We don’t see much about COVID-19 in the mainstream media these days. People are tired of thinking about it, and the folks who generate content for the mainstream media are tired of writing about it. More than tired, I would say. Fed up is more like it. But a quick look at the data is interesting.
At present, the prevalence of COVID is diminishing. During the week ending August 31, 2.3% of emergency department visits reported positive COVID tests. This was a big drop from the 10.2% of ED visits that reported positive COVID tests the previous week. There has been a major increase in ED visits due to COVID just in the past few months, from the low positivity rate of less than 0.5% reported in the week ending May 9th of this year. Since the onset of the pandemic, there have been 139.8 million positive COVID-19 emergency department visits.
Overall numbers of hospitalizations and deaths have also dropped significantly. Here are some COVID death figures for comparison:
Week ending 09/14/2024 – 563 deaths due to COVID
Week ending 08/31/2024 – 663 deaths due to COVID
Week ending 05/18/2024 – 336 deaths due to COVID
Week ending 01/27/2024 – 2,131 deaths due to COVID
Week ending 01/17/2023 – 3,870 deaths due to COVID
Week ending 01/15/2022 – 19,078 deaths due to COVID
Week ending 01/19/2021 – 25,974 deaths due to COVID
According to my calculator, the most recent COVID death rate is about 2.5% of what it was at the pandemic’s peak. It’s not over, but it really is way, way down.
In the US, there have been 111,820,092 confirmed cases of COVID, and 1,219,489 confirmed deaths. Over 95% of the US population has identifiable COVID antibodies, which reduces the impact in terms of hospitalizations and deaths. Compared to COVID in its peak period, what we’re looking at now is definitely minor. The disease has its ups and downs, but the overall trajectory is clear. In general, COVID is diminishing as a serious threat.
The long COVID story is somewhat more concerning. The CDC has taken its time in defining long COVID. They have settled on three months as the time span that defines long COVID – if symptoms persist three months or longer, the illness qualifies as long COVID. The prevalence of long COVID in adults over the age of 18 declined very slightly, from about 7.5% in June of 2022 to about 6% in the same period in 2023, and has not changed since 2023. Those percentages are not percentages of the general population, but of the population that was diagnosed with COVID – in other words, since 2023 about 6% of persons initially diagnosed with COVID went on to develop long COVID. According to the CDC, the persons most likely to be affected by long COVID are women, Hispanics, hospitalized individuals, persons with underlying health conditions, and the unvaccinated. In adults reporting previous episodes of COVID, the prevalence of long COVID is much higher. It reached 18.9% back in 2022 and has not changed much since that time.
Just a couple of months ago, we were reluctantly forced to pass on the information that Paxlovid doesn’t do much to alleviate the symptoms of long COVID.
The parade of variants continues. As of August 31, there were 26 variants in circulation in the US. The variant currently causing the most infections is designated as KP.3.1.1, which is associated with 42.2% of COVID cases that come to the attention of the healthcare system and are subject to blood tests. This is certainly a small fraction of all the COVID cases since the beginning of the pandemic, but that percentage is likely an accurate reflection of the current prevalence of that variant in the overall population. All of these current variants are in the Omicron class.
Speaking as Doc Gumshoe, I’m reluctant to stake a firm position on the current COVID outlook. Yes, hospitalizations and deaths are way down, but each new variant seems to be highly infective. Vaccines and previous infections boost cellular immunity and are effective protection against serious illness, but are not that protective against initial infection. My wife and I wear masks only when we think we need to. At the movies or the theater we wear masks, also at any healthcare facilities we visit. Will this be the way it is for the rest of the 21st century? We certainly hope not, but it’s hard to know.
Some but not all antidepressants may result in weight gain
This was based on a study that analyzed data from more than 183,000 persons between the ages of 20 and 80, with an average age of 48 years. Most subjects were overweight or obese at the start of the study. The majority of the subjects were women. The researchers analyzed participants’ electronic health records and body mass index. They gauged weight gain or loss at regular intervals of six, twelve, and twenty-four months, after people started taking antidepressants. (Petimar J. Medication-induced Weight Change Across Common Antidepressant Treatments. Ann Intern Med 2024 Aug; 177(8);993-10030
In connection with this study Dr. Roy Perlis, associate chief of psychiatric research at Massachusetts General Hospital, pointed out that it is important to understand that weight gain is a key reason that people might stop taking antidepressants, even if the antidepressants are working well, and also that some people may be reluctant to start taking antidepressants even if they feel depressed or anxious, because they are aware that weight gain is a likely side effect and they are more concerned about weight gain than about their feelings of depression or anxiety.
