Gravity IV safety · MEDHA 2026

DripFix

A device by Intervene

Watches every drip, so your nurses don't have to.

A clip-on monitor for ordinary gravity IV lines that counts every drop, alerts before a flow rate drifts dangerous, and, if no one gets there in time, pinches the line shut itself.

DRIP CHAMBER · LIVE READBED 04

DRIP RATE--:--

42drops/min

Flow steady, within target

This is the same readout that sits at the bedside. No dashboard required to trust it.

The problem

A gravity IV only works if someone keeps watching it.

Unlike infusion pumps, gravity lines don't monitor themselves. A nurse sets the rate by counting drops, then has to come back, again and again, to check it hasn't drifted, kinked, or stopped. In a busy ward, that math doesn't hold up.

<15%

of drop-rate readings taken from gravity IV lines on a hospital ward were within 10% of the rate actually prescribed. Measured across 509 observations on 86 patients.

Crass & Vance, Am. J. Hosp. Pharm. PMID 3976679 · in vivo study ↗

566

patient deaths involving monitoring device alarms were reported to the FDA over four years, with alarm fatigue named a major contributor. An alarm nobody answers is not a safeguard.

AHRQ, Making Healthcare Safer III FDA MAUDE, 2005–2008 ↗

~25%

of all medication errors recorded in an Indian paediatric tertiary-hospital study were incorrect IV infusion rate alone, the single largest cause.

Pediatric Review: Intl J. of Pediatric Research DOI 10.17511/ijpr.2018.i09.02 · 2018 ↗

Multiply that across a ward of twenty drips and one nurse, and the numbers stop being numbers.

What it does

Three jobs. Only one of them is new.

Most IV monitors on the market already do the first two. DripFix adds the third.

Monitor. An LED and light sensor sit on opposite sides of the drip chamber and count every drop as it detaches and falls. Nothing touches the fluid path, so sterility and regulatory approval aren't complicated by it.

Alert. When the flow drifts from the prescribed rate, DripFix raises it three ways at once: a light, a buzzer, and a dashboard flag. In a ward full of alarms, sound alone gets ignored.

Intervene. If a dangerous deviation isn't corrected in time, a small servo mechanically pinches the tubing shut, holding the line safe until a nurse can get there. Monitoring-only devices stop at the alert. This is the difference.

Every other device on the ward alarms and waits. DripFix alarms, and then stops the error.

Where it fits

Built first for the wards with the least room for error.

A 3 kg neonate on a paediatric drip has almost no tolerance for a rate that drifts. That is where a missed kink stops being an inconvenience.

Primary wards

Also built for

Standalone or connected

Works without Wi-Fi. Works better with it.

Every feature a nurse needs is on the device itself. The dashboard is an upgrade, never a dependency.

BED 04 · PARACETAMOL IV
38 drops/min
● within target range

Rate is set right on the device with a rotary dial. No app, no pairing, no network needed to get started.
If Bluetooth drops or the ward Wi-Fi goes down, the bedside display, alerts, and occlusion still work exactly as before.
The dashboard connection only adds a second, easier way to see the same thing. It never becomes a single point of failure.

Instead of walking a full round to check twenty beds, a nurse can see at a glance which ones actually need a visit.

Click a bed to see its live reading.

How the sensing works

Nothing touches the fluid path.

The IV monitors already on the market are expensive, bulky on the pole, and dependent on ward Wi-Fi to function at all. And when one does catch a problem, announcing it is the only thing it can do. DripFix answers each of those in turn: a small clip-on jaw that reads drops optically from outside the chamber, runs every core function on the device itself, and can close the line when nobody is free to.

DripFix · frontIllustration

Hover or tap a marker

One clip. No tools. No break in the line.

The jaw opens over the top of a standard drip chamber and holds there one-handed. The fluid path is never opened, never touched, and never re-sterilised.

FitsStandard 15 & 20 drop/mL chambers
SensingOptical, non-contact
AlertsLED · buzzer · dashboard
InterventionServo tubing clamp
Power18650 Li-ion · USB-C
LinkBluetooth, optional

STEP 1

Count the drop, not the fluid

An LED and a light sensor face each other across the top of the drip chamber, where drops are still discrete. Every falling drop breaks the beam once.

STEP 2

Compute the rate

An onboard microcontroller turns drop timing into drops-per-minute, checks it against the prescribed rate, and tracks how consistent the flow has been.

STEP 3

Alert, then intervene

A deviation triggers light, sound, and dashboard alerts together. If it isn't corrected, a servo pinches the tubing shut until a nurse resolves it.

