A Recovery Estimate Reflects Symptoms and Function, Not What the Patient Reports

September 16, 2026

A Stringent Definition Puts Recovery at 13.5%

Recovery in schizophrenia is usually reported as a single number, which leaves unsaid what was assessed. A 2025 review separates three outcome types: symptom-, functional-, and appraisal-based. Appraisal-based recovery covers quality of life and personal recovery together, so quality of life is an appraisal-based outcome rather than a measure of functioning.2 A meta-analysis of 50 studies that applied one stringent investigator-rated definition found a median recovery proportion of 13.5% (interquartile range: 8.1% to 20.0%); the included studies were naturalistic, so no conclusion about treatment can be made.1 That definition required symptomatic and functional criteria to be met together, so no appraisal-based outcome entered the definition.1 Functional recovery and personal recovery are assessed from different perspectives, and improvement in one does not establish improvement in the other.

Functional and Personal Recovery Are Rated by Different People

Functional recovery, observed functioning across self-care, social interactions, leisure time activities and educational or vocational activities, is rated from outside the patient, because self-reports converge only modestly with informant reports and observable milestones; the clinician-rated Personal and Social Performance scale showed an intraclass correlation of 0.79 on repeat rating of patients whose condition did not change.2 In contrast, personal recovery, the subjective sense of living a personally meaningful life whether or not symptoms continue, can come only from the patient; the five-item, patient-reported Brief INSPIRE-O takes two minutes and reports a Cronbach’s alpha of 0.83.2 The 2025 review treats functional and personal measures as complementary rather than interchangeable.2

Patient-Reported Scores Meet High Standards, With Limits

Weighing functional and personal measures together requires the patient-reported measures to meet high validation and replication standards, which a 2026 COSMIN and GRADE review set out to test for the Schizophrenia Quality of Life questionnaire. Across 14 studies it confirmed 84.5% of construct-validity and 89% of responsiveness hypotheses, with the 18-item version best balancing performance and feasibility.3 Those results should be read with caution, because ten of the 14 studies were French and cross-cultural certainty ranged from very low to moderate; floor and ceiling effects in the 18-item version reached 38% and 35.1% in some domains against a threshold of 15% to 20% for problematic score distribution.3 The instrument discriminated between groups differing in symptom severity, remission status, and functioning, though the authors caution against cross-national comparison. No threshold yet allows one patient’s change to be called clinically meaningful.

One Measure Does Not Predict Another

Among the functional milestones, competitive employment, independent living and marriage or equivalent correlate only weakly with one another, which the 2025 review interprets as evidence that averaging across them may not produce an accurate full-range functional assessment.2 The review also reports functional and personal recovery as empirically distinct constructs, and a personal recovery score does not closely track indicators outside itself, as evidenced by the Brief INSPIRE-O, which correlates only moderately with social contacts and self-reported general health.2 A gain on one measure does not establish a gain on another.

Recovery Estimates Depend on What Was Measured and Who Rated It

No single measure here is wrong. Each is partial. A recovery estimate determined from observable criteria and a patient-reported score answer different questions. Which outcome was measured, and who rated it, decides what a recovery estimate can be taken to mean.

Infographic contrasting functional and personal recovery in schizophrenia

Financial support was provided by Bristol Myers Squibb. Psychiatrist.com independently developed the content and maintained final editorial control.

References

  1. Jääskeläinen E, Juola P, Hirvonen N, et al. A systematic review and meta-analysis of recovery in schizophrenia. Schizophr Bull. 2013;39(6):1296-1306. doi:10.1093/schbul/sbs130
  2. Correll CU, Cortese S, Solmi M, et al. Beyond symptom improvement: transdiagnostic and disorder-specific ways to assess functional and quality of life outcomes across mental disorders in adults. World Psychiatry. 2025;24(3):296-318. doi:10.1002/wps.21338
  3. Rahmati M, Foiselle M, Fernandes S, et al. Quality of life assessment in individuals with schizophrenia: a COSMIN-based systematic review of S-QoL41 and S-QoL18 incorporating neurobiological support. eClinicalMedicine. 2026;99:104152. doi:10.1016/j.eclinm.2026.104152

A Rising Quality of Life Score Does Not Mean Improved Function

September 16, 2026

What a Self-Reported Score Actually Records

Quality of life instruments are widely used as an index of how a patient is doing, and a rise in the score is interpreted as a gain in functioning. What a self-reported score actually represents is the patient’s own judgment of their life. Self-reported quality of life tracks current mood more closely than other elements of everyday functioning.1,3 A rising score therefore leaves open what improved. Symptoms explain only a modest share of the variance in the score, and what does move with symptoms is largely mood. That influence of mood on self-assessment is seen in patients and healthy people alike.

