dat.hannum2020.RdResults from 35 studies measuring olfactory loss in COVID-19 patients using either objective or subjective measures.
dat.hannum2020The data frame contains the following columns:
| authorName | character | (first) author of study |
| DOI | character | article DOI number |
| ni | numeric | number of Covid-19 positive patients in the study |
| xi | numeric | number of Covid-19 positive patients in the study with olfactory loss |
| percentOlfactoryLoss | numeric | percent of the sample with olfactory loss |
| objectivity | character | objective or subjective measure used |
| measured | character | outcome measure |
| testType | character | type of test used |
| country | character | country where patients were treated |
| patientType | character | type of patient information and location where being treated |
One of the symptoms of COVID-19 infection is olfactory loss (loss of smell) either recently acquired anosmia (complete loss of smell) or hyposmia (partial loss of smell). One challenge to reaching this symptom is the wide range of reported prevalence for this symptom ranging from 5 percent to 98 percent. In this dataset studies were grouped into one of two groups based on the type of method used to measure smell loss (either subjective measures, such as self-reported smell loss, or objective measures using rated stimuli).
Ramirez VA , Hannum ME, Lipson SJ, Herriman RD, Toskala AK, Lin C, Joseph PV, Reed DR. 2020. COVID-19 Smell Loss Prevalence Tracker. Available from: https://vicente-ramirez.shinyapps.io/COVID19_Olfactory_Dashboard/ and https://github.com/vramirez4/OlfactoryLoss (accessed August 11, 2021)
Hannum, M. E., Ramirez, V. A., Lipson, S. J., Herriman, R. D., Toskala, A. K., Lin, C., Joseph, P. V., & Reed, D. R. (2020). Objective sensory testing methods reveal a higher prevalence of olfactory loss in COVID-19 positive patients compared to subjective methods: A systematic review and meta-analysis. Chemical Senses, 45(9), 865–874. https://doi.org/10.1093/chemse/bjaa064
medicine, covid-19, proportions
# copy data into 'dat' and examine data
dat <- dat.hannum2020
dat
#> authorName DOI ni xi percentOlfactoryLoss
#> 1 Vaira et al. 1 10.1002/hed.26204 72 44 61.111111
#> 2 Iravani et al. 10.1101/2020.05.07.20094516 16 13 81.250000
#> 3 Vaira et al. 2 10.1002/hed.26228 33 17 51.515152
#> 4 Merza et al. 10.1016/j.dsx.2020.04.047 15 2 13.333333
#> 5 Menni et al. 1 10.1101/2020.04.05.20048421 579 344 59.412781
#> 6 Levinson et al. 10.1101/2020.04.11.20055483 42 15 35.714286
#> 7 Haehner et al. 10.1101/2020.04.27.20081356 34 22 64.705882
#> 8 Speth et al. 10.1177/0194599820929185 103 63 61.165049
#> 9 De Maria et al. 10.1002/jmv.25995. 95 48 50.526316
#> 10 Menni et al. 2 10.1038/s41591-020-0916-2 7178 4668 65.032042
#> 11 Yan et al. 1 10.1111/alr.22579. 59 40 67.796610
#> 12 Luers et al. 10.1093/cid/ciaa525 72 53 73.611111
#> 13 Roland et al. 10.1111/alr.22602. 145 95 65.517241
#> 14 Boscolo-Rizzo et al 10.1007/s00405-020-06066-9 54 34 62.962963
#> 15 Liu et al. 10.3390/ijerph17093311 321 42 13.084112
#> 16 Paderno et al. 10.1111/alr.22610. 508 284 55.905512
#> 17 Lee et al. 10.3346/jkms.2020.35.e174 3191 488 15.293012
#> 18 Lechien et al. 10.1007/s00405-020-05965-1 417 357 85.611511
#> 19 Gelardi et al. 10.23750/abm.v91i2.9524 72 42 58.333333
#> 20 Giacomelli et al. 10.1093/cid/ciaa330 59 14 23.728814
#> 21 Shoer et al. 10.1101/2020.05.18.20105569.this 498 136 27.309237
#> 22 Vaira et al. 3 10.1002/hed.26269 345 241 69.855072
#> 23 Bocksberger et al. 10.1007/s00106-020-00891-4 63 26 41.269841
#> 24 Mao 10.1101/2020.02.22.20026500 214 11 5.140187
#> 25 Spinato 10.1001/jama.2020.6771 202 130 64.356436
#> 26 Beltran-Corbellini 10.1007/s00405-020-05999-5 79 25 31.645570
#> 27 Trubiano et al. 10.1093/cid/ciaa655 28 7 25.000000
#> 28 Moein et al. 10.1111/alr.22587. 60 59 98.333333
#> 29 Hornuss et al. 10.1101/2020.04.28.20083311 45 38 84.444444
#> 30 Yan et al. 2 10.1111/alr.22592. 128 75 58.593750
