Background
For newborns and infants wearing diapers the difficulties in characterizing the appearance of the stool are significant, since the changes in consistency, quantity, and color of the stool are higher than in other age groups [
1‐
4]. The gestational age, the degree of maturation of the gastrointestinal tract, the type of diet administered and the presence of possible congenital malformations, such as some hepatic diseases that cause alterations in the color of stool, influence the wide variation of the intestinal habit of children in these age groups [
1,
2]. Thus, in 2009, the Amsterdam Infant Stool Scale (AISS) was created and validated, providing a specific tool for the evaluation of the stool of children up to 120 days old [
1]. The AISS allows the evaluation of stool consistency, quantity, and color through the interpretation of a series of images of stool in diapers. The amount of stool should be analyzed from the percentage of the occupied diaper, which facilitates and standardizes the analysis [
1,
2]. It can be applied for stool evaluation by parents, caregivers, and healthcare professionals. The AISS proved to be more useful to evaluate the bowel pattern of children who still use diapers, compared to the Bristol Stool Form Scale (BSFS) [
2] and its use has also been increasing [
5‐
8]. However, to be used in clinical practice and scientific investigations in Brazil, it is mandatory to perform the translation and cross-cultural adaptation process for Brazilian Portuguese language [
9‐
11]. Therefore, we carried out translation and cross-cultural adaptation of AISS into Brazilian Portuguese and evaluated the psychometric properties of the translated version.
Methods
This was a single-center study, developed at the Botucatu Medical School, São Paulo State University (UNESP), between September 2017 and September 2019. First, the process of translation and cross-cultural adaptation of the AISS to Brazilian Portuguese was performed (Step 1). Subsequently, the evaluation of the psychometric properties of the translated version (Step 2) was performed through application and evaluation by five examiners utilizing 238 stool photographs of children under 120 days old.
The stool photographs were obtained from the stools of children up to 120 days old, including term and premature infants who were in the maternity ward and neonatal unit of a tertiary hospital, and healthy children who were in outpatient care at the Pediatric Outpatient Clinic. This study was approved by the local Research Ethics Committee (protocol number 69504517.9.0000.5411).
Step 1: translation and cross-cultural adaptation
Aiming to maintain the quality of the cultural adaptation process, the scale was translated and adapted according to the internationally recommended methodology [
9‐
11], incurring six phases:
Phase 1: translation
This phase consisted of two translations from the original language (English) into the target language (Brazilian Portuguese). These translations were carried out, independently, by two bilingual translators, whose mother tongue was Brazilian Portuguese.
Phase 2: summary of translations
The synthesis meeting was held with the participation of two translators, together with a committee of experts, composed of professionals with experience in the field of children’s health (3 doctors, 1 nurse, 1 psychologist) and a university professor, with experience in cross-cultural adaptation of health assessment instruments.
Phase 3: Backtranslation
The synthesized version was translated back into English by two translators who had not participated in the first stage and did not belong to the health field. These translators were mother tongue English speakers and were not informed of the concepts explored by the instrument. These two translations were done independently, without knowledge of the original version of the scale.
Phase 4: pre-final version
The pre-final version was built after evaluation and discussion by all translators and the expert committee. The backtranslations were confronted with the original version of the scale. The committee’s function was to analyze the translated versions and develop the pre-final version.
Phase 5: application of the pre-test and assessment of the degree of understanding
The pre-test was applied to a sample of 40 adults, 20 healthcare professionals and 20 adults who were literate and did not work in the health field [
9‐
14]. These participants each evaluated a stool photograph of a newborn by applying the translated version of AISS. A five-point Verbal Numerical Scale (VNS) was then applied to assess how easily the translated version of the scale as a whole and each of its three components (quantity, consistency, and color) was understood. The guiding question to evaluate of the translated scale as a whole was: “Did you understand what was asked and the differences between these types of stool?”, and to evaluate each of the components was: “Did you understand the differences between these types of stool according to this component of the scale?” The minimum ascribed value was zero (“I did not understand anything”) and the maximum value was five (“I understood perfectly and have no doubts”). Values below three were considered to indicate insufficient understanding [
11‐
13]. These data were tabulated and the median values (minimum/maximum) were calculated. The questions with more than 15% of values considered of insufficient comprehension would have to be reformulated by the expert committee and applied to new respondents [
11,
14] Potential differences between the two groups of participants in this phase were also analyzed.
