Measuring Clinical Quality in Remote Care: PROMs vs Biometrics

At a glance

Remote care quality needs both PROMs and biometrics: biometrics show physiological change, PROMs show whether patients feel and function better.
Sheba Medical Center uses Datos Health to lift adherence for cardiac rehab and CHF patients, with real-time patient communication.
Datos Health's hospital-in-the-home programs generally start post-discharge and run 12 weeks, per its hospital-at-home page.
One Datos Health platform replaces point solutions and typically cuts cost of care per patient by 30-50%, per its hospital-at-home page.
Automated assisted self-care surfaces only the patients who need clinical attention, instead of flooding teams with raw device alerts.

Datos Health

Published: 2026-10-26

Measuring clinical quality in remote care takes both PROMs and biometrics, because each answers a different question and neither is sufficient alone. Biometrics — blood pressure, oxygen saturation, weight, heart rate and the rest of the vital-sign data streamed from connected devices — tell you what a patient's physiology is doing between visits. PROMs, short for patient-reported outcome measures, are structured questionnaires in which patients report their own symptoms, function and quality of life; their companion measure, PREMs, captures patient-reported experience of the care itself. A credible remote quality programme pairs the two: biometric trends for physiological deterioration and early warning, PROMs and PREMs for recovery, adherence and the outcomes that value-based contracts actually pay on.

That pairing is operational, not theoretical. Datos Health's hospital-in-the-home programs generally begin post-hospital discharge and last 12 weeks, providing clinical oversight through biometric data collection and patient-reported outcome measures together, according to the company's hospital-at-home page — so the measurement design is built into the pathway rather than bolted on at audit time. Datos Health is an AI-driven remote and hybrid care platform that lets hospitals, HMOs and health organisations design, automate and deploy personalised care pathways, and it was named a Leading HealthTech Company of 2025 by TIME, as published on the company's about page. For clinical and operations leaders in Australia and New Zealand running Hospital in the Home and virtual wards as of 2026, the practical question is how to collect both measure types without adding a second system, a second login and a second queue of alerts for already-stretched nursing teams.

Adherence is part of that question: per Datos Health's case-studies page, Sheba Medical Center uses the Datos Health platform to increase program adherence for cardiac rehab and CHF patients and communicate with them in real time. The sections below cover what each measure type is good for, where biometrics miss the patient's experience, which measures suit which service lines, and how to combine both in one automated pathway.

How do PROMs and biometrics differ as evidence of clinical quality in remote care?

Patient-reported outcome measures (PROMs) and device-captured biometrics differ less in rigour than in what they can witness: PROMs are structured questionnaires in which the patient scores their own symptoms, function and quality of life, while biometrics are objective physiological readings — blood pressure, oxygen saturation, weight, heart rate — streamed from connected devices. Both are evidence of clinical quality. They answer different audit questions.

Before comparing them, it helps to fix the criteria a virtual ward team is actually judged on:

What it measures — physiological state versus lived experience and function.
Data provenance — who generates the reading, and how contestable it is.
Cadence and latency — continuous or on-schedule capture versus episodic capture at defined intervals.
Common failure mode — what makes the signal unusable.
Governance fit — which quality or value-based contract requirements it satisfies.

Criterion	PROMs	Device-captured biometrics
What it measures	Symptom burden, function, quality of life	Objective physiological values
Data provenance	Patient-generated, self-scored	Device-generated, instrument-bound
Cadence	Episodic, at defined intervals in the pathway	Scheduled or near-continuous
Failure mode	Survey fatigue and non-completion	Device non-use, poor technique, connectivity gaps
Governance fit	Value-based care, patient experience reporting	Early Warning Scores, escalation and safety review

PROMs are decisive where the clinical goal is function — cardiac rehabilitation, oncology supportive care, perioperative recovery — because no vital sign records whether a patient can climb stairs again. Biometrics are decisive where deterioration is physiological and time-sensitive, such as congestive heart failure or chronic obstructive pulmonary disease.

Mature Hospital in the Home programs run both streams inside one pathway rather than two disconnected systems. Datos Health supports that by collecting patient-reported measures and connected-device readings in the same interactive care plan, with the results flowing back through its EHR/EMR integration so the clinical record holds both forms of evidence side by side.

What do PROMs, biometrics and clinical quality actually mean in a virtual ward setting?

In a virtual ward, PROMs, biometrics and clinical quality measures sit in different data categories, and a Hospital in the Home team should define each one before choosing what to track. This section covers that narrow case only: hospital-level care delivered in the patient's home, where the care team never physically sees most of the data it acts on.

