From 12/31/2020 to 10/5/2021, 1,447 high-risk patients with positive COVID-19 diagnosis received mAb infusion at FHCSD – these patients were eligible to participate in the survey study. Surveys were conducted from 6/8/2021 to 11/1/2021. Interviewers contacted 635 patients and 411 chose to participate. All 411 patients completed the Baseline Survey, 199 patients completed both D14 and D28+, and 212 patients completed D28+ only. There was variation in follow-up times from symptom onset to D14 (mean 19 days; SD 3; median 19; range 13 to 29) and D28+ (mean 72 days; SD 52; median 43; range 21 to 186).
Demographic characteristics of the study population are shown in Table 1. The study population was 53% female and 47% male. Participants ranged in age from 16 to 92 (mean age: 50 years). In terms of race, 61% of participants were white and 10% were non-white. In terms of ethnicity, 56% of participants were Latino/Hispanic and 16% were non-Latino/Hispanic. Forty-four percent were FHCSD patients. Most participants lived in either Central (40%) or South (30%) San Diego. With respect to vaccination status pre-infusion, 34% of patients were fully vaccinated, 2% were partially vaccinated, and 64% were unvaccinated or had unknown vaccination status. The preferred language of most patients was English (52%) or Spanish (43%).
Table 1. Demographic characteristics among patients who completed both D14 and D28+ Surveys vs only D28+ Survey*
|
Day 14 and Day 28 (N=199)
|
Day 28 Only (N=212)
|
Total (N=411)
|
Sex Assigned at Birth
|
|
|
|
Female
|
102 (51.3%)
|
117 (55.2%)
|
219 (53.3%)
|
Male
|
97 (48.7%)
|
95 (44.8%)
|
192 (46.7%)
|
Age (years)
|
|
|
|
Mean (SD)
|
49.1 (15.9)
|
50.8 (15.5)
|
50.0 (15.7)
|
Median [Min, Max]
|
49.0 [16.0, 88.0]
|
51.0 [18.0, 92.0]
|
50.0 [16.0, 92.0]
|
Age Group
|
|
|
|
<18
|
2 (1.0%)
|
0 (0%)
|
2 (0.5%)
|
18-29
|
20 (10.1%)
|
19 (9.0%)
|
39 (9.5%)
|
30-39
|
38 (19.1%)
|
30 (14.2%)
|
68 (16.5%)
|
40-49
|
41 (20.6%)
|
43 (20.3%)
|
84 (20.4%)
|
50-59
|
48 (24.1%)
|
62 (29.2%)
|
110 (26.8%)
|
60+
|
50 (25.1%)
|
58 (27.4%)
|
108 (26.3%)
|
Race
|
|
|
|
Non-white
|
15 (7.5%)
|
24 (11.3%)
|
39 (9.5%)
|
Unknown
|
67 (33.7%)
|
56 (26.4%)
|
123 (29.9%)
|
White
|
117 (58.8%)
|
132 (62.3%)
|
249 (60.6%)
|
Ethnicity
|
|
|
|
Latino/Hispanic
|
93 (46.7%)
|
138 (65.1%)
|
231 (56.2%)
|
Non-Latino/Hispanic
|
36 (18.1%)
|
30 (14.2%)
|
66 (16.1%)
|
Unknown
|
70 (35.2%)
|
44 (20.8%)
|
114 (27.7%)
|
San Diego Region
|
|
|
|
Central
