Cross-cultural adaptation of the CRS-PRO questionnaire into Spanish

Abstract

Background

Chronic rhinosinusitis (CRS) is a prevalent disease responsible for significant morbidity. Diagnostic criteria are well established but response to clinical and surgical treatment need greater consistency. The objective of this paper was to perform a cross-cultural adaptation of the CRS-PRO to the Spanish language for its use as a routine questionnaire in the evaluation of CRS.

Methods

The CRS-PRO questionnaire was translated using a three-step procedure in 50 patients, following the recommendations of the International Society for Pharmacoeconomics and Outcomes Research (ISPOR). Native speakers of both European and South American Spanish participated in the translation.

Results

Of the initial 12 items, 4 were discordant among the three translators before a consensus was reached (Step 1). The item “breathe through the nose” (1/12) did not match between the back translation and the initial version of the CRS-PRO (Step 2). Step 3 generated a final revised Spanish version of the CRS-PRO. Fifty patients (average age 51.1, 48% male) completed the questionnaire. Significant gender differences were found in responses to “My face hurt” (P = .043). Significant differences in “My face hurt” also appeared for those completing the questionnaire in less than 80 s (P = .047). Differences in “problems with smell” (P = .039) were noted between chronic rhinosinusitis with and without polyposis.

Conclusion

This study presents the validated and Spanish-adapted version of the CRS-PRO questionnaire to evaluate CRS symptoms in the Spanish-speaking population.

Resumen

Antecedentes

La rinosinusitis crónica (RSC) es una enfermedad prevalente responsable de una morbilidad significativa. Los criterios diagnósticos están bien establecidos, pero la respuesta al tratamiento clínico y quirúrgico necesita una mayor consistencia. El objetivo de este trabajo fue realizar una adaptación transcultural del CRS-PRO al idioma español para su uso como cuestionario de rutina en la evaluación de la RSC.

Métodos

El cuestionario CRS-PRO se tradujo mediante un procedimiento de tres pasos en 50 pacientes, siguiendo las recomendaciones de la Sociedad Internacional de Farmacoeconomía e Investigación de Resultados (ISPOR). En la traducción participaron hablantes nativos de español europeo y latinoamericano.

Resultados

De los 12 ítems iniciales, 4 fueron discordantes entre los tres traductores antes de llegar a un consenso (Paso 1). El ítem «respirar por la nariz» (1/12) no coincidía entre la traducción inversa y la versión inicial del CRS-PRO (Paso 2). El paso 3 generó una versión final revisada en español del CRS-PRO. Cincuenta pacientes completaron el cuestionario (edad media 51.1, 48% varones). Se encontraron diferencias significativas de género en las respuestas a «me duele la cara» (P = .043). También aparecieron diferencias significativas en «me duele la cara» para los que completaron el cuestionario en menos de 80 segundos (P = .047). Se observaron diferencias en «problemas con el olfato» (P = .039) entre la rinosinusitis crónica con y sin poliposis.

Conclusiones

Este estudio presenta la versión validada y adaptada al español del cuestionario CRS-PRO para evaluar los síntomas de la RSC en la población hispanohablante.

Key points

• •

  Assesment of severity and post-treatment outcomes in CRSwNP and CRSsNP remains difficult and most questionnaires are only available in English, except for SNOT-22, DyNaChron and the new CRS-PRO.

• •

  The CRS-PRO questionnaire is a validated instrument that measures patient symptoms and quality of life and was created to assess the impact of treatments (medical or surgical) in CRS patients.

• •

  The CRS-PRO questionnaire was successfully translated following cross-cultural adaptation guidelines into Spanish language, ensuring its conceptual equivalence with the original questionnaire.