The five data types, and what each one is for

PROMs (patient-reported outcome measures) — structured questionnaires the patient completes about symptoms, function and quality of life, scored on validated instruments. They capture what no sensor records: breathlessness on stairs, pain, confidence managing medication.
PREMs (patient-reported experience measures) — the patient's rating of the care process itself: communication, access, feeling heard. Useful for value-based contracts and for explaining why patients drop out of a pathway.
Biometric telemetry — vital signs collected from connected devices, such as blood pressure, oxygen saturation, weight, pulse and glucose. This is what feeds Early Warning Scores (EWS) and escalation thresholds.
Adherence data — whether the patient actually did what the care plan asked: readings taken, questionnaires returned, education completed. A missed reading carries clinical information of its own.
Clinical quality measures — the program-level indicators your governance committee reviews, such as escalation response time, unplanned readmission and pathway completion.

Three attributes to fix per measure before go-live

For every item you plan to track, agree its source (device, patient, or the EHR/EMR record), its cadence (how often it is expected, and what counts as overdue), and its action threshold (the value or score that triggers a named response, and who owns that response). A measure with no named owner and no threshold simply adds to alert noise.

Datos Health carries both streams inside a single interactive care plan, so the definitions a team fixes at this stage become the configuration it builds and adjusts in the no-code Design Studio.

When does a biometric reading miss what the patient is actually experiencing?

A biometric reading — a measured physiological value such as blood pressure, oxygen saturation or weight captured by a connected device — misses what the patient is living through whenever the body compensates. Values sit inside threshold while symptom burden, function or quality of life quietly deteriorates. Part of the confusion is that "stable" is doing two jobs in home-based programs.

Physiological stability means the numbers hold: saturation above the alert floor, weight flat day to day, heart rate unremarkable. A COPD patient can hold saturation while becoming too breathless to climb the stairs she managed a fortnight ago.

Experienced stability means the person reports no worsening in symptoms, daily function or wellbeing. This is the territory of PROMs and PREMs — patient-reported outcome measures and patient-reported experience measures, structured questionnaires completed by the patient about symptoms, function and the care they received. An oncology patient on treatment may have textbook vitals alongside escalating fatigue, nausea and neuropathy.

Throughout this article, "stable" means both together, because device data alone can look unchanged while the patient's experience worsens.

Common scenarios where device data looks reassuring:

Cardiac rehab: exercise tolerance and confidence fall before any resting vital sign shifts.
Chronic heart failure: weight and blood pressure in range while breathlessness, sleep and swelling worsen.
Oncology: treatment toxicity presents as symptom burden with no vital-sign correlate.
Perioperative recovery: pain, wound anxiety and reduced mobility sit outside the biometric set entirely.
Diabetes: glucose within target while hypoglycaemia fear reshapes eating, driving and work.

Datos Health closes this gap by collecting device values and patient-reported measures inside the same automated pathway, so a worsening symptom or function score can surface the patient for clinical attention even when every connected-device value is sitting comfortably inside its threshold.

Which measures suit which service lines and care pathways?

When deciding which measures suit a given service line, the useful starting point is what the pathway is trying to influence and how quickly a change becomes visible. Each pathway carries its own mix of measure classes, and in Datos Health those sets are configured per pathway rather than applied uniformly across every program.

The attributes that define a measure set:

Biometric signals — values such as blood pressure, oxygen saturation, heart rate, weight, temperature and glucose, captured from connected devices. They matter when physiological drift precedes a clinical event.
PROMs (patient-reported outcome measures) — validated symptom, function and quality-of-life scales completed by the patient. They matter when the outcome that counts is not visible in a vital sign.
PREMs (patient-reported experience measures) — structured feedback on the care experience itself, used where retention, engagement and contract reporting are in scope.
Adherence and engagement signals — task completion, questionnaire response and session uptake within the care plan, useful as an early indicator that a patient is disengaging.

Pathway	Biometric emphasis	Patient-reported emphasis
Cardiac rehab	Heart rate, blood pressure, activity and steps	Functional capacity, exercise confidence, PREMs for retention
Oncology supportive care	Weight, temperature, pulse	Symptom burden and toxicity scales, quality of life
Chronic disease management (CHF, COPD, diabetes)	Weight, oxygen saturation, glucose or continuous glucose	Breathlessness, symptom diaries, self-management confidence
Post-surgical and perioperative recovery	Temperature, respiration, wound-related checks	Pain scores, recovery milestones, return-to-function
Hospital in the Home / virtual wards	Multi-vital sets feeding Early Warning Scores (EWS)	Daily symptom check-ins alongside clinical escalation rules

Hospital in the Home sits at the intensive end of this range: the measure set has to support day-to-day escalation decisions for an acutely unwell patient while still producing an end-of-program outcome view for the admitting service. High-risk pregnancy pathways follow a similar dual pattern, pairing scheduled biometric checks with structured self-report between appointments.

How do you combine PROMs and biometrics inside one automated pathway?

Combining PROMs and biometrics in a single pathway works best when both data types are designed together, not bolted together after launch. PROMs — patient-reported outcome measures, the structured questionnaires patients complete about symptoms, function and quality of life — tell you how the patient is doing. Biometrics, the device-captured vital signs such as blood pressure, oxygen saturation and weight, tell you what the body is doing. The steps below are written for teams at the build-and-launch stage, who have already selected a platform and are configuring a live service line.

