Cross-cultural adaptation and psychometric validation of the Spanish version of the Nasal Polyposis Quality of Life Questionnaire (NPQ)

Abstract

Introduction

To perform the cross-cultural adaptation into Spanish and the psychometric validation of the Nasal Polyposis Quality of Life Questionnaire (NPQ) for use in patients with chronic rhinosinusitis with nasal polyps (CRSwNP).

Methods

A total of 140 adult patients diagnosed with CRSwNP were included. The original NPQ was adapted from Italian into Spanish following established international guidelines. Construct validity was assessed through exploratory factor analysis (EFA) with principal components. Internal consistency was analyzed using Cronbach’s alpha coefficient. Test–retest reliability was determined by the intraclass correlation coefficient (ICC). Concurrent validity was examined using Pearson correlations with the SNOT-22 and NOSE-E questionnaires. Discriminant validity was analyzed using ROC curves.

Results

EFA revealed a five-factor structure explaining 67% of the total variance. The Spanish NPQ showed excellent internal consistency (Cronbach’s α = 0.917) and high test–retest reliability (ICC = 0.991; 95% CI: 0.983–0.996). The standard error of measurement (SEM) was 2 points, and the minimal detectable change (MDC95) was 5 points. Strong correlations were observed with the NOSE-E (r = 0.802) and SNOT-22 (r = 0.870). The NPQ total score demonstrated high discriminative ability for severe symptoms (AUC = 0.947), with a cutoff point >41 providing 88.9% sensitivity and 93.3% specificity.

Conclusions

The Spanish version of the NPQ is a valid, reliable, and clinically useful instrument for assessing symptom burden in patients with CRSwNP. Its multidimensional structure allows capturing the specific impact of this phenotype and supports its use in both clinical practice and research.

Resumen

Introducción

Realizar la adaptación transcultural al español y la validación psicométrica del cuestionario Nasal Polyposis Quality of Life Questionnaire (NPQ) para su uso en pacientes con rinosinusitis crónica con pólipos nasales (RSCcPN).

Metodología

Se incluyeron 140 pacientes adultos diagnosticados de RSCcPN. El cuestionario original NPQ fue adaptado del italiano al español siguiendo las directrices internacionales establecidas. La validez de constructo se evaluó mediante análisis factorial exploratorio (AFE) con componentes principales. La consistencia interna se analizó con el coeficiente alfa de Cronbach. La fiabilidad test–retest se determinó mediante el coeficiente de correlación intraclase (CCI). La validez concurrente se examinó mediante correlaciones de Pearson con los cuestionarios SNOT-22 y NOSE-E. La validez discriminativa se analizó mediante curvas ROC.

Resultados

El AFE reveló una estructura de cinco factores que explicaban el 67 % de la varianza total. El NPQ en español mostró una excelente consistencia interna (α de Cronbach = 0,917) y una alta fiabilidad test–retest (CCI = 0,991; IC 95 %: 0,983–0,996). El error estándar de medida (EEM) fue de 2 puntos, y el cambio mínimo detectable (CMD95) de 5 puntos. Se observaron correlaciones fuertes con el NOSE-E (r = 0,802) y el SNOT-22 (r = 0,870). La puntuación total del NPQ presentó una alta capacidad discriminativa para síntomas graves (AUC = 0,947), con un punto de corte >41 que ofreció una sensibilidad del 88,9 % y una especificidad del 93,3 %.

Conclusiones

La versión española del NPQ es un instrumento válido, fiable y clínicamente útil para evaluar la carga sintomática en pacientes con RSCcPN. Su estructura multidimensional permite captar el impacto específico de este fenotipo y respalda su uso en la práctica clínica y en investigación.

