This population-based study of older women highlights the new finding that the probability of frailty increased in a dose-response fashion as the number of chronic diseases that are associated with inflammation increased, accounting for age, race, education, and smoking. Many of the most frequent inflammatory-related disease combinations in frail older adults were also considerably less prevalent in non-frail older adults. Altogether, these results suggest not only that a higher inflammatory-related disease count is a risk factor for frailty, but also that specific disease combinations, possibly through their shared mechanistic pathways, further contribute to the onset of frailty. It is known from prior large cohort studies that multi-morbidity commonly co-exists with frailty (Barzilay et al., 2007
; Fried L.P. et al., 2004
; Woods et al., 2005
). However, these observations have been based primarily upon prevalence percentages and do not examine the extent that multi-morbidity is related to frailty. This work provides evidence that there is a dose-response relationship between the sum count of inflammatory-related diseases and the likelihood of frailty in older women.
A key clinical implication of these findings is that a higher sum count of inflammatory-related diseases in older adults should alert clinicians to the possibility that their patients, if not already frail, have an increased probability of becoming frail. Similarly, the likelihood of frailty should be considered in older adults with multi-morbid inflammatory-related diseases that include specific disease combinations, which are most frequently associated with frailty. Identification of these vulnerable older adults is an essential first step for maximizing strategies that would prevent or reduce sarcopenia, improve energy, and benefit physical function. Factors known to precipitate or accelerate frailty, such as decreased physical activity and function, poor nutritional intake, and polypharmacy, may be addressed with resistance exercises, nutritional supplementation, and a routine review of medications to ensure that the benefits outweigh their side effects (Crentsil et al., 2010
; Fried et al., 2005
). One recent randomized clinical trial on frail, older women indicated that regular low-impact exercise supplemented by daily oral dehydroepiandrosterone, an anabolic steroid hormone which decreases with aging, may modestly improve lower extremity strength and physical performance over a 6-month interval (Kenny et al., 2010
). While the long-term clinical impact of such an intervention in frail older adults has not yet been determined, these findings are still promising.
In this current study, we identified 1) CKD and depressive symptoms, 2) CVD and depressive symptoms, and 3) CKD and anemia as the three most prevalent, significant disease combinations in frail older adults, respectively. Characterization of the specific chronic conditions most frequently associated with frailty adds to the growing literature on multimorbidity in frail older adults. Previous research has linked depression in CKD to a 68% prevalence of frailty among incident dialysis patients. (Johansen et al., 2007
). Major depressive episodes in patients with CKD have been found to increase the risks of dialysis initiation and frailty-related outcomes, such as hospitalizations and death, even after adjusting for CKD severity and co-existing medical conditions (Hedayati et al., 2010
). Depression also commonly co-occurs across the clinical spectrum of CVD and has been similarly associated with major frailty-related outcomes, including disability, as well as heightened rates of CVD events in patients with pre-existing disease (Celano et al., 2011
; Hance et al., 1996
). Decreased physical activity associated with depressive symptoms and the net deficit in skeletal muscle protein synthesis associated with a heightened catabolic state and underlying inflammatory changes in CKD and CVD can facilitate muscle mass atrophy, which ultimately may lead to a frail state (Adams et al., 2006
; Persinger et al., 2003
; Rantanen et al., 2000
; Schapp et al., 2006
). Furthermore, advanced stages of CKD have been known to cause anemia from impaired renal synthesis of erythropoietin with poorer kidney function corresponding with lower hemoglobin levels (Astor et al., 2002
). Because lower hemoglobin levels diminish oxygen delivery to major organ systems, anemia can negatively impact the ability of CKD patients to engage in physical activity. As a consequence, patients with co-existing CKD and anemia may experience accelerated loss of muscle mass, subsequently resulting in sarcopenia and the development of frailty. While it is not surprising that having more than one chronic condition is worse than having just one, we have identified the importance of these specific disease patterns based on both their prevalence rates and associations with frailty.
Another striking result in our study was that depressive symptoms, CVD, and anemia were among the most common of the eight single inflammatory-related diseases that were found in significant disease patterns of frail older adults. This provides further evidence suggesting that frailty often co-exists with disease combinations that include depressive symptoms, CVD, or anemia. However, whether frailty is a cause or consequence of specific disease combinations remains uncertain. To determine the role of disease-related mechanisms in the development of frailty, the temporal relationship between frailty and significant disease combinations, such as CKD or CVD and depressive symptoms, needs further elucidation as does the extent to which frailty affects important health outcomes in older adults with these specific co-existing conditions.
