TY - JOUR
T1 - Characterising patterns of COVID-19 and long COVID symptoms
T2 - evidence from nine UK longitudinal studies
AU - the CONVALESCENCE Study
AU - Bowyer, Ruth C.E.
AU - Huggins, Charlotte
AU - Toms, Renin
AU - Shaw, Richard J.
AU - Hou, Bo
AU - Thompson, Ellen J.
AU - Kwong, Alex S.F.
AU - Williams, Dylan M.
AU - Kibble, Milla
AU - Ploubidis, George B.
AU - Timpson, Nicholas J.
AU - Sterne, Jonathan A.C.
AU - Chaturvedi, Nishi
AU - Steves, Claire J.
AU - Tilling, Kate
AU - Silverwood, Richard J.
N1 - Publisher Copyright:
© 2023, The Author(s).
PY - 2023/1/21
Y1 - 2023/1/21
N2 - Multiple studies across global populations have established the primary symptoms characterising Coronavirus Disease 2019 (COVID-19) and long COVID. However, as symptoms may also occur in the absence of COVID-19, a lack of appropriate controls has often meant that specificity of symptoms to acute COVID-19 or long COVID, and the extent and length of time for which they are elevated after COVID-19, could not be examined. We analysed individual symptom prevalences and characterised patterns of COVID-19 and long COVID symptoms across nine UK longitudinal studies, totalling over 42,000 participants. Conducting latent class analyses separately in three groups (‘no COVID-19’, ‘COVID-19 in last 12 weeks’, ‘COVID-19 > 12 weeks ago’), the data did not support the presence of more than two distinct symptom patterns, representing high and low symptom burden, in each group. Comparing the high symptom burden classes between the ‘COVID-19 in last 12 weeks’ and ‘no COVID-19’ groups we identified symptoms characteristic of acute COVID-19, including loss of taste and smell, fatigue, cough, shortness of breath and muscle pains or aches. Comparing the high symptom burden classes between the ‘COVID-19 > 12 weeks ago’ and ‘no COVID-19’ groups we identified symptoms characteristic of long COVID, including fatigue, shortness of breath, muscle pain or aches, difficulty concentrating and chest tightness. The identified symptom patterns among individuals with COVID-19 > 12 weeks ago were strongly associated with self-reported length of time unable to function as normal due to COVID-19 symptoms, suggesting that the symptom pattern identified corresponds to long COVID. Building the evidence base regarding typical long COVID symptoms will improve diagnosis of this condition and the ability to elicit underlying biological mechanisms, leading to better patient access to treatment and services.
AB - Multiple studies across global populations have established the primary symptoms characterising Coronavirus Disease 2019 (COVID-19) and long COVID. However, as symptoms may also occur in the absence of COVID-19, a lack of appropriate controls has often meant that specificity of symptoms to acute COVID-19 or long COVID, and the extent and length of time for which they are elevated after COVID-19, could not be examined. We analysed individual symptom prevalences and characterised patterns of COVID-19 and long COVID symptoms across nine UK longitudinal studies, totalling over 42,000 participants. Conducting latent class analyses separately in three groups (‘no COVID-19’, ‘COVID-19 in last 12 weeks’, ‘COVID-19 > 12 weeks ago’), the data did not support the presence of more than two distinct symptom patterns, representing high and low symptom burden, in each group. Comparing the high symptom burden classes between the ‘COVID-19 in last 12 weeks’ and ‘no COVID-19’ groups we identified symptoms characteristic of acute COVID-19, including loss of taste and smell, fatigue, cough, shortness of breath and muscle pains or aches. Comparing the high symptom burden classes between the ‘COVID-19 > 12 weeks ago’ and ‘no COVID-19’ groups we identified symptoms characteristic of long COVID, including fatigue, shortness of breath, muscle pain or aches, difficulty concentrating and chest tightness. The identified symptom patterns among individuals with COVID-19 > 12 weeks ago were strongly associated with self-reported length of time unable to function as normal due to COVID-19 symptoms, suggesting that the symptom pattern identified corresponds to long COVID. Building the evidence base regarding typical long COVID symptoms will improve diagnosis of this condition and the ability to elicit underlying biological mechanisms, leading to better patient access to treatment and services.
KW - COVID-19
KW - Clustering
KW - Long COVID
KW - Longitudinal studies
KW - Symptom patterns
UR - http://www.scopus.com/inward/record.url?scp=85146656809&partnerID=8YFLogxK
U2 - 10.1007/s10654-022-00962-6
DO - 10.1007/s10654-022-00962-6
M3 - Article
AN - SCOPUS:85146656809
SN - 0393-2990
VL - 38
SP - 199
EP - 210
JO - European Journal of Epidemiology
JF - European Journal of Epidemiology
IS - 2
ER -