The study specifically compared one antidepressant, sertraline (Zoloft, now marketed by Viatris, formerly marketed by Upjohn and Pfizer) with several other antidepressants – escitalopram (Lexapro, from Forest), paroxetine (Paxil, from GlaxoSmithKline), duloxetin (Cymbalta, from Eli Lilly). citalopram (Celexa, also from Forest). fluoxetine (Prozac, also from Eli Lilly), venlaxafine (Effexor, from Pfizer), and bupropion (Wellbutrin, a drug with conflicting data concerning different generic versions, and whose manufacture and distribution involves many different firms).
The biggest weight gains were associated with escitalopram and paroxetine. Each of these was linked with a 1.4 pound weight gain at 6 months. For escitalopam, the weight gain rose to 3.6 pounds at 24 months, and for paroxetine it reached 2.9 pounds at 24 months. Sertraline was associated with a small weight gain, just 0.5 pounds, at the 6 month mark, but the weight gain increased to 3.2 pounds at the 24 month marker. Citalopram, fluoxetine, and venlaxafine were much in the same range as sertraline. The only drug that was associated with any weight loss at all was bupropion, just 0.25 pounds at 6 months. But that trend switched directions at 24 months, when bupropion was associated with an average weight gain of 1.2 pounds.
The study’s author points out that weight gain and weight loss can be associated with depression itself. Some individuals can actually lose weight as a result of depression, which can make people lose their appetite. Perhaps when people start taking an antidepressant, their appetite improves, and they regain the weight they had lost.
The study was observational, meaning it cannot prove that antidepressants cause weight changes, only that they were linked with them. It wasn’t a randomized, controlled trial and the participants taking antidepressants weren’t compared to a control group not taking the medications.
Further limitations of the study were pointed out by the authors. One was that only about one third of the subjects were taking their prescribed antidepressant medication just six months after the study started, which makes it difficult (if not impossible) to link later weight changes with a specific drug. And, of course, the study was not randomized, so the authors could not tell if the differences in the weight outcomes were due principally to the differences between the medications or other differences between the study subjects.
I am in accord with the underlying principle of the study, namely that the subject of weight gain associated with antidepressants is definitely important. Excess weight and obesity is a huge health concern in the US and around the globe, and also, for many persons, weight control borders on an obsession. And huge numbers of persons take antidepressants. The CDC reported that during 2025 to 2018, 13.2% of American adults used antidepressants – about twice as many women (17.7%) as men (8.4%). And women are reportedly more concerned about weight gain than men. Thus, the subject that this study was addressing is of concern to a great many individuals.
But in terms of results that concerned individuals might act on, the study came up rather short. The study appears to have been designed in such a way as to come up with an advantage for sertraline, but no such advantage emerged. The only antidepressant that demonstrated any weight loss was bupropion, and then only at the six-month interval.
It’s hard to say if there was any overall conclusion emerging from the study. I grant the authors’ good intentions. It would clearly be worthwhile to puzzle out which – if any – of the available antidepressants had no adverse effects at all in terms of weight gain. The study did not accomplish this, and it seems likely that meeting that particular objective is unrealistic. The best I can say is, “Nice try!”
Should all persons aged 70 or older take statins?
Despite acknowledged skepticism on whether statins do more good than harm, the consensus is that persons over 40 who have any of several cardiovascular risk factors can reduce their risk of major adverse cardiovascular events by taking statins. About 47 million individuals in the US take a daily statin pill. It has been suggested that far more than those 47 million persons are in fact at major heart attack risk. According to the CDC, 86 million Americans have elevated cholesterol, and about 121 million have heart disease. According to cardiovascular practitioners, these individuals would all benefit from statin therapy.
But now it is being suggested that all individuals 70 years old or older would benefit from statins, regardless of whether they had any specific cardiovascular risk factors.
