Mental health clinicians were spending an hour every day searching records before each review. We moved that information work to the system, and gave the hour back to clinical care.
Whole-of-person mental health reviews require pulling threads from multiple records — medication, presentations, social history, recent contacts. Each review takes an hour of preparation, every week, for every clinician on the team. That hour comes out of clinical care.
A weekly clinical review for someone with a complex mental health presentation requires the clinician to reconstruct what has happened across the system since the last review. Medication changes recorded by one prescriber. ED presentations logged in another system. Phone contacts and home visits noted in case-management notes. Recent allied-health touchpoints, if they exist, in a third system.
Most of these systems do not talk to each other. The clinician searches each one, manually. Compiles a picture in their head. Spots — or misses — what has changed since the last review.
The result is predictable: clinicians arrive at reviews under-prepared, not because they don't care, but because the system asks them to be detectives before clinicians. Care gaps are missed not for lack of skill but for lack of time.
AutoReview surfaces curated, prioritised review lists into the case manager's workflow each week — built from EHR scans, dispensing data, presentations and contact records — so the clinician arrives at every review prepared.
Clinicians manually scanned records each week to identify which clients needed urgent review. Often discovered care gaps after a crisis had already started.
Algorithms scan EHR weekly and surface clients with detected risk markers — missed medication, recent ED contact, social-history changes — before the clinician's review preparation begins.
Clinicians reconstructed clinical pictures from fragments across systems. Pattern recognition relied on experience and time-pressured judgment.
Each surfaced client comes with a structured timeline showing changes since last review, integrated across systems, with risk flags highlighted.
Time pressure forced informal triage. Highest-risk clients sometimes received the same preparation depth as those needing routine review.
Risk-stratified workflow: review intensity follows where the algorithmic detection shows it is most needed. Clinical judgment shapes the final call.
Reviews proceeded with whatever preparation time allowed. Care planning decisions were sometimes delayed until next contact.
Clinical decisions stay with the clinician — that doesn't move. They land earlier, with better evidence, on the right clients, with care plans adjusted before issues escalate.
AutoReview is deployed across six SA Community Mental Health Centre regions as part of an MRFF national infrastructure investment. The platform integrates with state EMR systems and dispensing data feeds, surfacing prioritised review lists directly into the case manager's workflow.
The system supports — and does not replace — clinical authority. It does not generate clinical recommendations and does not act on patients without clinician review. All operations sit under formal information governance, clinical safety and HREC approval.
The numbers below are projected at full pilot scale across six CMHC regions — not yet finalised in published evaluation. Where outcomes are projected rather than verified, we say so.
The same configuration logic that worked in hospital pharmacy — moving information gathering to the system in AutoMedic — works in community mental health. Different setting. Different workforce. Different decision being made. Same method.
Each Sandpit configuration applies the same four-operation decomposition to a different capacity problem. What stays constant is the method.
Bring it to the Sandpit. We'll diagnose where capacity is lost, configure what should change, and test it in a governed live setting before you commit at scale.