General operational and educational information for corrections professionals. Not legal, medical, or compliance advice, and not a certification of compliance with any law or standard. Policies and standards vary by agency and jurisdiction; follow your facility's policy and your own legal, medical, and professional advisors.
Detox watch is two practices running in parallel: a clinical protocol owned by medical and an observation practice owned by custody. Most facilities do both reasonably well and document the seam between them poorly, which is exactly where the case gets decided.
Withdrawal deaths are among the most litigated events in local corrections, and they rarely involve a facility that did nothing. They involve a facility that started a watch, ran it through the first night, and then let the intensity fall away on day two and three, which is when alcohol and benzodiazepine withdrawal is typically most dangerous.
The structural reason is attention decay. A person who arrives visibly unwell gets watched closely. Forty hours later they are a known quantity, staff have changed twice, and the person who set the watch level is off shift. Nothing was decided to reduce the watch. It simply eroded.
The second reason is that withdrawal presents as behaviour. Agitation, confusion, refusal, and aggression get read as attitude by people who are not clinicians. The line between difficult and deteriorating is not visible from a cell door.
A stated level and interval, with an author and a time, and a stated basis for that level. Watch levels that appear in the record with no author are indefensible.
Observation across the full window, not concentrated at the front. If the clinical risk peaks between twenty-four and seventy-two hours, the observation record should show that period as the most careful part, not the thinnest.
Specific observations rather than notations. A column of identical entries through a period of escalating withdrawal is among the most damaging records a facility can produce.
Vital signs and clinical contact on your medical authority's schedule, recorded where custody can see that they happened.
Hydration and intake where your protocol addresses it. It fails early and it is easy to evidence.
An explicit escalation trigger: what custody does, and who they call, when something changes at three in the morning. Ambiguity here is what produces a delayed call.
Both sides keep records. Neither side routinely records the traffic between them, and the traffic is what the review will ask about.
Record every handoff as an event on both sides against the same clock: who passed, who received, what was observed, what was requested, and what came back. A notification with no recorded response is an unanswered message.
Record what custody was told to do and evidence that it reached the post. A clinical order that changes only in the chart will not reliably be executed by an officer who does not read the chart.
Carry the person by name across every shift change for the duration of the watch, with current stage and what has changed since the last handoff.
Reconcile clocks. Custody and clinical records that disagree by minutes produce contradictions that read as concealment rather than as configuration.
Detox cells are frequently the least observable rooms in the building and hold the people whose condition can change fastest. Virtual Patrol applies continuous attention to those designated areas and raises observable changes, such as loss of movement or a posture consistent with collapse, for a person to review immediately.
It also produces the coverage record across the twenty-four to seventy-two hour window, which is precisely the period facilities are least able to evidence afterwards, and the period where attention decay does its damage.
It makes no clinical assessment and diagnoses nothing. A person reviews every raised event before action and no facial recognition is used. Clinical protocol belongs to your medical authority.
Commonly between twenty-four and seventy-two hours for alcohol and benzodiazepines, which is usually days two and three. Follow your medical authority's protocol, and check whether your observation record actually reflects that curve.
The tendency for a watch to erode without anyone deciding to reduce it, as the person becomes familiar and the staff who set the level go off shift. It is the most common failure in a detox watch.
Evidence of what was communicated between custody and clinical, and specific observations rather than repeated identical entries.
Per your policy, with clinical input where it contemplates it. What matters for the record is that the level has an author, a time, and a stated basis.
Write it down explicitly rather than leaving it to judgment. Ambiguity about who to call and when is what produces the delayed call.
No. It surfaces observable changes for a person to look at. Clinical assessment belongs to clinicians.