Introduction

The prevalence of chronic rhinosinusitis with nasal polyps (CRSwNP) has been estimated to be around 3%, with twice as many males affected as females.¹ After the publication of EPOS2012,² treatment algorithms emerged for the use of nasal washes, intranasal corticosteroids (INCS), systemic corticosteroids (SC), antibiotics (ATB) and surgery. However, the adherence and ability of these treatments to control sinonasal disease is variable, resulting in somewhat unpredictable therapeutic results and symptom control.³ Comparison of the EPOS2012 publication² and EPOS2020⁴ reveals substantial changes that have been introduced in the management of CRS, especially with the advent of biological therapy (BT), although there seems to be a gap between these recommendations and clinical practice.⁵ Recently, these changes have also been described in the POLINA2.0 document.⁶

The new therapeutic paradigm in CRS emphasizes patient-reported outcome measures for treatment decision-making. The most commonly used instrument is the Sinonasal Outcome Test 22 (SNOT-22). However, this questionnaire can be improved in some areas. It has 22 items, which can be lengthy for some patients; questions regarding dizziness, sadness, frustration and ear pain can cause confusion regarding causality; significant improvements in the sinonasal domain may be offset by other concurrent health issues and, lastly, it is proprietary with a cost for clinical use.^7,8

The recently developed 12-item Patient Reported Outcomes in Chronic Rhinosinusitis (CRS-PRO)^10,11 has been validated and found to be responsive to medical and surgical therapy in both CRSwNP and chronic rhinosinusitis without nasal polyps (CRSsNP). The CRS-PRO has demonstrated better correlation with radiographic changes after medical management or endoscopic sinus surgery (ESS) when compared to the SNOT-22.^9–11 Much like the SNOT-22, the CRS-PRO is divided into distinct subdomains: rhino-psychologic, facial discomfort, and cough.

The adaptation of the CRS-PRO to other languages, ensures its conceptual equivalence with the original questionnaire.¹² This study presents the validation process of the Spanish version of the CRS-PRO, to be used as a routine questionnaire in the assessment of patients with CRS.

Methods

Permission was obtained from Bruce K. Tan, MD, MS to use the CRS-PRO questionnaire. All patients who participated in the study gave their consent for inclusion and use of their data anonymously, following the regulations in accordance with the Declaration of Helsinki.

The original language in which the questionnaire was developed was. U.S English with Spanish as the target language. The CRS-PRO questionnaire was translated according to the recommendations of the International Society for Pharmacoeconomics and Outcomes Research (ISPOR).¹³ A local professional translator, native in the target language and bilingual in the source language was recruited (XMR), as was a QoL expert in the target language. Four rhinologists, from both Spain and South America, who spoke the target language were also involved (JMS, AGLL, XMR, AB). The three-step procedure translation process was as follows:

Step 1 (forward translation): Three independent rhinologists produced a forward translation from the source language (English) into the target language (Spanish). All were native speakers of Spanish and fluent in English. They all discussed the translations and agreed upon a single reconciled version (the combined version). The goal was to achieve a conceptually equivalent translation of the original questionnaire; the language used had to be colloquial and easy to understand. By combining the three translations, the Spanish version 1 was obtained.

Step 2 (backward translation): Spanish version 1 was translated into English by a native English-speaking translator without knowledge of the original CRS-PRO. Items containing discrepancies were re-translated. The backward translation and a report of the translation were sent to Dr. Tan for review and comment. This version was compared with the original CRS-PRO questionnaire and revised until a satisfactory translation was obtained. Comparison of the backward version was made with the original source version to detect any misunderstandings, mistranslations, or inaccuracies in the intermediary forward version of the questionnaire. The development of a consensus and translation by means of this committee methodology reduced the cultural and social bias that may result when only one or two translators are responsible for the translation. This resulted in Spanish version 2. This version was submitted to a panel of six health care professionals (5 otolaryngologists) and 3 patients to determine whether it was understandable and easy to use. Further corrections led to the production of Spanish version 3.

