In this study, we used the modified Delphi method to develop the Chinese version of the LTCRL assessment scale. After one expert meeting, three rounds of Delphi process, and a pilot assessment of the LTCRL performance of home care workers in Taiwan, the scale was found to be feasible and reliable in the pilot test. The higher the education levels of home care workers in Taiwan, the better was the LTCRL performance, especially in the process/appraise and apply/use categories.
In the Chinese version of the LTCRL assessment scale, we incorporated both the concept of health literacy and reablement literatures [17, 24] and then modified the scenario-based questionnaires to fit the common scenario of reablement in Taiwan. The scale was completed through a three-round Delphi process by 10 respondents from five professional fields. Latif and colleagues stated that a good result of the Delphi process can be achieved even with a small panel of 10 individuals [26]. We invited a panel of experts who had experiences in reablement, and most of them work as supervisors for reablement in both private and public healthcare settings; thus, the scale met face/content validity. Regarding reliability, the coefficient of K-R-20 was 0.53, which is acceptable (>0.5) for clinical practice [28]. The reason for the low reliability could be that the proportion of the responses to the items was highly correct. The average score of the scale is 25.19, and the highest score obtainable is 29. The dichotomy of the responses designed for the scale might be the other reason for the low reliability. Nevertheless, after a structured expert consensus was obtained throughout the modified Delphi method and the pilot assessment, the Chinese version of the LTCRL scale appears to be feasible for assessing LTCRL among home care workers in Taiwan.
In our pilot test, we recruited the participants with purposive sampling, and most were willing to join the study. Only 1 of 120 participants declined to join the study (recruitment rate, 99.17%). Throughout the pilot test, none of the participants reported difficulties in reading the question items, and all participants completed the questionnaires within 15 minutes as expected. Based on the framework of Medical Research Council guidance, any problems that undermine the feasibility of the evaluation process should be addressed [29]. Previous studies have reported that participant recruitment, retention, acceptability, and compliance, as well as the expected sample size for predicting the study outcomes through pilot tests, need to interpreted carefully [30]. Although the present study showed it is feasible to assess LTCRL among home care workers in Taiwan, we suggest that future studies should be interpreted carefully, and we present our findings with caution in terms of the feasibility issues.
One of the roles of home care workers is being a personal trainer for elderly individuals who are involved in reablement [31]. Herein, the ability to absorb knowledge as well as apply it to the clinical setting is essential. We found that the Taiwanese home care workers were good at the access/obtain and understand categories but not in the later skillful and advanced competencies of process/appraise and apply/use categories. Furthermore, the education levels were related to the process/appraise and apply/use categories, which is consistent with previous studies in which people with higher education levels performed better on health-related literacy [18, 19, 21]. In addition, our findings shed light on the existing educational program for home care workers in Taiwan. The teaching strategies for the current in-service education in Taiwan are mostly lectures and case reports [16], which are more likely linked to basic literacy (access/obtain and understand). We recommend that the education programs further emphasize on scenario-based exercises related to advanced competencies of literacy. Thus, home care workers involved in reablement would have better skills in evaluating caring problems and in making optimal decisions to help the elderly re-learn or regain their ability to perform ADLs independently.
According to this study, the roles of education and self-rated general Chinese proficiency were important in LTCRL. Clearly, individuals with higher education levels would perform better and better utilize knowledge for practical use than those with lower education levels. This finding is consistent with results in the literature, which showed that education had a positive effect on health-related literacy [18, 19, 21]. Surprisingly, the working experience of home care workers was not significantly associated with literacy in this study, possibly because we measured the variable by the number of working years and not the status of employment (full-time, part-time, and unemployed), as in other studies [20]. Although not statistically significant in the present study, the gradual decrease in the LTCRL scores was notable. A qualitative study reported that LTC professionals with abundant experience of conventional home care services faced more challenges to this new approach of caring for the elderly [32]. Hence, we recommend that in-service education for home care workers in Taiwan should be more focused on reablement in the future.
This study had several limitations. First, the pilot assessment for the scale recruited participants with purposive sampling. Although we recruited participants from northern, central, and southern Taiwan, this is not representative of the overall home care worker population in Taiwan. Second, each category has an unbalanced number of questionnaire items. One of the expert panel members suggested the merging of the categories of access/obtain (5 items) and understand (4 items), resulting in a relatively equal number of items to the categories of process/appraise (11 items) and apply/use (9 items). However, considering the original conceptual framework for the literacy definition [17], we decided to maintain four categories in the study. Third, the Delphi method was used to obtain data from expert consensus, and it is considered a validation process. However, given the dichotomy response design for the scale and a highly correct response rate in the pilot test, the indicators of construct validity were not as good as the criteria used in testing construct validity through confirmatory factor analysis. We recommend that in future studies, developing an assessment scale through the Delphi method should also include testing the enhanced validity, such as with construct validity. Fourth, the insufficient number of participants enrolled in the pilot study could be a bias in testing reliability and construct validity for the scale items, and it is suggested to include 5–10 participants per question item for testing the scale. The sample size for testing the self-developed questionnaires should be considered carefully in the future. The limitations of the study were mostly qualitative in nature, and the Delphi method is generally recognized as an effective tool for determining expert consensus when there is limited or no definitive evidence and measurement tools have not been developed [23].