|
75 (37.7%)
|
90 (42.5%)
|
165 (40.1%)
|
East
|
24 (12.1%)
|
33 (15.6%)
|
57 (13.9%)
|
North Central
|
28 (14.1%)
|
14 (6.6%)
|
42 (10.2%)
|
North Coastal
|
1 (0.5%)
|
2 (0.9%)
|
3 (0.7%)
|
North Inland
|
7 (3.5%)
|
4 (1.9%)
|
11 (2.7%)
|
Outside County
|
3 (1.5%)
|
1 (0.5%)
|
4 (1.0%)
|
South
|
59 (29.6%)
|
66 (31.1%)
|
125 (30.4%)
|
Unknown
|
2 (1.0%)
|
2 (0.9%)
|
4 (1.0%)
|
FHCSD Patient Status
|
|
|
|
FHCSD
|
51 (25.6%)
|
128 (60.4%)
|
179 (43.6%)
|
Non-FHCSD
|
148 (74.4%)
|
84 (39.6%)
|
232 (56.4%)
|
Primary Care Provider
|
|
|
|
FHCSD
|
51 (25.6%)
|
128 (60.4%)
|
179 (43.6%)
|
Kaiser Permanente
|
49 (24.6%)
|
17 (8.0%)
|
66 (16.1%)
|
Other
|
73 (36.7%)
|
41 (19.3%)
|
114 (27.7%)
|
San Ysidro Health
|
3 (1.5%)
|
3 (1.4%)
|
6 (1.5%)
|
Scripps
|
8 (4.0%)
|
12 (5.7%)
|
20 (4.9%)
|
Sharp
|
9 (4.5%)
|
10 (4.7%)
|
19 (4.6%)
|
UCSD
|
1 (0.5%)
|
1 (0.5%)
|
2 (0.5%)
|
VA
|
5 (2.5%)
|
0 (0%)
|
5 (1.2%)
|
Vaccination Status Pre-infusion
|
|
|
Fully Vaccinated
|
98 (49.2%)
|
43 (20.3%)
|
141 (34.3%)
|
Partially Vaccinated
|
1 (0.5%)
|
8 (3.8%)
|
9 (2.2%)
|
Unvaccinated/ Unknown
|
100 (50.3%)
|
161 (75.9%)
|
261 (63.5%)
|
Language
|
|
|
|
English
|
114 (57.3%)
|
98 (46.2%)
|
212 (51.6%)
|
Spanish
|
72 (36.2%)
|
103 (48.6%)
|
175 (42.6%)
|
Unknown
|
13 (6.5%)
|
8 (3.8%)
|
21 (5.1%)
|
Other
|
0 (0%)
|
3 (1.4%)
|
3 (0.7%)
|
* All patients also completed the Baseline Survey.
Abbreviations: VA, Veterans Affairs; SD, standard deviation; FHCSD, Family Health Centers of San Diego; UCSD, University of California, San Diego.
At baseline, the five most prevalent symptoms were cough (70%), fever (50%), headache (44%), fatigue (40%), and “other” (40%) (Figure 1). At D28+, seven symptoms had a prevalence over 20% including cough (29%), headache (28%), myalgia (26%), anosmia (24%), dyspnea (23%), post-exertion polypnea (21%), and ageusia (20%) (Table 2). At D28+, the most common mild symptoms were cough (20%), anosmia (20%), ageusia (18%), headache (17%), and myalgia (15%). The most common moderate symptoms were headache (9%), myalgia (9%), cough (8%), dyspnea (8%), and post-exertion polypnea (7%). The most common severe symptoms were anosmia (3%), diaphoresis (3%), hair loss (2%), diarrhea (2%), ageusia (2%), and nausea (2%) (Table 2).
Data on the presence of the following symptoms were also collected, but no patients reported having them at baseline: thromboembolism, chronic kidney disease, need for oxygen, loss of hearing, and depression.
