Understanding functional and social risk characteristics of frail older adults: a cross-sectional survey study

Adults aged 65 and older are expected to comprise 21.7% of the U.S. population in 2040, up from 14.5% in 2014 [1]. With the exponential growth of the aging population, frailty has become a syndrome of increasing public health concern due to the projected impact on utilization of healthcare resources and age-related services [2]. Frailty in older adults is a clinical syndrome described as a loss of “physiologic reserve” from an accumulation of physical, environmental and psychosocial factors [3‐5]. These factors are often referred to as deficits that increase vulnerability to undesirable health outcomes [6].

There is currently no clear consensus on how to define and assess frailty. Some studies have defined frailty by using phenotypic characteristics such as weakness, unintentional weight loss, slow walk speed, exhaustion, and low physical activity [4], whereas other studies have identified a series of health-related deficits to calculate a Frailty Index (FI) and determine degree of frailty [7, 8]. To qualify as a deficit, the prevalence of the deficit must increase but not become universal with age and must include more than one organ system [8, 9]. Despite the differences in the operationalization of frailty, increasing age, female sex, lower educational attainment levels, and adverse health outcomes including falls, disability, hospitalizations, and mortality have been consistently associated with frailty in older adults [4, 6, 10].

Frailty can also be conceived as a dynamic process in which older adults transition between different states of frailty over time [11]. By screening for modifiable risk factors, interventions can potentially alter frailty status, prevent adverse health outcomes, and improve quality of life [12]. Recognizing these modifiable and even non-modifiable risk factors among community-dwelling frail older adults continues to remain a topic of interest and study.

Research suggests that approximately 30% of the 65 and older patient population have health problems complex enough to require care by a geriatrician [13]. However, a recent report by the Health Services Resource Administration in the U.S. Department of Health and Human Services projects that, based on this need, there will be a national shortage of nearly 27,000 geriatricians in 2025 [14]. The implication of this shortage is that adult primary care clinicians will increasingly be called on to manage care for older adults with complex needs such as frailty, with fewer geriatricians available for consultation. Effective care of this population will require the development of geriatric health risk assessment tools and care management protocols to allow primary care clinicians to more effectively identify and intervene on social-economic and health issues in ways that improve quality of life and prevent adverse health outcomes.

In this study, we attempt to identify potentially modifiable health and social-economic risks in the older adult patient population which, if appropriately addressed, may reduce or slow the pace of health and functional decline and improve quality of life among those with frailty or complex needs. This study aims to encourage further assessments and interventions in the broader older adult population that might help reduce the prevalence of frailty in the aging population.

In this large cross-sectional study of community-dwelling adults ages 65 to 90 years who were members of a Northern California health plan with an integrated healthcare delivery system model, we examined the prevalence of frailty and identified potentially modifiable and non-modifiable health risks and social-economic determinants of health that healthcare providers can utilize to intervene and request additional geriatrician support. Using a deficits accumulation model created in line with previous guidelines, we found an overall prevalence of frailty of 14.3% with frailty increasing for both men and women starting in the 75–79 year-old group and trending toward greater prevalence among women as compared with men at higher ages. Consistent with prior studies, frailty was also more common among older adults with low levels of education and income [4, 7‐9]. In our study cohort, the prevalence of frailty did not differ by race/ethnicity.

The prevalence of frailty described by prior studies varied widely, ranging from 5 to 58% in one review article [18] to 4% to 59.1% in another review article [10]. This wide range highlights the lack of consensus on how to define and operationalize frailty. The prevalence of frailty calculated in this study population was lower than the 22.7% calculated in a previous study that also used a deficits accumulation model [9]. Factors that may have contributed to the lower prevalence in our study included exclusion of participants over 90 years of age and the fact that our findings were based on self-reported data obtained from a survey of community-dwelling older adults, thus excluding adults with dementia or in skilled nursing facilities who are more likely to be frail [19]. Additionally, care management programs for patients with diabetes mellitus, congestive heart failure and asthma, as well as cardiovascular risk-reduction programs using cardioprotective medications, may have resulted in an overall improved health status of KPNC senior members [20]. The significant increase in prevalence of frailty starting at age 75 reflects findings from other studies demonstrating that the burden of frailty is more prominent among the older aging population [21]. Prior studies have shown that older adults are not a homogenous group [22, 23], with the frail group contributing to higher medical costs even after controlling for comorbidities [24]. This highlights the need to create patient-centered screening and interventions for frail adults and older adults at elevated risk of becoming frail.

