A study on the quality evaluation index system of smart home care for older adults in the community ——based on Delphi and AHP

The current study focuses on the unique characteristics of “wisdom” of smart elderly care to construct evaluation indexes for the quality of smart home care for older adults in the community. To ensure that the complexity and multi-dimensionality of the service quality of smart senior care are correctly reflected, this study redefined the service quality model for the five categories of smart elderly care. Similar to the scale developed by Bo Y [21], we created five new dimensions, including tangibility, reliability, timeliness, empathy and ease of use. The study innovatively adds the “ease of use” dimension, which fully reflects the “elderly-centred” approach and take into account the reality of technology anxiety among the elderly in the context of smart care. Several studies have found that one of the primary reasons for limited use of smart senior care is the lack of simplicity with which smart equipment and systems can be learned and operated [20, 22]. Moreover, privacy and security are also key factors when it comes to accepting smart senior care products and services [23‐25]. Therefore, we proposed incorporating the aspect of privacy and security into the “reliability” dimension. Additionally, we redefined the dimension of “responsiveness” into “timeliness” to stress both responsiveness to elderly users’ needs and timeliness of the service delivery. Herein, the five dimensions and definitions of the service quality model in this study are as follows:

The Delphi method was then used to screen evaluation indicators. The results of related studies have concluded that the number of consultation experts should be 15 ~ 50 [26]. Accordingly, a total of 15 experts were invited from universities, research institutes, senior care units, associations and hospitals to participate in the construction of the quality evaluation index system of smart home care services for older adults in the community. The selected experts not only have profound knowledge of the health professions, but also have experience in the elderly care industry, thus assuring the reliability of the consultation findings.

The inquiry questionnaire for this study was self-designed and consisted of four parts: basic information of the experts, indicator evaluation questionnaire in accordance with the initial index system, experts’ opinion on the degree of authority, and experts’ judgement basis. Each item was rated on a Likert 5-point rating scale: not important, somewhat important, general, quite important, and very important [27]. The expert group rated on a scale of 1–5 according to its importance and provided suggestions. Microsoft Excel 2010 software was used to create a database, and SPSS software was used to conduct the analysis. The full score rate, importance mean, standard deviation, and coefficient of variation of each indicator were calculated. The criteria for the change of the indexes in the experts’ correspondence were: any one of the perfect score rates < 30%, mean < 4.0, and coefficient of variation > 0.4 in the 1st round; any two of the perfect score rates < 30%, mean < 4.0, and coefficient of variation > 0.25 in the 2nd round were satisfied at the same time. In addition, if there were two or more experts with the same opinion or suggestion, the opinion and suggestion made by one expert were adopted and eventually adjusted and amended after group discussion by combining the index structure, content, and expert background.

The index weight is a vital parameter of the quality evaluation index of intelligent elderly care. It reflects the relative importance of the index and has a direct impact on whether the result of the evaluation system is scientifically accurate. The service quality of smart home care for older adults in the community is a complex, multidimensional concept that is perceived in many ways. AHP is an effective method to convert experts’ subjective attitudes of different evaluation factors into quantitative weights, which is suitable for problems that need to be decomposed in a complex way. Hence, we employed the analytic hierarchy process to determine the weights of the index system in this study. The calculation process of the AHP method was illustrated as Fig. 1.

Five main themes were identified in the analysis of the interview data by thematic analysis: (1) catering services, (2) domestic services, (3) medical care services, (4) emergency services and (5) amusement services. As a result, we have concluded five major types of smart services, including smart meal assistance, smart cleaning assistance, smart medical care assistance, smart emergency assistance and smart amusement assistance. The specific service item classifications are shown in Table 3.

Two rounds of expert consultations were conducted in this study. The basic information concerning the experts is shown in Table 4. The participation rate of experts in both rounds was 100%, indicating a high motivation of experts to participate. The effective rate of both rounds of questionnaires was 100%. The authority coefficients of the experts were 0.883 and 0.867 respectively, both of which exceeded 0.7, showing that the experts had a high degree of authority [31]. Thus, the results are reliable and persuasive. In the first round of consultation, the mean importance of each index item was between 4.13–4.87, and the overall coefficient of variation was between 0.072–0.202. In this round, we revised four indicators based on expert opinion and group discussion. In the second round, the mean importance of each index item was between 3.87–4.87, and the overall coefficient of variation was between 0.072–0.236. In this round, additionally, 3 items with a mean importance < 4 were retained after discussion.

After the two rounds of expert consultation, the coordination coefficients of the indices increased from 0.137 and 0.233, and the significance test of the coordination coefficients was P<0.01, which indicated that the experts’ opinions converged. Thus, all indices were included in the evaluation system based on exclusion criteria. The confirmed service quality evaluation index system included five criterion layer index and 33 indicator layer index. All these results are shown in Table 5 and Fig. 3.

The weight coefficient of the whole quality evaluation index was built at all levels employing the analytic hierarchy process method. The weights of the 5 primary indices, from high to low were smart emergency assistance (0.332), smart meal assistance (0.272), smart medical care assistance (0.229), smart cleaning assistance (0.110) and smart amusement assistance (0.057). The weights of each level index are listed in Table 6.

