In this study with harmonized cohorts, internet use rates ranged from 5.56% in China, 39.37% in Mexico to 58.01% in the United States, 72.35% in England, and 83.46% in Denmark. We found developed countries are more digitalized than developing countries. Besides, middle-aged and older participants with internet use reported a lower prevalence rate of frailty compared to those excluded from digital in both developed (HRS, SHARE, ELSA) and developing countries (CHARLS, MHAS).
To our knowledge, this study was the first to explore the longitudinal association between internet use and frailty status in middle-aged and old adults. Previous research has indicated that internet use was inversely associated with frailty in postmenopausal women.[24] However, small sample size and cross-section design might limit the reliability and robustness of the finding. In addition, more studies focused on whether frailty status influenced internet use and reported inclusive conclusions, while it was not the concern of our study.[25–27] With the rapidly growing proportion of the population engaging in the internet over the past decade, researchers in the field of digital technology gradually explored the potential health effects of the internet.[28] Nevertheless, the rate of internet use in older adults remained relatively low, especially in developing countries such as China and Mexico. With technology advancing and population aging, the trend of the internet as a platform for providing health services and monitoring poor lifestyles was unavoidable in worldwide. A scoping review supported digital health interventions among people would be beneficial for the frailty status, although the initial purposes were various.[29] Population with frailty considered internet access maintained social connection with friends and relatives and engagement in their daily lives, which reduced social isolation and loneliness.[30] Utilizing the available multi-country population-based cohorts, we could investigate whether the effect of internet use on frailty differed in the context of developed and developing countries.
By including five longitudinal cohorts from 32 countries, internet use was discovered to be inversely associated with frailty in middle-aged and older adults. Previous studies found internet use had a protective effect on multiple chronic diseases (stroke, hypertension, eyesight, etc.), self-reported healthy status, functional limitations, depression, and cognition.[6, 9–12, 14, 31–33] However, few studies have combined those health deficits to evaluate the association of internet use and frailty. The present results indicated the protective effect was more pronounced in developed countries except for China. One possible explanation was that China experienced rapid economic development and increasing technological levels in 2011–2018. With the coverage of the internet, online health services such as purchasing medicines and asking for medical advice provided a chance to maintain self-health status in middle-aged and older adults.[34] Meanwhile, internet access as the common equipment is quietly changing the traditional channels of obtaining medical service and work information. We also found the most notable association was reported in participants from SHARE and the highest rate of internet use in Denmark. Compared to developing countries, more comprehensive health management systems and enriched online medical knowledge were implemented in developed countries. The findings suggested the requirements of enhancing internet access and refining online services to prevent the incidence of frailty, especially in developing countries. Moreover, bridging the digital divide in an aging society would reduce the health disparity in frailty status between developed and developing countries.
We further explored the potential mechanisms to explain the association of internet use and frailty status at population level. The internet as a virtual world played a role in providing the available resources to bring people closer together. Active internet users effectively maintained connections with friends and expanded social networks,[35] which could reduce loneliness and depression, especially in older adults.[36, 37] At the same time, loneliness and depression would be confirmed to accelerate the progression of frailty based on the population from ELSA.[38] Previous studies reported that the internet facilitated participation of social activities,[39, 40] which strengthened the social interaction and slowed the social disengagement.[41] Additionally, social activities, especially a high-frequency one, significantly decreased the frailty among middle-aged and older adults from CHARLS.[17] According to the mediation analysis in our study, the negative association of internet use on frailty was partially mediated by social activities. The internet also provided opportunities for online social engagement and support in special population with disability or function limitations. Besides, the attitude and purpose of using the internet also interpreted the current health effects on the prevention of frailty. Older adults expressed a positive mindset towards the internet infrastructure as an instrument of achieving convenience and relaxation.[42] Middle-aged and older adults tended to search for diet, exercise, online medical information, and remote health services because of the concern about their health.[43] For instance, short health videos on social media platforms were extremely popular among Chinese aging women.[44] Handy and unrestricted internet access encouraged people to recognize healthcare knowledge regardless of time and place, which would be an inexpensive way to improve the life of quality, particularly among population with health-related difficulties.[45] On the other hand, middle-aged and older adults expected to modify the unhealthy lifestyles associated with frailty through internet.[46] Regular internet users were more inclined to adopt healthy lifestyles including vegetable and fruit eating, not smoking, and not binge drinking from the unhealthy. Notably, to prevent the progression of frailty in this way needed a powerful self-management capability.
Although the present findings implicate that internet use was beneficial for slowing down the progression of frailty, what cannot be ignored is the situation of problematic internet use and internet addiction might spread among older people. Previous studies also supported that excessive time spent on the internet had a high risk of depressive symptoms.[47] In addition, there was a large proportion population with lower educational attainment in developing countries, who contributed to less help from the internet. In particular, our study has shown that the low educational level and high rate of internet use existed simultaneously in Mexico, which explained the protective effect of internet use was minimal compared to other cohorts to some extent. Under-educated people might be excluded from healthy information websites because a high school level or greater reading ability was required.[48] In addition, based on the population from CHARLS, the low-literacy older adults were less likely to search for health services.[49] So, relevant strategies on internet education and health literacy promotion to encourage effective internet access should be underscored in developed and developing countries.
In summary, our findings have some public health implications and practical value. First, internet had the potential to provide various opportunities for preventing frailty in middle-aged and older adults. Relevant workers in health field should promote internet penetration and provide more easy-to-read health information, especially for under-educated older adults. Moreover, social activities played a mediated role between internet use and frailty status, which mentioned volunteer activities and gatherings should be encouraged and publicized for middle-aged and older adults. Maintaining social connection and communication by the internet might be a crucial pattern, which increased the degree of participation in various outdoor activities and decreased loneliness. Well-established evidence showed that digital interventions have successfully emerged as a vital part of healthcare service, which facilitated the utilization of medical resources. For instance, a systematic review of randomized controlled studies demonstrated that the behavioral intervention using mobile health applications to manage chronic diseases had promising aspects.[50] Another recent review[51] revealed that e-health and m-health technologies increase physical activities, which could improve the frailty status.[52] Therefore, internet access among middle-aged and older adults would be helpful to build a healthy aging society.
Our study had some strengths. We included five national representative and longitudinal cohorts from developed and developing countries with large sample sizes. The five surveys were cross-cultural comparable at the beginning of the design. Besides, sufficient items from multiple repeated measurements to construct the frailty index and diverse sensitivity analyses enable the reliability and robustness of our results. Finally, The GEE models could effectively deal with the correlations from longitudinal panel data to reduce the error of the results.
Several limitations should be also aware in this study. First, recall biases were unavoidable because most variables selected for our study were determined based on self-reported information. Second, heterogeneity among HRS, CHARLS, SHARE, ELSA, and MHAS remained due to the different evaluations of exposure and outcome variable, especially between MHAS and other cohorts. Third, the dimensions of internet use were not considered as the lack of data availability. Fourth, unmeasured confounding including sleep and genetic susceptibility remained while the E-value was calculated. Fifth, we could not avoid the likelihood of opposite association between internet use and frailty, which might be statistically significant according to the previous studies.[25–27] Further research should be implemented to confirm the benefit of internet use based on other populations and determine the direction of causal association between internet use and frailty status. Additionally, the dynamic change in internet use and frailty status should be also explored, which contributed to healthy policy development.