Introduction

Chronic rhinosinusitis with nasal polyps (CRSwNP) is a persistent inflammatory disease of the nasal and paranasal mucosa. It is characterised by the presence of bilateral nasal polyps, as well as symptoms such as nasal obstruction, facial pressure, rhinorrhoea, and a significant decrease or loss of smell.¹ It is estimated to affect up to 4% of the general population, significantly impacting quality of life due to its association with sleep disturbances, fatigue, and social and emotional impairment.²

The assessment of disease burden in CRSwNP increasingly relies on patient-reported outcome measures (PROMs), which provide essential insight into the patient's subjective experience beyond endoscopic or radiological findings. Tools such as the 22-item Sinonasal Outcome Test (SNOT-22) or the NOSE scale are widely used in clinical practice and research. However, they are not specific to CRSwNP and may not reflect symptoms that are particularly relevant to this phenotype, such as anosmia, oral manifestations, or concerns about disease recurrence and surgical outcomes. These symptoms are more frequent and disabling in patients with nasal polyposis than in those with CRS without polyps.^3–5

To address this need, Baiardini et al. developed the 27-item Nasal Polyposis Quality of Life (NPQ) questionnaire, which is specifically designed to measure the physical, emotional, and functional impact of CRSwNP.⁶ The NPQ was originally developed and validated in Italian in 2022, demonstrating excellent psychometric properties and strong correlations with other general rhinopathology PROMs. However, despite the global relevance of Spanish-speaking populations, there is currently no validated version in Spanish.

The cross-cultural adaptation of PROMs requires systematic validation to ensure conceptual, semantic, and cultural equivalence, and is not merely a literal translation. International guidelines, such as those proposed by Beaton et al., the ISPOR working group, and the COSMIN initiative emphasise the need for rigorous direct and back translation, expert consensus, pilot testing, and formal psychometric validation.^7–9

The aim of this study was to perform the cross-cultural adaptation and psychometric validation of the Spanish version of the Nasal Polyposis Quality of Life (NPQ) questionnaire for use in patients with CRSwNP. The instrument is designed for use in clinical settings for diagnosis, treatment evaluation, and longitudinal follow-up, whether surgical or pharmacological.

Materials and methods

Study design

A cross-sectional validation study was conducted at the Public University Hospital of Jaén in Spain between July and December 2024. The study protocol was approved by the Andalusian Research Ethics Committee (SICEIA-2024-001376), and written informed consent was obtained from all participants. The study was conducted in accordance with the principles of the Declaration of Helsinki and current national regulations.

Translation and cross-cultural adaptation

The original NPQ was developed and validated in Italian. Although the results were published in English, there is currently no official, validated English version. Therefore, this study used the Italian version as the source document for its translation and cross-cultural adaptation into Spanish.

The translation process followed widely accepted international guidelines^7–11:

• 1

  Direct translation: Two native Spanish translators with medical knowledge produced independent translations of the original version.

• 2

  Synthesis: A multidisciplinary committee of experts, consisting of otorhinolaryngology specialists, a methodologist, and a linguist, produced a consensus version.

• 3

  Back translation: Two independent translators, native speakers of the original language and unrelated to the NPQ, undertook the back translation into Italian.

• 4

  Review by a committee of experts: All versions were compared to ensure conceptual, semantic, and idiomatic equivalence.

  A limited number of discrepancies were detected during the expert review and back translation phases. These were resolved by consensus and are detailed in Annex 1.

• 5

  Pilot test: The pre-final Spanish version was administered to 20 patients with CRSwNP to assess clarity, comprehension, and cultural appropriateness. No modifications were necessary after this step.

  To ensure conceptual and metric equivalence with the original instrument, the original NPQ scoring system (Likert scale from 1 to 5 per item, total score from 27 to 135) was retained.

Participants

A total of 140 Spanish-speaking adults with CRSwNP were consecutively recruited for the study. The selection process is summarised in Fig. 1. The diagnosis was confirmed by a senior ENT specialist in accordance with the EPOS 2020 criteria.¹

Figure 1.

Flow chart of the patient recruitment process.

