Visual analogue scale for sino-nasal symptoms severity correlates with sino-nasal outcome test 22: paving the way for a simple outcome tool of CRS burden

Chronic rhinosinusitis (CRS), with or without nasal polyps, is defined as an inflammation of nose and paranasal sinuses lasting for at least 12 weeks [1]. It is characterized by two or more symptoms, one of which should be either nasal blockage/obstruction or nasal secretions (anterior/posterior nasal drip). Other symptoms might be facial pain/pressure and hyposmia/anosmia. The prevalence of CRS in the European adult population is estimated by GA(2)LEN study to be around 11.9%, while in the USA it is considered even higher [2, 3]. The burden of disease and the impact on patients’ every day activity, work productivity, and overall Quality of Life (QoL) cannot be underestimated, especially in difficult-to-treat cases [4]. The direct cost of CRS in the United States is estimated at $8.6 billion/year, while societal indirect costs from productivity loss are approximately $10,077 per patient each year [5]. Interestingly, general health of CRS patients was worse compared to patients with congestive heart failure, Chronic Obstructive Pulmonary Disease (COPD), and Parkinson disease, using generic health-state utility scores [6].

To accurately assess the burden of disease in CRS patients, multiple disease-specific QoL questionnaires were designed and validated over the past years [7‐10]. These questionnaires focus on symptoms and how they affect patients’ daily life, emotional condition, and overall QoL. These instruments are designed to have a strong association with principal disease characteristics and the ability to reflect response to treatment. Among different disease-specific outcome measurements in CRS, the Sino-Nasal Outcome Test (SNOT)-22 is widely accepted and has been used in several studies even before its validation by Hopkins et al., in 2009. SNOT-22 is a reliable questionnaire, can be used to facilitate clinical practice, and validated in multiple languages [8].

Visual analogue scale (VAS) is a psychometric measurement instrument widely used in the Rhinology field and beyond to subjectively quantify patients’ symptoms severity. Originally designed to evaluate workers productivity by senior personnel, VAS gained more attention in the sixties in medicine, social science, and market research [11]. It represents a horizontal line of 10 cm with word anchors at each end representing the extreme feelings. Patients are instructed to indicate the point on the line that best corresponds to their status for the particular characteristic being evaluated. In addition to its high sensitivity, reliability and reproducibility, VAS is easy and simple to use by patients and health care providers [11]. It also does not require training, making VAS a highly valuable tool not only for everyday clinical practice, but also for real-life studies [11].

In allergic disease, VAS was found to correlate well with the Allergic Rhinitis and its Impact on Asthma (ARIA) severity classification system and QoL measurement instruments such as Rhinoconjunctivitis Quality of Life Questionnaire (RQLQ) [12]. Additionally, VAS is utilized to monitor the course of the disease, to assess treatment outcomes, to obtain self-assessments, and to define the level of control in allergic patients by MACVIA-ARIA project [13]. Lately, VAS has been incorporated into the MASK Allergy Diary mobile app, a clinical decision support system assessing allergic rhinitis (AR) severity for feedback to the patient and the doctor on level of disease control [14].

In CRS, VAS for total nasal symptom score (TNSS) is part of routine clinical practice to classify disease as mild, moderate, and severe. In research, VAS for TNSS and individual symptoms are frequently incorporated into studies as an instrument for estimating symptoms severity and burden of disease [15]. In contrast to allergy, correlations between VAS scores with more complex instruments assessing disease-specific burden like SNOT-22 in CRS are scarce [16, 17]. Toma and Hopkins, demonstrated a strong correlation between VAS and SNOT-22 in 65 CRS patients and they further attempted to stratify SNOT-22 score based on disease severity [17]. Here, we aim to study VAS scores for TNSS and individual sino-nasal symptoms in relation to SNOT-22 scores in a larger randomly selected group of CRS patients. In addition, correlation between VAS and SNOT-22 scores is explored in different CRS phenotypes (with/without nasal polyps, controlled/partly controlled/uncontrolled disease).

Of over 400 patients who met the inclusion criteria, a postal questionnaire was sent to 325 individuals of which 202 returned a filled questionnaire (response rate 62.15%). Twenty-two patients were excluded due to incomplete filling of the questionnaire and/or missing informed consent. A total of 180 patients completed the VAS symptom severity scores and the SNOT-22 questionnaire and were eligible for analysis. Of the 180 subjects analysed, 60% were male and 40% were female. The mean age of the studied population was 51.7 ± 16.6, ranged from 16 to 88. About 35.8% had known allergy, 28.7% had asthma, and 12.4% were current smokers. Among patients, 53.9% had CRSwNP, 46.1% had CRSsNP. Using the above mentioned VAS cut-off points for defining the level of control in CRS, 10% were classified as well controlled, 28.3% partly controlled, and 61.7% as uncontrolled (Table 1).

A significant correlation between SNOT-22 and VAS scores for TNSS (Fig. 1a) and individual symptoms was found, with VAS-TNSS showing the strongest correlation, followed by headache/pressure on the face, postnasal drip, and obstruction (Table 2). The maximum value of Pearson’s rho test was estimated at r = 0.655, (p < 0.001) for VAS-TNSS, and the minimum value was r = 0.301, (p < 0.001) for VAS-Loss of smell score. Correlation between VAS scores and SNOT-22 remained statistically significant, irrespective of the distribution of VAS individual scores into subclasses based on symptoms origin (sino-nasal, ocular, bronchial). In this case, VAS for sino-nasal symptoms showed the strongest correlation with SNOT-22 (r = 0.738; p < 0.001), followed by bronchial (r = 0.683; p < 0.001), and ocular symptoms (r = 0.559; p < 0.001) (Table 2). Furthermore, we found VAS for TNSS to correlate significantly with SNOT-22 in both CRSsNP (r = 0.697; p < 0.001), and CRSwNP (r = 0.608; p < 0.001) phenotypes (Fig. 1b).

