Dutch–Flemish translation of nine pediatric item banks from the Patient-Reported Outcomes Measurement Information System (PROMIS)®

Purpose The Patient-Reported Outcomes Measurement Information System (PROMIS®) is a new, state-of-the-art assessment system for measuring patient-reported health and well-being of adults and children. It has the potential to be more valid, reliable, and responsive than existing PROMs. The items banks are designed to be self-reported and completed by children aged 8–18 years. The PROMIS items can be administered in short forms or through computerized adaptive testing. This paper describes the translation and cultural adaption of nine PROMIS item banks (151 items) for children in Dutch–Flemish. Methods The translation was performed by FACITtrans using standardized PROMIS methodology and approved by the PROMIS Statistical Center. The translation included four forward translations, two back-translations, three independent reviews (at least two Dutch, one Flemish), and pretesting in 24 children from the Netherlands and Flanders. Results For some items, it was necessary to have separate translations for Dutch and Flemish: physical function—mobility (three items), anger (one item), pain interference (two items), and asthma impact (one item). Challenges faced in the translation process included scarcity or overabundance of possible translations, unclear item descriptions, constructs broader/smaller in the target language, difficulties in rank ordering items, differences in unit of measurement, irrelevant items, or differences in performance of activities. By addressing these challenges, acceptable translations were obtained for all items. Conclusion The Dutch–Flemish PROMIS items are linguistically equivalent to the original USA version. Short forms are now available for use, and entire item banks are ready for cross-cultural validation in the Netherlands and Flanders.

(two items), and asthma impact (one item). Challenges faced in the translation process included scarcity or overabundance of possible translations, unclear item descriptions, constructs broader/smaller in the target language, difficulties in rank ordering items, differences in unit of measurement, irrelevant items, or differences in performance of activities. By addressing these challenges, acceptable translations were obtained for all items. Conclusion The Dutch-Flemish PROMIS items are linguistically equivalent to the original USA version. Short forms are now available for use, and entire item banks are ready for cross-cultural validation in the Netherlands and Flanders.

Background
The use of pediatric Patient-Reported Outcome Measures (PROMs) in pediatric research, clinical trials, and clinical practice has increased in the past years to obtain insight into the consequences of a chronic illness on a child's life, to monitor patient's health status and to assist with communication [1].
As summarized by Terwee et al. [2], PROMs are not free of problems and challenges. For example, for many constructs, several instruments of varying quality have been developed. Some PROMs are burdensome for patients because they are too long, or contain irrelevant, incomprehensible or poorly formulated questions. Therefore, the Patient-Reported Outcomes Measurement Information System (PROMIS) project was initiated to advance the science and application of PROMs [3,4]. The PROMIS project transformed existing PROMs into a more optimal assessment system for measuring health-related quality of life (HRQOL). This new system has shown to have two important advantages. Firstly, the system has a higher validity, reliability, and better responsiveness than the existing PROMs [5][6][7]. Secondly, patients need to answer fewer items compared to traditional questionnaires, because the PROMIS consists of a set of item banks. An item bank is a set of questions that have all been statistically calibrated to the same underlying construct. The items from an item bank can be administered in brief, fixed questionnaires, or more efficiently through computerized adaptive testing (CAT) [8]. A CAT is a computer-administered test in that, after the first item, the presentation of items is determined by a person's response to previous ones. After about 5-7 items, the computer stops presenting questions [9].
The pediatric PROMIS project focused on the development of self-report item banks across several health domains for youth aged 8-18 years. The primary focus was on the measurement of generic health domains that are important for children across a variety of illnesses [10][11][12][13][14][15].
It is expected that PROMIS will be implemented worldwide, and that PROMIS instruments will rapidly replace existing PROMs [16][17][18]. In 2011, the Dutch-Flemish pediatric PROMIS Group was established with the aim to implement pediatric PROMIS instruments in the Netherlands and Flanders (northern Belgium). We strived to obtain one uniform Dutch-Flemish translation for all items. The goal of a universal approach to translations was to result in one version for multiple countries instead of country-specific versions of the same language. This approach was chosen for practical reasons and to avoid unnecessary language bias introduced by multiple translations. There is also policy support for one official language [19]. The first step concerned the translation and cultural adaptation of the PROMIS items into Dutch-Flemish. This paper describes the translation of nine item banks for children.

Methods
The translation was obtained using a universal approach based on the established Functional Assessment of Chronic Illness Therapy (FACIT) Multilingual Translation Methodology [20,21]. The goal of this methodology was to attain five dimensions of cross-cultural equivalence: 1. Semantic/linguistic: The meaning of the item is the same in the source and translated language; 2. Content: The item is relevant to both cultures; 3. Conceptual: The translated document measures the same theoretical constructs as the source; 4. Criterion: When compared to a known or standardized measurement, the translation exhibits similar measurement properties of the source; 5. Technical: The method of assessment results in comparable measurements in both cultures [22].
We addressed the first three dimensions. The last two will need to be checked by additional psychometric validation. We strived to obtain one uniform Dutch-Flemish translation for all items, but separate translations for Dutch and Flemish were produced when necessary.

