I ran across a fantastic article today (this was written almost a month ago, sorry)  in the Canadian Medical Association Journal.  Drs. Rockwood, Song, and Mitnitski, from Dalhousie University in Halifax, analyzed some long term data that obtains in the Canadian National Population Health Survey.  The title of the article says it all:  Changes in the relative fitness and frailty across the adult lifespan.  They employed the data from seven 2 year cycles (starting in 1994-5, ending in 2008-9) of that study.  The key performance indicators they examined were death, use of health services, and changes in health status.  What made the report intriguing (to me, of course) was that the authors developed what they called the “frailty index”.

Others have used such indices, but they were nowhere as comprehensive.   Moreover, most of those studies were clearly aimed at assessing the health of older adults.  This study involved a comprehensive range of ages and health status.

The Frailty Index (as defined by the authors) examined 42 self-reported parameters, to enable the researchers to evaluate the overall health of each subject.  Any deficit in a given parameter counted as a 1; the total deficits were divided by the maximum number of parameters (42), thereby providing a normalized grade (0 to 1) for the health of the subject.  Among the 42 parameters are disease conditions such as cancer, hypertension, asthma, arthritis; conditions such as long term disabilities, the requiring of help in preparing a meal or effecting chores, cognitive problems, body pains, feeling tired, feeling unhappy, speech problems, and the like.  As you can see, this list is fairly extensive.  (Please note that Dr. Rockwell is planning to develop a commercial version of this index.)

Not surprisingly, the Frailty Index increased with age (on a semi-logarithmic scale, it rose 3% annually; an equation is provided below). Moreover, the Frailty Index seemed to “top off” at 0.67 (less than 1% of the population exceeded this value).  Moreover, if the index were below 0.03, the subjects were neither clinical nor phenotypically frail; once the index value exceeded 0.21, there was little chance (≤5%) of being considered “robust”. What these numbers provided was the panoply of conditions from fit with only one deficit (≤0.03), less fit (<0.1 [4 deficits or fewer]), least fit (<0.21 [9 deficits]), frail (≥0.21), to the most frail (≥0.45 [19 deficits]).

As noted initially, there were seven cycles (every 2 year period from 1994-5 through

2008-9) that were available in the longitudinal study.  At each of the seven cycles, the mean Frailty Index value increased exponentially with age.  (The actual equation is Frailty Index = 0.029 [log(Age)] – 3.90 .)  Moreover, the number of “robust” individuals decreased monototically with age, while the number of frail people increased exponentially. Obviously, the most fit people were the younger subjects, more likely to be married, and more educated.  (The latter is, in my opinion, more related to the fact that younger people have been attending and graduating from college and graduate schools in much larger degrees than the preceding generations.  Hence, the younger are more educated.)   Those that were considered robust at the baseline period stayed healthy, while those that were frail tended to die off.  Once the subjects reached 65 years of age, the change in the Frailty Index began to increase substantially.  (Notice on the accompanying figure how coincident each of the cycle’s data appears; this is a fairly significant relationship.)

Not surprisingly, the more frail the subject, the more extensive the use of health services. Both age and the Frailty Index were predictors of mortality, but frailty at a younger age clearly rendered one more likely to die.

That deficits accumulate with age is not surprising, since this is how the nature of cells and their organelles- it’s called cell aging.  Age is closely related to mortality, because aging is closely related to the accumulation of deficits (which means an elevated Frailty Index). It should be noted that some 25% of the population moved in and out of the relatively fit category in at least one of the 7 interval cycles.  (The older the subject, the less likely it was the subject would return to his original Frailty Index.)

Since the frailer one is, the more likely one is to employ health services, Dr. Rockwell’s index has tremendous utility for health planning purposes.  Given a certain population,  we can determine the demand for health care if we know the Frailty Index spectrum for that group.

I look forward to seeing the developments Dr. Rockwell makes with this concept.

Share this: