Abstract
Background
The clinical consequences of repeated severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection are not clear, especially as they relate to long-term symptoms after infection. We analyzed data collected for the Innovative Support for Patients with SARS-CoV-2 Infections Registry (INSPIRE) to determine whether reinfection changes the likelihood of symptoms 3–6 months after reinfection compared with the likelihood in individuals experiencing a single infection.
Methods
Individuals reporting a single SARS-CoV-2 infection or a single reinfection were included in this analysis. A positive SARS-CoV-2 test occurring ≥90 days after a first infection was considered a reinfection. Outcomes included severe fatigue (fatigue severity score ≥25) and the presence of organ system symptoms 3–6 months after the last infection.
Results
The analysis included 886 individuals, 415 (46.8%) of whom experienced reinfection. For individuals who experienced their first infections in either the pre-Delta or Delta periods, the odds of having ≥3 symptoms 3–6 months after their most recent infection was lower in those reinfected than those with a single infection (weighted adjusted odds ratio, 0.45 [95% confidence interval, .21–.95] and 0.51 [.32–.79], respectively). However, in individuals reporting their first infection during the Omicron wave, the odds of reporting ≥3 symptoms after the most recent infection was higher in those reinfected than in those with a single infection (weighted adjusted odds ratio, 1.54 [95% confidence interval, 1.02–2.34]).
Conclusions
The timing of initial infection, reinfection, and the variants involved may play important roles in longer-term clinical outcomes. Repeated infection with Omicron variants may increase the risk of long-term symptoms.
Keywords: COVID-19, reinfection, SARS-CoV-2, long COVID, post-COVID-19 syndrome
The clinical consequences of repeated severe acute respiratory syndrome coronavirus 2 infection are unclear, especially as they relate to long-term symptoms. The timing of initial infection, reinfection, and the variants involved may play important roles in longer-term clinical outcomes.
Reinfections with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) have increased since the emergence of Omicron variants [1, 2]. Despite this, the importance of reinfection in both acute presentations of coronavirus disease 2019 (COVID-19) and post–COVID-19 syndrome, or long COVID, is poorly understood. While studies have suggested that there is a lower risk of severe disease during the acute phase of reinfection with SARS-CoV-2, infection-induced protection wanes over time and may be partially evaded by emerging variants [3–5]. It is not well understood how reinfection affects the likelihood of long COVID, defined as signs, symptoms, and conditions 3 months after the inciting SARS-CoV-2 infection [6, 7].
Given that most of the US adult population has had ≥1 infection [8], and new variants continue to emerge [9], reinfections will remain an ongoing issue. The Innovative Support for Patients with SARS-CoV-2 Infections Registry (INSPIRE) was designed to prospectively assess the longitudinal symptoms of adults who test positive for SARS-CoV-2 [10]. Collecting data on symptoms and repeated infections, INSPIRE allows for the characterization of the effect of reinfection on long-term symptomatic outcomes.
METHODS
INSPIRE Study Design and Data Collection
This was a secondary analysis of the INSPIRE dataset focusing on outcomes after reinfection. INSPIRE, described in detail elsewhere [10], is a longitudinal study (NCT04610515) conducted across 8 healthcare systems in the United States. Enrolled adults had symptoms suggestive of acute SARS-CoV-2 and were tested within the preceding 42 days with a Food and Drug Administration–approved/authorized molecular or antigen-based assay. Participants were followed up using a quarterly questionnaire for up to 18 months. Enrollment was between 7 December 2020 and 29 August 2022, with quarterly follow-up through 28 February 2023 and the final questionnaire completed on 15 March 2023. Symptomatic outcomes were assessed using the US Centers for Disease Control and Prevention (CDC) Person Under Investigation symptom list as well as the CDC Short Symptoms Screener [10]. Data on repeated SARS-CoV-2 testing and reinfection were collected as part of each questionnaire administration. For both initial infection and reinfection, variant periods were defined by dates with ≥50% of dominant strain, as described elsewhere (Supplementary Table 1) [11].