Designed for the user, not the engineer

Three decisions the ward will feel.

A gravity drip fails quietly, and the person expected to catch it is already carrying nineteen other beds. These are the choices made for that reality.

Alarm fatigue mitigation

Fewer alarms, and each one earns it

Between 72% and 99% of hospital monitoring alarms are false or non-actionable, which is exactly why staff stop hearing them. DripFix escalates in stages. The LED turns first, silently, and sound only follows if the deviation persists past a dwell window, so a patient shifting in bed never triggers a buzzer. Mute silences the sound but leaves the light red, so a muted problem stays a visible one.

Paine et al. / AHRQ, on non-actionable monitor alarms

Fail-safe by default

What happens when nobody comes

Every alarm assumes a responder. This one does not. If a dangerous rate stands uncorrected past the escalation window, the servo closes the line and holds it, turning a runaway infusion into a paused one. The clamp holds position without power, so a flat battery locks in the safe state rather than releasing it.

Error prevention

The mental math goes away

Counting drops for fifteen seconds and multiplying under interruption is where rate errors are born. The target is dialled in once and the device does the arithmetic continuously, catching drift while it is still drift, before it becomes a dose.

Ergonomics

What the hand and eye actually meet

One-handed clip. Opens over a standard chamber with no tools and no reason to open the sterile path.
Glove-friendly controls. Raised silicone buttons and a detented dial, sized for a gloved fingertip.
Readable at distance. The number up close, the colour across the bay.
Wipes clean. Sealed shell, no reprocessing, no consumable to reorder.

Hospital impact calculator

What a ward gets back.

Drag to your ward size. Every assumption behind these numbers is listed underneath, so you can challenge the model rather than trust it.

Ward bed capacity 10 beds

250100150200

Assumes every bed is running a gravity line, rechecked about 6 times per 12-hour shift.

Flow-error mitigation score 92 / 100

Watched today: 8% of infusion time With DripFix: 100%

A manual recheck verifies a line for a few minutes. The rest of the infusion runs unobserved, and that gap is where a kink or a drifting rate goes unnoticed.

2.7 h

Nurse rechecking time returned per day

about 980 hours a year

108

Manual rechecks no longer walked, per day

only flagged beds get a visit

220 h

Infusion time moved under continuous watch, per day

across 10 lines running 24 hours

How these numbers are calculated

90 seconds per manual recheck: walking to the bedside, counting for roughly 30 seconds, adjusting the clamp and charting it.
Two 12-hour shifts per 24 hours.
Rechecks taper with ward size, from roughly 6.5 per line per shift on a small unit down to 3 on a 200-bed ward, since attention per line falls as the ward grows.
15% of rechecks remain after deployment, covering genuine alerts and checks folded into clinical rounds. Time saved is calculated on the other 85%.
10-minute confidence window per manual check: the period after a check during which the line can reasonably be considered verified. Everything outside those windows counts as unobserved time.
The mitigation score is the proportion of unobserved infusion time per line that continuous monitoring removes.

This is an operational model for planning conversations, not a clinical outcome claim. It estimates recovered nursing time and monitoring coverage. It does not predict adverse event rates, and DripFix has not yet been through clinical validation.

A nurse's round, step by step

Same ward. Same nurse. What changes at each step.

Nothing about the patient changes. What changes is how much of the round is spent finding the one bed that actually needs attention.

Checking the rate

Without DripFix

Stand at the bedside, count drops for 15 to 30 seconds, then do the math to check the rate is still right.

With DripFix

An optical sensor counts every drop and shows the live rate on its own screen. No counting, no mental math.

The "just in case" walk

Without DripFix

A bag that looks low from across the room means a walk over to check, most of the time for nothing.

With DripFix

The dashboard shows every bed's status in one view, so a nurse only walks to the ones that actually need her.

The silent kink

Without DripFix

A kinked or slipped line can go unnoticed for many minutes, simply because no one happened to be standing there.

With DripFix

The sensor catches the drop in flow immediately and raises a light, sound, and dashboard alert together.

When an alert isn't enough

Without DripFix

An alarm still depends on a nurse being free to respond right away, and in a busy ward that is not guaranteed.

With DripFix

If the flow stays dangerous, a servo gently pinches the tubing shut, holding the line safe until someone arrives.

Losing the network

Without DripFix

Devices that depend on Wi-Fi or Bluetooth become dead weight the moment the connection drops.

With DripFix

Every core feature runs on the device itself. The dashboard is a bonus view, never a dependency.