Caring Contacts added nothing beyond good usual care in active suicidality

Design

Blinded, two-group randomised clinical trial

Population

709 American Indian and Alaska Native adults with active suicidality

Primary outcome

Suicidal ideation, attempts and hospitalisations over 12 months

Effect

No significant between-group difference; ideation about halved in both arms

The CARE trial randomised 709 adults from five American Indian and Alaska Native communities, all with active suicidality, to culturally adapted Caring Contacts plus optimal available resources, or to those resources alone. Caring Contacts are brief, undemanding messages expressing care, sent by text or card on a fixed schedule over 12 months. Most participants (85%) had a previous suicide attempt.

Suicidal ideation scores roughly halved in both arms over 12 months, and the proportion reporting a past-year suicide attempt fell from about 45% to about 15% in each. There was no significant difference between groups on ideation, attempts or hospitalisations. Only 64% provided 12-month data.

Caring Contacts have been widely adopted on the strength of earlier trials. This result does not show they are useless — the comparison arm received compassionate baseline engagement and strong community resources — but it suggests the messages add little when good care is already in place. Services with limited staff time may get more from the first contact and follow-through than from a message programme.

• Do not rely on scheduled caring messages as a substitute for active follow-up
• Prioritise a strong first contact, safety planning and access to community resources
• Caring messages remain reasonable as a low-cost addition where follow-up is otherwise thin
• Expect improvement in many patients over a year; keep reviewing those who do not improve

Why it matters

Challenges the assumption that a cheap message programme reliably reduces suicidal behaviour on top of decent care.

Don't overread it

The trial was in specific Indigenous communities with strong comparison care; it does not show Caring Contacts fail where follow-up is absent.

The statistics, in plain English

Both groups improved by similar amounts, so the improvement cannot be credited to the messages. Some of it is likely regression to the mean: people enrolled at a peak of distress tend to improve anyway. With 36% missing at 12 months, a small true difference could have been missed.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.