Research Article: Association of dysfunctional breathing with health-related quality of life: A cross-sectional study in a young population

Date Published: October 11, 2018

Publisher: Public Library of Science

Author(s): Ji-Myung Ok, Young-Bae Park, Young-Jae Park, Iratxe Puebla.


Symptomatic hyperventilation (SH) is a pathological condition that manifests with breathlessness, dyspnea, light-headedness, anxiety, and paresthesia. However, little is known about the prevalence of SH and its association with health-related quality of life (HRQoL) in a young population. The Nijmegen questionnaire (NQ), which measures severity of SH, had not previously been cross-culturally translated into Korean. In this study, the NQ was cross-culturally translated into Korean (KNQ), using translation and back-translation methods. To examine the reliability and validity levels of the KNQ, as well as its association with HRQoL, 237 college students (21.38 ± 2.45 years) were asked to complete the KNQ, the Korean version of the general health questionnaire (K-GHQ-30) and the short form-36 (K-SF-36). The KNQ showed satisfactory reliability (Cronbach’s α = 0.878). In the construct validity test, four factors (neuropsychological, respiratory, neurogastrointestinal, and neuromuscular) were extracted (% of total variance = 59.8). Using a KNQ cut-off score of 23 points, the prevalence of SH was 22.8%. Physical and mental HRQoL levels estimated by the K-GHQ-30 score and the 8 subscale scores of the K-SF-36 were lower in the SH group than in those of the non-SH group. It is concluded that the cross-culturally translated KNQ is reliable and valid, and management of SH may prevent a reduction in physical and mental HRQoL in a young population.

Partial Text

Hyperventilation is a condition in which arterial carbon dioxide (CO2) is eliminated excessively by respiration beyond the demands of metabolism, resulting in a wide range of symptoms [1]. Hyperventilation is associated with a variety of pathological conditions, including respiratory and cardiovascular diseases and psychiatric disorders [1]. Although a low arterial CO2 level is regarded as the primary criterion for hyperventilation [2], some studies have reported hyperventilation cases with normal CO2 levels [3]. Therefore, together with arterial CO2 levels, the presence of hyperventilation-related symptoms, including breathlessness, dyspnea, light-headedness, anxiety, and paresthesia, are now regarded as hyperventilation parameters.

A flowchart of the study design is presented in Fig 1. Two groups of volunteers participated in this study, groups A and B. Group A consisted of 45 apparently healthy participants: 13 males and 32 females with an age range of 24 to 55 years. The inclusion criterion for group A was the ability to read and understand the KNQ. Participants in group A were asked to complete the KNQ, and their responses were used to determine the face validity of the instrument. Group B consisted of 237 college students: 130 male (mean age of 21.35 ± 1.87 years) and 107 female students (mean age of 21.41 ± 3.02 years). The inclusion criteria for group B were being a college student and having no impediments to daily life caused by psychological or respiratory problems. Participants in group B were asked to complete the KNQ and the Korean versions of the General Health Questionnaire-30 (GHQ-30) [18], and Short Form-36 (SF-36) [19]. Data collected from group B were used to determine the reliability and construct validity of the KNQ, and to examine the association between SH and HRQoL levels estimated by the GHQ-30 and SF-36. The protocol for this study was approved by the Kyung Hee University Institutional Review Board (KHSIRB-15-010RA). Informed consent was obtained from all participants.

Table 1 lists descriptive characteristics of the KNQ, K-GHQ-30, and eight K-SF-36 subscale scores. Table 2 presents the prevalence of SH and non-SH by gender. Based on the 23-point cut-off for the NQ [4], 54 college students were categorized in the SH group, indicating a prevalence of 22.8%. There was no difference in SH prevalence based on gender. Table 3 provides the results of the reliability test of the 16 symptoms listed in the KNQ, including their Korean translations. In the final revision of the KNQ, five items (“Dizzy spells”, “Faster or deeper breathing”, “Short of breath”, “Bloated feeling in stomach”, and “Tight feelings round mouth”) were determined to contain words in both native Korean and Sino-Korean, since words from both these languages are regularly used inter-changeably in Korea [21]. The overall Cronbach’s α value was 0.878, indicating that the KNQ demonstrated satisfactory internal consistency [20]. This score was consistent with the reliability values of previous cross-cultural translations into other languages, which ranged from 0.70 to 0.9 [3,15]. Although Cronbach’s α increased after removing “tingling fingers” and “cold hands or feet” (0.879 and 0.884, respectively), the effect was slight (0.001–0.006). Therefore, the following analysis was conducted with all 16 terms included. In the construct validity test by PCA, four factors were extracted, and the percent of total variance was found to be 59.761% (Table 4). The four extracted factors were as follows: neuropsychological (factor 1), respiratory (factor 2), neurogastrointestinal (factor 3), and neuromuscular (factor 4).

In this study, we developed a cross-cultural translation of the NQ in Korean, with translation, back-translation, and face validity testing [17]. The “feelings of anxiety” term, which was removed from the construct validity test of the original NQ [4], had a high factor loading value (0.730) in this study, indicating that it should be included in the KNQ. Among the four KNQ factors, the “neuropsychological” and “respiratory” factors corresponded to the “shortness of breath” factor in the original NQ, and the “neuromuscular” and “neurogastrointestinal” factors corresponded to the “peripheral tetany” and “central tetany” factors of the original NQ [4]. Therefore, it appears that the factorial structure of the original NQ was largely replicated in the KNQ.




0 0 vote
Article Rating
Notify of
Inline Feedbacks
View all comments