Table 2. Severity and prevalence of symptoms at D28+ Survey (N=411)
Symptom
|
Mild
|
Moderate
|
Severe
|
Overall prevalence*
|
Cough
|
20.4%
|
7.8%
|
1.2%
|
29.4%
|
Headache
|
17.3%
|
9.0%
|
1.7%
|
28.0%
|
Myalgia
|
15.3%
|
8.8%
|
1.9%
|
26.0%
|
Anosmia
|
20.2%
|
NA
|
3.4%
|
23.6%
|
Dyspnea
|
14.6%
|
7.8%
|
0.7%
|
23.1%
|
Post-exertion polypnea
|
12.7%
|
7.4%
|
1.0%
|
21.1%
|
Ageusia
|
18.0%
|
NA
|
2.2%
|
20.2%
|
Cognitive disturbances
|
11.2%
|
6.6%
|
1.7%
|
19.5%
|
Chest pain/tightness
|
12.4%
|
4.4%
|
0.7%
|
17.5%
|
Arthralgia
|
7.1%
|
6.1%
|
2.0%
|
15.1%
|
Hair loss
|
6.6%
|
5.1%
|
2.4%
|
14.1%
|
Palpitations
|
6.3%
|
5.1%
|
0.7%
|
12.2%
|
Nausea
|
6.1%
|
3.2%
|
2.2%
|
11.4%
|
Diaphoresis
|
4.6%
|
3.4%
|
2.7%
|
10.7%
|
Rhinorrhea
|
8.0%
|
2.2%
|
0.2%
|
10.5%
|
Pharyngitis
|
7.5%
|
1.5%
|
0.5%
|
9.5%
|
Ear pain
|
6.6%
|
2.2%
|
0.7%
|
9.5%
|
Diarrhea
|
3.7%
|
2.2%
|
2.4%
|
8.3%
|
Rash
|
3.9%
|
2.0%
|
0.5%
|
6.3%
|
Chills/shivering
|
4.6%
|
1.2%
|
0.2%
|
6.1%
|
Need for oxygen
|
2.4%
|
2.4%
|
1.0%
|
5.9%
|
Loss of hearing
|
4.0%
|
NA
|
0.0%
|
4.0%
|
Fever
|
2.4%
|
0.2%
|
0.5%
|
3.2%
|
Vomiting
|
1.9%
|
0.0%
|
0.5%
|
2.4%
|
Thromboembolism
|
0.5%
|
1.0%
|
0.2%
|
1.7%
|
CKD
|
0.7%
|
0.2%
|
0.2%
|
1.2%
|
* Overall prevalence represents the sum of mild, moderate, and severe categories.
For diarrhea and vomiting, 1-2 was classified as mild, 3-4 as moderate, and 5 or more as severe. For loss of hearing, smell, and taste, same as usual was considered none, less than usual was considered mild, and no sense was considered severe.
Abbreviations: CKD, chronic kidney disease; NA, not applicable
Longitudinal findings
Among patients who completed all three surveys (N=199), at D14, the most prevalent symptoms were cough (55%), anosmia (45%), ageusia (41%), headache (32%), and dyspnea (31%). At D28+, the most prevalent symptoms were cough (32%), headache (24%), cognitive disturbances (23%), myalgia (23%), ageusia (22%), and post-exertion polypnea (22%) (Figure 2). The largest changes in prevalence over time (from Baseline to D28+) occurred in the following symptoms: fever (-50%), cough (-39%), cognitive disturbances (+22%), and post-exertion polypnea (+22%). The percentage of patients with no symptoms increased from D14 (16%) to D28+ (40%) (Figure 3). The percentages of patients with mild and moderate symptoms each decreased from D14 (33% and 41%) to D28+ (26% and 21%). At D14, 10% of patients had severe symptoms compared to 13% at D28+.
Among patients who completed the Baseline, D14, and D28+ surveys (N=199). Follow-up times for the surveys varied.
At baseline, no patients reported having the following symptoms: thromboembolism, CKD, palpitations, arthralgia, loss of hearing, need for oxygen, post-exertion polypnea, and rash.
Abbreviations: CKD, chronic kidney disease.
Among patients who completed the Baseline, D14, and D28+ Surveys. Follow-up times for the surveys varied.
“None” defined as no mild, moderate, or severe symptoms; “Mild” defined as ≥1 mild symptom (and no moderate or severe symptoms); “Moderate” defined as ≥1 moderate symptom (and no severe symptoms); “Severe” defined as ≥1 severe symptom.
The following symptoms were excluded from calculations because of how their severity levels were classified: vomiting, diarrhea, loss of hearing, anosmia, and ageusia.
Additional longitudinal outcomes included the change in PHQ-2 Total and PROMIS-29 Domain Scores over time. Mean PHQ-2 Total and PROMIS-29 Domain scores at D28+ are shown in Table 3.
Table 3. PHQ-2 and PROMIS-29 Domain Scores at D28+ Survey (N=411)
Measure*
|
Mean (SD)
|
PHQ-2 Total Score
|
0.5 (1.1)
|
PROMIS-29 Domain ScoresƗ
|
|
Physical Function
|
18.1 (3.2)
|
Fatigue
|
7.9 (4.2)
|
Pain Interference
|
6.2 (3.7)
|
Anxiety
|
4.4 (2.5)
|
Sleep Disturbance
|
9.3 (4.0)
|
Depression
|
5.1 (2.6)
|
Social Roles
|
13.6 (2.8)
|
* PHQ-2 Total Score range is 0 to 6; PROMIS-29 Domain Score ranges are 0 to 20.