With the shortage of geriatricians, care management, including screening and interventions, for frail older adults will increasingly become the responsibility of adult primary care physicians (PCPs). Although there is no gold standard for frailty measurement in the primary care setting, prior studies have shown screening tools useful to direct appropriate interventions [25, 26]. As PCPs already screen for a multitude of chronic conditions within a limited amount of time [27], efforts should be utilized to improve screening tools with an additional focus on training ancillary and administrative staff members to use these tools to target the appropriate older adult populations. A single screening question would help alleviate some of these constraints. Our findings support the need for future studies to determine whether a patient’s perception of how well they can take care of themselves identifies older adults with very low, moderate, or very high likelihood of being frail. The latter two groups, which in our study comprised about half the cohort, could then be further assessed for specific frailty characteristics and risks. Most health-related risks found in this study have potential interventions. About two-thirds of our frail group had frequent or ongoing pain, which likely interfered with their daily activities and participation in exercise. Prior studies have shown physical training, including resistance training, may provide some benefit in preventing frailty, but that exercise alone may not improve functioning [28, 29]. A multidisciplinary approach including behavioral changes, nutritional interventions, and cognition exercises coupled with exercise could be the interventions needed to prevent further development of frailty [29]. Over a third of our frail individuals had problems with insomnia (falling or staying asleep), about 11% are “short sleepers” (≤5 h/day), and about 15% are “long sleepers” (> 9 h/day). Prior studies have shown associations with worsening sleep quality and increasing frailty [30], with the primary causes being treatable conditions such as insomnia and sleep apnea [31, 32]. Insomnia and “short sleep” have also been shown to increase risk of falls among community-dwelling older adults [33‐35], and “short sleep” with an increased risk of cognitive impairment [36]. As expected, our frail group had a higher prevalence of falls compared to the non-frail group, but also a higher prevalence of use of medications that increase the risk of falling. Frail patients who are chronically using these types of medications or having problems for which these medications might be prescribed could potentially be tried on non-pharmacological approaches to managing insomnia, pain, depression, and anxiety, such as cognitive behavioral therapy, mindfulness meditation, yoga or gentle movement therapies, and exercise.

Dentition and other oral health problems should be a priority for frail older adults because these can lead to poor oral intake and involuntary weight loss [37], as well as systemic inflammation from gum disease that can increase risk of cardiovascular disease [38]. We found that over 30% of frail adults in our cohort had not had a dental check-up in the past 12 months, suggesting that oral health problems are likely to go undetected unless an examination of the mouth, teeth, and gums becomes part of the annual primary care physical examination for people who do not report having seen a dentist within the prior year.

Counterintuitive to the thought that frailty is a wasting disorder [4] but consistent with prior studies [39], we found a higher prevalence of obesity among the frail group, which could be explained by the obese population having limited mobility (or conversely that limited mobility lead to an obese state), requiring more assistance with ADLs/IADLs and having more chronic conditions [40], all factors included in our FI. Many frail individuals were also found to require more assistance with ADLs/IADLs including household chores, grocery shopping, preparing meals, and doing their laundry, which may highlight the need for continual guidance in regard to community resources and commercial programs that provide these services rather than depending on family and friends alone.

The social determinants of health analyzed in this study may or may not be modifiable among the frail population but are important to take into account in care planning. In line with prior studies [4, 6, 8, 9], the prevalence of frailty was highest among lower education levels, which likely represents a lower health literacy that would contribute to the poor lifestyle choices seen in frailty [41]. Financial strain and lower income level can impact seeking of needed medical and dental care, adherence to medical and medication regimens, and engagement in healthy self-care practices such as eating ≥5 servings of fruits and vegetables per day. Community services, such as home meal delivery programs to address the need for healthier diets, are available in certain areas, though the limitations of cost, availability of the programs, and sparse availability of research regarding outcomes with these programs still remain a concern [42]. Prior studies have also shown that hearing problems and frailty are associated with social isolation and depression [43]. As such, PCPs will increasingly need to involve instrumental social support systems of the patients including unpaid caregivers, social workers, or patient navigators to help provide necessary resources for the patients and families. Interestingly, nearly three-fourths of male seniors who are frail would depend on their spouses for help if they got sick or injured or needed special care. Assuming that the spouses are of similar age, that puts a great caregiving burden on female spouses who may also be dealing with limitations of their own.

This study has some limitations, including the cross-sectional study design and exclusion of non-English speaking members (the MHS questionnaire is only available in English), those living in nursing care or other institutional facilities, and those with cognitive deficits or severe dementia. Other studies have shown that ethnic minorities including Hispanic populations, and individuals in residential settings have a higher prevalence of frailty [22, 44]. Furthermore, this study was unable to validate the frailty index using participants' health records, but did follow the guidelines suggested by Searle et al. [8] for establishing a frailty index and chose deficits based on prior validated frailty indices including the Canadian Study of Health and Aging 70-item Frailty Index [6] and the Frailty Indices used by Searle et al. [8] and Song et al. [9]. There is also possible incorporation bias for the single-item screening question given that mobility is a component of the frailty index; however, future studies can compare this single question to other frailty indices for further examination. One of the strengths of this study is the large number of culturally diverse participants, which provides a more stable estimate of the prevalence and characterization of the frail subgroup. The frail cohort found in this study had similar demographic characteristics to all survey respondents aged 65–90 in regarding to mean age, sex, relationship status, level of education, income status, number of chronic medical conditions, self-perception of physical health, and history of falls in the past 12 months. Importantly, this study characterizes modifiable and non-modifiable risk factors in a contemporary and community-dwelling population of older adults, where those at highest risk for adverse health outcomes and subsequent utilization of healthcare resources can be further identified in clinical practice.

The frail older adult population is heterogeneous and thus frail patients and those at risk for becoming frail require a patient-centered assessment of their circumstances by primary care providers to try to improve their quality of life, avoid adverse health events such as falls, and slow physical and mental decline. The same set of characteristics identified in this study can proactively be used to assess all older adults to improve primary prevention of frailty and patient health and quality of life.