RPM’s value. New reimbursement models and quality measures will encourage broader adoption, especially in value‑based care arrangements.

Social Determinants and Health Equity – Future RPM platforms will integrate non‑medical data (social determinants, environmental factors) to personalize care. Programs will need to address biases and ensure equitable access to technology.

Conclusion: Harnessing RPM for Better Health

Remote patient monitoring represents a paradigm shift toward continuous, personalized care. Evidence shows that RPM reduces hospitalizations, improves patient outcomes and lowers costspmc.ncbi.nlm.nih.govpmc.ncbi.nlm.nih.gov. Market trends suggest widespread adoption in the coming years, driven by advances in wearable technology, AI analytics and supportive policy reformspmc.ncbi.nlm.nih.govintuitionlabs.ai. To unlock its full potential, health systems must thoughtfully design RPM programs—assessing patient needs, selecting appropriate devices, planning staffing and workflows, and incorporating human touch alongside technology. Ensuring privacy, equity and usability will be crucial as RPM becomes a mainstream component of healthcare. As future innovations emerge, RPM will continue to reshape medicine, empowering patients and providers to deliver smarter, more responsive and more humane care.

Frequently Asked Questions (FAQ)

What is remote patient monitoring (RPM)?

RPM is the use of digital devices to collect health information from patients at home and transmit it securely to healthcare providers. It enables continuous monitoring of vital signs, symptoms, and behaviors, allowing clinicians to detect problems early and adjust care. RPM is distinct from telehealth video visits because it focuses on ongoing data collection (telehealth.hhs.gov).

Which conditions benefit most from RPM?

RPM is particularly useful for chronic conditions where daily measurements can guide care, including heart failure, hypertension, diabetes, COPD, asthma, perinatal care, post‑surgical recovery, obesity, vertigo, cancer and neurological disorderstelehealth.hhs.gov. Hospital‑at‑home programs also use RPM to monitor moderate‑acuity patients at homepmc.ncbi.nlm.nih.gov.

What devices are commonly used in RPM programs?

Common devices include digital scales, heart monitors (smartwatches, ECG patches), continuous glucose monitors, blood pressure cuffs, pulse oximeters, spirometers and fetal monitorstelehealth.hhs.gov. Advanced tools like digital stethoscopes (Littmann CORE, Eko DUO, Thinklabs One, TytoCare) and ambient sensors for fall detection are increasingly usedjscimedcentral.com.

How do healthcare providers implement an RPM program?

Implementing RPM involves assessing patient needs and program goals, choosing appropriate devices, planning staffing and workflows, onboarding patients and caregivers, monitoring data and refining the program. Key considerations include usability, HIPAA compliance, reimbursement requirements and integration with electronic health recordstelehealth.hhs.govtelehealth.hhs.gov.

What evidence supports the benefits of RPM?

Multiple studies show that RPM reduces hospital admissions and readmissions, shortens hospital stays, lowers costs and improves patient outcomes and adherencepmc.ncbi.nlm.nih.gov. Hospital‑at‑home programs demonstrate improved survival and satisfaction, while AI‑enabled RPM programs achieve substantial reductions in readmissions and costspmc.ncbi.nlm.nih.govintuitionlabs.ai.

What challenges does RPM face?

Challenges include data overload and alarm fatigue, integration complexity, digital divide and equity concerns, changing regulations, privacy and security risks, and patient resistance. Addressing these requires careful program design, robust technology, patient and clinician education, and ongoing evaluationpmc.ncbi.nlm.nih.govpmc.ncbi.nlm.nih.gov.

How will RPM evolve in the future?

RPM will integrate more AI and predictive analytics, incorporate smart textiles and implantable sensors, adopt interoperable data standards and blockchain for secure data sharing, and merge with digital therapeutics. As policies and reimbursement structures mature, RPM will become a core component of value‑based care and personalized medicinepmc.ncbi.nlm.nih.govpmc.ncbi.nlm.nih.gov.