Long-Term Functional and Remission Outcomes After CHR-P

confidence: high · 4 sources

⚠
Working research — not a recommendation.

This page records what the evidence says a policy does, not what anyone should do about it. It has been synthesized from the cited sources but has not been independently reviewed.

What the evidence says

The sources on this page measure how people at clinical high risk for psychosis (CHR-P) fare when they do not transition to psychosis. Remission from the at-risk state is reported at very different rates. One prospective cohort followed 60 CHR patients who had not transitioned to psychosis (CHR-NT) for an average of 8 years. In that sample, 85% had remitted from their high-risk status. A meta-analysis of non-transitioned individuals found that 48.7% no longer met CHR-P criteria at a mean follow-up of about 30.7 months. A second source says fewer than half of CHR-P individuals achieve remission from their at-risk status over time, and gives no exact rate.

Remission from the at-risk label does not mean full recovery. In the 8-year cohort, 48% showed poor functional outcome and 35% had at least one axis I diagnosis apart from the CHR state. Only 28% had fully recovered clinically and functionally. The denominator for all of these figures is the 60 CHR-NT patients who took part in direct follow-up. The meta-analysis source adds that subthreshold symptoms persisted in a large share of non-transitioned individuals. It also reports diverse trajectories, including persistent symptoms and new mood, anxiety, or substance use disorders. It does not quantify the share on each trajectory.

Two baseline characteristics are linked to better outcomes, both on associational evidence. Better psychosocial functioning at baseline was associated with a higher likelihood of remission from CHR during follow-up (p = 0.003). Longitudinal findings suggest that CHR-P youth with preserved mentalising capacities are less likely to experience symptom exacerbation. That claim has no effect size and is hedged in the source.

A biological marker was tested and found nothing. Glx levels, a glutamate-related brain measure, did not predict clinical or functional outcomes in CHR-P participants. Two subgroup comparisons agree. Left anterior-dorsal cortex (LAUD) Glx did not differ between persistent and remitted attenuated psychotic symptoms (p = 0.400). It also did not differ between good and poor 12-month functional outcome (p = 0.286). Both subgroup tests are small and uncorrected.

The sources contain no treatment or intervention effects on long-term remission or function. They also contain no outcome data from follow-up beyond about 8 years.

Where the evidence splits

The remission rate among non-transitioned CHR-P individuals differs widely across sources. The 8-year cohort reports 85% remission. The meta-analysis reports 48.7% at a mean follow-up of about 30.7 months, and a second source says fewer than half remit. The follow-up lengths differ, and the cohort is a single sample of 60 patients who took part in direct assessment. The sources on this page do not test whether follow-up length, sample selection, or remission definition explains the gap. The split is unexplained by the sources on this page.

Thresholds and dose-response

No threshold established in the current sources.

Open empirical questions

Findings

Remission from CHR status among non-transitioned patients differs by source, from 85% to fewer than half

confidence — direction: highconfidence — magnitude: mediumconsensus: contested — the sources diverge on the rate, with no stated reason
mechanism
not established; the sources describe remission rates but do not explain what produces remission
effect
85% of 60 CHR-NT patients had remitted at a mean of 8 years. A meta-analysis found 48.7% no longer met CHR-P criteria at a mean of about 30.7 months. A second source states that fewer than half achieve remission.
conditions / moderators
not established. The cohort used the BPRS-E or SOPS to assess remission. The Nagasawa ARMS cohort was followed for a mean of 964.1 days (range 172–3830 days). The "fewer than half" figure has no exact rate.

Non-transition to psychosis coincides with poor functioning, other diagnoses, and persistent symptoms in many patients

confidence — direction: highconfidence — magnitude: high for the cohort figures, medium for the trajectory claimconsensus: thin — one cohort supplies the numbers, and the other source cites prior work without quantifying it
mechanism
not established
effect
at a mean of 8 years, 28% of 60 CHR-NT patients had fully recovered clinically and functionally. 48% showed poor functional outcome (PSP ≤70). 35% had at least one axis I disorder apart from the CHR state. Another source reports persistent symptoms and emergent mood, anxiety, or substance use disorders, without proportions.
conditions / moderators
not established. Assessment used the PSP and SCID-I in a single prospective cohort. The design is associational.

Better baseline psychosocial functioning is associated with a higher likelihood of remission from CHR

confidence — direction: highconfidence — magnitude: mediumconsensus: thin — one cohort
mechanism
not established
effect
higher remission rates with better baseline functioning (p = 0.003); no effect size given
conditions / moderators
the association was found in the overall CHR sample during follow-up. The source flags that a baseline predictor may not persist at ultra-long-term follow-up.

Preserved mentalising capacity is associated with a lower likelihood of symptom exacerbation

confidence — direction: mediumconfidence — magnitude: lowconsensus: thin — one source, no effect size
mechanism
not established
effect
decreases likelihood of symptom exacerbation; not quantified
conditions / moderators
not established; the source hedges with "suggest"

Glx levels show no measured effect on clinical or functional outcomes in CHR-P

confidence — direction: high for the overall null, medium for the subgroup nullsconfidence — magnitude: high for the overall null, low for the subgroup nullsconsensus: thin — one study, with consistent nulls across its outcome measures
mechanism
not established
effect
no measured effect. Glx did not predict clinical or functional outcomes. LAUD Glx did not differ between persistent and remitted attenuated psychotic symptoms (17 vs 20 participants, p = 0.400). It also did not differ between good and poor 12-month functional outcome (GAF ≥65, n = 23, vs GAF <65, n = 32; p = 0.286).
conditions / moderators
the subgroup tests were post hoc, small, and uncorrected for multiple comparisons