#> 31 Klopfenstein et al. 10.1016/j.medmal.2020.04.006 114 54 47.368421
#> 32 Gudbjartsson 10.1056/NEJMoa2006100 1113 119 10.691824
#> 33 Wee et al 10.1007/s00405-020-05999-5 154 35 22.727273
#> 34 Dawson et al 10.1101/2020.05.13.20101006.t 42 18 42.857143
#> 35 Ji Yun Noh et al 10.1016/j.jinf.2020.05.035 199 52 26.130653
#> comments
#> 1 In the abstact of the paper it says 60 cases with varying degree of hyposmia and 2 patients with anosmia -- Results showed 44
#> 2 Preprint, only small part of whole paper
#> 3 Changed to 17 cases, peer reviewed
#> 4 Changed cases to 2; 4 patients had "smell and taste disorders", of which 2 had both and 2 had only taste
#> 5 looks good
#> 6 Changed cases to 14, looks good
#> 7 Changes case numbers to 22 and total to 34
#> 8 Looks good, measured olfactory dysfunction over phone
#> 9 Studied "loss of smell or taste", looks good
#> 10 Smartphone app tracking, Studied "Loss of smell and taste"
#> 11 looks good
#> 12 looks good
#> 13 Changed cases to 95
#> 14 Of those who were tested, I found 34/54 had an altered sense of taste/smnell
#> 15 Studied "loss of smell or taste"
#> 16 looks good
#> 17 studied "anosmia or ageusia"
#> 18 looks good
#> 19 I could not find this paper
#> 20 Changed to 14 cases
#> 21 Changed cases to 136
#> 22 Looks good, Measures Mild-servere hyponosmia + Anosmia
#> 23 Check on this; some english
#> 24 Good study; subjective
#> 25 Subjective
#> 26 Subjective
#> 27 looks good
#> 28 good
#> 29 Good
#> 30 68 ambulatory + 7 admitted of all covid positive; are we sure this does not overlap with Yan 1?
#> 31 Good
#> 32 Good
#> 33 <NA>
#> 34 <NA>
#> 35 <NA>
#> objectivity measured
#> 1 Objective Taste and Smell, Smell only
#> 2 Objective Smell only
#> 3 Objective Smell only
#> 4 Subjective Smell only
#> 5 Subjective Taste and Smell
#> 6 Subjective Smell only
#> 7 Subjective Smell only
#> 8 Subjective Smell only
#> 9 Subjective Taste and Smell
#> 10 Subjective Taste and Smell
#> 11 Subjective Smell only
#> 12 Subjective Smell only
#> 13 Subjective Taste or Smell
#> 14 Subjective Taste or Smell
#> 15 Subjective Taste or Smell
#> 16 Subjective Smell only
#> 17 Subjective Taste or Smell
#> 18 Subjective Smell only
#> 19 Subjective smell only, taste and smell
#> 20 Subjective Taste and Smell, Smell only
#> 21 Subjective Taste or smell
#> 22 Objective Smell
#> 23 Subjective Taste and/or Smell
#> 24 Subjective Smell only
#> 25 Subjective Taste or Smell
#> 26 Subjective Smell only
#> 27 Subjective Taste and Smell, Smell only
#> 28 Objective Smell
#> 29 Objective Smell
#> 30 Subjective Smell
#> 31 Subjective Smell
#> 32 Subjective Taste or Smell
#> 33 Subjective Smell or Taste
#> 34 Subjective Smell
#> 35 Subjective Smell only
#> testType
#> 1 Connecticut Chemosensory Clinical Research Center orthonasal olfaction test
#> 2 5 odor smell panel - no indication of validation - used test-retest to measure reliability
#> 3 Connecticut Chemosensory Clinical Research Center orthonasal olfaction test, self administered olfactory test
#> 4 Unknown, Hospital Reported
#> 5 Self Reported
#> 6 Self Reported
#> 7 Self Reported
#> 8 Self Reported
#> 9 Self Reported
#> 10 Self Reported
#> 11 Self Reported
#> 12 Self Reported
#> 13 Self Reported
#> 14 Self Reported
#> 15 Unknown, Hospital Reported
#> 16 Self Reported
#> 17 Self Reported
#> 18 Self Reported, Survey based on NHANES and sQOD-NS
#> 19 Self Reported, Not specified
#> 20 Self Reported
#> 21 Self Reported Survey
#> 22 Connecticut Chemosensory Clinical Research Center orthonasal olfaction test, home discrimination test, validated test from previoud Vaira et al study
#> 23 Sniffin Sticks Screening 12 Test and Taste Strips for some patients
#> 24 Self Reported, EHR Records
#> 25 SNOT-22
#> 26 Self Reported
#> 27 Self Reported
#> 28 University of Pennsylvania Smell Identification Test
#> 29 Sniffin Sticks
#> 30 Self Reported
#> 31 Self Reported
#> 32 Self Reported
#> 33 Self Reported, Unspecified
#> 34 Self Reported
#> 35 Self Reported
#> country
#> 1 Italy
#> 2 Sweden
#> 3 Italy