Inclusion criteria: Age ≥18 years; native Spanish speakers; and bilateral nasal polyps confirmed by endoscopy or imaging. Exclusion criteria: Central nervous system disorders; active neoplasms; acute infections; previous head/neck radiotherapy; or significant systemic diseases.

A trained researcher recorded sociodemographic data (age, sex, weight, height, and BMI). A subgroup of 35 participants completed the NPQ again after 7–14 days to assess test–retest reliability.

Variables and instruments

The sociodemographic variables collected included age (in years), gender (classified as male, female, or non-binary), weight (in kilograms), height (in metres, measured with a Detecto Model 2391 mechanical medical scale with height gauge), body mass index (BMI, calculated as weight in kilograms divided by height in metres squared), and medical diagnosis. These data were coded by a researcher who was not involved in the clinical diagnosis or selection of participants, in order to reduce bias.

The Nasal Polyposis Quality of Life (NPQ) questionnaire⁶ is a 27-item PROM designed specifically to assess the impact of CRSwNP on health-related quality of life (HRQoL). Each item reflects a symptom or functional limitation experienced in the previous two weeks and is scored from 1 ('none') to 5 ('severe'). Higher scores indicate greater impairment.

To clarify the response anchors, 'no symptoms' was equated to 1 and 'maximum intensity' to 5; therefore, the theoretical minimum NPQ score (27 points) corresponded to an asymptomatic patient, maintaining metric equivalence with the original instrument.

Statistical analysis

Statistical analyses were performed using IBM SPSS Statistics for Windows, version 27 (IBM Corp., Armonk, NY, USA) and MedCalc® Statistical Software, version 22.021 (MedCalc Software Ltd., Ostend, Belgium). A two-tailed p-value < .05 was considered statistically significant, and all confidence intervals (CI) were calculated at 95%.

Descriptive statistics were used to characterise the sample. Continuous variables were expressed as means and standard deviations, and categorical variables were expressed as frequencies and percentages. Normality was assessed using the Kolmogorov–Smirnov test.

Floor and ceiling effects were calculated as the percentage of patients who obtained the minimum (27) or maximum (135) score; a value >15% was considered relevant.¹²

Construct validity was examined using exploratory factor analysis with principal components and Varimax rotation. Sample adequacy was assessed using the correlation matrix determinant, the KMO index, and the Bartlett's sphericity test.

As domains such as sleep, fatigue, and emotional impact may be correlated, this potential dependency was considered when interpreting the structure, and will be addressed in the confirmatory phase.

In PROMs, it is often recommended to extract factors and allow correlation between dimensions (oblique rotations). For parsimony and stability in moderate samples, principal components with orthogonal rotation (Varimax) were chosen; however, this choice was recognised as a limitation, so a confirmatory phase using confirmatory factor analysis (CFA) was planned. This will report CFI, TLI, and RMSEA.

Internal consistency was assessed using Cronbach's alpha. Values <.70 were considered unacceptable; values between .70 and .90 were considered acceptable or good; and values >.90 were considered excessively redundant.¹³

Test–retest reliability was calculated using the intraclass correlation coefficient (ICC), employing a two-way random effects model with absolute agreement. The ICC thresholds were: <.40 poor; .40–.75 moderate; .75–.90 substantial; >.90 excellent. The standard error of measurement (SEM) and the minimum detectable change with 95% confidence (MDC95) were also calculated.¹⁴

The Bland–Altman method was included for the total score, estimating mean bias and limits of agreement (±1.96 SD), together with SEM and MDC95.

Concurrent validity was assessed by correlating NPQ scores with SNOT-22 and NOSE-E scores using Pearson's r coefficient. Correlations >.50 were interpreted as strong; between .30–.50 as moderate; and <.30 as weak<.30.¹⁵

Discriminative validity was analyzed using ROC curves,¹⁶ using the total SNOT-22 score as a reference.¹⁷ The area under the curve (AUC) quantified the NPQ's ability to identify severe symptomatic burden. AUC values >.90 were interpreted as excellent; between .70 and .90 as good; between .50 and .70 as low; and ≤.50 as non-discriminative.¹⁸

Based on the ROC analysis, an optimal cut-off point was established by calculating the corresponding sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV).