Potential approaches to delay or prevent the development of frailty may include the following: 1) periodic screening for depressive symptoms and improved treatment of depression in older adults who have CKD or CVD, 2) multi-pronged management of vascular risk factors that contribute to the manifestation of renal insufficiency and cardiovascular disease, and 3) more rigorous evaluation for potentially treatable causes of newfound anemia in older adults, particularly in anemic patients with progressing CKD who may clinically benefit from erythropoietin-stimulating agents. Subsequent studies which compare the risk reduction of frailty using different strategies or treatments for co-managing specific inflammatory-related chronic diseases in older adults would serve as the basis for developing improved interventions for frailty.
A plausible mechanism for the association between multi-morbidity and the likelihood of frailty is the induction of multiple overlapping pro-inflammatory pathways associated with many disease-related systemic changes. This activated pro-inflammatory cascade is theorized to dysregulate a critical mass of metabolic and physiologic systems, which undermine homeostatic adaptive capacity, precipitating frailty (Fried et al., 2009
; Gruenwald et al., 2009). Elevations in systemic inflammatory biomarkers have been associated with the frailty measures and its potential risk factors (Hubbard et al., 2009
; Payette et al., 2003
), which include not only inflammatory-related chronic conditions, but also socio-demographic factors and health-associated behaviors, such as advancing age and smoking (Bruunsgaard et al., 2001
; Levitzky et al., 2008
). To this effect, a heightened pro-inflammatory state may play a key role in the multi-systemic dysregulation associated with frailty. Although our study population did not demonstrate any statistically significant differences for age group (70–74 vs. 75–79 years old) and smoking status (current, former, and never) between the frail and the non-frail women (both p-values= 0.06), general trends for these characteristics indicated that higher frequencies of frail, compared to non-frail, participants were older and current or former smokers. These trends in prevalence rates appear to support prior studies showing that advancing age and active smoking are associated with frailty (Hubbard et al., 2009
; Song et al., 2010
Thus far, assessment of the total pro-inflammatory burden in older adults and its relationship with the progression of frailty has remained a methodologic challenge (Bandeen-Roche et al., 2009
). Perhaps, the sum count of inflammatory-related chronic diseases in older adults may reflect the magnitude of an induced pro-inflammatory state. As a next step, it will be important to assess whether a sum count of inflammatory-related diseases may have prognostic utility in the development of a prediction index for incident frailty and its associated clinical outcomes, including disability and death. Results from these prospective cohort studies would demonstrate whether a greater inflammatory-related disease burden is indeed causally-related to the risk of frailty and its associated health outcomes. These studies may also reveal the extent to which current treatment strategies that effectively co-manage multi-morbid inflammatory-related diseases affect the risk of frailty.
Several caveats limit the conclusions that can be drawn from this study. First, we cannot establish a casual relationship between a higher count of inflammatory-related chronic diseases and the risk of frailty because of the study’s cross-sectional design. Our study population was also exclusively composed of community-dwelling older women. As a result, further research is needed in the minority of older adults who are men, since gender influences body composition and size, which may lead to differences in prevalence rates of inflammatory-related chronic diseases and in the progression rate of frailty (Cawthon et al., 2007
). The internal validity of this study, however, remains preserved. Next, the data were lacking in chronic diseases and conditions which were non-inflammatory in origin. Therefore, we could not feasibly perform parallel analyses investigating whether the sum count of non-inflammatory diseases is associated with frailty. An insignificant association would have further strengthened the specificity of our findings. We also could not capture the inflammatory contribution of undiagnosed, subclinical diseases. This may render the inflammatory-related disease count conservative and thereby result in underestimation of frailty in older adults with underlying inflammatory-related chronic conditions. Perhaps, this is one reason why participants who did not have any of the eight inflammatory-related diseases were found to have a statistically significant predicted probability of frailty (0.02, 95%CI=0.01–0.04), albeit extremely low. Finally, we could not evaluate the relationship of cognitive impairment or dementia to frailty. Underlying inflammatory mechanisms also may contribute to the development of cognitive impairment, which has been shown to independently increase the risk of frailty (Raji et al., 2010
). Because participants who were initially enrolled into WHAS were cognitively intact with a MMSE score ≥24, data were not available to study older women who had baseline cognitive impairment or dementia.
We demonstrated that the likelihood of frailty increases in a dose-response fashion as the chronic inflammatory-related disease count increases. Shared mechanisms among specific disease combinations may further contribute to this risk and facilitate the progression of frailty. Longitudinal ascertainment of chronic inflammatory-related diseases with respect to incident frailty in older adults will help establish a temporal relationship between a higher total disease count and the risk of frailty. Additionally, future research using larger cohorts to determine the extent that specific disease combinations effect frailty would lay the groundwork for evaluating likely disease mechanisms for therapeutic interventions. This, in turn, could lead to the development of improved treatment strategies to manage co-morbid diseases more effectively and potentially prevent or delay the progression of frailty in older adults.