Step 3 (patient testing): Spanish version 3 was tested on 50 patients from 3 different Spanish regions through face-to-face interviews, to determine whether instructions and items were understandable and unambiguous. Patients were all native Spanish speakers and belonged appropriate age-group, according to the prevalence of sinonasal disease, for the CRS-PRO questionnaire. Patients were asked to complete the questionnaire in the presence of a practitioner who noted their reactions regarding the understandability of each item. The number of subjects interviewed, their age, the time they took to complete the questionnaire, the difficulties encountered, and the suggested solutions were collected. Based on observations, we modified the questionnaire and obtained the Spanish version 4. This version was proof-read leading to Spanish version 5.

Statistical analysis

Demographic and clinical characteristics were reported as frequencies and inter-quartile range for categorical variables, and means, standard deviations and range for continuous and symptom variables. Statistical analysis was performed using R software (v. 1.3.10703, R Foundation for Statistical Computing, Vienna, Austria, www.r-project.org). Significance was determined at P < .05. Tests used were two-sided, with Mann Whitney U test for categorical variables and Students t-test for continuous variables when normality was demonstrated, while non-parametric variants were used if non- normal distribution was found. Normality was assessed using Shapiro-Wilk test.

Results

Four of the 12 items were initially discordant among the three translators before achieving a consensus (Step 1). The item “breathing through my nose” was discordant between the backward translation and the initial CRS-PRO version (Step 2). Another item was modified according to healthcare professionals and patient feedback leading to Spanish version 3. The definitive version (Spanish version) is presented in Fig. 1.

Figure 1.

CRS-PRO Spain.

A total of 50 participants completed the questionnaire with an average age of 51,1 years old and an approximately equal gender distribution (48% male). Completion of the questionnaire took an average of 81,1 s; being the fastest 30 s while the longest, an outlier, was 200 s (Table 1). When comparing gender, there was a significant difference between responses to item 3 “My face hurt” (P = .043) (Table 2). No significant differences were noted in any individual question when age > 50 (Table 3). When comparing the time to complete the questionnaire in less than 80 s, statistically significant differences were also found in the responses to item 3 “My face hurt” (P = .047) (Table 4). Also, when comparing CRSwNP with CRSsNP, there was only a significant difference between responses to question 8 “problems with smell” (P = .039) (Table 5). Eighteen (52%) had CRSwNP as the underlying diagnosis, and the remainder had CRSsNP.

Table 1.

Baseline.

	Female (n = 26)	Male (n = 24)	Total (n = 50)
Age (years old mean)	52.3 (SD = 11.84)	49.8 (SD = 13.22)	51.1 (SD = 12.44)
Time to complete (seconds mean)	80.5 (SD = 29.65)	81.9 (SD = 35.23)	81.1 (SD = 32.05)
Faster response (seconds)	40	30
Slower response (seconds)	150	200
CRSwNP patients (percent)	9 (34.7%)	17 (65.3%)	26 (52%)
Number of previous surgeries	13 (43.3%)	17 (56.7%)	30 (60%)

SD = Standard deviation.

Table 2.

Comparison of responses between Female and Male participants.

	Female (n = 26)	Male (n = 24)	Statistics and P-value
General data
Age (years)	52.3 (10.42)	49.8 (13.94)	chi2 = 0.87; P = .357
Time to complete (seconds)	80.5 (30.14)	81.9 (41.93)	chi2 = 0.94; P = .332
Answers to the CRS-PRO
Item 1 “nasal breathing”	2.62 (1.20)	2.78 (1.33)	chi2 = 0.39; P = .564
Item 2 “pressure”	2.35 (1.53)	2.00 (1.21)	chi2 = 1.09; P = .122
Item 3 “face hurt”	2.27 (1.48)	1.43 (0.81)	chi2 = 4.91; P = .043*
Item 4 “blow nose”	3.19 (1.37)	3.04 (0.98)	chi2 = 0.34; P = .652
Item 5 “coughing”	2.31 (1.32)	2.17 (1.47)	chi2 = 1.98; P = .177
Item 6 “mucus in throat”	2.81 (1.21)	2.43 (1.26)	chi2 = 2.63; P = .100
Item 7 “mucus in nose”	2.85(1.27)	3.26 (0.87)	chi2 = 0.13; P = .776
Item 8 “smell problems”	2.81 (1.75)	3.30 (1.26)	chi2 = 2.71; P = .084
Item 9 “awake at night”	2.35 (1.45)	2.35 (1.45)	chi2 = 0.41; P = .423
Item 10 “fatigued”	2.35 (1.45)	2.48 (1.23)	chi2 = 0.14; P = .698
Item 11 “worried will worsen”	3.38 (1.69)	3.04 (1.21)	chi2 = 0.79; P = .412
Item 12 “frustrated by condition”	2.50 (1.31)	2.43 1.27)	chi2 = 0.79; P = .454