Phase 6: evaluation of results and obtaining the final version
This phase consisted of the analysis of the results obtained in the pre-test, by the members of the expert committee. From the discussion of the items that still had some difficulty of understanding by the population evaluated, with minimal modifications, the final version of Brazilian Portuguese AISS (BP-AISS) was created.
Step 2: psychometric properties assessment
A total of 238 photographic images were taken of stools from children up to 120 days old, who had no metabolic disorders, congenital malformations, or gastrointestinal disorders and who had not undergone gastrointestinal surgery. The photographs were taken during the daytime period, by three researchers, with the same digital camera (zoom lens, original magnification × 4 and × 7.2 megapixels) [
1]. The diapers with the stool were positioned at 20 cm from the digital camera. The camera’s macro function was applied to all photos. To photograph fresh stool, nurses informed researchers every four hours about the bowel movements of all children in the hospital or an outpatient clinic.
The evaluation of the psychometric properties of the BP-AISS included tests to assess the reliability and validity of criteria. For this, BP-AISS was applied for evaluation of the 238 photographs obtained by five examiners: Examiner 1 was a pediatric surgeon; Examiner 2 was a neonatologist; Examiner 3 was a literate adult woman with completed higher education but without professional experience of child healthcare; Examiner 4 was a nurse working in the neonatal unit, and Examiner 5 was a last year graduate medical student. Examiners who were specialists in children’s health (Examiners 1, 2, and 4) had at least 10 years of professional experience.
The reliability of the translated scale was investigated by comparing the results of the evaluations of the photographs performed by each of the five examiners (inter-examiners reliability), and by the agreement between the evaluations performed by Examiner 5, at two different moments after 3 months (intra-examiner reliability), to investigate the reproducibility of the scale. The validity of the criterion was investigated through correlation analysis between the classifications determined by the non-specialist examiner (Examiner 3) and by the expert examiners, with professional performance in the child health field (Examiners 1, 2, and 4), whose evaluations were considered the “gold standard”.
Statistical analysis
The sample size for the evaluation of psychometric properties of the BP-AISS was calculated from the highest value of agreement between examiners (78%), reported in the study of Bekkali et al. (2009) [
1], considering a zero value of kappa of 0.50, with test power estimated at 90%, to detect differences of up to 70% for the zero value of kappa.
The agreement values were determined using the kappa statistic, using the kappa estimator with quadratic weights (Fleiss-Cohen), considering the predominantly ordinal character of the scale [
15]. The correlation analysis between the responses obtained by the different examiners was performed by Kendall’s correlation coefficient.
Continuous numerical data were expressed as median (minimum/maximum). Continuous numerical variables of non-parametric distribution were evaluated by the Mann-Whitney and Kruskall-Wallis tests, followed by the Dunn post-test. The comparison between the responses in the evaluation of a stool photograph, was performed by the Kolmogorov-Smirnov test. The significance level was 5% and the analysis was performed in the SPSS 22.0 for Windows.
Discussion
This was the first time that AISS went through the process of translation and cross-cultural adaptation to a language other than English. The values obtained during the pre-test phase for investigating the degree of understanding were considered satisfactory [
9‐
11]. The pre-final version also proved to be applicable for healthcare professionals and lay adults, with no significant differences between the classifications determined by these two groups of participants.
The evaluation of the psychometric properties of the BP-AISS showed agreement indicators considered satisfactory among the different combinations of examiners [
16]. Moreover, we observed a high percentage of identical responses, determined by different examiners, for the same stool photograph evaluated by the translated scale. The percentage of responses that varied more than two classifications on the scale was limited, demonstrating that the same images, when evaluated by the scale, by different individuals, provide close responses. The BP-AISS also proved reproducible, with a substantial agreement, in the analysis of stool photographs by the same examiner at different times. Thus, the tests developed for the investigation of reliability proved that the BP-AISS is reliable, by providing similar results for the same respondent at different times, characterizing stability, and for different examiners, characterizing equivalence, composing the two axes of external reliability [
17,
18].
