Validation of the Turkish version of the quality of recovery-40 questionnaire

After the approval of the Gaziosmanpasa University Medical School Ethics Committee (13 − KAEK − 093), patients were informed about the study and written consents for participation were obtained. Patients who were admitted to the outpatient anesthesia unit for preoperative examinations associated with a scheduled elective surgery in general, orthopedic, and otorhinolaryngological surgery under general anesthesia, from April 2013 to July 2013, were included in the study. Those under 18 or over 80 years of age, as well as those who had an alcohol or substance addiction problem, refused to participate in the study, had an American Society of Anesthesiologists (ASA) score of III or IV, or were non-cooperative were excluded from the study.

The QoR-40 is a self-rated questionnaire that consists of 40 items to evaluate the pain levels and physical and emotional state of patients, comprising five subscales: emotional state (n = 9), physical comfort (n = 12), patient support (n = 7), physical independence (n = 5), and pain (n = 7). Each item is scored on a five-point Likert scale ranging from 1 to 5. The scores for the subscales are given by the sum of the corresponding items, and the total score is computed by a summation of all items. The total score ranges from 40 to 200. In order to design the QoR-40 T, we first obtained development authorization from the original author.

The original version of the QoR-40 was translated from English to Turkish by three individuals, including a native English speaker (university graduate, living in Turkey for 3 years) and two academics from the Department of English Language. The Turkish version of the QoR-40 was back-translated into English by two different English linguistic academicians from the Department of English Language. Thereafter, each question was rendered in its most comprehensible form by a committee of three people which consisted of one health professional fluent in English, a Turkish linguist, and an English linguist.

After obtaining the approval of the original author for the backward translation, a pilot test was conducted on 10 randomly selected patients. The pilot testing revealed that all items in the questionnaire were understandable to Turkish patients. With this result, the Turkish version of the QoR-40 was finalized.

Patient characteristics, including age, gender, marital status, and education, were recorded. To evaluate the health status, the QoR-40 T was used during the preoperative period, on the third day after surgery, and one month after surgery. The post-discharge quality of life from the third day and one month after surgery were assessed using the Turkish version of the SF-36 questionnaire. The SF-36 consists of 36 items and standardized response choices, organized into eight multi-item scales, including physical functioning (PF), role limitations due to physical health problems (RP), bodily pain (BP), general health perceptions (GH), vitality (VT), social functioning (SF), role limitations due to emotional problems (RE), and general mental health (MH). It is summarized in two component summary scores, the Physical Component Summary (PCS) and the Mental Component Summary (MCS). All subscale scores range from 0 to 100, with higher scores indicating higher levels of functioning or well-being.

The Kaiser-Meyer-Olkin measure of sampling adequacy and Bartlett’s test of sphericity were used to confirm the assessment of the factor analysis for these items. A principal components analysis was conducted for the construct validity of the QoR-40 T. For the basis of this analysis, the components were extracted with an eigen value of greater than 1. Varimax rotation with Kaiser normalization was selected, and the rotation converged in 7 iterations, with values below 0.40 being suppressed. Additionally, the correlation between the QoR-40 T score on the third-day and the duration of hospitalization, and the QoR-40 T score on the third-day between men and women, were compared to support the construct validity. The internal consistency of the QoR-40 T was also evaluated by using Cronbach’s alpha (α) for the global QoR-40 T and subscale scores. A Cronbach’s alpha coefficient of 0.70 to 0.95 was considered to be acceptable [12]. Item-total correlation coefficients were used to assess the association between the scores. To measure the test-retest reliability, the QoR-40 T was performed in the same patients on the third day and one month after their operations. The postoperative third-day total QoR-40 T score was compared with the recovery visual analog scale (VAS) to assess the concurrent validity (Additional file 1). In addition, predictive validity was determined by assessing the association between the related dimensions of the questionnaires: physical comfort, emotional state, physical independence, pain dimensions of the QoR-40 T and the physical component summary, mental health, physical functioning, and the bodily pain subscales of the SF-36, respectively. This was evaluated by using the Spearman correlation coefficients for each measurement period.

