healthmedicinet

In this representative study of Canadian adults 18–79 years old, the overall prevalence of frailty was 6.6% using a modified Fried model and 7.6% using a FI. Frailty was more common in older individuals, but the prevalence in each age category differed depending on the model used [11, 12]. When the frail only participants were examined, the prevalence of individual frailty criteria differed between the younger and older age groups (Tables 3 and 4). These findings suggest that frailty can be prevalent at any age and may present differently in younger versus older adults.

The findings of previous studies using older cohorts demonstrate that the prevalence of frailty is higher with the FI versus the Fried criteria are further supported by the present study [4, 25–27]. Similarly, in a systematic review, the prevalence of frailty was 12% using the Fried model and 24% using the Accumulation of Deficits Model in community-dwelling adults 65 years or older [4]. In addition, our findings are similar to those in the Canadian National Population Health Survey which showed an increase in the prevalence of frailty from 2 to 22% in those aged younger than 30 and 65 years or older, using an FI [7]. Our findings indicate that frailty generally was not significantly different between males and females (except for the Fried frailty model amongst the entire cohort), which contrast previous studies [28–31]. It is possible that included health deficits, which captured mostly chronic conditions and laboratory values, may underestimate frailty in women. For example, women tend to have higher rates of depression and anxiety [32], which were excluded in the FI in the present study. Furthermore, some of the chronic conditions within the FI tend to develop later in life in women (e.g., cardiovascular diseases) – given our relatively young sample, this could be another possible explanation for the mostly non-significant findings in frailty prevalence between men and women.

The agreement between the two frailty measures in our study was low, which is common when comparing frailty scales. In fact, eight frailty scales were compared in the Survey of Health, Ageing and Retirement in Europe cohort, which included the FI and Fried criteria, showed that all the scales categorized less than 3% of participants as frail [27]. Although there was a higher agreement between the FI and the Fried criteria in that study compared to the present study, our inclusion of a younger cohort compared to the aforementioned study could account for this difference. There is also a possibility that the two frailty measures used in this study could be capturing groups of individuals who may or may not be vulnerable to poor health outcomes, which needs to be explored in further studies.

The evidence generated from our study indicates that it is feasible to measure frailty in younger adults. Further research is needed to explore the feasibility and value of frailty screening for adults of all ages [33]. However, given the paucity of evidence describing the prevalence and health impact of frailty in younger adults, it is still too early to recommend frailty screening in this age group. Possible benefits to frailty screening, at least in the older population, is providing additional risk assessment for those requiring invasive procedures. For example, frailty is shown to increase one’s risk for postoperative cardiac-surgical outcomes [3]; identifying someone who is frail could lead to more conservative approaches (e.g., trans-catheter aortic valve replacement) that could help maximize a frail older adult’s quality of life. On the other hand, frailty screening could lead to incorrectly identifying someone as frail who is not and might result in withholding beneficial treatments in favor of more conservative approaches.

Considering that our data shows that younger adults (18+) can be frail, it warrants future investigation to provide clarity as to whether frailty similarly impacts poor health outcomes in the young versus the old. More specifically, further research is needed to determine the prevalence of frailty, its potential health impact in the young, and whether frailty in younger age groups is associated with the use of more healthcare resources, as compared to their non-frail peers. While both the FI and Fried criteria are predictive of mortality in older adults [4], our study warrants that studies should be conducted to determine the health impact of frailty in younger age cohorts, and which tool could be used to most accurately assess their future health risk. Evidence suggests that the FI has better prognostic ability for predicting mortality in older adults than the Fried criteria in the short to medium term [27, 34], but to our knowledge, there has been no comparison of their prognosis in younger adults. This nascent evidence must be strengthened, with further investigation needed in younger population-based cohorts.

In order to determine the health impact of frailty in younger populations, proponents of research in frailty should consider linking CHMS data with administrative health databases. Furthermore, while the FI is suggested to provide better prognostication than the Fried model in older adults [34], it is unclear if the FI or Fried criteria (or other frailty tools) have better predictive validity for determining adverse outcomes in younger adults. Our data also show that the Fried model estimated a higher frailty prevalence in the 18–34 age group (5.3%) compared to the FI (1.8%), which needs further exploration. Among the individual Fried frailty criteria, it is interesting that approximately half of the younger cohorts scored positive on exhaustion and unintentional weight loss which suggests that targets to prevent or treat frailty might differ compared to the older age groups with an increasing prevalence of impaired mobility. Collectively, the evidence base of frailty in younger age groups must be strengthened.