Introduction

Olfactory Dysfunction (OD) may affect 1 to 20% of the general population [1]. The primary causes of OD are sinonasal disorders, post-viral olfactory dysfunction, neurological diseases and post-traumatic lesions of the olfactory nerve [1]. The prevalence of OD has substantially increased since the onset of the coronavirus disease 2019 (COVID-19) pandemic, reaching 30 to 86% of patients infected by Alpha, Delta or Omicron variants [2,3,4]. The OD may include anosmia, hyposmia or parosmia throughout the clinical course of the disease. According to several studies, the OD may persist in a significant number of patients more than 6-month post-infection and may affect their quality of life (QoL) [4, 5]. The evaluation of OD has to involve psychophysical tests and patient-reported outcome questionnaires that provide additional insight into the impact of OD on patients QoL [6].

In 2005, Frasnelli et Hummel developed the Olfactory Disorder Questionnaire (ODQ), which is a patient-reported outcome questionnaire reporting the features of the olfactory disorders and the related impact on QoL [7]. ODQ was validated in English [6], Korean [8] and Chinese [9]. To date, there is no validated version for French speaking countries, which include more than 400 million inhabitants.

Methods

Ethical statement

The study protocol was approved by the Institutional Ethics Committee (n° CHUSP20032020)). Patient and healthy individual informed consent was obtained.

Questionnaire development

A multidisciplinary team composed of two otolaryngologists, a psychologist and a linguist worked on the French adaptation of ODQ (Fr-ODQ) from the English ODQ [6]. Members of the team were native French speakers. Prior to the validation of Fr-ODQ, the draft was sent to 5 patients to detect potential misunderstandings and was improved to have the final version (Additional file 1: Appendix 1).

Setting

The study was conducted in two medical centers (University of Mons and Dour Medical Center, Dour, Belgium) between January 2021 to April 2022. Irrespective to the etiology, patients with OD were asked to participate to the study. Patients with severe neurological diseases limiting the understanding of the study protocol or those who were not native French-speaker were excluded.

A control group of healthy individuals was composed, matching age and gender of the study group. To be included, healthy individuals had to have no neurological, otolaryngological (sinonasal) or general disorders that may impact the olfactory function. Individuals with a history of COVID-19 were not included in the control group.

Olfactory and nasal evaluations

Participants completed Fr-ODQ, while nasal complaint evaluation was performed using the French version of the sinonasal outcome tool-22 (SNOT-22) [10, 11]. Psychophysical olfactory evaluations were performed with Threshold, Discrimination and Identification test (TDI; Medisense, Groningen, Netherlands), which is a standardized and validated evaluation of olfaction. The sum of the scores of threshold, discrimination and identification subtests was used clinically to assess olfactory performance. Participants were considered normosmic or dysosmic when TDI ≥ 30.75 and < 30.75, respectively [12].

Validity, reliability and responsiveness to change

The statistical analyses were performed with Statistical Package for the Social Sciences for Windows (SPSS version 22.0; IBM Corp, Armonk, NY, USA). A level of significance of p < 0.05 was used.

The Fr-ODQ was completed twice over 7-day period to evaluate the test–retest reliability (Spearman correlation coefficient). The internal consistency was evaluated with Cronbach’s alpha. External validity was assessed by a correlation analysis between Fr-ODQ and Fr-SNOT-22 (Spearman correlation coefficient). The internal validity was evaluated with a comparison of the Fr-ODQ scores between patients and healthy subjects (Mann–Whitney U test.). The responsiveness to change of Fr-ODQ was evaluated in a subgroup of patients with reversible OD, excluding patients with > 2-year post-traumatic loss of smell. Only patients with an improvement of TDI scores 3 months after the initial evaluations were selected for the responsiveness to change analysis. For these patients, authors evaluated the reduction of Fr-ODQ. Patients were invited to adhere to an olfactory training twice daily.

Results

Eighty-nine patients (67 females) with OD and 65 healthy individuals (50 females) completed the study. The characteristics of patients are reported in Table 1. The most prevalent comorbidities of patients included hypothyroidism, hypertension and hypercholesterolemia. The causes of OD consisted of post-viral (N = 80), post-traumatic (N = 4), neurological (N = 2) and idiopathic (N = 3) ODs. The mean duration of OD was 15.7 ± 8.2 months. Parosmia concerned 52 patients (58.4%) patients. Five patients reported TDI > 37 but they suffered from severe parosmia. The mean Fr-SNOT-22 score of patients was 31.80 ± 17.69 (Table 2).

