Introduction

Type 1 diabetes is one of the most common chronic diseases seen in children and adolescents [1, 2]. Individuals with type 1 diabetes constitute 5–10 % of all diabetics in the world [3]. There are no recent studies which determine the incidence and/or prevalence of T1DM in Turkey [4, p.187]. However, according to the incidence study performed in 1996 with children between the ages of 0–15 who had Type 1 diabetes (The National Children's and Adolescents' Diabetes Data Collection Study (UÇADİVET)), this rate was found to be 2.52 in 100.000 yearly [5]. Local studies indicate, just like studies worldwide, that the incidence of Type 1 diabetes is increasing and that the starting age of Type 1 diabetes, which is usually adolescence, is sliding towards the ages between 5–9. In our country, the total number of minors under the age of 18 who have diabetes is estimated to be around 15,000 [4, p.187]. Self-monitoring constitutes a major part of diabetes management [1]. When diabetes is managed effectively, long-term complications are prevented or delayed. However, many children/adolescents with diabetes cannot comply with diabetes management due to various reasons such as wanting to be independent, avoiding the responsibilities regarding diabetes management and complexity of diabetes management [1, 6]. As a result, their quality of life is adversely affected [1, 7, 8].

According to the definition of the World Health Organisation (WHO) “Quality of life is defined as individuals' perceptions of their position in life in the context of culture and value systems in which they live and in relation to their goals, expectations, standards and concerns.” [9, p.5]. Assessment of quality of life via psychological tests play an important role in determining the impact of the treatment process on an individual's quality of life and in determining related health policies. Treatment and assessment of diseases and disease approaches in adolescents is different from those in children and adults. These differences should also be considered when measuring the quality of life in adolescents [10].

Improving the quality of life of an adolescent with diabetes is a dynamic process that differs in characteristics with each individual and that requires different personal approaches [11]. In providing the best glycemic control with the least amount of medical help in diabetes management, holistic care and an interdisciplinary team approach increases success. In such a team, members of various disciplines such as a medical doctor, nurse, dietician, pharmacist, psychologist, social services professional, and a podiatrist would be present [11, 12]. However, diabetes nurses spend more time with adolescent with diabetes and his/her parents. Therefore, they play a key role in increasing the adolescent's quality of life. In diabetes nursing, interventions aimed at maintaining quality of life provide appropriate nursing care for adolescents with diabetes, which, in turn, lead to relief and improvements in physical, emotional and psychosocial functioning [13, 14]. Measuring quality of life in adolescents with diabetes creates the opportunity to assess the impact of treatment and care practices on all aspects of life from the adolescent's point of view. Furthermore, measuring adolescents' quality of life guides nursing studies which evaluate nursing practices and the quality of care as well as contributing to the development alternative diabetes management strategies [14].

No specific quality of life scale is available for adolescents aged 13–18 years with diabetes in Turkey. Thus, the necessity of adapting the Pediatric Quality of Life InventoryTM (PedsQL) Diabetes Module 3.0 Teen Report (Ages 13–18) and Parent Report for Teens (Ages 13–18) to the Turkish population became apparent. Mean Cronbach's alpha values of the PedsQLTM Diabetes Module 3.0 scales, which include 28 items, is 0.71 for child report and 0.77 for parent report [15]. Validity and reliability studies of the PedsQLTM Diabetes Module 3.0 scales were conducted in many countries including Sweden, Iran, Greece, and Hungary [7, 8, 16–18].

The aim of the study was to examine the psychometric properties of the PedsQLTM 3.0 Diabetes Module Teen and Parent Report (Ages 13–18) in Turkish adolescents with type 1 diabetes.

Methods

Participants

A total sample of 104 adolescents aged 13–18 years with type 1 diabetes and their parents who applied to a university hospital were invited to participate in this study. This hospital in the city of Istanbul in Turkey has a sizeable diabetes center where many children with diabetes are followed-up and monitored. Volunteering adolescents with duration of diabetes of 1 year or more who regularly attended routine check-ups (every 3 months) were included in the study since adaptation to a chronic disease takes approximately 6 months-1 year [19]. Therefore, adolescents with duration of diabetes of 1 year or more were deemed to be appropriate for participation in the study, considering the adaptations process of adolescents with diabetes. The mean age of the participants was 14.65 ± 1.51; where 51.9 % (n = 54) were girls and 48.1 % (n= 50) were boys. For 36.5 % (n = 38) the duration of diabetes was 1–3 years, for 36.5 % (n = 38) it was 4–6 years, and for 26 % (n = 27) it was 7 years or above. The mean HbA1C values were 8.67 ± 2.04.

