The Economist - USA (2021-07-17)

TheEconomistJuly17th 2021
Graphic detail Long covid

77

Fitbit for purpose

T

hegravestcostsofthecovid­19pan­
demic, in lives lost and economies par­
alysed,  have  been  plain  to  see.  But  one  of
its  most  durable  consequences  is  also
among  the  hardest  to  measure:  the  plight
of “long­haulers” with chronic symptoms.
Doctors have not yet agreed on a defini­
tion of long covid. Some use it to describe
symptoms lasting at least four weeks; oth­
ers  for  six  months  or  more.  Nor  is  there
consensus  on  which  symptoms  qualify.
Most accounts cite exercise intolerance or
shortness  of  breath;  changes  to  involun­
tary functions like heart rates; or cognitive
impairment  (“brain  fog”).  But  the  severity
needed for a diagnosis is open to debate.
Covid­19 is too new for long­run studies
of  its  impact  to  be  complete.  However,  a
new  paper  by  researchers  at  Scripps  Re­
search  Translational  Institute  and  CareE­

volution, a health­care firm, suggests that

heart  rates  can  distinguish  some  long­

haulers from patients who recover fully.

In March 2020 Scripps began obtaining

data  from  users  of  wearable  fitness  track­

ers.  So  far,  38,000  people  have  enrolled.

The study covers 875 participants who had

symptoms  of  respiratory  illness  and  were

tested for covid­19, of whom 234 were posi­

tive. The rest tested negative and probably

had other ailments, letting the researchers

compare covid­19 with different diseases.

Covid­19  was  distinctively  disruptive.

Although  patients  with  symptoms  who

tested negative did increase hours of sleep

and reduce the number of steps they took,

these  changes  were  modest  and  brief.  In

the  covid­positive  group,  the  effects  were

2.5 times larger, and lasted for a month.

The most striking differences involved

heart rates. An elevated pulse can leave tis­

sue short of oxygen, causing lightheaded­

ness  or  shortness  of  breath.  Whereas  the

resting heart rates of people with non­co­

vid ailments rose only slightly, those of co­

vid­19  patients  oscillated.  In  most  cas­

es—86% of people testing positive—pulses

rose  on  average  by  1.4  beats  per  minute

(bpm), and then fell by 2.9bpm. As patients

recovered,theirheartsspedupagain,to

0.5bpmabove  the  starting  point.  They  re­

turned to normal after about 65 days.

However,  the  remaining  14%  of  people

with covid­19—a share similar to the inci­

dence  of  long  covid  found  in  some  other

studies—faced graver changes. Their puls­

es followed the same rise­fall­rise­fall pat­

tern, over a similar time period. However,

the impact was larger, with an average peak

increase of 8.8bpm. Moreover, after 70 days

their  heart  rates  levelled  out  at  5.4bpm

above  the  pre­covid  pace.  Their  pulses

stayed in that range until the study ended.

Long­haulers  often  bemoan  unending

symptoms.  Although  the  study  did  not

track  symptoms  after  the  acute  phase  of

covid­19,  it  confirms  that  the  disease  can

affect vital signs for months. It also identi­

fies risk factors for long covid: participants

with  highly  elevated  pulses  were  unusu­

ally likely to suffer coughs, body aches and

trouble breathing early in their illnesses.

Many  young  people  think  they  do  not

need covid­19 vaccines because their risk is

low. However, youth did not reduce partic­

ipants’ chances of having elevatedpulses.

Greater awareness of long covidmight re­

lieve such sceptics of this fallacy.n

A new study using wearable devices
could help to define long covid

→ Changes in resting heart rates could provide a firm basis for diagnosing a common form of long covid

875 peoplewhoreporteda respiratoryillnessandweretestedforcovid-19
UnitedStates,March25th2020-January24th 2021

Changeinsleepduration,minutes

Changeinrestingheartrate,beatsperminute

Changeinstepsperday

Symptomsduringearly,acutephaseofcovid-19

Bychangeinheartrate4-8weeksintoillness,%ofgroup

*Averagerestingheartrate4-8weeksafteronsetofsymptomsat

least5 beatsperminuteabovepre-covidlevel Source:“Assessment

ofprolongedphysiological& behavioralchangesassociatedwith

covid-19infection”,byJ. M.Radinetal.,JAMANetworkOpen, 2021

Statisticallysignificantdifferences

Nostatisticallysignificantdifference

100806040200

100806040200

Cough

Bodyache

Di  culty

breathing

Stomachache

Congestionor

runnynose

Headache

Fatigue

Sorethroat

Fever,chills

orsweating

0

50

100

-50

Covid-negative

0525 075100 125 0 25 50 75 100 125

-4 000

4,000

-2,000

0

2,000

Dayssincefirstsymptoms Dayssincefirstsymptoms

-2

-4

0

2

4

6

8

10

12

0 10 20 30 40 50 60 70 80 90 100 110 120 130

Dayssincefirstsymptoms

95%confidence

Covid-negative

Covid-positivewith

highly elevatedpulse*

Covid-positivewith

highlyelevatedpulse*

Covid-positivewith

highly elevatedpulse*

Covid-positivewith

highlyelevatedpulse*

OtherOthercovid-positivecovid-positive

OtherOthercovid-positivecovid-positive

Share of participants with respiratory symptoms, %

Covid-negative

Covid-positive with highly elevated pulse*

73 23 4

Other covid-positive

↓Othermeasuresreturnedtonormal

onemonthaftertheonsetofcovid-1

Long-haulers’heart ratesstopped

falling after 70 days,andremained

elevated afterwards

↓ Long-haulers’ heart rates stopped

falling after 70 days, and remained

elevated afterwards