Definitions of Single Infection and Reinfection
Individuals who had a verified positive SARS-CoV-2 test result at the time of enrollment into the INSPIRE cohort and who completed survey questions allowing for the evaluation of repeated infection and symptomatic outcomes 3–6 months after their last infection were eligible for inclusion in this analysis. We did not have adequate follow-up time to evaluate outcomes in participants with multiple reinfections.
A participant was considered to have a single infection if they had a positive test result at enrollment and no repeated positive results >30 days after the initial first positive result. The first positive test date was considered the start date of the initial disease episode. A participant was considered to have a reinfection if they had a positive SARS-CoV-2 test result at enrollment and a self-reported positive result ≥90 days after the date of their first positive test result. The reported test date for the repeated infection was considered the start date of the reinfection episode.
To account for ongoing viral shedding [12], any repeated positive test within 30 days of a previous infection was considered to represent the preceding infection. Individuals reporting another positive test result >30 days but <90 days after the initial infection were excluded, as it was unclear whether the new positive test reflected prolonged shedding after their first infection or an early reinfection. However, these individuals were included in the reinfection group if they reported a positive test ≥90 days after their initial infection. The 90-day cutoff for reinfection is consistent with other studies on reinfection [13].
Outcomes
Because symptoms were assessed every 3 months and reinfection could occur at any time, the evaluated outcome was the presence of symptoms in the survey corresponding to the time window 3–6 months after a single infection or reinfection (Figure 1). Our primary outcomes were the presence of severe fatigue, defined as a fatigue severity score ≥25, which was calculated as described elsewhere [14], and the presence of multiple organ system symptoms (≥3 total symptoms). Secondary outcomes included the presence of individual symptoms.
Figure 1.
Schematic of study survey timing and comparison groups. This analysis lined up the first reinfection episode in reinfected individuals with the sole infection for those with a single infection episode and compared outcomes in the 3–6 months following the last infection. To compare outcomes, we stratified by the variant period of the first infection. Comparison groups consisted of individuals with a single variant for those with a single infection versus the variant of the first infection followed by (designated by an arrow) the variant of the second infection for those in the reinfection group. This created comparison profiles as summarized in the figure: single pre-Delta infection versus initial pre-Delta infection followed by reinfection with pre-Delta, Delta, or Omicron (pre-Delta vs pre-Delta → pre-Delta || Delta || Omicron); single Delta infection versus initial Delta infection followed by reinfection with Delta or Omicron (Delta vs Delta → Delta || Omicron), and single Omicron infection compared with 2 Omicron infections (Omicron vs Omicron → Omicron). We also compared outcomes in individuals who had a single Omicron infection with those in individuals who had an initial Delta infection followed by an Omicron reinfection (Omicron vs Delta → Omicron).
Analytic Methods
We conducted a bivariate analysis to assess whether participant characteristics in the single-infection and reinfection groups differed significantly. We used the Mann-Whitney U test for continuous variables and the χ2 test for categorical variables. For categorical variables with ≥3 categories, multiple comparisons were not conducted because the significance of the overall association was sufficient for both descriptive purposes and selecting variables for the subsequent analytic steps. To evaluate the balance in the baseline characteristics between infection groups, we calculated the standardized mean difference to assess the difference between the 2 group means for each characteristic (Supplementary Figure 1). To mitigate bias arising from significant differences in certain baseline characteristics between groups that were associated with outcomes simultaneously, we used inverse propensity score weighting techniques with trimming to balance the baseline characteristics between infection groups [15].
Because the variant at the time of initial infection was a strong predictor of outcome in our exploratory analyses and showed a strong imbalance between infection groups, we stratified by the variant period of first infection, creating 3 models, one for each of the 3 variant periods (Figure 1). These comparison models are (1) single pre-Delta infection versus initial pre-Delta infection followed by reinfection with pre-Delta, Delta, or Omicron (referred to throughout as “pre-Delta vs pre-Delta → pre-Delta || Delta || Omicron”); (2) single Delta infection versus initial Delta infection followed by reinfection with Delta or Omicron (Delta vs Delta → Delta || Omicron); and (3) single Omicron infection versus initial Omicron infection followed by reinfection with Omicron (Omicron vs Omicron → Omicron). To further assess the impacts of Omicron reinfection, we created a fourth model comparing outcomes 3–6 months after the last infection in individuals who had a single Omicron infection and those with a first infection during the Delta period and a reinfection during the Omicron period (Omicron vs Delta → Omicron).