Design

A methodological design was used. Methodological design is a process used to determine the validity and reliability of instruments in order to measure constructs used as variables in research [20, p.114]

Instruments

The Pediatric Quality of Life Inventory (PedsQL) 4.0 Generic Core Scales (GCS)

The parent/child should first complete the PedsQL™ Generic Core Scales and then complete any additional PedsQL™ Module [21]. Therefore, PedsQLTM 4.0 Generic Core Scales (GCS) Teen Report (Child Self Report-CSR and Proxy Parent Report-PPR) was used to assess concurrent validity. The Pediatric Quality of Life Inventory (PedsQL) is a modular instrument designed to measure health-related quality of life (HRQOL) in children and adolescents aged 2–18 years. The PedsQL 4.0 Generic Core Scales are child self-report and parent proxy-report scales developed as the generic core measure to be integrated with the PedsQL disease-specific modules [15]. Child self-report includes ages 5 to 7, 8 to 12, and 13 to 18 years. Parent proxy-report includes ages 2 to 4 (toddler), 5 to 7 (young child), 8 to 12 (child), and 13 to 18 (adolescent), and assesses parent's perceptions of their child's HRQOL. The 23-item PedsQL GCS 4.0 included physical functioning (8 items), emotional functioning (5 items), social functioning (5 items) and school functioning (5 items). The instructions ask how much of a problem each item has been during the past 1 month on a five-point Likert-type response scale. (0 = never a problem; 1 = almost never a problem; 2 = sometimes a problem; 3 = often a problem; 4 = almost always a problem). Items are reverse-scored and linearly transformed to a scale ranging from 0 to 100 (0 = 100, 1 = 75, 2 = 50, 3 = 25, 4 = 0). Total scores and subscale scores are computed as the sum of the items divided by the number of items answered. Higher scores indicate better quality of life. If more than 50 % of the items on the scales are missing, the scale score is not computed [22]. The validity study for the Turkish version of the scale's form for ages 13–18 was performed in 2007 by Çakın Memik et al. The Cronbach's alpha value was 0.82 for the adolescent form and 0.87 for the parent form [10].

Pediatric Quality of Life Inventory (PedsQL) 3.0 Diabetes Module

The PedsQL 3.0 Type 1 Diabetes Module was designed to measure diabetes-specific HRQOL. Child self-report includes ages 5 to 7, 8 to 12, and 13 to 18 years. Parent proxy-report includes ages 2 to 4 (toddler), 5 to 7 (young child), 8 to 12 (child), and 13 to 18 (adolescent), and assesses parent's perceptions of their child's diabetes-specific HRQOL. The 28-item multidimensional PedsQLTM 3.0 Diabetes Module included five subscales, which are diabetes symptoms (11 items), treatment barriers (4 items), treatment adherence (7 items), worry (3 items), and communication (3 items). The format, instructions, Likert type response scale, and the scoring method are identical to the PedsQLTM GCS 4.0, with higher scores indicating fewer symptoms or problems [15].

Data collection

Data was collected by the researchers between October 2011 and November 2012 during routine controls of adolescents at the diabetes polyclinic. Adolescents and parents filled the scales themselves. Researchers accompanied adolescents and parents during data collection. Therefore, there is no missing data. Data collection took approximately 10 to 15 min.

Validation

Permission was received from the MAPI TRUST Institute. The linguistic validation process consisted of 2 phases: forward translation and backward translation. PedsQLTM 3.0 Diabetes Module Scales Teen Report was first translated from English to Turkish separately by three bilingual linguistic, medical and nursing professionals and then the scales were translated back from Turkish to English by two different bilingual language experts. The back-translated and original scales were compared and found to be highly similar. The Turkish translation of the scales was presented to the opinion of seven people who were experts in nursing and health sciences in order to test understandability and cultural suitability. Content validity was achieved by making necessary corrections on the basis of minor suggestions such as word or sentence corrections. A preliminary administration was carried out with 25 adolescents with type 1 diabetes in order to ensure face validity.

Data analysis

NCSS (Number Cruncher Statistical System) 2007 & PASS (Power Analysis and Sample Size) 2008 Statistical Software (Utah, USA) were used for statistical analyses. Validity and reliability analyses were assessed by calculating reliability statistics and Cronbach's Alpha coefficients in addition to descriptive statistical methods (mean, standard deviation, frequency and ratio). Results obtained by applying principal component analysis and varimax rotation were used in the explanatory factor analysis of the Diabetes Module. Spearman Correlation Analysis was used for the assessment of relations between scale scores.

