Background
Functional decline and frailty are common in community dwelling older adults [
1‐
3] and influence the risk of adverse outcomes. Identifying those likely to develop adverse outcomes is important in order to target limited healthcare resources in an efficient and effective manner. Risk assessment utilizing different risk prediction models is increasingly being used in the community [
4]. Risk describes the amount of potential harm that can occur in a set period of time due to a specific event, or series of events and is the product of the probability that harm will occur and the magnitude of its severity [
5‐
7]. Rational decision-making in healthcare requires reliable and valid quantitative ways of expressing risk that balance the potential costs and benefits of different management strategies [
8].
Multiple factors including cognitive impairment, depression, medical comorbidities, low levels of physical activity and social isolation are associated with an increased risk of adverse outcomes [
9‐
11]. Many but not all of these predispose to the development of frailty [
12]. These factors can be grouped into three main categories or domains: mental state, activities of daily living (ADLs) and medical state. The ability of each individual’s caregiver network and social supports to manage the person’s care deficit also affects their level of risk. Inadequate social or caregiver networks predict mortality and contribute to other poor healthcare outcomes [
13,
14], including institutionalization [
15,
16].
A variety of different methods have been used in an attempt to identify community dwellers at risk of adverse outcomes. Many focus on the identification of frailty [
17], acting as short surrogates for Comprehensive Geriatric Assessment [
18]. They include direct (home assessment) [
19] and indirect (postal survey) [
20‐
22], targeted and non-targeted assessment strategies [
23]. Indirect and non-targeted community screening is less efficient, suggesting that rapid screening, followed by triage and appropriate management of high-risk individuals is most effective [
23]. Although the stratification of risk scores using these instruments is associated with clinically meaningful gradients of adverse outcomes [
4], most risk prediction models have poor predictive ability [
4]. This is especially true at an individual level and may relate to a failure to incorporate important personalized social and demographic data [
4].
Community healthcare nurses known also as public health nurses (PHNs), visit patients in their home and may be in the best position to screen older people, both opportunistically and proactively. PHNs play a key role in all areas of healthcare delivery in the community, including assessment of care needs [
24] and can be trained to deliver specific interventions from psychosocial strategies [
25] to interventions for chronic medical conditions [
26] in the home environment. In some countries, people with chronic illnesses such as dementia are more likely to be attended by PHNs than other healthcare professionals [
27]. Studies have found high levels of frailty related risk factors among patients under PHN follow-up in the community [
28]. Despite this, few studies have examined the role of PHNs in the care of frail and functionally impaired community dwelling older adults. In particular, few studies consider the factors that influence PHNs’ decision-making or that contribute to their interpretation of risk.
The purpose of this study was first to establish the prevalence of risk factors for frailty and functional decline in a sample of community dwelling older adults monitored by PHNs in Ireland, second to identify factors associated with perceived risk of adverse outcomes and third to investigate their distribution according to the severity of that perceived risk using a new risk stratification model, the Risk Instrument for Screening in the Community (RISC).
Discussion
This paper presents the prevalence of several established predictors of functional decline and frailty among community dwelling older adults being monitored by their PHN. It also presents the type and prevalence of factors that contribute to PHNs’ perceived risk of adverse outcomes, scored using a new screening and assessment instrument, the RISC. There was a high prevalence (54%) of frailty, as defined by CFS scores of 5 to 9 (
mildly frail-terminally ill)[
34], among this cohort of community dwellers. This was expected, given that patients being followed by PHNs are a selected sample of older adults and are more likely to have medical and other co-morbidities than a cross sectional sample of all community dwelling older adults.
Several variables correlated strongly with frailty. These included function as determined by the BI and notably PHNs own “opinion” of frailty. The close correlation of the CFS to the BI likely reflects the scoring mechanism of the CFS, which depends on the functional stage of the patient. That the opinion of PHNs in this sample correlated strongly and significantly with a validated measure of frailty, the CFS [
34], suggests that healthcare workers, familiar with their patients, accurately predict risk without the need for standardized assessment instruments. Several established factors such as age [
36] and female gender [
37], that might increase the likelihood of frailty, were not found to correlate with the CFS. Of the other risks identified, CI in particular, correlated with the degree of functional impairment (BI). Age itself did not impact upon function or frailty. However, several challenges remain in separating frailty as a concept from the individual factors that are associated with it. It is not established whether these factors, including markers of cognition and functional impairment, cause or merely reflect the development of frailty. Future studies should include established objective markers, such as those included in the Fried criteria [
38] like weight, grip strength and walking speed, which may help clarify the interaction between the components of frailty and the frailty phenotype as a whole.
In this sample, only a small percentage of the total older adult population was perceived to be living at risk. Perceived risk of adverse outcomes, within one year of assessment was generally low. This corresponds with low prevalence rates of institutionalisation and death from the community. Previous analysis of a risk register of community dwelling older adults in County Cork, Ireland, found a similar prevalence of perceived risk [
39]. In that sample the composite risk of all adverse outcomes was measured at 7%. In this study, cognition, functional level and frailty correlated with the perceived risk of institutionalisation, although gender and social isolation did not. In particular, cognitive impairment increased the perceived risk of all adverse outcomes, suggesting that these patients should receive particular attention when they live in the community. Pooling results, into low, medium and high or indeed into minimum and maximum increased the number of patients perceived to be at increased risk. Perceived risks were significantly greater for those rated as frail on the CFS compared with those rated as non-frail or frail. Perceived risk of hospitalisation was higher than for the other adverse outcomes. Most patients however, even in this highly selected community sample, were regarded as being low risk of adverse events.
