When the participant wakes at 2am, who's counting?

A support worker is on a sleepover. At 2:10am the participant wakes, needs help to the bathroom, is unsettled for twenty minutes. At 4:30am it happens again, briefly.

By 9am, when the worker signs off, what's the record of that night?

For most providers, it's the worker's memory. Sometimes a message sent to the team WhatsApp group. Sometimes nothing at all, and the question gets asked at the end of the fortnight when someone is preparing the pay run: did anything happen on Tuesday night?

The rule is clear. The record isn't.

Under the SCHADS Award, a sleepover is a rest period covered by an allowance. But if a worker has to get up and provide support during it, that's work — and it's paid separately, at overtime rates, with a minimum of one hour no matter how short the interruption was.

So a five-minute settle at 3am is an hour's overtime. Three separate interruptions totalling half an hour is still an hour's overtime, because the Award looks at total time worked during the sleepover period.

The rule isn't the hard part. Knowing what happened is.

Where it actually goes wrong

Nobody sets out to underpay a support worker for a broken night. It happens because the information arrives too late and too vague to act on.

The record is a recollection. A worker asked on Friday what time a participant woke on Tuesday gives an honest answer that is also a guess. Ten minutes becomes “about twenty”. Two interruptions become “a couple”.

It's scattered across the wrong tools. Times sit in WhatsApp messages, shift notes, or a text to the coordinator. Someone has to scroll back through a group chat to reconstruct a night that happened nine days ago.

It relies on the worker raising it. A worker who's been up twice overnight and is tired at 7am may simply not mention it. The interruption pay is an entitlement, but nothing triggers it unless somebody speaks up.

And it's invisible in the pay run. The sleepover allowance appears because it's attached to the shift. Interruption pay doesn't, because there's nothing in the system that knows it happened.

The result is quiet, consistent underpayment of exactly the work that's hardest to do — being woken in someone else's home at 3am and staying calm.

What it should look like

The record should be made at the moment, by the person who was there, and it should flow into pay without anyone re-entering it.

In Nuvela, a worker on a sleepover marks the participant as awake when it happens, and marks them settled when it's over. That's two taps, on a phone that's already in the room. The times are captured then and there rather than reconstructed later.

Those timestamps become the interruption record for that shift. When the pay run is prepared, the interruption is already there — the total time during the sleepover, the one-hour minimum applied, at the right overtime rate for that time of night. Nobody has to remember, ask, or scroll back through a chat.

The same record answers the other question a provider eventually gets asked, which is what actually happened overnight. A night with four interruptions on a participant's record is information — about their support needs, about whether the sleepover is the right arrangement at all, and about whether an active night shift is closer to what's really being provided.

The thing worth checking in your own business

Whatever software you use, one question is worth asking this week: if a participant woke twice last Tuesday night, where would you find that now?

If the answer is a group chat, a worker's memory, or nowhere, then your interruption pay is being estimated rather than calculated. It probably runs short, and it's the kind of shortfall that accrues quietly across every sleepover, every fortnight.