Translation
The translation team implemented specific steps in order to develop precise and culturally appropriate translations 10  staff assessed the equivalence of the final translation and verified that documentation of the decision-making process was complete. The Dutch language coordinator was consulted again for additional input when necessary; 9. Formatting, typesetting, and proofreading of the final questionnaire or item forms were performed by two proofreaders working independently.

Testing of translations
The target language version resulting from the described translation process was pretested in a pilot study in a convenience sample of native Dutch-or Flemish-speaking children aged 8-17 years. Exclusion criteria were (1) unable to read and speak Dutch or Flemish and (2) unable to provide verbal informed consent. Each item was debriefed using a standardized cognitive debriefing interview to ensure that the meaning of the item remained equivalent after translation. After the interviews, FACITtrans staff analyzed subjects' comments to empirically determine the linguistic validity and acceptability of the questionnaire.

Translation
• Differences between Dutch and Flemish For four out of nine item banks (Table 1) and seven out of 151 items (5 %), it was necessary to have separate translations for Dutch and Flemish for some items (see asterisks Table 1). For example, the word ''walking'' was translated as ''lopen'' in Dutch, but had to be translated as ''stappen'' in Flemish because ''lopen'' means running in Flemish (''hardlopen'' in Dutch) and ''stappen'' means going out in Dutch.

Pilot testing
The Dutch-Flemish PROMIS item banks were tested in the Netherlands and Flanders. In total, 24 children (12 from Belgium, 12 from the Netherlands) participated, of which 58 % were female (n = 14) and 42 % were male (n = 10), with an average age of 14 years. All participants were from the general population, with the exception of six children recruited for the Peer Relationships and Asthma Impact item bank, who were all diagnosed with asthma. Five slight changes in the wording of questions were made after testing. For example, the phrase ''Ik voelde me aanvaard door andere kinderen'' was changed to ''Ik voelde me geaccepteerd door andere kinderen…'' in which ''aanvaard'' as well as ''geaccepteerd'' both means accepted.

Discussion
Nine pediatric PROMIS item banks for children were translated from English into Dutch-Flemish. Some difficulties were found in the translation of some items, but eventually an acceptable translation was obtained for all items. The translation was produced using a strong standard PROMIS methodology. The PROMIS translation methodology was developed through substantial research in the HRQOL field to ensure that translations reflect conceptual equivalence with the English source and are rendered in language that is culturally acceptable and relevant to the target population. We strived to obtain one uniform Dutch-Flemish translation for all items. Nevertheless, for seven items, it was necessary to have separate translations for Dutch and Flemish.
The use of pediatric PROMIS item banks has clear advantages over traditional PROMs [23], and therefore, we highly recommend to use these PROMIS instruments. PROMIS instruments have been found to have less measurement error and better responsiveness than existing measures, which leads to smaller sample sizes required in clinical studies [24,25]. PROMIS scores are easy by using item response theory (IRT) methods; scores on interval level are obtained. In addition, all PROMIS instruments are scored on a common metric: scores are expressed as T-scores with a mean score of 50 (representing the mean score of the reference population) and a standard deviation of 10.
PROMIS instruments can be administered in short forms or through CAT (or a combination of both). CAT has great advantages over traditional paper questionnaires. In children, only one study has been performed with mixed results, so more studies are needed [25].
A limitation of this study is that the questionnaires are translated to Dutch-Flemish, but the content validity of the items banks was not yet analyzed. Before PROMIS can be considered valid for use in the Netherlands and Flanders, several steps should ideally be taken, as described by Terwee et al. [2]. Cross-cultural validation studies are needed to evaluate the IRT model fit in Dutch-Flemish children and to test for possible differential item functioning (DIF) between language groups. For example, the use of different units of measurement may have affected the item difficulty and may therefore introduce DIF. All persons at a given trait level should answer an item in the same way regardless of the language version completed. If an item functions differently in the original and translated versions, the item exhibits DIF with regard to translation. If important language DIF is found, language-specific item calibrations may need to be developed. Further research on these translations will increase confidence in their use. To make the first step in validating the pediatric item banks in the Netherlands, we will start a psychometric study in children with Juvenile Idiopathic Arthritis.
In conclusion, the translated versions of the Dutch PROMIS items are linguistically acceptable and are ready for cross-cultural validation studies in the Netherlands and Flanders.
Acknowledgments The Dutch-Flemish translation of the PROMIS item banks was financially supported by a grant from the Dutch Arthritis Association. This study was supported by Pfizer BV Pharmaceuticals. We would like to thank Hedy van Oers, Perrine Limperg, The number of items with a Dutch and Flemish version; a = 1, b = 2 and c = 3 and Juliette van der Linden for their help with reviewing some of the PROMIS item banks.

Conflict of interest
The authors declare that they have no conflict of interest.
Open Access This article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.