This was based on an observational study in which researchers analyzed data from the UK Biobank and Whitehall II studies in more than 20,000 adults 70 years and older with or without previous cardiovascular disease. This data was then used in a cardiovascular disease simulation model to estimate their heart disease risk, survival rate, quality-adjusted life years, and healthcare costs with and without lifetime standard or higher intensity statin therapy. (Mihaylova B, et al. Heart 2024;0:1–10. doi:10.1136/heartjnl-2024-324052)
On analysis, the researchers found that participants who continued with standard statin treatment for their lifetime increased their quality-adjusted life years by 0.24 – 0.70 years and those on higher-intensity statin therapy raised their quality-adjusted life years by another 0.04 – 0.13 years.
The study’s lead author, Borislava Mihaylova, DPhil, University of Oxford, said “The effects of statin therapy reported here across people 70 years of age and older are, as expected, a bit smaller but sizeable. These results complete the picture of likely substantial health benefits with statin therapy across the continuum of age and risk levels in the population.”
A question inescapably arises: are these small increases in life expectancy worth the increased risks that accompany statin treatment? It looks to me that the maximum benefit, on average, is a bit less than one “quality-adjusted” year of increased life. Some individuals will get more than that and some will get less.
But we have to consider the trade-off. The list of possible statin side effects is long and concerning, viz: headache, nausea, dizziness, gastrointestinal bloating, diarrhea, constipation, muscle or joint ache, confusion, memory loss, kidney damage, liver damage, muscle breakdown (rhabdomyolysis), elevated blood sugar and type 2 diabetes.
Of these, rhabdomyolysis is probably the one of the greatest concern. This condition affects about 26,000 persons in the US. It is generally uncommon, and uncommon among persons taking statins – about 1.5 in 100,000 statin users develop rhabdomyolysis. This would account for fewer than a thousand of the rhabdomyolysis cases in the US. Generally, rhabdomyolysis is a result of overexertion or trauma and results in a breakdown of muscle fiber. The toxic components of the muscle fiber can enter the circulation and cause kidney damage.
By itself, the risk of rhabdomyolysis should not be enough to dissuade a person from statin therapy if the heart disease risks were significant, in particular, elevated cholesterol. However, in a person with no specific cardiovascular risks, the benefits of statin therapy look dubious. Speaking for myself, I am not the least disposed to start statin therapy in the dubious hope of gaining a small fraction of an additional year of life.
In case you didn’t notice, the study population included both persons with and without diagnosed cardiovascular risk factors. In the US, as I pointed out above, the number of people with elevated cholesterol is almost double the number currently taking statins. If the UK proportion is similar, it is possible (and even likely) that the statins conferred that benefit only to the study participants who actually had cardiovascular risk factors, and not to the subjects without any risk factors. Why would it be otherwise? Why would statins, which lower cardiovascular risk by reducing cholesterol transport, bring any benefit to individuals whose cholesterol levels were normal? After all, as we have several times pointed out in past discussions, cholesterol at normal levels is an essential physiologic factor. If levels are in the normal range, why would there be any benefit in bringing these levels lower?
My skepticism extends not only to the results of the study, but to the motives of the study in the first place. Were the authors looking for a reason to prescribe statin therapy to an ever-increasing fraction of the population? The funding for the research came from the UK National Institute of Care and Health Research, so it wasn’t a pharmaceutical company looking to boost sales of its statin.
I need to repeat what I said at the beginning of this discussion, that statin therapy provides very significant benefits to persons with heart disease risks. The study under discussion was meant to determine whether statin therapy provided similar benefits to individuals who did not have these heart disease risks. In my opinion, this study did not accomplish that objective.
A notable AI accomplishment
AI can do many things, including things related to healthcare. AI can come up with compounds that, based on their configuration, have the potential to be employed as beneficial drugs. Whether AI can verify the efficacy of these potential compounds, in actually having beneficial effects when used in humans, needs to be determined on a case-by-case basis. There are estimated to be about 20 million organic compounds, any of which might be valuable as drugs. If AI could at least identify the possibles out of this huge group, it would be a major step in the right direction
The NY Times recently reported what I would judge to be a notable, and highly promising, AI accomplishment. A man with amyotrophic lateral sclerosis (ALS), which used to be known as Lou Gehrig’s disease, had totally lost his ability to use his voice. He could not make a single sound, much less utter a word or take part in a conversation.
Physicians at the University of California, Davis, were able to construct a three-dimensional printed model of his brain and use it as a way to determine where to implant electrodes that connected the patient’s brain with a computer. Then they sank four electrode arrays into his brain’s outer layer. Each array had 64 spikes and looked like a tiny bed of nails. Each spike detected impulses from the neurons that fired when the patient attempted to form a word by moving his tongue, lips, or jaw. The computer was able to translate those attempts into sounds.