Ethical considerations

A written consent was obtained from the hospital where the study was conducted. After the adolescents and their families were informed about the purpose of the study, the types of scales that would be used, how long data collection would take, how all information pertaining to them would be confidential, and how they could leave the study any time they want, all adolescents and their parents volunteering to participate in the study separately provided written informed consents. We did not obtain a written document from the ethical committee because the hospital did not consider it necessary. The study was carried out in accordance with the principles of the Declaration of Helsinki.

Results

A total sample of 104 adolescents aged 13–18 years with type 1 diabetes participated in the Turkish validity and reliability study of PedsQLTM 3.0 Diabetes Module (DM) Teen Report (Child Self Report-CSR and Parent Proxy Report-PPR).

Validity

Content Validity

The content validity of the scales were assessed by an expert panel of seven academicians and health professionals working on diabetes (one professor in internal diseases nursing, one professor in community health nursing, one professor in pediatric nursing, one assistant professor in pediatric nursing, one assistant professor in community health nursing, one psychologist in pediatrics, and one pediatric diabetes nurse). Minor corrections were applied in accordance with the experts' recommendations. Kendall's W concordance analysis was conducted in order to assess the content validity of CSR and PPR of The PedsQLTM 3.0 DM and it was found that there was concordance among experts (Kendall's Wa = 0.206 ,df = 27, x2 = 38.89, p = 0.065). A subtype of content validity is face validity, which is a rudimentary type of validity that basically verifies that the instrument gives the appearance of measuring the concept [23, 24]. Face validity of the scales were assessed with a pilot study conducted on 25 adolescents with type 1 diabetes and their parents, and no negative feedback was received.

Construct Validity

The adequacy of the items was assessed in terms of item content. The original PedsQLTM 3.0 DM contains five subscales: diabetes symptoms, treatment barriers, treatment adherence, worry, and communication. Although the original PedsQLTM 3.0 DM contained five subscales in the explanatory factor analysis, the treatment barriers and treatment adherence subscales were evaluated as the Treatment subscale because of their compatibility with regard to the contents of their questions and, the Turkish validity and reliability study of PedsQLTM 3.0 DM was continued with four subscales: diabetes symptoms, treatment, worry, and communication. The four factors explained 57.99 % of the variance in CSR and 56.96 % of the variance in PPR. The Kaiser-Meyer-Olkin measure was 0.79 in CSR and 0.75 in PPR, indicating the applicability of explanatory factor analysis. Barrlett's test of sphericity was statistically significant in both CSR (χ2 = 1503.65; p = .001) and PPR (χ2 = 1445.83; p = .001). The acceptable level for scales items was above 0.30 [25] in the results of explanatory factor analysis (Table 1).

Table 1 Characteristics of Subscales in PedsQLTM 3.0 DM child-self-report (CSR) and parent proxy-report (PPR) (N = 104)
Full size table

Concurrent Validity

Concurrent validity was assessed through an analysis of the intercorrelations between the PedsQLTM 4.0 GCS total scores and the PedsQLTM 3.0 DM Teen Report subscales. Intercorrelations are shown in Table 2.

Table 2 Intercorrelation between PedsQLTM 3.0 DM and PedsQLTM 4.0 GCS (N = 104)
Full size table

Reliability

The total-item and internal consistency reliability values of the subscales are shown in Table 3. The PedsQLTM 3.0 DM had a Cronbach's alpha of 0.92 in CSR and 0.88 in PPR. Item-total correlations are shown in Table 1 and ranged from 0.33 to 0.70 in CSR and from 0.33 to 0.73 in PPR.

Table 3 Internal consistency reliability for PedsQLTM 3.0 DM total-items and subscales in child-self-report (CSR) and parent proxy-report (PPR) (n = 104)
Full size table

Discussion

The main characteristics sought for in a good measurement tool are reliability and validity [20, p.114]. The psychometric properties of the Turkish version of PedsQLTM 3.0 DM were measured with a wide range of methods and statistical analyses. This improved the strength of validity and reliability.

Validity

Content validity represents the universe of content or the domain of given constructs [23, 24]. Kendall's W concordance analysis was conducted to assess content validity in this study. As a result of the test, no meaningful difference was found among the views of the experts. Accordingly, the terms in the scale are appropriate for our culture and represent the field that is to be measured.

