# Design a remote patient monitoring system: system design interview question

> Patient vital signs: delayed readings, uncertain device clocks, clinical rules, and urgent alerts that require clinician acknowledgment.

Source: https://sysround.com/questions/design-remote-patient-monitoring-system
Updated: 2026-11-05


Difficulty: hard. Levels: Senior, Staff+.

## How the interviewer opens it

"Design a system that collects patient vital signs from connected medical devices and flags concerning readings. Devices can lose their connection, and some alerts need urgent attention."

## What to clarify first

- Fleet and traffic
- Read workload and retention
- Connections and time quality
- Clinical rules
- Alert deadlines
- Absent data
- Alert operations
- Consistency, durability, and availability
- Patient identity and access
- Clinical limits and exclusions

## Scope of a strong answer

Core requirements:

- Collect and display vital signs from enrolled devices.
- Accept delayed readings without data loss or duplicate incidents.
- Apply approved rules for concerning readings, uncertain times, and absent data.
- Deliver urgent alerts and track clinician acknowledgment.

Usually out of scope:

- Develop medical devices or sensors.
- Choose clinical thresholds or diagnose patients.
- Provide local alarms without a connection.
- Dispatch emergency services.
- Integrate with hospital record systems.
- Support active service in multiple regions.

## What interviewers probe

### Retried uploads

A lost response can repeat an accepted batch. A large reconnect wave can also delay the new readings that need urgent action.

### Delayed and out-of-order readings

Arrival order and device time can disagree. The service must preserve urgent evidence without treating unknown or old measurements as current clinical facts.

### Urgent alert delivery

Provider acceptance does not establish clinical responsibility. Lost responses, missed deadlines, and concurrent acknowledgments must not erase required action.

### Device identity and patient records

Clinics reuse devices, and delayed readings can belong to a previous patient. A wrong assignment can expose records or alert the wrong team.

## Common mistakes

- Promises timely cloud alerts for readings that remain on a disconnected device.
- Uses upload time as measurement time or lets uncertain readings postpone absence alerts.
- Creates one urgent incident for every abnormal packet.
- Treats provider acceptance as clinician acknowledgment.
- Acknowledges readings before durable storage or restarts alert deadlines after a failure.
- Assigns delayed readings to the device's current patient instead of its historical assignment.

## The bar by level

- **Senior:** Defines persistence evidence, delayed-data behavior, uncertain-time handling, and per-device absence deadlines. Handles durable acceptance, duplicate inputs, provider timeouts, ownership races, and persistent escalation. Calculates storage and replay capacity.
- **Staff+:** Defines safety boundaries across devices, the service, providers, and clinicians. Resolves clock uncertainty without inventing fresh evidence or hiding critical values. Connects deadline semantics to admission limits and recovery behavior. Defines reassignment controls, rule-version policy, regional recovery, and measurable failure limits.

Practice this question with an AI interviewer: https://sysround.com/practice/new?problem=remote-patient-monitoring


---
SysRound (https://sysround.com) is an AI practice tool for system design interviews. It has a voice or typed AI interviewer and a shared whiteboard. At the end, you get feedback on each skill and a model answer. The AI feedback can be wrong.
