INTRODUCTION

Dry eye disease (DED), a multifactorial disorder, is among the most prevalent ocular conditions encountered by ophthalmologists. This condition is characterized by changes in the tear composition and volume, as well as the ocular surface integrity, and manifests with a broad range of symptoms and associated risk factors⁽¹⁾. Symptoms of DED largely affect patients' quality of vision and may therefore have profound effects on daily activities^(1,2). The chronic and heterogeneous nature of DED, with respect to both etiology and clinical presentation, introduces many challenges in terms of patient evaluation⁽³⁾. In addition, multiple studies have reported poor correlations between clinical tests and with associated symptoms related to the severity and progression of the disease^(4,5).

Surveys are an important tool used in health-related research, as well as research pertaining to other fields, such as human and social sciences. For example, symptom questionnaires have been increasingly implemented to assess the quality of life related to a specific disease, quantify symptoms, evaluate natural disease courses, and determine the impacts of treatment strategies. The widespread use of validated questionnaires in distinct populations warrants translated and language-adapted versions. To expedite this task, many translation guidelines have been published in an attempt to ensure equivalence between the original and translated versions^(6,7).

As DED can occur without major clinical signs of tissue damage, it is considered to be a symptom-based condition, especially in mild to moderate cases. The prevalence of DED varies considerably among different populations, and a complete understanding of the global and South American prevalence is lacking^(8-13). Hence, validated questionnaires that fully investigate and refine patients' reports of DED symptoms are important for the objective measurement of symptoms and the evaluation of general prevalence in populationbased studies⁽¹⁴⁾.

In this study, we aimed to validate the Portuguese language version of a simple, comprehensive and patient-reported dry eye symptom questionnaire. This short questionnaire comprises of two direct questions regarding DED symptoms and a third question concerning a previous clinical diagnosis of DED. According to previous studies, this questionnaire has been applied to various populations and has yielded a sensitivity of 77% and specificity of 83% for clinical DED diagnosis, using a Schirmer 1 value cutoff point of ≤10 mm or tear breakup time <10 seconds⁽¹⁵⁾. In other words, this short DED questionnaire is sensitive, repeatable, and easily administered in large epidemiologic studies, as well as during clinical research^{(11,12,16,17)}. We hope that our validated Portuguese language version will facilitate a better understanding of the prevalence of DED symptoms in the Brazilian population.

METHODS

This study was conducted in the Department of Ophthalmology at the University of Campinas, after receiving approval from the Faculty of Heath Sciences Ethics Committee. To obtain a scientifically accurate translation and transcultural validation of the original English version of the questionnaire into the target Portuguese language version, we followed a three-phase process initially proposed by Beaton and Gjersing^(6,7). First, the initial translation and transcultural adaptation of the English version to the Portuguese language was performed by two independent translators, followed by an interdisciplinary panel evaluation of the translated version. Second, the Portuguese version was back translated into English by two independent native speakers, followed by evaluation and comparison with the original English version by the same interdisciplinary panel. Third, the final version of the questionnaire was applied to a selected population to verify inter- and intra-observer concordance. Our validation process, listed below, was based on the above guidelines. We ensured that the following tasks were completed:

Kappa coefficients were calculated by evaluating the agreement between responses to each question at two distinct times of application. Continuous variables were compared using the Mann-Whitney U test, categorical variables were assessed using the chi-square test, and correlations were determined using Spearman correlation coefficients. Differences were considered significant at p<0.05. All analyses were performed using Prism statistics software, version 5.0 (GraphPad, Inc., San Diego, CA, USA). The study design is summarized in figure 1.

Figure 1 Study design. 

The selected DED symptom questionnaire included two questions about symptoms: How often do your eyes feel dry? And how often do your eyes feel irritated? Regarding the incidence of symptoms, scores of 0, 1, 2, and 3 indicated never, sometimes, often, and constantly, respectively. DED symptoms were classified as absent if "never" was the response to both questions, and mild, moderate, or severe if "sometimes", "often" or "constantly" was marked as a response, respectively, as previously described⁽¹⁵⁾. A third yes or no question addressed previous DED diagnoses: Have you ever been diagnosed (by a clinician) as having dry eye syndrome? (Figure 2).

Figure 2 Original English and Portuguese versions of the symptom questionnaire. 

RESULTS

The Portuguese version of this DED symptom questionnaire was applied twice by two different observers to a population comprising 30 medical students (mean age, 22.93 ± 2.81 years). No difficulties emerged during the translation and adaptation steps because the questionnaire comprised simple and direct queries. Similarly, the remaining steps of the process did not raise any controversies. The overall results are presented in table 1 and table 2. The correlation coefficient values were consistently high and statistically significant for all three items: 0.92 for dry eye symptoms; 0.90 for irritation; and 1.0 for previous diagnosis of dry eye (Table 1). The Kappa concordance values exceeded 0.79, indicating substantial to excellent concordance (Table 2).

DISCUSSION

Despite the frequent incidence of DED, its prevalence in the Brazilian population is poorly understood. The present study indicates that the Portuguese translation and adaptation of this DED symptom questionnaire yielded a reliable tool, as evidenced by the high internal consistency of the answers obtained and the high correlation coefficients. To develop this tool, we followed guidelines used previously in similar endeavors, including a scientifically rigorous translation and adaptation process. Although the chosen questionnaire comprised simple and direct questions, the use of two independent translators for each language translation was very useful because it allowed the multidisciplinary assessment panel to validate the questionnaire through comparisons and discussion. The inclusion of members with distinct areas of expertise in this assessment panel was crucial as it allowed comparisons from different points of view and solved discrepancies with the aims of consensus and proper adaptation. Accordingly, our results demonstrate that this validated questionnaire is both reliable and reproducible and can be applied in future population-based studies to determine the prevalence of DED symptoms.

In conclusion, there remains a significant lack of knowledge about the current prevalence of DED in Brazil and the distinct regions within this very large country. This simple questionnaire appears to be a feasible, powerful instrument that will contribute epidemiological information about this common ocular condition on both local and more widespread levels.