Subsequently, we intended to evaluate the association between VAS and SNOT-22 in different levels of disease control. A significant association was found for SNOT-22 and VAS-TNSS in well controlled (r = 0.337; p = 0.031) and uncontrolled (r = 0.455; p < 0.001) but not in partly-controlled (r = 0.224; p = 0.095) CRS patients (Fig. 1c). No significant differences were found in outcomes of the 20 non-responders to our questionnaire compared to those who responded.

To date, SNOT-22 is used in different research fields and in clinical practice to determine the burden of disease, the outcome of medical or surgical intervention, and to improve candidate selection for surgery [8, 9]. However, a large number of CRS patients primarily visit their pharmacist or general practitioner to seek advice for their disease or they self-medicate. It is of vital importance to use a simple and reliable tool that can be used by all healthcare providers and patients for self-assessment [16, 18]. As such, VAS has been incorporated into different mHealth tools. Emerging technologies could increase patient participation in treatment decision-making [19]. Consequently, these new tools might improve compliance to treatment, increase level of control, and facilitate doctor-patient communication [19]. Recently, VAS was validated to assess allergic disease control on smartphone screens for the MASK-rhinitis project [14].

We found a significant association between SNOT-22 and VAS for all symptoms (p < 0.001), with VAS for TNSS showing the best correlation (r = 0.655). Interestingly, nasal obstruction which is considered to have the highest average severity among patients with CRS, did correlate with SNOT-22 but headache/facial pressure, and postnasal drip showed stronger association. Although the importance of headache/facial pressure as a cardinal symptom in CRS has been questioned, it appears to have the second strongest association (r = 0.607). VAS for loss of smell was found to have the minimum value of Pearson’s rho test (r = 0.301). These findings, yet atypical, suggest that patients’ perspective of symptoms severity is the result of an interaction between many factors. Age, gender, socio-economic status, psychological profile, and other comorbidities may modify patients’ perception of symptoms burden appraisal. In line with this hypothesis, several studies in the literature demonstrated lack of correlation between subjective and objective measures assessing symptoms severity [20, 21].

The strong association observed between VAS for bronchial symptoms and SNOT-22 (r = 0.683; p < 0.001) is justified on the basis of current knowledge on upper and lower airways interaction [5, 22]. CRS and asthma are common manifestations of an inflammatory process within the contiguous upper and lower airways. Furthermore, it is well established that asthma and CRS frequently coexist, and treatment of one condition could alleviate the coexisting one [5]. Concerning association between allergy and CRS, there is conflicting data in the literature. Nonetheless, allergy testing and treatment remain an option in CRS [5]. Herein, we show a significant correlation between VAS for ocular and bronchial symptoms and SNOT-22. Our results could be explained as patients with allergic, non-allergic rhinitis, and asthma patients were not excluded from the study, and may further support the link between upper and lower airways as stated in the “concept of united airway disease” [1, 5, 22].

Interestingly, a significant association was found for SNOT-22 and VAS-TNSS in well controlled (r = 0.337; p = 0.031) and uncontrolled (r = 0.455; p < 0.001), but not in partly-controlled (r = 0.224; p = 0.095) CRS patients. As this was a postal questionnaire survey, applying the current EPOS criteria for defining the level of control was not feasible [1]. Utilization of specific VAS–TNSS cut-offs for level of control assessment was based on a previous real-life study, where VAS scores of CRS symptoms were compared for different levels of control according to EPOS criteria in 389 patients [15]. Recently, the same cut-off points were used in mySinusitisCoach, an app for patients with CRS developed by medical experts in the field, to assess the level of disease control [23]. The weak correlation observed in the intermediate intervals (VAS > 2 and ≤ 5) could be explained by the halo effect, which may be noticed when several items are to be evaluated with different types of scales [11].

In line with our hypothesis, VAS scores for TNSS and for individual symptoms correlate well with SNOT-22. Among different aspects, mean VAS scores for sino-nasal symptoms showed stronger association with SNOT-22 than with ocular and bronchial symptoms. Overall, VAS-TNSS can accurately predict disease severity, level of control, and burden of disease which is in accordance with the revised EPOS statement [1]. Our data confirmed that VAS for TNSS can be used as a first and easy system to evaluate the burden of CRS followed by the more extensive SNOT-22 questionnaire if more detailed analysis is required. The high response rate (62.1%) that was achieved and the fact that data obtained from telephonic interviews were fully in line with the written questionnaires, could allow us to overcome speculation for bias deriving from the nature of the study. Despite the selection of CRS patients being treated in a tertiary referral center, and the relatively small sample size, our data underline the value of a simple tool like VAS for TNSS. As VAS scores are nowadays being used in the novel digital Apps for allergic rhinitis but also CRS [24], our data support the use of VAS for evaluation of the disease burden in daily life.

In conclusion, we here show that VAS scores can be used in CRS to evaluate the burden of disease. Undoubtedly, our data are paving the way for a simple evaluation of CRS disease burden. Taking into account that VAS can be easily digitized, it can play a key role not only in everyday clinical practice but also in mHealth tools designed to monitor disease activity in CRS patients.