The Rhinoplasty Rosetta Stone: Using Rasch Analysis to Create and Validate Crosswalks between the NOSE and the SCHNOS Functional Subscale

van Zijl FVWJ, Declau F, Rizopoulos D, Datema FR.

PubMed 38551458 Publisher (DOI)Free full text (PMC)

What this paper says

Using data from 552 rhinoplasty patients who completed both questionnaires, the authors built conversion tables letting a NOSE score be translated into a SCHNOS breathing score and back, with agreement of 0.93.

Overview

Two questionnaires dominate the measurement of breathing outcomes after rhinoplasty: the NOSE scale and the obstruction subscale of SCHNOS. Because they score on different metrics, results from practices using one cannot be compared with the other. The authors used Rasch analysis, a statistical method that places different items on a common measurement scale, to link them.

Sections of note

Design: measurement study using 552 rhinoplasty patients who completed both instruments, plus a second independent sample for validation.
Separate Rasch models were run for each instrument, then the underlying construct was estimated from the calibrated item parameters.
Original scores from both instruments were anchored to the shared Rasch scale, linking them.
Total scores on the two instruments were strongly correlated.
The co-calibration model fitted the data adequately.
Back-and-forth crosswalk tables were produced in both directions.
For patients with moderate blockage, predicted SCHNOS scores were slightly higher for a given NOSE score.
Intraclass correlation between predicted and actual scores was 0.93 in both directions, described as adequate for group level comparisons.

What it means for a patient

This does not change treatment. It allows results from different practices and studies to be pooled.
The stated adequacy is for group level comparison, so converted scores should not be read as precise for an individual.
The two instruments diverge slightly at moderate blockage, which is where most rhinoplasty patients sit.
Validation in a second independent sample is a methodological strength.

Why this paper matters

Rhinoplasty outcome research is fragmented partly because practices use different questionnaires, so results cannot be aggregated. A validated conversion between the two dominant instruments removes that barrier for breathing outcomes. No equivalent link exists yet for the appearance subscales.

Terms

NOSE scale: a questionnaire measuring nasal blockage, where higher means worse.
SCHNOS: a validated questionnaire with separate subscales for nasal appearance and blockage.
Rasch analysis: a statistical method placing questionnaire items on a common interval scale.
Crosswalk: a table converting a score on one instrument to the equivalent on another.
Intraclass correlation coefficient: a measure of agreement between two sets of values.
Patient-reported outcome measure: a questionnaire completed by the patient rather than the clinician.

Summary written by rhinoplasty.cc from the abstract, 2026-09-09; not medical advice. The authors' own abstract follows.

Abstract

Background: The Nasal Obstruction Symptom Evaluation (NOSE) and the Functional subscale of the Standardized Cosmesis and Health Nasal Outcomes Survey (SCHNOS-O) are widely used patient-reported outcome measures to measure functional outcomes of rhinoplasty. However, as different instruments produce scores on different metrics, results of these instruments cannot be linked directly, thus hindering comparison and aggregating of rhinoplasty outcome data from practices using either instrument. The aim of this study was to develop and validate crosswalks between the NOSE and the SCHNOS-O.

Methods: In a sample of 552 rhinoplasty patients who completed both instruments, the NOSE and SCHNOS-O scales were co-calibrated onto a common interval-scaled metric using Rasch analysis. Separate Rasch models were run per instrument, and the latent constructs were estimated using the calibrated item parameters. By anchoring original patient-reported outcome measure scores of both instruments to this Rasch computed measurement scale, the scores of both instruments were linked. A second independent sample was used to validate the created crosswalks.

Results: Total scores on the NOSE and SCHNOS-O were strongly correlated. The Rasch-based co-calibration of the NOSE and SCHNOS-O items resulted in a model that adequately fitted the data. Back-and-forth crosswalk tables were created from the NOSE to the SCHNOS-O. For patients with moderate nasal obstruction, predicted SCHNOS-O scores were slightly higher for a given level of the NOSE. Intraclass correlation coefficients between predicted and actual scores were 0.93 for both directions, indicating adequate agreement for group-level comparisons.

Conclusions: This study developed and validated Rasch-based crosswalks from the NOSE to the SCHNOS-O and vice versa. The provided crosswalks enhance comparison and harmonization of functional rhinoplasty outcomes.

Abstract as indexed by PubMed; the article is open access (PubMed Central).