Table 1 Epidemiological and clinical outcomes of patients
Full size table
Table 2 Comparison of Olfactory Questionnaire between patients and healthy individuals
Full size table

The Cronbach’s alpha value was 0.827 for the items of Fr-ODQ, which indicated a high internal consistency. The test–retest reliability was high for total Fr-ODQ scores (rs = 0.944, p = 0.001), Parosmia and Life Quality Statement scores and moderate-to-high for Sincerity Statement score (Table 3). The correlation between Fr-ODQ and Fr-SNOT-22 total score was high (rs = 0.498; p = 0.001), indicating adequate external validity. The mean score of Fr-ODQ was significantly higher in patients compared with controls, which supported an adequate internal validity (Table 4). Seventy-five patients adhered to a 12-week olfactory training protocol for a post-viral OD, consisting of twice daily smell of odor according to the Hummel protocol [13]. The recovery of smell sense occurred after 3.4 ± 4.3 months. Focusing on the 32 patients who had an improvement at the TDI 3 months after the start of the olfactory training, the mean TDI score significantly increased from 19.12 ± 9.08 to 29.50 ± 8.56 (p = 0.028); while the Fr-ODQ significantly increased from 45.05 ± 19.25 to 50.82 ± 12.71 (p = 0.008), which suggests an adequate responsiveness to change property (Table 4).

Table 3 Test re-test reliability findings
Full size table
Table 4 Responsiveness to Change property
Full size table

Discussion

The recent increase of the prevalence of OD makes the development of olfactory patient-reported outcome questionnaires important. To date, only the Fr-sQOD-NS was adapted for French speaking countries [14] but this questionnaire only reports limited QoL items. In the present study, we developed a French version of the olfactory disorder questionnaire, which reports high internal consistency and test–retest reliability.

The internal consistency of the Fr-ODQ was comparable with those of the German and English versions. Indeed, the internal consistency of the German version of ODQ was 0.54 and 0.93 for negative and positive statements, respectively [7], while the Cronbach value of the English version of ODQ was 0.90 [6].

The test-restest reliability of the Fr-ODQ was particularly high (rs = 0.944) compared with the data of the Frasnelli et Hummel version (0.71 and 0.78 for negative and positive statements) [7]. In the study of Langstaff et al. the test–retest reliability ranged from 0.56 to 0.77 according to the sub-scores of ODQ [6]. In the present study, the external validity was investigated through a correlation analysis between SNOT-22 and Fr-ODQ. Frasnelli et Hummel assessed the external validity through a correlation analysis between Beck Depression Inventory, Mood Inventory and German ODQ questionnaire [7]. They reported significant association between all questionnaires, supporting an adequate external validity. Langstaff et al. did not find significant external validity comparing TDI and ODQ scores [6]. The use of SNOT-22 in the present study limits us in the comparison of our results with those of the literature. We have chosen SNOT-22 because that was the closest questionnaire to the ODQ regarding symptoms and QoL impact. Indeed, there is no other olfactory questionnaire validated in French at the exception of Fr-sQOD-NS [14], which included items of the ODQ. The high internal validity outcomes of the Fr-ODQ corroborated those of the German version in which Frasnelli et Hummel observed significant higher score of ODQ in hyposmia/anosmic patients compared with normosmic individuals [7].

Responsiveness to change is another important parameter in the validation of a patient-reported outcome questionnaire [15]. The originality of the validation of the Fr-ODQ was the assessment of the ‘responsiveness to change’ property. Focusing on patients who reported improvement at the TDI scores (at least 1 point increase), we observed a similar significant improvement of Fr-ODQ total score, suggesting an adequate ‘responsiveness to change’ property.

The primary limitation of this study are the low number of patients and the high proportion of post-viral OD individuals. The characteristics of patients with post-viral OD may be different from the OD of patients with chronic rhinosinusitis with nasal polyps or other etiologies. The high proportion of post-viral OD was related to the onset of the COVID-19 pandemic and the increase of the number of anosmic, hyposmic and parosmic patients in the general population. The German and English version of ODQ were both validated before the pandemic and, therefore, included more patients with non-COVID-19 OD. In future studies, authors may use the smell identification test (UPSIT) in addition to TDI to strengthen the olfactory evaluation. The Fr-ODQ could be used in many future French studies from France, Belgium or Canada in which authors will investigate smell function in common conditions, including chronic rhinosinusitis with or without polyps [16, 17], or COVID-19 [18]. The main strengths of this study are the realization of TDI evaluations, allowing the confirmation of the olfactory dysfunction and the evaluation of ‘responsiveness to change’ parameter.

Conclusion

The Fr-ODQ is the first patient-reported outcome questionnaire validated for French speaking patients. Fr-ODQ is reliable and valid for the evaluation of the olfactory dysfunction and the related impact on QoL of French-speaking patients.