We summarized the weighted prevalences for each symptom by infection group stratified by variant of first infection. We ran weighted multivariable logistic models incorporating inverse propensity score weighting for each symptom stratified by initial infection variant. All models adjusted for the following baseline characteristics: demographic characteristics (age, sex/gender, education, and family income), social determinants of health, tobacco and substance use, preexisting health conditions, and COVID-19 vaccination status (vaccinated or not vaccinated before the index SARS-CoV-2 test). When the variant for the second infection in the reinfection group included >1 possibility, we controlled for the variant period at time of reinfection (pre-Delta, Delta, or Omicron). We included interactions between infection groups and variant periods to account for the difference in the effect of reinfection among variant periods. Variable subgroups were collapsed to create larger subgroups, as described elsewhere [16]. The weighted adjusted odds ratios (w-aORs) were computed to assess the risk ratio of each symptom between infection groups by variant period. We used SAS 9.4 software (SAS Institute) for statistical analyses. Given the exploratory nature of this study, multiplicity adjustments were not performed when testing the statistical significance between infection groups across symptoms. All tests were 2 sided with a significance threshold of P = .05.
RESULTS
Study Population
Of the 4547 participants who tested positive for SARS-CoV-2 at enrollment, 2330 submitted follow-up surveys in the 3–6-month period following their last infection and were evaluated for inclusion in the analysis (Figure 2). Of these, 1444 were excluded because they did not complete survey questions to allow for evaluation of symptoms (characteristics of included and excluded individuals are compared in Supplementary Table 2). Of the 886 included individuals, 415 (46.8%) experienced a reinfection.
Figure 2.
Study flow chart showing numbers of individuals enrolled in the Innovative Support for Patients with SARS-COV-2 Infections Registry (INSPIRE) cohort, those evaluated for this analysis, and final numbers of individuals included in this analysis. Variant profiles are shown for both single-infection and reinfection groups. Abbreviation: SARS-CoV-2, severe acute respiratory syndrome coronavirus 2.
Characteristics of Individuals Reporting Single Infections and Reinfections
Individuals with a single infection were enrolled later in the study timeline: 77.9% of single infections (367 of 471) occurred during the Omicron period (Table 1 and Supplementary Figure 2). Most individuals with reinfection experienced their first infection during the Delta period (249 of 415 [60.0%]) and their second during the Omicron period (388 of 415 [93.5%]); 19.5% of individuals with reinfection (81 of 415) had both their initial infection and their reinfection during the Omicron period (Figure 2 and Supplementary Figure 3). Single Omicron infections occurred slightly earlier than Omicron reinfections (Supplementary Figure 4). In the reinfection group, the average intervals between first and second infections were 353, 250, and 165 days for those who experienced their first infection in the pre-Delta, Delta, and Omicron periods respectively (P < .001) (Supplementary Table 3).
Table 1.