The validity of a criterion represents the relationship between scores for a given instrument and some widely accepted measure, i.e. an instrument or criterion considered to be the “gold standard” [
17]. For the evaluation of this psychometric measure, we consider as the “gold standard” measure the expert examiners’ classifications of the stool photographs according to the BP-AISS. We observed that there was a statistically significant correlation between the classifications of the expert examiners and the non-specialist examiner, for the three components of the BP-AISS, suggesting that the scale can provide a measure considered adequate since its results agree with the results of the “gold standard” evaluations.
In the evaluation of the indicators of agreement obtained in the tests performed between different examiners, the “Consistency” component obtained the lowest values, with a statistically significant difference for the “Quantity” and “Color” components. This result is like that described in the original validation study of the scale, in which the “Consistency” component was the one that also presented the lowest rates of agreement [
1]. Since the evaluation of consistency is a fundamentally important parameter in evaluating the stool’s aspect, being directly related to the colonic transit time, this can be considered as a limitation of AISS. Possibly, this limitation is related to the evaluation of stool present in diapers that make it difficult to determine the consistency, especially when compared to the stool present in toilets—the scenario that is commonly measured by BSFS. Especially in stool with a softer consistency, contact with the buttocks, the dispersion over the surface of the diaper, and the time interval between both the bowel movement and the evaluation are factors that can substantially alter the evaluation of consistency [
3]. This potential limitation can be minimized by performing a direct evaluation of the stool in the diapers without the use of photographs. Wojtyniak et al. (2018) [
8] obtained indicators of agreement among examiners with higher values, for the three components of AISS, when the analysis was performed directly in the diapers and not through the evaluation of photographs. These limitations related to the evaluation of the consistency of stool by AISS have been one of the arguments used by authors who propose new graphic scales for the evaluation of stool in children in this age group. Recently, Huysentruyt et al. (2019) [
3] described a new scale, called the Brussels Scale, which proposes the use of seven types of stool for the determination of consistency, like that proposed by the BSFS. Although the authors found high indicators of agreement between different examiners when comparing the images of the seven types of stool on this scale with the images of the seven types of stool on the BSFS, we believe that AISS allows a more global assessment of the appearance of the stool and the pattern of bowel movement, so peculiar in children of these age groups. In addition to the evaluation of consistency, AISS allows the evaluation of the amount of stool, of relevance in clinical follow-up, for example, in patients who are recovering from intestinal transit after surgical approaches or in treatment for allergic enterocolitis and other gastrointestinal pathologies. The AISS also allows for the evaluation of the stool color, information that is very relevant in the clinical assessment of children of these age groups. For example, for the identification of acholic and hypocholic stool related to obstructive jaundice. In this sense, even healthcare professionals may present difficulties in the identification of acholic or hypocholic stool [
19] which reinforces the indication of the clinical use of graphic scales for systematic evaluation of the stool of newborns and infants, to diagnose potential alterations early [
20].
Two main limitations of this study should be considered. First, the study was conducted in a single center, which limits generalizations and may bring biases related to the social, economic, and cultural context of the sample. Second, stools were analyzed in photographic images and not directly in the diapers, which can influence the interpretation of the stool’s consistency [
8]. However, the analysis of stool photographs is commonly used in validation studies of visual stool form scales since it allows evaluations by different examiners at different times [
1,
21‐
24]. Furthermore, this limitation was minimized by obtaining the photographic images of fresh stools always taken in less than four hours after the bowel movements, according to the methodology used in the AISS development study. Conversely, some strengths of the study can be highlighted, such as the significant number of photographs of diapers analyzed, the evaluation carried out by five different examiners, including healthcare professionals and the lay public.
Open AccessThis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit
http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (
http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.