Methods

This systematic review was carried out following the COSMIN guideline for systematic reviews of PROMs (version 2.0, 2024 update), and reported following the PRISMA-COSMIN for Outcome Measurement Instruments reporting guideline.17,18 The review protocol was prospectively registered in the International Prospective Register of Systematic Reviews (PROSPERO; CRD420261287006). A comprehensive summary of the methodological framework is provided in Table 1.

Evaluation stage | Analytical component | Purpose of assessment | Outcome classification
Stage 1. Evidence identification | Systematic literature screening | Identification of all studies reporting measurement properties of S-QoL instruments | Eligible studies selected
Stage 2. Methodological appraisal | COSMIN Risk of Bias tool | Appraisal of methodological rigour using a four-level rating system | Very good/Adequate/Doubtful/Inadequate
 | Instrument development | Assessment of construct definition, conceptual framework, target population, context of use, and qualitative development procedures | Qualitative judgement
 | Content validity | Evaluation of relevance, comprehensiveness, and comprehensibility from patient and expert perspectives | Qualitative judgement
 | Structural validity | Examination of factor structure or IRT/Rasch model fit and sample adequacy | Qualitative judgement
 | Internal consistency | Assessment of inter-item coherence (Cronbach's alpha), conditional on sufficient structural validity | Qualitative judgement
 | Cross-cultural validity/Measurement invariance | Evaluation of translation procedures and measurement invariance across language or cultural groups | Qualitative judgement
 | Reliability | Stability of scores over time or between raters using ICC estimates | Qualitative judgement
 | Measurement error | Analysis of measurement error indices relative to meaningful change thresholds | Qualitative judgement
 | Criterion validity | Correlation with an established reference standard, when available | Qualitative judgement
 | Hypotheses testing for construct validity | Testing predefined hypotheses for convergent and known-groups validity | Qualitative judgement
 | Responsiveness | Ability to detect change over time using hypothesis-based or subgroup approaches | Qualitative judgement
Stage 3. Evaluation of measurement performance | COSMIN criteria for good measurement properties | Judgement of each measurement property | Sufficient (+)/Insufficient (−)/Inconclusive (?)
Stage 4. Certainty of evidence | Modified GRADE approach | Overall confidence in the evidence considering bias, inconsistency, and imprecision | High/Moderate/Low/Very low
Stage 5. Instrument recommendation | PROM endorsement decision | Classification of S-QoL instruments for use in research or practice | PROM recommendation based on evidence synthesis

Search strategy and selection criteria

A comprehensive literature search was conducted across PubMed, PsycINFO, and Web of Science from database inception through 10 December 2025. The search strategy was developed in accordance with COSMIN recommendations for systematic reviews of PROMs and was structured around the target construct, instrument, and measurement properties. The search strategy combined terms for the schizophrenia-related QoL, S-QoL instruments (including S-QoL18, S-QoL41, S-QoL-MCAT, and related variants) with keywords related to measurement properties. The detailed search queries for each database are provided in Supplementary Table S1. Two reviewers (MR and MF) independently screened all records, and discrepancies were resolved by discussion with additional reviewers (LB and PA). Additionally, reference lists of included studies were manually checked to identify further relevant publications.19 No language restrictions were applied as long as an English abstract was available.

Eligibility criteria were established based on COSMIN guidelines for systematic reviews of PROMs. We included empirical studies that evaluated at least one measurement property of the S-QoL18, S-QoL41 or S-QoL-MCAT in populations with schizophrenia. There were no restrictions regarding participants’ age, clinical setting, or language of administration. All validated formats of the instruments were considered eligible, whether self-reported, interviewer-administered, or proxy-reported. Although neuroimaging studies are not part of the COSMIN taxonomy for measurement properties, studies conducted by the original instrument developers that employed neuroimaging or related imaging modalities were additionally identified and reviewed. These studies were considered as providing neurobiological support for the theoretical constructs underlying the S-QoL dimensions, rather than as evidence of formal psychometric properties. Their inclusion aimed to contextualise the conceptual robustness of the instruments and to illustrate potential neurobiological coherence between subjective QoL domains and objective brain correlates, without contributing to the COSMIN-based evaluation or evidence synthesis.

Non-original publications (e.g., narrative reviews, editorials, conference abstracts) and studies evaluating only other QoL instruments without S-QoL18 or S-QoL41 were excluded; however, reference lists of narrative reviews were screened to identify additional relevant studies.