Values correspond to numbers (proportions) for categorical variables and means (±standard error) for quantitative variables.

Table 3.

Comparison of responses between age.

	Under 50 years old (n = 24)	Over 50 years old (n = 26)	Statistics and P-value
General data
Time to complete (seconds)	82.04(48.19)	86.12 (43.84)	chi2 = 0.02; P = .870
Answers to the CRS-PRO
Item 1 “nasal breathing”	3.17 (1.35)	2.24 (1.07)	chi2 = 0.11; P = .469
Item 2 “pressure”	2.67 (1.57)	1.72 (1.32)	chi2 = 0.03; P = .280
Item 3 “face hurt”	2.04 (1.52)	1.72 (1.26)	chi2 = 0.07; P = .741
Item 4 “blow nose”	3.29 (0.98)	2.96 (1.02)	chi2 = 0.54; P = .882
Item 5 “coughing”	2.54 (1.49)	1.96 (1.21)	chi2 = 0.10; P = .823
Item 6 “mucus in throat”	3.00 (1.33)	2.28 (1.49)	chi2 = 0.94; P = .641
Item 7 “mucus in nose”	3.13 (0.97)	2.96 (1.34)	chi2 = 0.13; P = .362
Item 8 “smell problems”	3.04 (1.49)	3.04 (1.03)	chi2 = 0.41; P = .534
Item 9 “awake at night”	2.50 (1.34)	2.20 (1.27)	chi2 = 0.13; P = .793
Item 10 “fatigued”	2.67 (1.55)	2.16 (1.41)	chi2 = 0.37; P = .450
Item 11 “worried will worsen”	3.50 (1.14)	2.96 (1.64)	chi2 = 0.04; P = .874
Item 12 “frustrated by condition”	2.92 (1.21)	2.14 (1.53)	chi2 = 0.63; P = .511

Values correspond to numbers (proportions) for categorical variables and means (±standard error) for quantitative variables.

Table 4.

Comparison of responses between completion time.

	Under 80 s (n = 28)	Over 80 s (n = 22)	P-value
General data
Age (years)	48.60 (11.05)	54.35 (13.63)	chi2 = 0.06; P = .831
Answers to the CRS-PRO
Item 1 “nasal breathing”	2.75 (1.23)	2.65 (0.98)	chi2 = 1.22; P = .197
Item 2 “pressure”	2.14 (1.65)	2.25 (1.26)	chi2 = 1.53; P = .216
Item 3 “face hurt”	1.78 (0.98)	2.73 (1.67)	chi2 = 4.27; P = .047*
Item 4 “blow nose”	3.00 (1.01)	3.35 (0.96)	chi2 = 2.31; P = .091
Item 5 “coughing”	2.10 (1.15)	2.40 (1.69)	chi2 = 0.27; P = .789
Item 6 “mucus in throat”	2.46 (1.61)	2.90 (1.39)	chi2 = 0.30; P = .592
Item 7 “mucus in nose”	2.96 (1.02)	3.20 (1.11)	chi2 = 1.61; P = .161
Item 8 “smell problems”	3.25 (1.56)	2.85 (1.33)	chi2 = 0.17; P = .534
Item 9 “awake at night”	2.28 (1.12)	2.50 (1.14)	chi2 = 1.14; P = .343
Item 10 “fatigued”	2.39 (1.11)	2.45 (1.32)	chi2 = 0.33; P = .822
Item 11 “worried will worsen”	3.17 (1.32)	3.40 (1.25)	chi2 = 0.23; P = .441
Item 12 “frustrated by condition”	2.46 (1.56)	2.55 (2.21)	chi2 = 0.22; P = .655

Values correspond to numbers (proportions) for categorical variables and means (±standard error) for quantitative variables.