Ɨ Higher scores indicate better functioning on Physical Function and Social Roles Domains. Higher scores represent worse symptomology on Fatigue, Pain Interference, Anxiety, Sleep Disturbance, and Depression Domains.
Abbreviations: PHQ-2, Patient Health Questionnaire-2; PROMIS-29, Patient-Reported Outcomes Measurement Information System-29.
Mean PHQ-2 Total Scores decreased from D14 to D28+ (Figure 4). For the two PROMIS-29 Domains where higher scores indicate better functioning, mean scores increased from D14 to D28+ for Physical Function but decreased for Social Roles (Figure 5). For four out of five of the PROMIS-29 Domains where higher scores represent worse symptomology (Anxiety, Fatigue, Sleep Disturbance, and Pain Interference), mean scores decreased from D14 to D28+. Mean scores for the PROMIS-29 Depression Domain increased from D14 to D28+.
PHQ-2 Total Score range is 0 to 6.
Abbreviations: PHQ-2, Patient Health Questionnaire-2.
PROMIS-29 Domain Score ranges are 0 to 20. Higher scores indicate better functioning on Physical Function and Social Roles Domains. Higher scores represent worse symptomology on Anxiety, Depression, Fatigue, Sleep Disturbance, and Pain Interference Domains.
Abbreviations: PROMIS-29, Patient-Reported Outcomes Measurement Information System-29.
Time to mAb infusion
Results of the tests of association between TTI and outcomes of interest at D28+ showed that TTI was not statistically significantly associated with any outcome (symptom count category, symptom severity category, PCC status [by any of 6 PCC definitions], PHQ-2 Total Score, or any PROMIS-29 Domain Score (Table 4).
Table 4. Tests of association between TTI and outcomes at D28+ Survey (N=409)
Predictor variable
|
Outcome variable
|
P-value
|
Pearson’s Chi-squared tests
|
Time to infusion
|
Symptom count category
|
0.6569
|
Time to infusion
|
Symptom severity category
|
0.3205
|
Time to infusion
|
PCC status (Definition 1)
|
0.8362 Ɨ
|
Time to infusion
|
PCC status (Definition 2)
|
0.7404 Ɨ
|
Time to infusion
|
PCC status (Definition 3)
|
0.8200 Ɨ
|
Time to infusion
|
PCC status (Definition 4)
|
0.3765 Ɨ
|
Time to infusion
|
PCC status (Definition 5)
|
0.1582 Ɨ
|
Time to infusion
|
PCC status (Definition 6)
|
0.7731 Ɨ
|
Independent 2-sample t-tests
|
Time to infusion
|
PHQ-2 Total Score
|
0.2991
|
Time to infusion
|
PROMIS-29 Physical Function
|
0.4223 ƗƗ
|
Time to infusion
|
PROMIS-29 Fatigue
|
0.1553 ƗƗ
|
Time to infusion
|
PROMIS-29 Pain Interference
|
0.02793 ƗƗ
|
Time to infusion
|
PROMIS-29 Anxiety
|
0.8561
|
Time to infusion
|
PROMIS-29 Sleep Disturbance
|
0.07808 ƗƗ
|
Time to infusion
|
PROMIS-29 Depression
|
0.9668
|
Time to infusion
|
PROMIS-29 Social Roles
|
0.335 ƗƗ
|
Ɨ P-value for Pearson's Chi-squared test with Yates' continuity correction.
ƗƗ P-value from Welch’s t-test due to unequal variances. Otherwise, Student’s t-test (assumes equal variances) performed.
Pearson’s Chi-squared tests performed for categorical outcomes; 2-sample independent t-tests performed for continuous outcomes. Significance level (alpha) of 0.05 for Pearson’s Chi-squared tests. Bonferroni correction used for 8 independent 2-sample t-tests – significance level (alpha) of 0.00625.
Symptom count categories defined by total number of symptoms at Day 28 Survey (0, 1-2 , 3-4 , or 5-20).