Screening of titles, abstracts, and full texts was conducted independently by two reviewers (MR and MF). Discrepancies were resolved through discussion, with additional reviewers (LB and PA) consulted if consensus could not be reached.

Data were independently extracted by two reviewers (MR and MF) using standardized COSMIN forms. Extracted information included study characteristics such as author, year, country, population characteristics, sample size, mean age (SD), and language of administration.

In addition, details on measurement properties were collected, content validity, including structural validity (factor analysis), internal consistency, cross-cultural validity, reliability (intraclass correlation, Pearson, or Spearman coefficients), measurement error (standard error of measurement, smallest detectable change), construct validity (e.g., convergent validity), responsiveness, and floor and ceiling effects. Information required for evidence synthesis and grading was also collected where available. Discrepancies between reviewers were resolved by consensus, or, when necessary, in consultation with additional reviewers (LB and PA).

Evaluation of the quality of studies and PROMs

The evaluation of measurement properties was conducted in accordance with the COSMIN framework, encompassing the Risk of Bias (RoB) checklist, the updated criteria for good measurement properties, grading the quality of evidence, and development of recommendations. Moreover, the lack of a recognised gold standard precluded assessment of criterion validity. These assessments were performed independently by two reviewers (MR and MF), and any disagreements were resolved through discussion with additional reviewers (LB and PA).

The measurement properties of all PROMs were assessed individually in accordance with the recommended sequence outlined in the COSMIN guidelines. Initially, content validity was evaluated for each PROM. In addition to studies specifically reporting on content validity, we also retrieved and considered information related to the development process of PROMs.

Each study was first assessed using the COSMIN Risk of Bias checklist, with methodological quality rated as very good, adequate, doubtful, or inadequate. In line with COSMIN guidance, only critical methodological flaws were considered sufficient to downgrade risk of bias, thereby avoiding over-penalisation of minor limitations. The distinction between standards and criteria was carefully applied when interpreting results. Subsequently, each PROM was evaluated according to COSMIN criteria for good measurement properties, resulting in ratings of sufficient (+), insufficient (−), or indeterminate (?) (Supplementary Table S2).

The results from all available studies for each measurement property were then synthesized at the PROM level and judged against the same criteria. When findings across studies were consistent, they were summarized as sufficient (+), insufficient (−), or indeterminate (?). In cases of inconsistency between study results, COSMIN recommendations for handling heterogeneity were followed, which led either to an inconsistent (±) rating or to separate ratings for relevant subgroups of studies (+, −, ?). Indeterminate findings were excluded from the summary when other studies reported sufficient or insufficient results. The final stage of evaluation for each measurement property and PROM involved grading the overall quality of evidence using the GRADE approach. Evidence quality was categorized as high, moderate, low, or very low, based on considerations of risk of bias, inconsistency, imprecision, and indirectness.17,20 To assess construct validity and responsiveness, predefined hypotheses were established a priori and subsequently compared with the empirical findings for each PROM, as presented in Supplementary Tables S3 and S4, respectively.1,2,6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16,21,22

Data synthesis and analysis

Owing to the availability of several language versions and the substantial heterogeneity in study populations, instrument versions, outcome definitions, and statistical indices reported across studies, conducting a pooled meta-analysis of internal consistency (Cronbach's alpha) and test–retest reliability (ICC) was not methodologically justified. Alternative quantitative synthesis approaches were considered; however, the degree of clinical, methodological, and statistical heterogeneity across studies was judged to be too substantial to support meaningful aggregation of estimates. Instead, a structured best-evidence synthesis was undertaken, incorporating risk of bias, methodological quality, and consistency of findings. Results were summarised narratively and stratified by measurement property (reliability, validity, and responsiveness) and instrument type. In accordance with COSMIN recommendations. Instead, results were descriptively reported by presenting the range of values for each instrument, providing an overall summary of the available evidence.17

Formulation of recommendations

Based on the overall evaluation of content validity, other measurement properties, and the quality of evidence, a final conclusion was drawn regarding the suitability of each PROM for use. PROMs supported by sufficient content validity and adequate evidence for relevant measurement properties were considered appropriate for recommendation, whereas instruments with high-quality evidence of insufficient measurement properties were not recommended. When the available evidence did not allow a clear judgement, no definitive recommendation was made. Recommendations for future research were formulated with a focus on the most promising PROMs. If more than one instrument was considered suitable for use, feasibility and interpretability were additionally taken into account to identify the most appropriate PROM.17

Ethics statement

Current study was performed based on previously published studies. The ethical approval and informed consent were not required.

Role of the funding source

No funding was received for this work.