Age shall not wither them — because HIV really doesn’t kill
Posted by Henry Bauer on 2009/02/04
“Age shall not wither…” seems an attractive sentiment as we begin to notice signs of ageing. The saying is variously employed to enlist, flatter, reassure:
“Age shall not wither them. . . . More men over 50 are taking up youthful challenges. We meet the golden not-so-oldies who refuse to settle down with their pipe and slippers”
“Age shall not wither her . . . Some of the most exciting artists of our time are women over 60.”
“Age shall not wither them — The elderly are America’s most reliable voters; . . . America’s older population already has considerable clout. . . . The old will play a main role in choosing the next president, and future ones.”
“Age shall not wither them — Not Dead Yet: a Manifesto for Old Age . . . It is highly encouraging that someone as authoritative as Rabbi Baroness Neuberger has written such a militant book on age discrimination”
That age doesn’t mean loss of physiological capabilities is most remarkably illustrated by the mortality among older “people living with AIDS” (PWAs): their death rate doesn’t increase significantly with age.
I’ve mentioned that before, but almost as an afterthought and with numbers given for only a few selected years [Table II, “Living with HIV; Dying from What?” , 10 December 2008]. Table 1 below has the numbers for all the years (1982-2004) for which the National Center for Health Statistics has published figures for new diagnoses of “AIDS” and for deaths from “HIV disease”. [The death rates were calculated as described in full detail in earlier posts: “HAART saves lives — but doesn’t prolong them!?”, 17 September 2008; “Living with HIV; Dying from What?” , 10 December 2008.]
According to these data, “HIV” is not significantly more deadly for people beyond age 40, or 50, or 60, than it is for people in their 30s:
Though these calculations are shown to two significant figures, no such accuracy is claimed, nor is it necessary to make the single point that the variation of death rate with age is unlike that for any other condition, or for all-cause mortality. That’s easier to see when the rates in each calendar year are normalized around age 25 or 30:
By stark contrast, with all (other?) diseases and illnesses the risk of death increases rapidly, indeed exponentially, with age (Table 3), reflecting the fact that the human organism copes less and less effectively with diseases, illnesses, or health challenges in general, as it ages:
To forestall some foreseeable quibbles, I show data for eight years spanning more than half a century. With almost no variation, there is something like a doubling of the death rate for every decade of age, despite considerable changes in the ethnic composition and age structure of the US population over that period. An exponential increase in mortality with age has long been taken for granted by demographers (Gompertz or Gompertz-Makeham Law). That progression also fits common knowledge and common sense. Although sound-bites and jokes vary somewhat over the age at which humans are judged to be at peak physiological condition, it would be unusual to suggest that it is later than the twenties or early thirties — think of performances in the most physically demanding sports, for example.
But the death rate among PWA (Tables 1, 2) shows nothing like an exponential increase with age. Within the reliability of the numbers and to a first approximation, the rate remains virtually unchanged with age.
On closer examination and to a second approximation, during the 1980s there does seem to be an increase in death rate with age. That trend peters out in later years and is gone by the late 1990s, indeed in the most recent years the death rates even seem to be somewhat lower at the higher ages. Although these second-order trends are at best weak, if they are real then they make sense. Up to the mid-1980s, the criterion for “PWA” was actually manifest illness, a clinical diagnosis of advanced illness within months, at most a couple of years, of death; “HIV-positive” became a criterion only after the mid-1980s, since “HIV” was officially “discovered” only in April 1984. With an overall mortality so high, above 50%, one could not expect an exponential increase with age, so the observed changes are in reasonable accord with normal expectation. In later years, however, and especially after 1993, the criterion for “PWA” was primarily an “HIV” test, and it is during this period that there is no significant increase of mortality with age. Note too that in the most recent years the mortality is so small — a few percent — that there’s no mathematical bar to exponential increases with age.
If “HIV-positive” together with low CD4 counts — the criteria for classifying PWA — signified a disease, then the mortality would increase with age in some manner approaching the exponential rise seen with other illnesses and diseases. Since actual experience is anything but that, those criteria for diagnosing “AIDS” are wrong. They don’t diagnose a fatal condition * (see footnote).
I made the comparison with all-cause mortality because direct comparison with known infectious and fatal diseases isn’t possible: only with “HIV/AIDS” are data available for numbers who were infected and didn’t die as well as for deaths. Perhaps the closest comparison would be with influenza, because so large a swath of the population is exposed to it that changes with age in the death rate per total population are a reasonable approximation to changes in the death rate per numbers infected, or at least would not be vastly different in terms of age distribution. As shown in an earlier post, the death rate from flu illustrates the usual rapid increase with age:
The scale on the y axis is logarithmic, so the approximately linear increase of y with increasing x represents an approximately exponential increase in death rate with age, just as with all-cause mortality.
From a poster presentation, “Death and Aging in the Time of Influenza: United States, 1960-2002”
by Nobuko Mizoguchi, MPH/MPP Department of Demography, University of California at Berkeley] http://www.popassoc.org/files/public/MizoguchiPoster.pdf
PWA do not experience a risk of death that increases with age
in the manner seen with flu or with all-cause mortality.
Therefore PWA were not selected according to any specific or general risk of death.
In other words, “HIV-positive” does not identify people suffering from a specific fatal illness.
This is yet another demonstration that “HIV” is not a pathogenic agent.
What explains this absence of any great variation of death rate with age? That the death rate among PWA varies little with age is actually just another way of expressing the fact remarked on earlier, that the age distributions for testing “HIV-positive”, and for being diagnosed with actual “AIDS”, and for the population of PWAs, and for deaths from “HIV disease”, are virtually superposable [“HIV, AIDS, and age: HIV/AIDS theory is wrong”, 23 January 2009, and other posts cited therein]. That indicates some rather fundamental relationship tying together “HIV-positive”, PWA, and death from “HIV disease”. Thereby one possible way of resolving the mortality conundrums is eliminated, namely, that those dying don’t belong to the same population as those surviving because— for example — the latter benefited from antiretroviral drugs whereas the former didn’t. [That last possibility is also eliminated by the fact that the median ages of PWAs and of those dying changed over the years at exactly the same rate — “Living with HIV; Dying from What?” , 10 December 2008].
Caveat for “HIV-positive” people: Although in general “HIV” tests do not detect the presence of a fatal pathogen, in some cases a positive “HIV” test does coincide with a serious illness, because “HIV” tests are subject to innumerable cross-reactions. Unfortunately, individuals caught in the “HIV-positive” circumstance are likely to be treated with antiretroviral agents, with their awful “side”-effects, instead of according to what the specific condition of each individual might call for in medical reality.