What are the build steps?

Define the cohort and the quality question. Name the population (cardiac rehab, CHF, COPD, post-surgical) and decide which validated PROM instrument and which vital signs actually answer it.
Start from a pre-built program, not a blank canvas. In the Datos Health no-code Design Studio, clinical teams modify an existing care program themselves instead of queuing a change request with IT.
Set two separate cadences. Connected devices push readings continuously or daily; questionnaires are scheduled at clinically meaningful intervals — baseline, mid-program and discharge.
Write one escalation rule covering both streams. Combine threshold breaches, Early Warning Scores and questionnaire responses into the same logic so only patients needing clinical attention surface.
Integrate with the EHR/EMR so results land in the patient record rather than a second portal nobody opens.
Pilot, then iterate in the studio. Adjust thresholds and question sets once real data arrives.

Where both streams feed a single escalation rule, a reported symptom can confirm or discount a borderline reading before anyone is paged, so alert volume becomes a property of how the pathway logic is written, and sensor count alone predicts little about it.

Put a standing review of that logic in the governance plan: thresholds drafted before go-live rarely survive contact with a real cohort, and the teams that revise them keep escalation volume inside what existing staff can absorb.

Frequently Asked Questions

What is the difference between PROMs and biometrics when measuring clinical quality in remote care?

Measuring clinical quality in remote care draws on two complementary data streams: PROMs and biometrics. PROMs — patient-reported outcome measures — are structured questionnaires in which patients report symptoms, function, pain and quality of life in their own words; PREMs, their sibling, capture the experience of care itself. Biometrics are device-captured physiological readings such as blood pressure, oxygen saturation, weight and heart rate. PROMs describe how the patient is living; biometrics describe how the body is performing, and quality programs typically track both.

How do teams collect both without drowning in alerts?

Alert fatigue is the usual reason quality programs stall, so the collection method matters as much as the measures. Datos Health uses interactive care plans and automated assisted self-care — patients self-manage parts of their pathway through guided, automated steps — to raise adherence and surface only the patients who need clinical attention. Questionnaire prompts, education and device readings run on the pathway's own schedule, so the clinical team reviews a filtered, prioritised list rather than a raw feed of every reading.

Which vital signs can be captured alongside patient-reported measures?

Datos Health's published integrations table on its integrated-devices page lists 19 connected devices and platforms, spanning glucose, continuous glucose, blood pressure, oxygen saturation, temperature, respiration, pulse, heart rate, weight, workout, steps and sleep. That breadth matters for chronic care management, because a heart-failure pathway, a COPD pathway and a diabetes pathway each need a different biometric mix paired with condition-specific questionnaires. Datos Health is device-agnostic, so each pathway's device mix can be chosen to fit its condition.

How long does a program need to run before quality data becomes meaningful?

Program duration determines whether trend data is interpretable. Per Datos Health's hospital-at-home page, its hospital-in-the-home programs generally begin post-hospital discharge and last 12 weeks, providing clinical oversight through biometric data collection and patient-reported outcome measures. A window of that length lets teams establish a baseline, observe change across repeated questionnaire cycles, and compare physiological trends against reported symptoms — something a handful of isolated readings cannot support. Other pathways can be built and adjusted in the no-code Design Studio to suit their own clinical goals.

Does measuring quality remotely actually pay for itself?

It can. Per Datos Health's hospital-at-home page, its hybrid care platform typically reduces the cost of care per patient by 30-50% in hospital-in-the-home programs. One device-agnostic Datos Health platform with EHR/EMR integration also replaces multiple point solutions, and it is licensed per patient as SaaS with no change fees. In the US, Datos Health supports RPM and RTM reimbursement, which are US billing code families, alongside value-based care contracts; for Australian and New Zealand Hospital in the Home and virtual ward services, the cost-of-care case is the relevant one. Where value-based agreements are in place, PROMs and PREMs are often their reporting currency.

What should a governance review cover before scaling a measurement program?

Three things, in practice. First, measure definitions: every tracked item needs an agreed source, cadence and action threshold, with a named owner for the response. Second, integration: results from questionnaires and devices should flow into the EHR or EMR so they sit in the clinical record rather than a parallel system. Third, external validation: per Datos Health's about-us page, TIME named Datos Health a Leading HealthTech Company of 2025, and per its KLAS Research page, KLAS Research published an Emerging Technology Spotlight report on the platform. Both are reference points among the due-diligence materials a procurement team will gather.

About this article

Datos Health publishes this article under its own name and is responsible for its accuracy. Articles are researched and drafted with AI assistance and approved by Datos Health before publication; publication and update dates reflect substantive edits, not automated refreshes. Last updated: 2026-10-26