This paper has several limitations. The data collection was based upon a retrospective review of the patients’ PHN records and the analysis depended upon these being accurate and up to date. Some demographic data including 21 RISC and 19 CFS scores were not available for patients that had not been reviewed within the last six months. Sampling PHN records may also have led to selection bias in that patients followed by their PHN are inherently at higher risk of adverse outcomes compared to the general older adult population. The method of sampling may also have created bias. However, with quota sampling investigators are less concerned with having sufficient numbers to match the proportions in the entire population, but instead aim to sample enough patients to ensure that even small subgroups are adequately represented. Additionally, it is not certain that all PHNs correctly classified patients according to their risk. The study was however, conducted in conjunction with each patient’s PHN, who know their patients well and provide care for them over several years. Another limitation is that since the RISC is still being validated, the ability to generalize and prognosticate on the significance of the risk factors associated with the perceived risk identified are reduced. Once validated, further analysis will be required to investigate the association of these markers of frailty and perceived risk with the outcome data. Likewise, the optimal cut-off point for each adverse outcome is not yet established. The CFS has also not been validated with nurses, which may have lead to bias. Although we assessed the inter-rater reliability of the RISC, the reliability of the CFS was not determined. The subjectivity of the frailty assessment is another limitation. Both the PHNs own assessment of frailty and the CFS are subjective measures, and the inclusion of an objective observer rated assessment instrument would have reduced potential bias. Furthermore, the screening tools used for cognition and function in this study are not gold standard instruments and may have under-estimated the true prevalence of cognitive and functional impairment in this population. The AMTS is less accurate than many other short cognitive screens such the Quick mild cognitive impairment screen [
40] and the Montreal Cognitive Assessment [
41] and is particularly insensitive at differentiating mild cognitive impairment from normal cognition and early dementia. Likewise, the BI is a crude gauge of function and does not score instrumental ADLs. These instruments are however, widely used and are the prescribed instruments in use in the community in this region. The Charleston Co-morbidity Index is criticized for its poor predictive validity, particularly among older adults [
42]. In particular, it fails to incorporate medical conditions like Parkinson’s disease, multiple sclerosis and inflammatory bowel disorders, which may contribute to comorbidity.
The strengths of this paper are the comprehensive nature of the review of the PHN records, conducted in busy health centres and the inclusion of a large cross-sectional and representative community sample, increasing the generalizability of the results. The prevalence of comorbidities reported in this study are similar to other community samples of other PHNs in Ireland [
28]. A retrospective cross sectional clinical audit carried out in Dublin City, Ireland, investigating the prevalence of four frailty-related risk factors identified similar rates of suspected cognitive (16.4%) and functional impairment (BI score ≤15, 23.5%) albeit they were slightly higher in this study at 16.7% and 30.6% respectively [
28].
The RISC was developed to measure risk patients' levels and is in the process of being validated. It was designed with PHNs for use in a community setting and shows excellent inter-rater reliability (IRR) [
31]. This research is ongoing, and a follow-up is underway to ascertain the current status of the patients and evaluate the predictive validity of the RISC to these adverse outcomes in this population. This prospective cohort study will investigate if risk scores, described by the RISC, predict these adverse outcomes. Targeting limited resources, to medium and high-risk individuals (maximum risk), could improve efficiency in the use of limited healthcare resources [
43]. Future studies will investigate if the RISC, aligned to tailored intervention programmes or care bundles, can reduce risk and incidence of adverse outcomes in community dwelling older adults.
Conclusions
In this study the majority of community dwelling older adults were perceived by their PHN to be at minimum risk of adverse outcomes. This may facilitate targeting of these patients to prevent or postpone adverse outcomes. On the other hand, there was a large proportion of frail older adults (54%; ≥5 on the CFS) found in this study, and it remains to be seen if simple, albeit multidimensional, risk scores like the RISC may be more efficient in targeting patients for Comprehensive Geriatric Assessment than instruments like the CFS. Frailty (subjective or objective), cognitive impairment and functional status were markers of perceived risk. PHNs opinion appears to correlate with a validated frailty scale, supporting the utility of nurses in triaging patients in the community. Several factors traditionally associated with frailty such as age, gender and social isolation did not correlate with the CFS, RISC or with PHNs’ perceived risk of adverse outcomes, suggesting that despite their high prevalence, they may not be useful indicators for triaging community dwellers.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
Study concept and design: RO’C, GO’K, DWM. Subject information data (clinical and demographics): EH, EO’C and GO’K. Acquisition of data: EH, EO’C and RO’C. Data analysis and interpretation: YG and RO’C. Drafting of the manuscript: RO’C, DWM, UC, EO’H. AS provided statistical support and review. Editing and reviewing the final manuscript: RO’C, DWM, NC, EO’H. All authors read and approved the final manuscript.