Soon after implantation, the device – implants plus the computer, termed a neuroprosthesis – was able to recognize a 50 word vocabulary with 99.6% accuracy. Here’s how the article in the New England Journal of Medicine summarized the results of the treatment:
“Twenty-five days after surgery, on the first day of system use and following 30 minutes of collection of cortical recordings and processing while the participant attempted to speak, the neuroprosthesis achieved 99.6% accuracy with a 50-word vocabulary. On the second day, after 1.4 additional hours of system training, the neuroprosthesis achieved 90.2% accuracy using a 125,000-word vocabulary. With further training data, the neuroprosthesis sustained 97.5% accuracy for self-paced conversations for over 248 cumulative hours over 8.4 months after surgical implantation.
“In an individual with ALS and severe dysarthria, an intracortical speech neuroprosthesis reached a level of performance suitable to restore naturalistic communication after brief training.” (N Engl J Med. 2024 Aug 15; 391(7): 609–618. doi: 10.1056/NEJMoa2314132)
I’m aware that lots of people view AI as a serious and looming threat to human existence. If machines can do all that, why do we need humans? The example above is a clear indication of why we do need humans. It was humans who figured out that AI might constitute an answer to that patient’s highly concerning problem. AI can process the brain scans and come up with where exactly to implant the electrodes, but a human has to instruct it to do so. And it’s a whole lot more practical and convenient to have a human being actually implant the electrodes. We’re not antiquated and useless – not yet, anyway!
We all know that high blood pressure – hypertension, as it is known to the healthcare world and most of the rest of the world as well – is a clear and concerning health risk. When we go to any medical practitioner, almost always the first things the nurse does is check our height, weight, and blood pressure. It feels routine and almost trivial. If your blood pressure is a bit high on that initial reading, chances are the nurse, or the physician, will check it again. It’s important to do the blood pressure measurement correctly.
Here’s what the Harvard Health Letter says about taking your own blood pressure.
“Whether you are at the doctor’s office or checking your own blood pressure, it’s important to take certain steps to get accurate readings.
 Don’t drink a caffeinated beverage or smoke during the 30 minutes before the test. Sit quietly for five minutes before the test begins.
ď‚· During the measurement, sit in a chair with your feet on the floor and your arm supported so your elbow is at about heart level.
ď‚· The inflatable part of the cuff should completely cover at least 80% of your upper arm, and the cuff should be placed on bare skin, not over a shirt.
 Don’t talk during the measurement.
ď‚· Have your blood pressure measured twice, with a brief break in between. If the readings are different by 5 points or more, have it done a third time.
It’s a good idea to have your blood pressure measured in both arms at least once, since the reading in one arm may be higher (usually the right, since there’s more direct blood flow from the heart on that side). The higher number should be used to make treatment decisions.”
Why did Harvard Health think it was a good idea to post those instructions? My guess is, because many more people are now doing their own blood pressure readings. There was a time when a person would have his/her blood pressure checked only in the context of a healthcare visit of some kind. But nowadays, simple home blood pressure measuring devices are widely (and inexpensively) available. Mistakes in using those devices, as well as possible defects in the devices themselves, could easily lead to erroneous readings, and erroneous readings – whether on the high or low side – could have harmful consequences. Perhaps the Harvard Health instructions will make a real difference in reducing the frequency of wrong blood pressure readings, and, also perhaps, lead to improved health outcomes.
* * * * * * *
The next time I put fingers to the keyboard (in my capacity as Doc Gumshoe, anyway) I’m going to take a look at urinary tract infections. These are the most common infections in the US, and they are much more than a nuisance.
As I’ve said many times, I welcome your comments – keep them coming! Also, please let me know of any areas in the healthcare orbit that I should scrutinize.
Be well, and thanks again! Best, Michael Jorrin (aka Doc Gumshoe)
[ed note: Michael Jorrin, who I dubbed “Doc Gumshoe” many years ago, is a longtime medical writer (not a doctor) and shares his commentary with Gumshoe readers once or twice a month. He does not generally write about the investment prospects of topics he covers, but has agreed to our trading restrictions.  Past Doc Gumshoe columns are available here.]
