The aim of this study was to translate and adapt the interRAI AC to the Flemish hospital setting and Belgian care context in compliance with interRAI regulations. We used a combination of translation techniques [8, 11, 12] to address the complex nature of the interRAI Suite, which was to be applied in the complex Belgian context. This context is complex because three interRAI instruments of three care settings are used conjointly in one web application and because Belgium has three official languages. Therefore, the process of adapting the Flemish interRAI AC could not be done in isolation. We had to approach the adaptation process, appreciating its particular background/perspective and taking into account the implications of adaptation at multiple levels.
On a hospital level, the assessment file needs to contain all the items that are usually assessed by routine CGA in Belgian geriatric wards. For this reason, we involved clinicians of multiple disciplines (e.g., nurse, geriatrician, case manager). Both academic clinicians and non-academic clinicians reviewed the administrative sections and indicated how these sections should be adjusted to the Belgian care setting. Furthermore, they suggested which topics (e.g., informal caregiver) should be added in order to be comparable with routine acute assessments (steps 3, 5, 6).
On an assessor level, the wording needs to fit the jargon currently used in Flemish geriatric literature and in clinical practice. The phrasing must be fluent and unambiguous to the assessors in clinical practice  (steps 2, 3, 5, 6). The scoring of clinical situations sometimes requires explanations or examples that cannot be fully documented in the interRAI AC itself. Therefore, the manual should complement the instrument. Within the Belgian software (BelRAI), this additional information is near at hand, as the assessment items are all linked to specific pages of the manual.
On a patient level, various caregivers will consult interRAI assessments carried out in hospitals, as patients move across several care facilities. In order to track a patient’s health and functional status longitudinally, the adaptation procedures had to guarantee that the core set of items in the interRAI HC, interRAI LTCF, and interRAI AC instruments remained identical, both formal and semantically (step 8).
The interRAI AC is part of the Belgian interRAI (BelRAI) portfolio, which will be implemented in three language regions: Flanders, Wallonia, and the German community. On a national level, the aim was to centralize patient data. Therefore we recursively evaluated interim versions to harmonize the draft and final versions across the three official languages throughout the adaptation process (steps 4, 9). In addition to centralization, harmonization across languages is beneficial for using the BelRAI portfolio in bilingual regions, so colleagues within the same hospital can complete a shared assessment in the preferred language. Furthermore, if a patient moves to another region or if a caregiver speaks another language, previous records can be consulted in the language of choice.
On an international level, there is a need for reliable, large datasets for cross-national comparison of geriatric care in order to increase geriatric knowledge. Therefore, the process of adapting the Flemish interRAI AC instrument was done rigorously. During the adaptation procedure, the official source instrument served as a reference and was consulted repeatedly and systematically every time an item was adapted (steps 1 to 9) .
On an interRAI level, the interest of different nations in using the interRAI Suite continues to grow. It is of utmost importance that the initial content is preserved. There are regulations for permitted adaptations; interRAI retains the copyright to the instrument. We followed the interRAI regulations and submitted the Belgian portfolio for careful examination and official approval (step 10).
The application of this systematic and iterative 10-step approach (Figure 2) produced the Flemish version of the interRAI AC. We are confident that the adapted instrument closely resembles the content in the standard version. This conclusion, however, must be qualified, with the understanding that it is impossible to achieve 100% validation . Also, one can always argue that significant differences in cross-national use could be the result of methodological flaws rather than actual differences . We believe that the careful step-by-step process of validation described in the present study reduces the latter possibility to an acceptable minimum. However, the procedure described in this paper is only a first step of a larger process, involving extensive psychometric research aimed at obtaining a wide and diverse body of evidence about various aspects of validity , reliability , and responsiveness.
Thus far, psychometric evidence on the original version of the interRAI AC is scarce and is limited to draft versions [2, 6, 21]. The results of the current research must be interpreted within this context. Furthermore, this process resulted in a first Flemish version of the interRAI AC. Belgium is the first country to test and use multiple instruments of the interRAI portfolio simultaneously in transitional care. The wording of some specific items was different across the interRAI HC, interRAI LTCF, and interRAI AC instruments (e.g., nausea versus vomiting). Our approach in comparing these instruments in a meticulous process revealed these differences. More research is needed to harmonize all instruments of the interRAI portfolio. InterRAI considers the development of these instruments to be dynamic: These instruments can be optimized and revised in upcoming years as more clinical experience is gained .
At this stage, we noticed that the desired adjustments did not always match the possible adjustments. In other words, the suggestions made by experts and clinicians on how the interRAI AC instrument would best fit the acute context could not always be put into practice. There were constraints. Uniformity with the InterRAI HC and InterRAI LTCF was a priority, because small differences in wording or scoring would imply problems in the reliability of transmural data transfer. Also, since no overall scores are calculated in the interRAI method, the items are regrouped into clinical assessment protocols (CAPs) and scales defined by interRAI. Altering items would affect the clinical algorithms of the output. Moreover, some adjustments are unavoidable in the perspective of instrument integration, even if the clinicians did not mention these. For example, the word ‘patient’, which is common in the acute care sector, was changed to ‘client’ due to practical reasons having to do with the BelRAI software architecture. Another example is intake data, with a more administrative character, which need to be uniform across the interRAI portfolio.
There is no gold standard for translation techniques [8, 10–12]. Rather than performing a back-translation, we used multiple expert panels of differing constitution for pre-pilot evaluation and subsequent field-testing to carefully control the quality of the translation. According to Geisinger  and Cha et al. , this technique is more effective for ensuring that the translation and adaptation is conducted appropriately . During each step, problematic items were identified. But before adjusting the instrument, the items were compared with their original counterparts and, when necessary, revised by the instrument adapter or a committee. Independent back-translation could be used in future studies to further validate the interRAI AC in the Belgian acute care context. Although the current translation and adaptation process was time-consuming, all the different steps were necessary. Because the goal was not merely to guarantee that items on the interRAI AC tap into the same construct but also to have confidence that each item and each scoring option across the instruments tap into the same construct. This procedure (Figure 2) might be used by others facing similar challenges of complex translation and adaptation situations in which multidimensional instruments will be used across multiple languages in multiple care settings. As the use of the interRAI Suite continues to grow worldwide and as the interRAI Suite expands to other care settings and populations, this procedure can guide future translations.