Symptoms Explain a Small Share of the Variance

Eack and Newhill quantified how much of the score symptoms account for. Pooling 56 studies from 1966 to 2005, they found only modest associations with composite quality of life: general psychopathology at r = -0.34, negative symptoms at -0.25, and positive symptoms at -0.20. General psychopathology, the strongest of the three, explained no more than 12% of the variance, and under 8% in longitudinal studies.1 Most of what a quality of life score reflects lies outside symptom severity.

Self-Reports Track Mood More Closely Than Function

One candidate for the remainder is the patient’s mood at the time of rating. Oliveri and colleagues asked whether depression shapes how patients rate their own social functioning, and included healthy controls to test whether the effect is specific to schizophrenia. Stable outpatients (n=218) and controls (n=154) self-reported their social functioning on the Specific Levels of Functioning scale. After entering sex, diagnosis, and depression as predictors, only depression entered the model and accounted for 23% of the variance in self-reported functioning. Depression correlated with self-reported functioning at r = -0.44 in patients and -0.55 in controls.2 The association held in people without the diagnosis, and the authors concluded that the effect is not confined to schizophrenia. Both measures were self-reported, so shared method variance cannot be excluded.2 A transdiagnostic review reports the same pattern at the instrument level.3

Affective Symptoms Predicted the Most Domains

Repeated assessment of the same patients separates the symptom changes that accompany quality of life change from those that do not. Murphy and colleagues analyzed an 18-month randomized trial with active psychosocial treatment in both arms, in which 102 early-course patients were assessed on the WHOQOL-BREF at baseline, 9 months, and 18 months. Increases in affective symptoms predicted lower total, physical, psychological and social relationships quality of life, and increases in positive symptoms predicted lower environmental quality of life; no other symptom class predicted any domain at P < .05.4 Results should be read with caution, because antipsychotic use was an inclusion criterion and missing data reached 29.7%.4 Affective change was the only class associated with the total score, reaching three of the four domains.

Pair the Quality of Life Score With a Symptom Measure

Quality of life scores respond to more than function. Symptoms account for only a small share of the variance in these scores, and most of that share comes from mood and anxiety symptoms rather than psychosis. An improvement in the total may reflect reduced depression or anxiety rather than gains in daily functioning. Pairing the instrument with a depression or anxiety scale separates the two.

Infographic on QoL-Score; affective symptoms impact WHOQOL-BREF domains

Financial support was provided by Bristol Myers Squibb. Psychiatrist.com independently developed the content and maintained final editorial control.

References

  1. Eack SM, Newhill CE. Psychiatric symptoms and quality of life in schizophrenia: a meta-analysis. Schizophr Bull. 2007;33(5):1225-1237. doi:10.1093/schbul/sbl071
  2. Oliveri LN, Awerbuch AW, Jarskog LF, Penn DL, Pinkham A, Harvey PD. Depression predicts self assessment of social function in both patients with schizophrenia and healthy people. Psychiatry Res. 2020;284:112681. doi:10.1016/j.psychres.2019.112681
  3. Correll CU, Cortese S, Solmi M, et al. Beyond symptom improvement: transdiagnostic and disorder-specific ways to assess functional and quality of life outcomes across mental disorders in adults. World Psychiatry. 2025;24(3):296-318. doi:10.1002/wps.21338
  4. Murphy SM, Flores AT, Sarpal DK, Keshavan MS, Eack SM. Quality of life in schizophrenia: symptomatic markers of improvement over time. Psychiatr Rehabil J. 2026;49(3):247-252. doi:10.1037/prj0000690

Learn More About Psychedelics in Psychiatry

Explore the Collection
Neural network illustration symbolizing psychedelic research advancements

Subscribe to read
more articles

Subscribe