Results

The Spanish version of the NPQ was adapted in accordance with the international guidelines for cross-cultural adaptation described in the Methods section. Pilot testing confirmed the comprehensibility and cultural appropriateness of the pre-final version.

A total of 140 adult patients diagnosed with CRSwNP participated in the validation study, with a mean age of 46.36 years (SD ± 15.02); 60% were male. The mean body mass index (BMI) was 26.8 kg/m² (SD ± 4.9). All participants were native Spanish speakers who met the EPOS 2020 diagnostic criteria, including the presence of bilateral nasal polyps, as confirmed by endoscopy or imaging. To assess test–retest reliability, a subgroup of 35 patients completed the NPQ twice, with an interval of 7–14 days between assessments.

No significant floor or ceiling effects were observed.

Factor structure

Exploratory factor analysis with Varimax rotation and Kaiser normalisation revealed a five-factor structure that explained 67.65% of the total variance (see Table 1). The determinant of the correlation matrix was close to zero, and the suitability of the data for factor analysis was confirmed by both the KMO index and Bartlett's test.

Table 1.

Variance explained by the main factors of the NPQ.

Factor	% Variance	Cumulative %
1 (Emotional/Social)	15.94	15.94
2 (Sensory/Concerns)	15.75	31.69
3 (Oral/pharyngeal symptoms)	13.20	44.89
4 (Sleep/Fatigue)	12.74	57.63
5 (Disease control)	10.02	67.65

Note: Five factors with eigenvalues >1 were retained, explaining 67.65 % of the total variance.

The rotated component matrix (Table 2) showed a consistent factorial distribution. The five extracted factors grouped items corresponding to:

• 1

  Emotional and social aspects (e.g., embarrassment, stress, and surgical concerns).

• 2

  Sensory symptoms (e.g., disturbances in smell and taste).

• 3

  Oral and pharyngeal discomfort (e.g., dry mouth and bad taste).

• 4

  Sleep disturbances and fatigue.

• 5

  Disease control and symptom management.

Table 2.

Rotated component matrix (main loadings per item).

Item	Main factor	Loading
Item 8. Social situations	Factor 1: Emotional/Social	.76
Item 14. Recurrence of the problem	Factor 2: Sensory/Concerns	.77
Item 19. Difficulty smelling	Factor 2: Sensory/Concerns	.89
Item 21. Bad taste in mouth	Factor 3: Oral symptoms	.83
Item 1. Sleep problems	Factor 4: Sleep/Fatigue	.84
Item 11. Poor control of the disease	Factor 5: Disease control	.69

Note: Only loadings >.30 are shown. The items are grouped into five conceptually coherent factors.

Internal consistency

The overall internal consistency of the NPQ was excellent, with a Cronbach's alpha coefficient of .917. Corrected item-total correlations ranged from .24 to .76. The removal of any individual item did not significantly improve the overall alpha value (see Table 3). Item-by-item analysis revealed that items 4 and 11 had slightly lower item-total correlation values, although these were within an acceptable range.

Table 3.

Item-by-item analysis (summary).

Item	Corrected item-total correlation	α if eliminated
Item 4. Bad breath	.24	.918
Item 11. Poor control	.35	.922
Item 17. Snoring at night	.35	.917
Item 16. Stress	.76	.911
Item 26. Symptom control	.70	.912

Note: Items 4, 11, and 17 had lower, but acceptable, correlations. No item improved the overall alpha when eliminated (α = .917).

Test-retest reliability

The actual interval was 7–14 days, and there were no clinical or therapeutic changes between administrations.

The intraclass correlation coefficient (ICC) for test–retest reliability was .991 (95% CI: .983–.996), indicating excellent reproducibility. The standard error of measurement (SEM) was 2 points, and the minimum detectable change (MD95) was 5 points.