In sum, the quality evaluation index system for smart home care services for older adults in the community identified in this study has the following characteristics. First, the index system better reflects the connotation of smart care for the elderly, and highlights the characteristics and advantages of “wisdom” in the process of aging service provision. Secondly, the evaluation indices are more optimized and have stronger realism. In the process of index development, we also took into account the “fear of technology” among the elderly and their concern about the leakage of personal privacy in the process of smart senior care development in the era of big data. Rooted in China’s national conditions, the final secondary indicators are directly measurable indicators based on the actual situation in Guangzhou and Shenzhen, which can be used to intuitively measure the evaluation of elderly users on the quality of smart senior care services.

From the criteria level, the weighted value of the five established first-level indices is shown in Table 6: smart emergency assistance (0.332), smart meal assistance (0.272) and smart care assistance (0.229). This, indicates that these three indices are the key parts used to measure the quality of community smart home care, whereas the weight values of smart cleaning assistance and smart amusement assistance are relatively lower, namely, 0.110 and 0.057 respectively. The results from the current study suggest that when constructing the smart senior care service system, the development, design, popularization and service quality improvement of equipment related to emergency assistance, meal assistance and medical care assistance must be prioritized. Actually, through wearable devices and intelligent monitoring service devices, intelligent emergency assistance and intelligent medical assistance can realize the functions of positioning, monitoring and assisting elderly individuals. This can help the elderly achieve independence and effectively improve the quality of intelligent elderly care and help the development of smart elderly care. Smart assistance for emergency and smart assistance for medical care was affirmed by foreign researchers earlier. They have also advocated the expansion of the coverage of information technology in home care services [32] and fully exploit smart devices and technologies such as remote intelligent control to respond quickly to remote calls for assistance from elderly individuals. This will lower the cost of medical care integration, and alleviate the shortage of care services [33]. Similarly, a study has also proposed that at present, the fields with a high embedding degree of Internet technology largely focus on delivering meals, life care and medical services. However, due to the inherent deficiency of Internet technology in providing spiritual comfort [34], the embedding degree of Internet technology is limited, and it is difficult to fully rely on relevant devices or terminals. They can only play a supplementary role [35]. It is noteworthy that, as the development of smart elderly related industries is still in the preliminary stage, the content of smart elderly services still has limitations and lacks uniformity. In the future, with the continued growth of smart senior care, the weight of smart amusement assistance has more room to rise, which is also in harmony with the idea of some foreign scholars. For instance, Godfrey [36] suggested that science and technology should be integrated into the traditional elderly care model to innovate the social interaction of the elderly through information technology. Similarly, Khaksar [10] argued that nursing homes should use social robots to help older adults overcome social barriers and innovate smart senior care services from theory to practice.

From the program level, the indicators with higher weight values are mostly focused on the three dimensions of timeliness, reliability and ease of use. In other words, efforts must be made in the following areas during the construction of smart home care services for older adults in the community: guarantee timely response to user needs and timely provision of services; secure satisfactory services provided by professionals [37], enhance the ability of the platform to make adjustments based on user feedback and to stress confidentiality of user information [24, 38, 39]; ensure that the design of smart devices and systems is age-friendly and eliminate the digital divide as much as possible [37]. Additionally, the finding from the weight values of the secondary index, it shows that the weights of the indicators under the “empathy” dimension is generally the lowest. The major reason for this is that the Internet has not yet been deeply integrated into the senior care industry in China. This, makes it difficult to provide personalized and customized services of different types and pricing in accordance with the situation, needs and preferences of elderly users.

The combination weights are able to reflect the importance of the secondary indicators for all service components. In the combination weights, there are 15 indicators that exceed the average weights. The high indicator weights are largely distributed under the three dimensions of timeliness, reliability and ease of use of each service content. This is basically consistent with the results of the weights of the primary and secondary indicators, also implying that these three dimensions have a significant status and role in the evaluation of smart home care services for older adults in the community. Yang Bo in China [21] first tried to construct a scale for the quality of smart home care services, and concluded that the five service quality dimensions in descending order of weight were: reliability>empathy>timeliness>ease of use>tangibility. Our study concludes that timeliness, reliability, and ease of use are more important, which is consistent with previous findings to some extent [21]. However, studies have produced a conflicting result in regard to the importance of “empathy” [21]. The inconsistent findings could be caused by the heterogeneity of perspectives and cultural backgrounds of those interviewed and the consultation participants. Therefore, further studies are needed to confirm the importance of the different dimensions.

This study has established a system for evaluating the service quality of intelligent home care services for seniors in the community. To begin with, this index system can be used to evaluate the current situation of smart home senior care service quality, summarize the existing problems, explore the key factors and effective paths to achieve high quality. This index system, furthermore, may also be used as a regulatory benchmark to guide the government, senior care enterprises and communities to take measures to advance the quality. However, there are still some shortcomings in the study. First, the reliability, differentiation and representativeness of the indicators in the index system need to be further tested by specific applications. Accordingly, in the future, we will use this index system to conduct field research among elderly users, so that the evaluation index system can be continuously improved. Furthermore, since the constructed evaluation index system is based on the specific practices in Guangdong Province, the external probability validity of the service evaluation index system constructed in this study may have certain limitations. Thus, further in-depth research is needed to expand its empirical scope. Next, we will use this index system to conduct field research among the users of smart home care services, so that the evaluation index system can be continuously improved.