Table 5.

Comparison of responses between chronic rhinosinusitis phenotipe.

	CRSwNP (n = 26)	CRSsNP (n = 24)	Statistical and P-value
General data
Age (years)	53.03 (11.33)	48.95 (15.11)	chi2 = 1.64; P = .201
Answers to the CRS-PRO
Item 1 “nasal breathing”	2.69 (1.22)	2.69 (1.17)	chi2 = 0.19; P = .604
Item 2 “pressure”	2.15 (1.31)	2.21 (1.08)	chi2 = 0.22; P = .811
Item 3 “face hurt”	1.80 (1.33)	1.95 (1.62)	chi2 = 1.41; P = .230
Item 4 “blow nose”	3.23 (0.86)	3.00 (1.14)	chi2 = 0.09; P = .666
Item 5 “coughing”	1.88 (1.11)	2.65 (1.26)	chi2 = 3.21; P = .052
Item 6 “mucus in throat”	2.42 (1.35)	2.86 (1.45)	chi2 = 1.37; P = .209
Item 7 “mucus in nose”	3.42 (1.21)	2.60 (1.24)	chi2 = 0.17; P = .098
Item 8 “smell problems”	3.84 (1.38)	2.13 (1.07)	chi2 = 8.11; P = .039*
Item 9 “awake at night”	2.46 (1.26)	2.21 (1.35)	chi2 = 0.09; P = .706
Item 10 “fatigued”	2.53 (1.27)	2.26 (1.45)	chi2 = 0.03; P = .818
Item 11 “worried will worsen”	3.50 (1.21)	2.91 (1.51)	chi2 = 1.87; P = .123
Item 12 “frustrated by condition”	2.69 (1.52)	2.21 (1.88)	chi2 = 0.20; P = .763

Values correspond to numbers (proportions) for categorical variables and means (±standard error) for quantitative variables.

Discussion

In this study, we describe the cross-cultural adaptation of the CRS-PRO questionnaire to measure the symptoms and psychosocial impact of CRS with and without nasal polyps.^9–11 This evaluation methodology has already been used by other authors, identifying important cultural differences between populations that should be comparable.^14,15 Thus, the translation/back-translation process we used is essential to obtain a translation that captures the true meaning of the original document.¹⁶ Our aim was to obtain a translation of the idea or concept, rather than a literal translation of each item in order to obtain an adapted and well-accepted instrument for assesing CRS symptoms and quality of life, such as the one that has recently been adapted for the French population.¹²

ISPOR recommendations were chosen for the cross-cultural adaptation process, which made it possible to obtain a culturally adapted Spanish version of the CRS-PRO equivalent to the English version.¹³ The ISPOR recommendations stipulate that at least two translators should perform an English‐Spanish translation independently, followed by a consensus meeting. Therefore, we chose three independent translators from different Spanish-speaking regions of the world to translate the concept on which the questionnaire is based as accurately as possible and ensure reproducibility. One major strength of this cross-cultural adaptation is the use of the translation/back-translation single blinded process, which allows the designer of the English questionnaire to correct residual misunderstandings.

As recommended by cross-cultural adaptation guidelines, it is advisable to assemble a panel of experts to test the questionnaire. During step 2, Spanish version 2 was submitted to a panel of six health care professionals (5 otolaryngologists) and 3 patients. This step resulted in many changes. It was essential to include non-medical professionals (herein patients) since the surgeon’s point of view are sometimes biased and lacks comprehension of technical terms for the general population. It is recommended to include patients themselves in the translation process during step 3. Indeed, since the questionnaire is ultimately intended for them, it increases the fluidity of the questionnaire by gathering their opinion on each item. Thus, we opted to include some patients immediately in step 2 and their remarks gave rise to important changes leading to Spanish version 3.

