"Deaths of Despair" in the U.S.

Covid Deaths, Not Lockdown Deaths, part 4

"Deaths of Despair" in the U.S. 

November 18, 2020

As part 3 explained (will explain), any sizeable deaths caused by "lockdown" would be found under natural causes like heart disease, but mainly what we see there is a lot of apparently misdiagnosed COVID-19 fatalities, These are equal to around 50% of the virus deaths reported, or one in three goes unnoted, and the U.S. death toll may be over 400,000 already. Lockdowns, mask mandates, and other measures have kept that number from being even higher, but that's not to say they don't also have a heavy cost, including lives lost due to denied medical care and screening for a range of serious conditions. 

Yet attention has been grabbed by a smaller, more optional class of deaths - people who died from lockdown-related unemployment or social distancing, via suicide or fatal overdose. The former is highlighted as driven by loneliness and sheer desperation, while those who drink themselves to death or overdose might do so out of boredom and idiocy. All of them have a hand in their own deaths, yet previously existing mental conditions don't matter to the critics like medical ones do with covid - people who kill themselves died OF, not WITH lockdown, and they're sure the number is immense and set to grow.

There is little reason to doubt that deaths from both suicide and over-intoxication have increased with idleness, isolation, and depression, and there's various evidence to support it. Overdoses especially seem easy these days with all these opioids making the rounds. But the proportions are unclear, and can hardly compare to the other scales of fatality involved here; all non-natural deaths combined (suicide, homicide, other violence, accidents) are usually less than 10% of a yearly total, while >90% are from natural causes and, as part 3 showed, those are all in a general and massive elevation this year. 

But as an RT article warned back on May 7 Lockdown-inspired suicides on course to DWARF coronavirus deaths in Australia & in time, even in US – studies. This noted how President Donald Trump had "warned early on in the pandemic that there would be “suicides by the thousands” if prolonged economic shutdowns were imposed" over what he considered a harmless "kung-flu" that would soon vanish via "herd mentality." With such advice widely refused, the article foresaw "fallout looming on the horizon", citing an Australian professor's predictions of up to 1,500 extra suicides there by year's end - a 50% increase - and a dramatic increase already underway in Tennessee. 

Martin Armstrong / Armstrong Economics reported back on developments on September 6, claiming "there have been far more deaths from suicides and drug overdoses than from COVID-19," a claim he falsely attributes to the head of the U.S. Centers for Disease Control (CDC). I'll come back to that below, but first, Armstrong whined "the leftist media will no doubt claim the CDC is wrong. The social media firms will scrub any mention that this COVID has been a hoax." With that set and the CDC's backing secured, Armstrong turned on the people "allegedly responsible for the rising death toll from destroying everyone’s future. ... Everyone should write to the Attorney General and demand a criminal investigation into Zucker, Gates, Soros, Schwab, and of course Fauci." 

On June 16, James Lucas at the Federalist alerted readers Research Finds Lockdowns Are Far Worse For Health And Lives Than Coronavirus. Mainly, he was worried about unemployment caused by all the stay-at-home orders and business closures:

"[U]nemployment has very deleterious effects on public health, particularly due to an increased risk of cardiovascular disease and stroke. ... Unemployment also adversely affects mental health ... directly increases mortality by increasing suicides. ... Extended unemployment can reduce average life span by up to two years. Applying that to the more than 40 million unemployed in the United States right now and the results are frightening."

Of course circumstances will vary, and it will be quite stressful for some workers, and especially for many business owners. But usually such job-related anxiety comes from loss of income, while in many if not most cases, pandemic relief ensured unemployment benefits at 100% for lockdown-related temporary layoffs. Some people including myself enjoyed a fully paid "staycation" and have been back to work for a while now, no worse for the wear. 

75,000 "Deaths of Despair" in the U.S. - a desperate distortion

One study Mr. Lucas cited found that being unemployed was historically “associated with a twofold to threefold increased relative risk of death by suicide,” while the current situation was worse, not better; "an even more recent study has put the increase in lockdown-related “deaths of despair” at 75,000 in the United States alone." That's an astounding claim. An increase over normal levels of 75,000 would account for all the non-covid excess deaths to that time (around 50,000, non-adjusted) and then account half of them a second time, with no help from any other cause.

Let's examine the claim. This would be suicides and drug-alcohol overdose, with "despair" likely just presumed as universal. This data is usually delayed 1-2 years before release, but for reference: 

* In 2018, some 48,000 people died from suicide in the United States (CDC)

* In 2019, almost 71,000 people died from drug overdoses. (CNN) 

These numbers should be fairly similar year to year, so we could plug them into 2020 and have a decent idea what to expect otherwise. I hear both are rising each year, but let's skip that here to let lockdown have a try at explaining any increase.  They total 119,000, with about 40% suicide and 60% OD. In context, that 75,000 would be about 30,000 EXTRA suicides by early July (atop some 48,000 expected all-year), and 45,000 people who drank/shot up too much and died in that same span, beyond the usual yearly number around 71,000. 

