The relationship is the intervention.
Beheld pairs each member with a caller they come to trust — and members pick up, again and again. Randomized trials prove that relationship improves depression, loneliness, and glycemic control, and surfaces what members never tell the system.
Published in JAMA Psychiatry · JAMA Network Open · Psychiatric Services — covered by The New York Times · Scientific American · STAT · AMA — recognized by the Christensen Institute
THE PROBLEM
Who needs help, and when? Everything you've built runs on that question.
You've assembled real talent — clinicians, care managers, CHWs — but knowing who and when still runs on cold outreach: staff dialing lists, now AI bots dialing the same lists. Members don't pick up for protocols, human or automated — so capacity is spent prospecting, and the members who most need you surface in crisis.
Beheld inverts it: a relational layer members actually answer — one that knows who needs help and when, and delivers them to your teams ready to act.
THE MODEL
Everyone, held. Your teams, where they're needed most.
Every at-risk member gets a consistent, empathetic caller — no checklist, no agenda, no clock. For most, the relationship itself is enough: they feel steadier, act sooner, and stay out of crisis.
One layer covers your whole at-risk population — at less than half the cost of extending your care teams to reach it.
One layer covers your whole at-risk population — at less than half the cost of extending your care teams to reach it.
IT WORKS
They pick up. Again and again.
App opens and call attempts measure activity. A member choosing to answer their caller — month after month — measures trust.
3+
Conversations a month — the member picks up for a conversation they want.
75 - 85%
Still engaged at six months, in some of the hardest-to-reach populations.
98
Net Promoter Score — at a 65%+ response rate, itself a measure of engagement
RESULTS
Outcomes Follow
13 → 9
Depression (PHQ) in four weeks, vs. no change in controls
-1.1%
A1c among patients with subthreshold or higher depressive symptoms, relative to usual care.
26% vs. 38%
Members with one or more ED visits, vs. non-enrolled members in the same plan
20%
Lower pre-term birth rate, vs. non-enrolled members in the same plan.
Clinical results from randomized controlled trials — JAMA Psychiatry (2021), Psychiatric Services (2023), JAMA Network Open (2024). Program results from Medicaid claims. Full detail on the Evidence page.
Who We Serve
-
Pay for Quality, Performance I Projects tied to HEDIS targets.
An always open channel for members in the context of evolving policy requirements.
-
We are loneliness experts, and our trials not only demonstrate reductions in loneliness but also how loneliness interventions improve physical health. MORE.
-
As you bring on populations we help you understand them fast, rapidly building out your presence and understanding of patients for whom you are at risk but barely know until too late. MORE.
-
We rapidly help you understand who’s attributed to you. With Medicare FFS we bring relational layer to patients who you’re not in touch with but will affect your bottom line. MORE.
-
At a fraction of the cost, be in touch with everyone you’re looking after, making sure those who need to come in are carefully served, and ready to utilizatio. MORE.
INSIGHTS
Know the members you've never met
What you don’t know, will hurt. A member had stopped grilling — something his caller knew he loved. Asked why, he shared what he'd told no one: embarrassment about a smell. A possible foot ulcer, escalated that day — invisible in his chart, his claims, and every screening tool.
Survey-less surveys. Every conversation deepens a picture no survey captures: comorbidities, social needs, mood, motivation — naturally prioritized by what matters to the person. Just took on thousands of members you know nothing about? This is the fastest way to understand them.
