US: Long Covid prevalence from Household Pulse Survey

The US National Center for Health Statistics has added a question about Long Covid to their Household Pulse Survey.

Nearly One in Five American Adults Who Have Had COVID-19 Still Have “Long COVID”

New data from the Household Pulse Survey show that more than 40% of adults in the United States reported having COVID-19 in the past, and nearly one in five of those (19%) are currently still having symptoms of “long COVID.”

The data were collected from June 1-June 13 by the U.S. Census Bureau and analyzed by CDC’s National Center for Health Statistics (NCHS). The Household Pulse Survey is an ongoing partnership between the Census Bureau, and CDC and other federal agencies. NCHS recently added questions to the survey to assess the prevalence of post-COVID-19 conditions, sometimes called “long COVID.”

For all U.S. adults, the new data show:

• Overall, 1 in 13 adults in the U.S. (7.5%) have “long COVID” symptoms, defined as symptoms lasting three or more months after first contracting the virus, and that they didn’t have prior to their COVID-19 infection.
• Older adults are less likely to have long COVID than younger adults. Nearly three times as many adults ages 50-59 currently have long COVID than those age 80 and older.
• Women are more likely than men to currently have long COVID (9.4% vs. 5.5%).
• Nearly 9% of Hispanic adults currently have long COVID, higher than non-Hispanic White (7.5%) and Black (6.8%) adults, and over twice the percentage of non-Hispanic Asian adults (3.7%).
• Bisexual adults and transgender adults (7.5%) were more likely to have current long COVID symptoms than adults of other sexual orientations and gender identities. 12% of bisexual adults have current long COVID symptoms, compared to 7% of straight and gay and lesbian adults. An estimated 15% of transgender adults have current long COVID symptoms, compared to 5% of cis-gender male adults and 9% of cis-gender female adults.
• The prevalence of current long COVID symptoms differed between states. The states with the highest percentage of adults who currently have long COVID symptoms were Kentucky (12.7%), Alabama (12.1%), and Tennessee and South Dakota (11.6%). The states with the lowest percentage of adults who currently have long COVID symptoms were Hawaii (4.5%), Maryland (4.7%) and Virginia (5.1%).

More details

These are the relevant questions:

Have you ever tested positive for COVID-19 (using a rapid point-of-care test, self-test, or laboratory test) or been told by a doctor or other health care provider that you have or had COVID-19? Answer Choices: yes, no

Did you have any symptoms lasting 3 months or longer that you did not have prior to having coronavirus or COVID-19? Answer choices: yes, no

Long term symptoms may include: Tiredness or fatigue, difficulty thinking, concentrating, forgetfulness, or memory problems (sometimes referred to as “brain fog”), difficulty breathing or shortness of breath, joint or muscle pain, fast-beating or pounding heart (also known as heart palpitations), chest pain, dizziness on standing, menstrual changes, changes to taste/smell, or inability to exercise.

Do you have symptoms now? Answer choices: yes, no

 

Two things jump out at me about this:

1. Long Covid prevalence from this survey is 3 times higher than the UK ONS survey (looking at the comparable number for people 17+ who have LC 12+ weeks after infection). That's strange.
2. Only 40% of Americans know they've had Covid, but that's probably roughly half of the real number, so a lot of asymptomatic infections. I wonder how many of those would also report they have LC symptoms.

[Edit: Redid these numbers for adults only in the UK.]

 

Looking at this closer, I think the US question overestimates LC prevalence (because there will be some people who have new symptoms that are not connected to Covid, and are more likely to report symptoms if you prime them while asking with a list of common symptoms). UK ONS survey probably underestimates LC prevalence because they ask about LC (defined as still experiencing symptoms that aren't explained by something else) and some people won't connect Covid-related symptoms to prior Covid infection or won't believe them to be Long Covid specifically. So probably reality is somewhere between the two.

 

Or they might underestimate how certain symptoms might be explained by something else than corona.

 

(Tried to make some overview but I haven't been following these estimates closely, so please correct me if I'm wrong):

The US data show that 18.9% of adults who ever had COVID are currently experiencing long COVID.

A previous ONS Survey from last year showed that "Among a sample of over 20,000 study participants who tested positive for COVID-19 between 26 April 2020 and 6 March 2021, 13.7% continued to experience symptoms for at least 12 weeks." This estimate, however, might be outdated?

A major difference seems to be that the US survey estimates that only 40% of people thinks they ever had COVID-19 while the ONS study in the UK previously gave an estimate of around 70%.

The US survey estimates 7.5 % of adults having long COVID while the ONS came with figure of 3.1%.

I assume that the 13.7% UK estimate wasn't on a representative sample of the population otherwise they would have an estimate of 70*0.13 = 9% instead of 3%. Anyone who can help to make sense out of these data?

 

If you download the data for the latest ONS Long Covid survey and look at Table 2, you can see the total number of those 17 years old or older with LC for 12+ weeks, then divide by an estimate of the total UK population 17+ in private households, you end up with roughly 2.6% (the overall 3.1% figure includes those with LC of 4-12 weeks duration as well and includes kids who are less likely to have LC, so we have to take those out to compare to US data).

There will also be a small difference because the UK data is as of May 1 and the US data is as of early June, so the US data will include more people who are now 12+ weeks from the tail end of the BA.1 wave.

ONS data definitely shows less likelihood of LC per infection now compared to the first year of the pandemic, due to vaccination and now Omicron. They recently showed about 8% of vaccinated adults have symptoms 4-8 weeks after an Omicron infection, which probably means about 5-6% for 12+ weeks. That would be under 5% once you include kids.

I don't think you can compare the ratio of LC to people infected between the two surveys, because the ONS participants have frequent PCR tests and so are likely to know about asymptomatic infections, while the US survey participants evidently do not. Probably fair to assume there isn't a major difference between the two countries in total number of infections, though we don't really know, but there is a significantly higher vaccination rate in the UK (especially among adults).

That 70% figure from the UK is very outdated. They haven't provided an update, but based on the very large number of infections since then, it must be much higher unless only the previously infected are catching it.

 

Thanks @LarsSG !

 

I've wondered this too. The "ravages of old age" is in my list of things that could conceivably be helped if we actually understood the mechanisms behind ME.

A bit off-topic but one other related point, if symptoms and objective data adequately overlapped. Perhaps there's an arbitrary, practitioner-dependent age where symptoms that might in a younger person be considered as "ME" are instead labelled "old-age". Once over that age threshold, the elderly are not subjected to diagnostic models and treatment paradigms that consider it is due to abnormal illness beliefs and amenable to CBT. Why not? "Old age" is not a diagnosis (despite it appearing on certain recent death certificates).

ETA: See this new paper in Cell (2023) on the role of re-awakened HERVs in cellular aging