That seems faintly possible, but something says they calculated high or, as it turns out, the findings were misrepresented, maybe by accident. As the linked press release explains "new research released by Well Being Trust (WBT) and the Robert Graham Center for Policy Studies in Family Medicine and Primary Care" predicts "as many as 75,000 more people will die from drug or alcohol misuse and suicide." Lucas made it sound like that was an estimate of the actual deaths so far, but it's the high end of an estimated range of future deaths over an unclear timespan, anticipating ... hold on ... 

Here's the report, direct PDF link. A key issue is increase in mortality per one point rise in unemployment (is it 1%, 1.3% or 1.6%?) and rate of economic recovery. "the negative impact of isolation and uncertainty" suggested the highest increase in mortality (1.6%) would apply, as it did in the Great Depression. With a slow recovery (Same as Great Depression) they estimate that would cause up to 154,000 deaths of despair, but spaced out over a span of 10 years, and apparently with little mitigation possible in the meantime. 

A fast recovery would take 4 years, as the report considers it, and at just 1% mortality increase, might drive as few as 27,644 to the grave in the end. The number 75,000 doesn't come up, but is similar to medium recovery at +1.6% mortality and 80,735 deaths of despair over 7 years. 14,932 of those were estimated by the end of 2020. Now that sounds plausible, as far as I know. 

119,000 despair deaths annually could be split into 29,000 per quarter, except for the last quarter getting 32,000 (holidays increase suicide). The lockdown period in question is roughly second quarter, so around 29,000 deaths normally. I could see a full doubling of the rate for an extra 29,000 deaths, though that seems high. I'll defer to the WBT/Graham Center report that predicted a total of 14,932 such deaths by the end of 2020. Even if we put all those in the acute lockdown span in question, that would be a roughly 50% increase over the expected 29k - 15,000, not 75,000. I'll plot this out below.

A 50% increase is apparently kind of extreme. The cited RT article noted "In the best-case scenario, suicide rates will increase 25 percent, Professor Ian Hickie predicted, observing that 40 percent of those would be among young people. If the Australian economy continues to deteriorate, suicide rates could increase 50 percent." Someone more expert than me trying to raise alarm predicts 25% and maybe 50% increases, and the WBT/Graham Center report doing the same proposes something close to or less than 50%. So in my plotting below, I'm more than fair in considering 50% and 75% increases in the U.S. and still finding the alarmist claims fail even my lenient plausibility test. First though - why I "claim the CDC is wrong" or, as it turns out, was misrepresented.

Redfield's Comments: True?

Center for Disease Control Director Robert Redfield has been widely cited for his statements on July 14 that "We’re seeing, sadly, far greater suicides now than we are deaths from COVID. We’re seeing far greater deaths from drug overdose that are above excess that we had as background than we are seeing the deaths from COVID. " That's a direct quote from a fairly balanced article with some detail, which I'll mainly cite below: CDC Director Compares Rate of Suicides to COVID-19 Deaths, by Micaela Burrow, Town Hall, July 28. See also National Review.

Armstrong Economics for one was less balanced, reporting sloppily: "The Director of the CDC has come out and bluntly stated that there have been far more deaths from suicides and drug overdoses than from COVID-19," as in comparative totals to date. But Redfield was clear it's something "we're seeing ... now," at a moment when covid deaths had been brought much lower than they had been or would be. In fact he may have meant to make a dual point that covid was brought so low it was even lower than the increased self-endings now on par.

His exact meaning is otherwise a bit unclear. For example, were there "far greater suicides" and "far greater deaths from drug overdose" - as it sounds - or did the two combined outstrip covid? As I'll show below, that makes more sense by the data. Michael Thau seems to get the "now" aspect, but decided "Even Redfield admits lockdowns are now killing at least two times the number of Americans as COVID-19," exceeding it "by far" once for suicides, once again for ODs. (Head of CDC Admits Lockdown Killing Way More Americans Than COVID! Urges Masks. When Will These People Be Held Accountable for the Carnage They've Caused? Redstate.com, July 27, 2020) 

COVID-19 was killing around 500-700 a day at the time, so that's at least 1,000-1,400 lockdown deaths per day suggested between the two causes. The stats used here yield a normal rate of ~128 suicides and 192 ODs per day, or ~320 combined, as an expected norm. Thau's reading would be around four time the normal rate, while data and realistic predictions suggest more like 1.5 times at most. 