Construct validity was supported by the results of the explanatory factor analysis. The factor analytical approach is a procedure that gives the researcher information about the extent to which a set of items measures the same underlying construct [23, 24]. In this study, acceptable factor loadings for the scales items were above 0.30 [25] and therefore, no items from the original scale was excluded. The original PedsQLTM 3.0 DM child and adolescent forms consisted of five subscales, whereas the Turkish version comprised of four subscales. Similarly, factor loadings of both child and parent's forms were gathered in four subscales since it was impossible to do so in five subscales in the study by Nansel et al. (2008) [26]. The PedsQLTM 3.0 DM scale is a valid scale for Turkish adolescents with type 1 diabetes which has four subscales.

Concurrent validity refers to the degree of correlation between two measures of the same concept administrated at the same time [23, 24]. The intercorrelation between PedsQLTM 4.0 GCS subscales scores and the PedsQLTM 3.0 DM subscales scores ranged from 0.36 to 0.71 in CSR and 0.36 to 0.62 in PPR. The intercorrelations had from medium to large effect sizes, which confirmed the concurrent validity of the PedsQLTM 3.0 DM. These findings are consistent with earlier studies [7, 8, 16–18]. No correlation between PedsQLTM 3.0 DM Communication subscale and PedsQLTM 4.0 GCS Physical Functioning Scale Score was found in both CSR and PPR forms because PedsQLTM 4.0 GCS Physical Functioning Scale Score only investigates behaviors related to physical activities such as running and swimming; whereas PedsQLTM 3.0 DM Communication subscale investigates communications with the healthcare personnel and other individuals. Therefore, it was expected that there will be no correlation between these two scales.

Reliability

The reliability of a research instrument is defined as the extent to which the instrument yields the same results on repeated measures [23]. A level of Cronbach's alpha at 0.70 or higher is considered to be an acceptable level of reliability [23, 25]. There are five major tests of reliability that can be used. The test used depends on the nature of the instrument. They are known as test-retest, parallel or alternate form, item-total correlation, split-half, Kuder-Richardson (KR-20), Cronbach's alpha, and interrater reliability [23]. In this study, item-total correlation and Cronbach's alpha was used for testing internal consistency reliability. The PedsQLTM 3.0 DM had a Cronbach's alpha of satisfactory internal consistency. Cronbach's alpha coefficients varied between 0.88 and 0.91 for CSR and PPR in Sand's (2012) [8] and Lukác's (2012) [18] studies and are similar to our findings. Furthermore, the Cronbach's alpha values in our study were higher compared to those in the studies of Jafari (2012), Varni (2003), Emmouilidou (2008), Abdoul-Rasoul (2012) and Nansel (2012) [7, 15–17, 26]. Item-total correlations of all items were above 0.30 in this study. Therefore, no items were excluded from the original scale.

Limitations

This study has some limitations. First, the study was carried out only in one diabetes center. For this reason, the sample size of the study is limited. Second, results of the study can only be generalized to Turkish adolescents aged 13–18 years with diabetes who has a duration of diabetes of 1 year or more. The scale is not appropriate for use in people with diabetes between the ages of 13–18 whose duration of diabetes is below one year. As a third factor, adolescents with other co-morbid disorders (celiac diseases etc.) were not included in the study.

Conclusion and Recommendations

The psychometric analyses of the Turkish version of PedsQLTM 3.0 DM Teen and Parent Reports indicate high reliability and good content and construct validity. Based on these findings, it can be concluded that the Turkish version of PedsQLTM 3.0 DM is a useful instrument for measuring diabetes-specific health related quality of life in Turkish teens aged 13–18 years with type 1 diabetes. The applicability of the scale should be tested for adolescents between the ages of 13–18 whose duration of diabetes is below one year and those who have other co-morbid disorders. This instrument can be easily used by pediatric diabetes nurses. In addition, it is thought that PedsQLTM 3.0 DM forms a basis for interventions which support adolescents' and their parents' quality of life in a positive way and improve the quality of nursing care. Nursing care provided for children and adolescents with diabetes is not the sole responsibility of pediatric diabetes nurses in Turkey. Pediatric diabetes nurses also care for children and adolescents with endocrinologic disorders. This, in turn, increases the work burden of diabetes nurses and limits the time spent for the care of adolescents with diabetes. Diabetes nurses will be able to detect problems which have a negative impact on diabetes-specific health related quality of life among adolescents with diabetes and to increase the quality of care provided for them by using this scale. This scale can provide a guideline for developing new scales in our country on the subject.