Cross-cultural validation. The first phase of the study was the cross-cultural validation of the Stoma-QoL. In the second phase, the cross-sectional study was conducted to analyze variables associated with the questionnaire point scoring. The study was conducted between September 2014 and August 2015.
Cross-cultural validation is a complex process that requires high methodological rigor; there is no consensus in the literature about the how it should be performed.12 For the cross-cultural adaptation of this questionnaire to the Brazilian variant of Portuguese, the following steps were carried out: validation, instrument translation to the new language, assessment by experts, and back-translation by a native English speaker.12,13 Initially, the questionnaire in its original English version was translated into Portuguese independently by a nutritionist and a nurse with at least 5 years of experience in their fields. They translated and analyzed each item selected to compose the instrument until they agreed to ensure the instrument’s applicability and whether all the aspects to be studied were accurately represented. This validation is not statistically determined and is not expressed by any correlation coefficient. The final consolidated version of this independent translation was submitted to a native English language translator, together with the original in English, to evaluate the adaptation of the items to the new language. The translator must be proficient in both languages of interest and familiar with the cultures associated with the language of the different groups. This characteristic allows for the translation process to consider the nuances of the language for which the instrument is targeted, facilitating a more culturally appropriate adaptation process.14
Participants. After translation, the Stoma-QoL was administered to patients served by the Stoma Patient Health Care Service in Juiz de Fora, a Brazilian city located in the Southeast region of the country with approximately 500 000 inhabitants and a health care hub for ~2 million people. This service provided care for 428 patients monthly; all were contacted by telephone and invited to participate in the survey. Study inclusion criteria stipulated participants should be at least 18 years of age, have an intestinal stoma (ileostomy or colostomy), and be physically and mentally capable of completing the Stoma-QoL along with a sociodemographic, clinical, and eating behavior questionnaire. Eating habit variables included eating comfort in the postoperative period, excluding foods for a period of time, fear of eating, and excluding foods for different reasons. Participation was voluntary. Researchers completed the paper-and-pencil questionnaires: Table 1, Table 2, and Table 3 show the questions and scoring.
The questionnaires were administered by trained researchers as part of their nutritional clinical care. All volunteers gave their written informed consent after being provided oral and written information about the aims and protocol of the study. The Stoma-QoL questionnaire takes 5 to 10 minutes to complete. The questionnaire with sociodemographic, clinical, and eating behavior characteristics takes 15 to 20 minutes to complete.
The present study was approved by the Ethical Committee in Human Research from Federal University of Juiz de Fora (protocol number 516.306).
Data collection. The first author trained a team of graduates in nutrition to participate in the patient assessment on the selected measurement day. All data were collected using a standardized form developed for the study and stored electronically. The identities of patients were kept anonymous, and patients had the option to refuse to participate during the assessment.
Statistical analysis. The reliability and accuracy of the Stoma-QoL were assessed through their internal consistency using Cronbach’s alpha coefficient. Internal consistency was assessed for the questionnaire as a whole, with a minimum value of 0.70 adopted as a satisfactory level.15 Agreement and reproducibility were assessed using the intraclass correlation coefficient (ICC). An ICC of 0.75 indicates excellent study reproducibility.16
Convergent validity. The last step of the study was establishing convergent validity, defined as the significant relationship between 2 or more measures of a single construct or of theoretically related constructs using different methods or evaluation instruments.17 This step of the validation process involved the Stoma-QoL questionnaire translated into Brazilian Portuguese and the 12-item version of the Short-Form Health Survey (SF-12) previously validated in Brazil.18 The SF-12 questionnaire evaluates 8 dimensions that influence quality of life and considers the individual’s perception of aspects of health over the previous 4 weeks. Each item has a group of responses distributed on a graduated Likert scale and evaluates the following dimensions: physical functioning, role-physical, pain, general health, vitality, social functioning, role-emotional, and mental health. Through an algorithm specific to the instrument, 2 scores are obtained: the Physical Component Summary (PCS) and the Mental Component Summary (MCS). In both, the total score ranges from 0 to 100, with higher scores associated with better quality of life.19-21 A sample of 11 (10%) of all study participants was randomly selected to respond to the SF-12 questionnaire. The Pearson correlation coefficient was determined between each item of the questionnaires and the final scores for the PCS and MCS components. In interpreting the Pearson correlation coefficient, a value (both positive and negative) >0.6 is considered a strong correlation, 0.35 to 0.6 indicates moderate correlation, and 0 to 0.35 indicates a weak correlation.22
Results from the Stoma-QoL and the sociodemographic/clinical/eating behavior questionnaire were entered, processed, and analyzed using SPSS software, version 15 (SPSS Inc, Chicago, IL). Sociodemographic, clinical, and eating behavior variables were descriptively analyzed. The normal distribution of Stoma-QoL total scores was assessed using the Kolmogorov-Smirnov test, the normality of the scores was rejected, and nonparametric tests (Mann-Whitney and Kruskal-Wallis) were used to analyze the associations among independent sociodemographic, clinical, and eating behavior variables. A 5% significance level was considered for analysis.
Among the 428 intestinal ostomy patients in the care of the stoma health care service, 111 of both genders met the inclusion/exclusion criteria and/or were permitted to participate. The majority (57, 51.4%) were men, most (94, 84.7%) were in the 50 years of age or older age group, 65 (58.6%) were married, and 14 (12.6%) had a college education (see Table 1).
Among all participants, 70 (63.1%) had a colostomy and 41 (36.9%) had a ileostomy; 66 (59.5%) had a permanent stoma, 72 (64.9%) had the stoma for more than 1 year, and 79 (71.2%) had the ostomy due to cancer of the colon and rectum (see Table 2).
Regarding eating behavior, 69 (62.2%) had felt comfortable with their normal eating habits within 1 day to 3 months of surgery, 52 (46.8%) were afraid to eat some type of food, and 52 (46.8%) excluded foods from their eating routine for fear of causing gas (see Table 3).
The internal consistency of the translated Stoma-QoL had a Cronbach’s alpha of 0.87, indicating the instrument is sufficiently reliable as a research tool. The reproducibility over time was estimated by an ICC of 0.85 (95% confidence interval [CI] 0.69–0.95), demonstrating good reproducibility. The convergent validity between the Stoma-QoL and the SF-12 using the Pearson coefficient confirmed a higher correlation among the emotional, mental health, social aspect, and vitality items, of the MCS (r = 0.52; P =.02) and lower values for correlations between the PCS and the questions that assess general health, functional capacity, physical aspects, and pain (r = 0.38; P = .04), although both were significant. A moderate and significant association was found between the questionnaires.
The mean overall Stoma-QoL scale score for study participants was 58.7 ± SD 12.0 (range 32.0–78.0). Scores were mostly comparable among the characteristics, and statistically significant differences were observed for gender and eating behavior (see Table 4). Women had lower quality of life than men (P = .02), as well as stoma patients who fasted (P =.05) and those who excluded foods from their eating routine for fear of repercussions (P <.001). Colostomy and ileostomy patients had a similar quality of life, as well as persons whose stoma was created due to colon or rectal cancer and other diseases.