#> 4 Duhok, Iraqi Kurdistan
#> 5 UK and US
#> 6 Israel
#> 7 Germany
#> 8 Switzerland
#> 9 Italy
#> 10 UK and US
#> 11 US
#> 12 Germany
#> 13 US/Online Global
#> 14 Italy
#> 15 Taiwan
#> 16 Italy
#> 17 Korea
#> 18 Belgium, France, Spain, Italy
#> 19 Italy
#> 20 Italy
#> 21 Israel
#> 22 Italy
#> 23 Germany
#> 24 China
#> 25 Italy
#> 26 Spain
#> 27 Australia
#> 28 Iran
#> 29 Germany
#> 30 US
#> 31 France
#> 32 Iceland
#> 33 Singapore
#> 34 US
#> 35 South Korea
#> patientType
#> 1 18+, nasal swab confirmed
#> 2 18+, 16 COVID-Confirmed
#> 3 18+, nasal swab confirmed, home-quarantined
#> 4 All ages, COVID-Confirmed
#> 5 16-90, COVID-Confirmed
#> 6 15+, Hospitalized, nasal swab confirmed
#> 7 Patient showing symptoms, only COVID positive patients used
#> 8 All ages, COVID-Confirmed, Hospitalized or Treated at hospital
#> 9 All ages, COVID-Confirmed, home-quarantined
#> 10 16-90, COVID-Confirmed,unspecified hospitalization/quarantine
#> 11 COVID-Confirmed, home-quarantined
#> 12 COVID-Confirmed, home-quarantined
#> 13 COVID-19 Confirmed, online presumably home-quarantined
#> 14 Home-quarantined, COVID-positive included, Household contacts of patients from Spinato paper
#> 15 Unknown
#> 16 Hospitalized and home-quarantined
#> 17 Home-quarantined
#> 18 Home-quarantined, not specified
#> 19 Not specified
#> 20 Hospitalized
#> 21 Home-quarantined, not specified
#> 22 Home-quarantined hospital staff, and hospitalized patients
#> 23 hospitalized
#> 24 Hospitalized
#> 25 Home-quarantined
#> 26 Hospitalized?
#> 27 Unspecified, Clinically evaluated
#> 28 Hospitalized
#> 29 Hospitalized
#> 30 Hospitalized and home-quarantined
#> 31 Hospitalized and home-quarantined
#> 32 Unknown
#> 33 Hospitalized
#> 34 Home-quarantined
#> 35 Unspecified, Likely Hospitalized and Home-quarantined
# load metafor package
library(metafor)
# compute effect size
dat <- escalc(measure="PR", xi=xi, ni=ni, data=dat)
# split data into objective and subjective datasets
dat_split <- split(dat, dat$objectivity)
dat_objective <- dat_split[["Objective"]]
dat_subjective <- dat_split[["Subjective"]]
# random-effects model all studies
res_all <- rma(yi, vi, data=dat)
print(res_all, digits=2)
#>
#> Random-Effects Model (k = 35; tau^2 estimator: REML)
#>
#> tau^2 (estimated amount of total heterogeneity): 0.06 (SE = 0.01)
#> tau (square root of estimated tau^2 value): 0.24
#> I^2 (total heterogeneity / total variability): 99.28%
#> H^2 (total variability / sampling variability): 138.10
#>
#> Test for Heterogeneity:
#> Q(df = 34) = 7878.14, p-val < .01
#>
#> Model Results:
#>
#> estimate se zval pval ci.lb ci.ub
#> 0.49 0.04 11.88 <.01 0.41 0.57 ***
#>
#> ---
#> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
#>
# random-effects model objective
res_objective <- rma(yi, vi, data=dat_objective)
print(res_objective, digits=2)
#>
#> Random-Effects Model (k = 6; tau^2 estimator: REML)
#>
#> tau^2 (estimated amount of total heterogeneity): 0.03 (SE = 0.02)
#> tau (square root of estimated tau^2 value): 0.16
#> I^2 (total heterogeneity / total variability): 94.83%
#> H^2 (total variability / sampling variability): 19.35
#>
#> Test for Heterogeneity:
#> Q(df = 5) = 132.69, p-val < .01
#>
#> Model Results:
#>
#> estimate se zval pval ci.lb ci.ub
#> 0.75 0.07 10.70 <.01 0.61 0.89 ***
#>
#> ---
#> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
#>
# random-effects model subjective
res_subjective <- rma(yi, vi, data=dat_subjective)
print(res_subjective, digits=2)
#>
#> Random-Effects Model (k = 29; tau^2 estimator: REML)
#>
#> tau^2 (estimated amount of total heterogeneity): 0.05 (SE = 0.01)
#> tau (square root of estimated tau^2 value): 0.22
#> I^2 (total heterogeneity / total variability): 99.23%
#> H^2 (total variability / sampling variability): 129.15
#>
#> Test for Heterogeneity:
#> Q(df = 28) = 6391.95, p-val < .01
#>
#> Model Results:
#>
#> estimate se zval pval ci.lb ci.ub
#> 0.44 0.04 10.51 <.01 0.36 0.52 ***
#>
#> ---
#> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
#>