The Bland–Altman plot (Fig. 2) was presented as a complement to the measurement error parameters, showing SEM = 2 and MDM95 = 5 points.

Figure 2.

Bland–Altman plot.

Concurrent validity

NPQ scores correlated strongly with those of the NOSE-E (r = .802; p < .001) and SNOT-22 (r = .870; p < .001), supporting the concurrent validity of the instrument.

Discriminative validity and cut-off point

The NPQ showed high accuracy in identifying patients with severe symptomatic impact. The area under the ROC curve (AUC) was .947 (95% CI: .896–.978; p < .001), indicating excellent discriminatory ability. An optimal cut-off point of >41 points was identified as the threshold for severe disability (see Fig. 3).

Figure 3.

ROC (Receiver Operating Characteristic) curve for the Spanish version of the NPQ.

With this threshold, sensitivity was 88.89% and specificity was 93.27%. Table 4 shows the predictive values and likelihood ratios derived from the ROC analysis.

Table 4.

Predictive values derived from ROC curve analysis.

Cut-off point	Sensitivity (%) (95 % CI)	Specificity (%) (95 % CI)	LR+	LR−	PV+ (%) (95 % CI)	PV− (%) (95 % CI)
> 41	88,89 (73.9–96.9)	93.27 (86.6–97.3)	13.21 (6.40–27.26)	.12 (.047–.30)	82.1 (68.9–90.4)	96.0 (90.6–98.4)

Discussion

The instrument's internal consistency, as measured by Cronbach's alpha, was found to be .917. This indicates a high degree of internal homogeneity, albeit with some redundancy. This result is consistent with the original Italian study (α = .93)⁶ and other validated instruments, such as the SNOT-22, which has an internal consistency ranging from .83 to .94 across different language versions.^4,19 This level of consistency suggests that the Spanish version preserves the conceptual cohesion of the original instrument while maintaining linguistic clarity in the target population.

Despite the high alpha value, the items cover different domains, and eliminating items did not substantially improve the coefficient (see Table 3), so the complete version was retained.

The five-factor structure extracted from the factorial analysis explained 67% of the total variance. The factorial distribution was conceptually consistent, grouping items into dimensions reflecting sensory alterations (smell and taste), psychological impact (stress and embarrassment), functional limitations (sleep and fatigue), oral symptoms (dry mouth and bad taste), and disease control (concerns about symptom management and medication). This multidimensional structure highlights the NPQ's ability to capture the complexity of the symptomatic burden in CRSwNP. While the original NPQ authors also identified multiple domains, our analysis refines the structure within the Spanish-speaking population, which suggests a potential cultural influence on item grouping.

The test–retest reliability analysis showed a CCI of .991, confirming the questionnaire's excellent temporal stability. This is one of the highest values reported in comparable validations: for example, the Italian version of the SNOT-22 showed a CCI of .85,¹⁹ the Spanish version a CCI of .87,⁴ and the Arabic version of the CRS-PRO a CCI of .91.²⁰ The Spanish version of the CRS-PRO also demonstrated excellent reproducibility.²¹ The SEM and MDC95, estimated at 2 and 5 points respectively, suggest excellent accuracy for monitoring clinical changes over time.

Concurrent validity was confirmed by strong positive correlations with the NOSE-E (r = .802) and SNOT-22 (r = .870), indicating that the NPQ in Spanish is highly consistent with other widely used quality-of-life measures in rhinology. While this demonstrates the convergent validity of the instrument with general tools, it also highlights the need for specific questionnaires for CRSwNP, such as the NPQ, which can reflect symptoms not captured by more general instruments.