Responsiveness, or sensitivity to clinical change, was measured by evaluating the standardized response mean (SRM). The SRM is calculated by dividing the mean change in the score by the standard deviation (SD) of the change [13, 14].

Normality and variance were tested using the One-Sample Kolmogorov-Smirnov, skewness and kurtosis, histograms, and Q-Q plots for each variable. Quantitative data were presented as the means and standard deviation, and qualitative data as the frequency and percentage. Associations were performed by using the Spearman rho correlation coefficient (ρ). Analyses were completed by using the Statistical Package for Social Sciences (SPSS Inc., Chicago, IL) version 20.0 program. The statistical significance for all analyses was set at p < 0.05.

Patient characteristics are presented in Table 1. Of the 180 patients approached in the present study, there were 16 refusals (recruitment rate = 0.91); another 27 patients which had been discharged from the hospital could not be reached to complete their questionnaires (completion and return rate = 85%). Thereafter, a total of 137 patients were accepted to participate in the study and completed the questionnaires. The SF-36 was administered to all patients one month after surgery, and all of them completed it. Most of the patients were able to complete the QoR-40 T questionnaire form without any assistance. The mean duration of hospitalization was 3.93 ± 1.29 (IQR 3–5 days).

The Cronbach’s alpha (α) values and the Spearman rho (ρ) correlation coefficients of physical comfort, emotion, physical independence, patient support, and pain, with the global QoR-40 score on the third day after surgery, were 0.902, 0.858; 0.878, 0.769; 0.858, 0.603; 0.926, 0.607; and 0.825, 0.680; respectively (Table 2). In addition, the Cronbach’s alpha of the global QoR-40 T on the third day after surgery was 0.936. The item-total correlation coefficients on the third day after surgery ranged from 0.299 to 0.687. Test-retest reliability (Spearman’s correlation coefficient) of the QoR-40 T three days after surgery was 0.658 (p < 0.01). A positive moderate correlation was obtained between the related subscales of the QoR-40 T and SF-36 on the third day and one month after surgery. The physical comfort, emotional state, physical independence, and pain dimensions of the QoR-40 T appeared to be related to the physical component summary, mental health, physical functioning, and bodily pain subscales of the SF-36, respectively (Table 3). Moreover, a positive significant correlation was observed between the total QoR-40 T score and that on the third day after surgery, and the recovery VAS (ρ = 0.468, p < 0.01).

The mean preoperative QoR-40 T was 182 (with an SD of 10.4), and after surgery it was 171 (with an SD of 20). Changes in the QoR-40 T and the subscales before and three days after surgery are presented in Table 4. The mean total QoR-40 T score before surgery was 182 ± 10.4, 171 ± 20 on the third day after surgery, and 186 ± 14 one month after surgery. An SRM of 0.62 was conducted between the preoperative and third day QoR-40 T mean scores. An SRM of 0.2 indicated a small effect of intervention, 0.5 a moderate effect, and 0.8 or greater a large effect [12, 13].

The findings of the principal components analysis revealed a five-factor structure in the QoR-40 T of physical comfort, emotion, physical independence, patient support, and pain, as in Myles et al. [1]. Additionally, the ratio of the explanatory importance of the factors by one factor (R²) was 0.268. The item loadings in the physical comfort component ranged from 0.532 to 0.804, the emotion component from 0.474 to 0.789, physical independence from 0.561 to 0.871, patient support from 0.620 to 0.845, and pain from 0.541 to 0.776 (Table 5). In addition, the construct validity was supported by a negative correlation with the duration of hospitalization (ρ = 0.265, p = 0.002), and a lower mean postoperative QoR-40 T score was seen in women (167.73 with an SD of 19.90) compared to men (175.03 with an SD of 19.72) (p = 0.015), as indicated by Myles et al. [1].