Outcomes Summary · September 2026

Six years of recovery support, measured.

Every Accountable member gets a peer recovery coach and daily remote monitoring, and that coach is one text away. We computed every figure on this page from our anonymized program data, covering everyone served from December 2020 through September 8, 2026.

-40%depression score (PHQ-9) among members who started at moderate or higher
-38%anxiety score (GAD-7) among members who started at moderate or higher
97.4%of submitted breathalyzer readings negative
82Net Promoter Score; 77% of responses a perfect 10

We prepared this summary for health plan and treatment-provider partners. It covers 10,110 members served since December 2020, and the Methodology section defines every measure.

Get the full report

13 pages, 11 figures, and the methodology behind every number. We'll email you the PDF.

We use your details to send the report and to follow up once about how Accountable works with partners. No newsletters unless you ask.

What six years of data show

Since December 2020, more than 10,000 people have gone through this model: a peer recovery coach, daily remote monitoring, and a coach who answers texts. This is what happened to them.

77%of members improved on depressionAmong members who started at moderate or higher on the PHQ-9; 53% finished below the moderate threshold.
78%of members improved on anxietyAmong members who started at moderate or higher on the GAD-7; 60% finished below threshold.
99.0%negative readings for the typical memberAmong members with 30 or more submitted breathalyzer readings; 35% have never submitted a positive.
32 minmedian reply time56% of member messages answered within one hour, 90% within 24 hours.
83%improved their recovery capitalMean score rose 10% from first to latest assessment, and every domain measured improved.
88.9%negative toxicology screensAcross all resulted screens; 96.1% of video-observed instant panels are negative.

Depression, anxiety, and craving all fall

First assessmentLatest assessment15.79.4PHQ-9 (depression)n = 281, 28+ days apart148.6GAD-7 (anxiety)n = 249, 28+ days apart4.42.5BSC (craving)n = 482, 28+ days apart

Mean first vs. latest score for members with two or more assessments at least 28 days apart. PHQ-9 range 0 to 27, GAD-7 0 to 21, BSC 0 to 12; lower is better.

Members get better. Among members who began with moderate-or-higher depression symptoms (PHQ-9 ≥ 10) and were re-assessed at least 28 days later, the mean score fell from 15.7 to 9.4 (a 40% reduction), and 53% finished below the moderate threshold.

Anxiety followed the same pattern: GAD-7 fell from 14.0 to 8.6 (38%). Self-reported craving among members who reported any craving at baseline dropped 44%, and 49% reported no craving at all at their latest check-in. Recovery capital rose in every domain measured.

The full report shows the week-by-week trajectories, the shift in PHQ-9 severity bands, and how recovery capital changes across every domain measured.

What's inside

Every figure is computed programmatically from Accountable's own raw program tables for the full reporting population; none were taken from summary dashboards.

Who we serveEnrollment growth, age and gender mix, and diagnoses on file.
The modelPeer coaching, remote breathalyzer and toxicology monitoring, messaging, groups, rewards, and care coordination, and what members work on.
Engagement and responsivenessMessage volume, reply times, off-hours coverage, and coaching session show rates.
Monitoring and testingEvery scheduled breathalyzer reading by result, toxicology screens, and what happens when a reading is missed.
Clinical outcomesPHQ-9, GAD-7, and craving change from first to latest assessment, week-by-week trajectories, and severity shifts.
Recovery capitalDomain-level change across a coach-administered recovery capital assessment.
Satisfaction and care coordinationNPS distribution and member comments, contacts with clinical teams and families, and escalation resolution.
MethodologySource tables, reporting population, every definition, and the limitations.

Ready to see these outcomes for your members?

We can report these same measures for your attributed population, on whatever cadence you need, and join them to your own claims and clinical data by member identifier.