In comparing the numerical data with the French-speaking CRS-PRO,¹² the French-speaking population enlisted 30 participants with a mean age of 49.2 years, whereas the Spanish-speaking cohort, involved 50 participants with an average age of 51,1 years old. The gender distribution was comparable, with 63.3% male participants in the former and 48% in the latter. Concerning the time taken to completion of the questionnaire, divergent findings surfaced. The French study reported an average completion time of 67 s, in contrast to the Spanish study's average completion time of 81,1. Notably, the French study noted no instances of participants exceeding a 2-minute completion mark, whereas the Spanish study identified one outlier who took 200 s to complete the questionnaire. When the results were evaluated according to demographic variables, disparities and similarities emerged. The French study observed significant gender-based differences in questionnaire responses (items 7 and 9), while the Spanish validation found significant differences in item 3 (My face hurt) compared to the French validation. Whereas the Spanish study did not find significant differences when comparing age (>50 years), statistically significant differences were also found in item 3 (My face hurt) parameter when comparing completion of the questionnaire greater or less than 80 s. It is striking that differences repeatedly appear in the “My face hurt” aspect, although it must be considered that sometimes the perception of headache tends to confuse patients with CRS when completing the questionnaires since these symptoms can have many non-synogenous causes.¹⁷ Interestingly, both studies reported no significant distinctions based on age categories for all questionnaire items. However, the Spanish study unveiled a notable discrepancy between patients diagnosed with chronic rhinosinusitis with polyps and those without polyps regarding a specific questionnaire item (item 8 “I had problems with my sense of smell”). Overall, even though both studies underline the importance of linguistic and cultural considerations in questionnaire validation, nuanced differences exist in participant demographics, completion time, and results interpretation that need to be kept in mind. These findings thus highlight the necessity of contextual understanding and cautious interpretation of results in cross-cultural research endeavors.

Test-retest reliability was used to assess the consistency and reproducibility of the results obtained from patients with Spanish version 5 and confirmed the reproducibility and efficiency of this questionnaire in its translated form. No translation is perfect; conceptual differences may always remain. When translated back into English, these differences were considered to be minimal and did not represent a significant change in the essence of the questionnaire.

The CRS-PRO questionnaire is a very useful tool for assessing quality of life in CRS. It has important advantages since, like Snot-22, it evaluates different subdomains such as the rhinopsychological sphere, facial discomfort and sleep, and because it is faster to answer due to its short length. In addition, it has shown better correlation with radiographic changes after medical management or ESS when compared to the SNOT-22.^9–11 Nevertheless, the use of the Snot-22 remains fundamental in CRS covering aspects of QOL not represented in the CRS-PRO, so it is to be expected that both questionnaires are ideal to be used in a complementary way or to use one or the other depending on the context of clinical practice or the purpose of the studies to be performed. More studies are needed that correlate the clinical and exploratory variables of patients with chronic rhinosinusitis with the results obtained in both questionnaires.

Conclusion

The Spanish version of the validated CRS-PRO questionnaire has been successfully translated to evaluate symptoms of CRS and QOL in the Spanish-speaking population. This tool can be especially valuable, as well as Snot-22, in follow-up examinations for measuring the outcome of medical and surgical treatment of patients with CRS, for comparison of results with the international literature, and its use in research and clinical trials. It provides a number of advantages, especially the short time required to be answered.

CRediT authorship contribution statement

Each of the authors contributed to the conception and design of the study, as well as to the writing of the article. Dr Maza-Solano and Dr Calvo-Henriquez were responsible for the statistical analysis of the data.

Patient consent statement

Not applicable.

Financial disclosure

None.

Data availability

Available upon reasonable request.

Acknowledgements

The authors would like to thank Dr. Bruce Tan, designer of the English questionnaire, for allowing us to make a cross-cultural adaptation of his questionnaire into Spanish.