To the Editor: There is a need for a unifying clinical description of frailty, an instrument that could be useful for the practicing physician and also in the research setting. The Cardiovascular Health Study (CHS) index developed by Fried and colleagues is a step forward.1 It uses objective and relatively easily measured criteria and has been shown to detect a population of patients with a high risk of falls, disability, and death. The Study of Osteoporotic Fractures (SOF) index, as described by Ensrud and colleagues,2 is an attempt to develop an instrument that detects this same population of at-risk individuals, albeit using a more easily applied diagnostic tool. We would like to make two comments.
The first is that the CHS index used in the SOF study was composed of items that were not the same ones used by Fried and colleagues. Even though the authors admit this limitation, they make no additional comments about how this could have interfered with the results or about the possible theoretical implications of this considerable change in the nature of the indicator.
Evidently, a fundamental postulate of this study is that the modified version of the CHS retains the same clinometric performance characteristics as the original, but the authors present no evidence in support of this. Table 1 highlights the differences between the original CHS index and the adaptation used in the SOF. These differences are arguably considerable, because the modified version of the CHS instrument uses different criteria in at least three of the five dimensions (weight loss, poor energy, and exhaustion). Take the measurement of poor energy, for example, in which the original CHS index uses the response to an item of the Center for Epidemiologic Studies Depression Scale, which has a Likert-scale format, whereas the SOF study adaptation of the CHS uses a simple yes or no answer to one of the questions of the GDS. In our view, these are different parameters.
|Item||CHS Frailty Index (Original)||CHS Frailty Index (SOF Adaptation)||SOF Frailty Index|
|Weight loss||Unintentional, of ≥10 pounds in prior year or, at follow-up, of ≥5% of body weight in prior year (by direct measurement of weight)||Shrinking as identified by an unintentional weight loss of ≥5% between the baseline and second examination (mean years between examinations 3.4 ± 0.5)||Irrespective of intention to lose weight of ≥5% between the baseline and second examination (mean years between examinations 3.4 ± 0.5)|
|Poor energy||Using the Center for Epidemiologic Studies Depression Scale, the following two statements are read: (a) I felt that everything I did was an effort. (b) I could not get going. The question is asked “How often in the last week did you feel this way?”||An answer of “no” to the question “Do you feel full of energy?” on the Geriatric Depression Scale||An answer of “no” to the question “Do you feel full of energy?” on the Geriatric Depression Scale|
|Weakness||Grip strength, stratified according to sex and BMI quartiles||Maximal grip strength in the lowest quintile stratified according to BMI quartile||Not applicable|
|Slowness||The slowest 20% of the population was defined at baseline, based on time to walk 15 feet, adjusting for sex and standing height.||Average walk speed in the lowest quintile stratified according to median standing height; walk speed: time in seconds to walk 6 m at usual pace (m/s)||Not applicable|
|Low physical activity level||Physical activity based on the short version of the Minnesota Leisure Time Activity Questionnaire. Kcal per week expended are calculated using standardized algorithm. This variable is stratified according to.|
Men: Those with 383 Kcal of physical activity per week are frail.
Women: Those with 270 Kcal per week are frail.
|Physical Activity Scale for the Elderly score in the lowest quintile||Not applicable|
|Ability to rise from chair||Not applicable||Not applicable||Inability to rise from a chair five times without using the arms|
Table 1 also shows the similarities between the parsimonious SOF index and the modified version of the CHS. They are so similar that it is difficult not to think that this is a case of a circular argument typical of information bias; the SOF index consists of three items, among which are weight loss and the complaint of exhaustion, which are measured in exactly the same way as in the modified CHS index. Thus 67% of the SOF index is identical to the modified CHS index. With this is mind, the coincidental 71% concordance in the classification of frailty between the two indexes is not surprising.
Second, we do not agree with the affirmation that “Like the CHS index, the three criteria of the SOF index reflect impairment in one or more physiological domains most frequently cited in frailty literature.” This supposedly implies that the components that make up these tests measure changes in the physiological parameters linked directly or indirectly to the complex multisystem physiological impairments that underlie the syndrome of frailty.3,4 Weight loss, reduction of caloric intake, decrease in grip strength, loss of walking speed, inability to rise repeatedly from a chair, and the complaint of exhaustion are related essentially to only one system: the muscular system, which is but one of the possible biomarkers of frailty. None of the items listed above could be considered adequate indicators of functional changes in the hormonal and immune systems, for example.5
The fact that the individuals who score positive on these tests have a greater risk of adverse health-related events means that these instruments are capable of detecting a population with greater risk but that probably constitutes only a subpopulation (those in whom sarcopenia is the chief manifestation) of the total group of frail individuals. Even allowing for the theory that loss of muscle mass might be a common final pathway in the frailty syndrome, equating frailty to sarcopenia is an oversimplification and may cause confusion as to the true pathophysiological basis of the former. To avoid this, it might be better to speak of a “phenotype of sarcopenic frailty” rather than a “phenotype of frailty.”
An unintended consequence of this semantic imprecision might be a weakening of the search for a more-comprehensive clinical diagnostic instrument for frailty, one that would include other possible biomarkers, thus widening the diagnostic net. The search for this comprehensive frailty detection instrument would ideally be done using prospective studies originally designed for this purpose and that would test a large array of variables.