Similarly, the NPQ in Spanish demonstrated high accuracy in identifying patients with severe symptomatic impact. The AUC was .947. A cut-off point of >41 points provided a sensitivity of 88.9% and a specificity of 93.3%, highlighting its clinical potential as a tool for screening severity and monitoring therapeutic outcomes. These values are comparable to or even higher than those reported by other CRS-specific instruments: for example, the French version of the SNOT-22 reported an AUC of .93,²² the Spanish version .91,⁴ the Chilean version .861,²³ and the CRS-PRO validated in Arabic (AUC = .85),²⁰ French,²⁴ and Spanish.²¹ A threshold of >41 points can serve as a practical reference in clinical protocols to identify candidates for surgery or intensive medical treatment.

From a practical perspective, the NPQ is concise (27 items), self-administered, and widely accepted by patients, as demonstrated by the absence of missing data or comprehension difficulties in the pilot test. Although it is slightly longer than the widely used SNOT-22, the NPQ covers domains that are particularly relevant to patients with nasal polyposis, such as loss of smell, medication side effects, and disease control. These are not covered by more general instruments. This makes it especially valuable in patients with advanced disease, surgical indications, or the need for longitudinal follow-up.

The NPQ-ES complements the SNOT-22 and NOSE-E because it focuses on specific aspects relating to quality of life with polyposis and is more sensitive to often-overlooked domains such as smell/taste, oral discomfort, perceived disease control, and concerns (recurrence, surgical outcome, medication). This more quality-of-life-focused approach is particularly useful when nasal patency is adequate but anosmia or hypogeusia persist, when selecting and prioritising biological therapies where the functional impact is disproportionate to the obstruction, and in postoperative follow-up to identify the residual burden on quality of life, among other clinical practice contexts.

This study has several strengths, including a rigorous translation process based on international guidelines such as those of Beaton et al.⁷ and the ISPOR working group,⁸ a robust sample size exceeding the standard recommendations (5 participants per item), and a comprehensive psychometric analysis. In addition, the inclusion of test-retest analysis and the ROC model reinforce the clinical applicability of the results.

However, some limitations should be acknowledged. Firstly, validation was carried out at a single centre in Spain, which may limit the generalisability of the results to other Spanish-speaking populations in different countries or cultural contexts. Secondly, although the NPQ demonstrated strong cross-sectional validity, its sensitivity to change and responsiveness to treatment were not evaluated. These aspects should be addressed in future research.

Validation was carried out at a single centre, which limits generalisation. We propose multicentre and multinational validation (in Spain and Latin America), including invariance analysis and a CFA with fit indices.

Future lines of research include external validation in other Spanish-speaking populations (e.g., Latin America), assessment of sensitivity to change in response to medical and surgical interventions, and the study of correlations with objective clinical markers (endoscopy, CT scores, olfactory tests). Cross-cultural invariance could also be explored through multigroup analysis and confirmatory factor analysis could be used to refine the domain structure.

Authors’ contributions

• -

  Montilla-Ibáñez MA: Participated in patient recruitment, coordinated data collection, and contributed to the initial drafting of the manuscript.

• -

  García-Fernández JM: Conceived the study, supervised the cross-cultural adaptation process, coordinated the research team, and led the final drafting of the manuscript.

• -

  Sánchez-Torices MS: Performed statistical analyses, prepared tables and figures, and reviewed the results and discussion sections.

• -

  Gómez-Gallego MP: Assisted in methodological development, supervised data management, and critically reviewed the manuscript for intellectual content.

• -

  Feliz-Fernández MA: Participated in the expert committee consensus and pilot test, collaborated in resolving translation equivalences, and reviewed the manuscript.

• -

  Lomas-Vega R: Provided methodological guidance throughout the validation process, supervised the statistical procedures, and approved the final version of the manuscript.

Ethical considerations

The study was approved by the Andalusian Research Ethics Committee (SICEIA-2024-001376) and written consent was obtained from all participants.

Funding

No specific support from public sector agencies, commercial sector, or not-for-profit organisations was received for this research study.

Data availability

The data supporting the findings of this study are available upon reasonable request to the corresponding author.