Several articles (including Thau, above) cite Redfield's comments as applying to all Americans, but that makes little sense even at early-mid-July's lower death rate, let alone compared to the masses of mortality seen in April. And he specified this is a "cost that we’ve seen, particularly in high schools," and perhaps only to there to the cited degree. At the time, schools were closed as normal for summer break, but had been closed for lockdown prior to that, and re-opening in the fall was a big question Redfield was addressing. 

Overdosing-overdrinking and suicide are always top killers for teenagers - a promising but moody segment with overall low mortality, and especially low COVID-19 mortality. It is logical that the situation had increased those tendencies, plausibly even doubling the relevant rates for this demographic. Yet all ages and all pandemic phases considered, COVID-19 has definitely killed far, far more people this year than all such deaths combined, normal and excess, and was probably outstripping them both even in early July. But "probably" is boring, so let's have a look. 

Suicide-OD estimated norm (as mentioned above) = ~29,000 each quarter (13 weeks). Lockdown measures began variously from the end of week 12 and more widely in week 14. If we start lockdown counting at week 14 to week 26, that's a full second quarter and covers the span to lowest covid deaths Redfield likely referred to. Plotting this in: 

* 29,000 divided by 13 weeks = 2,231 weekly average.

* at 50% elevation: 43,500 = 3,346 weekly av

* at 75% elevation: 50,750 = 3,904 weekly av

* of course the real numbers would vary more, maybe about normal in some spots, and up past 75% in others. The slight wiggle I added to the green lines was just for effect, so it looks like plotted data and not part of the scale.

This finds that if a 75% increase was in effect at weeks 25 and 26, they would exceed COVID-19 deaths slightly (by about 100 each week), for those two weeks only. And recall lockdown isn't responsible for all of those, just the smaller half in excess of normal (the 75% in a 175% situation). So the data suggests Redfield might be referring to a real but remarkable fact, covid deaths were briefly outstripped by elevated deaths of despair, underlining the need he was addressing - to relax controls, get schools re-opened, etc., while keeping the virus reined in so those covid deaths could stay low. 

Conclusion

Burrow at Town Hall noted "Where Redfield obtained his data is unknown ... health authorities will not have verified data regarding suicides and drug overdoses in 2020 for two more years." One CDC source says "For deaths due to external causes of death [including suicide, accident] or unknown cause, provisional data are highly unreliable and inaccurate in recent weeks, and it can take six to nine months to ensure sufficiently accurate estimates," while another says "Upward trends in other causes of death (e.g., suicide, drug overdose, heart disease) may contribute to excess deaths in some jurisdictions." But in general, these deaths are most likely not even reflected in the data we have, or only in small part - which would mean deaths of despair explain very little or none of the excess we have numbers to fret over. 

We may have yet to see how many extra thousands actually killed themselves due to lockdown. Even a 25% average increase for quarter 2 seems high to me, but if it ever hit 75 (as Redfield's comments suggest at weeks 25 and 26), then it may be too low. Let's say 30-35% = 8,700-10,150 extra deaths.  The 2nd quarter should be less acute in this regard as activity increased and much work resumed, But economic issues may still drive an increase of 10-20% = 2,900-5,800.  There may have been even more, or less, but a fair and broad estimated range might be 11,600 to 15,950 "deaths of despair" so far. 

Meanwhile, a somewhat reined-in COVID-19 has actually caused the bulk of overall U.S. mortality that, up to mid-November, is surely past 320,000, and may be as high as 405,000 (see here). All-causes mortality in the entire United States increased to at least 45% in the peak of deaths at week 15 (ending April 11), largely because New York State (including NYC) saw up to 416% its normal fatalities that same week (4,663 to 1,952 in 2019), because New York City saw 746% the usual deaths (7,863 vs. 1,054 in 2019). Importantly: that peak was kept from being higher and later because of lockdown, etc. As always, it gets its spike shape - narrow peak with a solid and sharp downslope - over the few weeks following a successful lockdown. Otherwise it would be just the lower upslope of a big, fat mountain, as shown above in orange and brown. The actual number is hard to say and depends on what else exactly would have happened. But it must be a comparable number - maybe much lower (50%?), equal, or twice as high. That puts it anywhere between 160,000 and 800,000 Americans who might have died by now but haven't. 

And I concede, these measures also contributed heavily to probably at least 11,600 additional self-cancellations in the last 6 months, and possibly over 16,000. Seeing that trade-off does give pause and reminds us of the need for balance, a chance the lockdown critics largely squandered.