Symptom severity categories defined by total number of moderate/severe symptoms at Day 28 Survey (0, 1, 2 , ≥3).
Time to infusion defined as the time between symptom onset and mAb infusion. Treated as a binary variable with groups: 1-6 days (n=203) vs 7-10 days (n=206).
Original N=411; tests exclude 2 patients with outlier times to infusion.
Abbreviations: PHQ-2, Patient Health Questionnaire-2; PROMIS-29, Patient-Reported Outcomes Measurement Information System-29; TTI, time to mAb infusion.
Prevalence of PCC
Based on Definition 1 (≥1 symptom at D28+), the prevalence of PCC in the study population was 69%; based on Definition 2 (≥3 symptoms), the prevalence was 44%; based on Definition 3 (≥5 symptoms), the prevalence was 30% (Table 5). Based on Definition 4 (≥1 moderate or severe symptom), the prevalence of PCC was 47%; based on Definition 5 (≥2 moderate or severe symptoms), the prevalence was 36%; based on Definition 6 (≥3 moderate or severe symptoms), the prevalence was 23%.
Table 5. Prevalence of PCC at D28+ Survey (N=403)*Ɨ
PCC Definition
|
Prevalence of PCC, n (%)
|
Definitions based on number of symptoms
|
|
PCC Definition 1 (≥ 1 symptom)
|
277 (68.7%)
|
PCC Definition 2 (≥ 3 symptoms)
|
178 (44.2%)
|
PCC Definition 3 (≥ 5 symptoms)
|
121 (30.0%)
|
Definitions based on severity of symptoms
|
|
PCC Definition 4 (≥ 1 moderate or severe symptom)
|
188 (46.5%)
|
PCC Definition 5 (≥ 2 moderate or severe symptoms)
|
145 (35.9%)
|
PCC Definition 6 (≥ 3 moderate or severe symptoms)
|
93 (23.0%)
|
* Based on 28 days or more post-symptom onset; follow-up times for D28+ survey varied.
Ɨ Original N=411; calculations exclude 8 patients with time from symptom onset to Day 28 Survey <28 days.
Abbreviations: PCC, post-COVID conditions.
Time to mAb infusion as a predictor of PCC
The Poisson regression produced crude and adjusted prevalence ratios of PCC based on TTI. In the crude (unadjusted) model, patients with TTI of 7 to 10 days had 1.02 times the risk of developing PCC compared to patients with TTI of 1 to 6 days (95% confidence interval [CI]: 0.90 to 1.16; P=0.75) (Table 6). In Model 1, which adjusted for demographic characteristics only, the prevalence ratio was 1.02 (95% CI: 0.89 to 1.16; P=0.79. In Model 2, which adjusted for both demographic and clinical characteristics, the prevalence ratio was 1.01 (95% CI: 0.89 to 1.16; P=0.84). Time to mAb infusion was not statistically significantly associated with PCC status at D28+ in any of the models.
Table 6. Prevalence ratio of PCC for TTI of 7-10 days versus 1-6 days (N=409)
Model
|
PR
|
95% CI
|
P-value
|
Crude (unadjusted)
|
1.02
|
0.90 to 1.16
|
0.75
|
Model 1: crude + demographic characteristics*
|
1.02
|
0.89 to 1.16
|
0.79
|
Model 2: model 1 + clinical characteristicsƗ
|
1.01
|
0.89 to 1.16
|
0.84
|
Note: Modified Poisson regressions with robust standard errors performed. Results are for time to infusion of 7-10 days (n=206) compared to the reference group, time to infusion of 1-6 days (n=203). PCC was defined by PCC Definition 1 (≥ 1 symptom at D28+ Survey). Original N=411; analysis excludes 2 patients with outlier times to infusion.
* Model 1 was adjusted for demographic characteristics (age, sex assigned at birth, race, and ethnicity) only.
Ɨ Model 2 was adjusted for demographic characteristics (age, sex assigned at birth, race, and ethnicity) plus clinical characteristics (vaccination status pre-infusion, mAb infusion type, and number of symptoms at Baseline Survey).
Abbreviations: CI, confidence interval; PCC, post-COVID conditions; PR, prevalence ratio; TTI, time to mAb infusion.