Characteristics of Single-Infection and Reinfection Groups
| Characteristic | SARS-CoV-2 Infection or Reinfection, No. (Observed Rate, %) | P Valueb | |
|---|---|---|---|
| Single Infection (n = 471) | Reinfectiona (n = 415) |
||
| Sociodemographic | |||
| Age at enrollment, y | |||
| 18–34 | 203 (43.1) | 189 (45.5) | .91 |
| 35–49 | 157 (33.3) | 131 (31.6) | |
| 50–64 | 77 (16.3) | 69 (16.6) | |
| ≥65 | 30 (6.4) | 25 (6.0) | |
| Missingc | 4 (0.8) | 1 (0.2) | |
| Sex/gender | |||
| Female | 280 (59.4) | 280 (67.5) | .03 |
| Male | 173 (36.7) | 123 (29.6) | |
| Transgender/nonbinary/other | 9 (1.9) | 4 (1.0) | |
| Missing | 9 (1.9) | 8 (1.9) | |
| Ethnicity | |||
| Not of Hispanic, Latin, or Spanish origin | 405 (86.0) | 353 (85.1) | .88 |
| Of Hispanic, Latin, or Spanish origin | 60 (12.7) | 55 (13.3) | |
| Missing | 6 (1.3) | 7 (1.7) | |
| Race | |||
| White | 355 (75.4) | 294 (70.8) | .02 |
| Black or African American | 19 (4.0) | 34 (8.2) | |
| Asian | 53 (11.3) | 42 (10.1) | |
| Other/multiple | 29 (6.2) | 37 (8.9) | |
| Missing | 15 (3.2) | 8 (1.9) | |
| Educational attainment | |||
| Less than high school diploma | 4 (0.8) | 4 (1.0) | <.001 |
| High school graduate or GED | 22 (4.7) | 27 (6.5) | |
| Some college but did not complete degree | 46 (9.8) | 80 (19.3) | |
| College degree (2 y) | 29 (6.2) | 29 (7.0) | |
| College degree (4 y) | 152 (32.3) | 135 (32.5) | |
| More than 4-y college degree | 208 (44.2) | 132 (31.8) | |
| Missing | 10 (2.1) | 8 (1.9) | |
| Marital status | |||
| Never married | 170 (36.1) | 144 (34.7) | .80 |
| Married/living with a partner | 262 (55.6) | 232 (55.9) | |
| Divorced/widowed/separated | 39 (8.3) | 39 (9.4) | |
| Family income (prepandemic) | |||
| <$10 000 | 17 (3.6) | 14 (3.4) | .06 |
| $10 000–$35 000 | 47 (10.0) | 47 (11.3) | |
| $35 000 to <$50 000 | 36 (7.6) | 51 (12.3) | |
| $50 000 to <$75 000 | 53 (11.3) | 59 (14.2) | |
| ≥$75 000 | 282 (59.9) | 223 (53.7) | |
| Prefer not to answer/missing | 36 (7.6) | 21 (5.1) | |
| Employment at time of initial infection | |||
| Not employed | 75 (15.9) | 57 (13.7) | .13 |
| Employed, identifies as not essential or non-healthcare worker | 212 (45.0) | 168 (40.5) | |
| Employed, identifies as essential or healthcare worker | 184 (39.1) | 190 (45.8) | |
| Employed before coronavirus outbreak | 396 (84.1) | 358 (86.3) | .41 |
| Working in a healthcare setting at time of initial infection | |||
| No | 309 (65.6) | 282 (68.0) | .87 |
| Yes | 87 (18.5) | 76 (18.3) | |
| Missing | 75 (15.9) | 57 (13.7) | |
| Self-identifies as healthcare-related essential worker | |||
| No | 294 (62.4) | 237 (57.1) | .02 |
| Yes | 102 (21.7) | 120 (28.9) | |
| Missing | 75 (15.9) | 58 (14.0) | |
| Has the COVID-19 pandemic caused you/your immediate family financial difficulties? | |||
| Not at all | 276 (58.6) | 214 (51.6) | .16 |
| A little | 134 (28.5) | 135 (32.5) | |
| Quite a bit | 35 (7.4) | 33 (8.0) | |
| Very much | 26 (5.5) | 33 (8.0) | |
| In the past month, how often were you worried that your food would run out before you got money to buy more? | |||
| Never | 413 (87.7) | 348 (83.9) | .25 |
| Sometimes | 40 (8.5) | 48 (11.6) | |
| Often | 18 (3.8) | 19 (4.6) | |
| In the past month, has the electric, gas/oil, or water company shut off service or threatened to shut off service in your home? | |||
| No | 456 (96.8) | 396 (95.4) | .37 |
| Threatened to shut off services | 15 (3.2) | 18 (4.3) | |
| Already shut off services | 0 (0.0) | 1 (0.2) | |
| Reports challenges with reliable transportation in past months | |||
| Medical transportation | 9 (1.9) | 12 (2.9) | .46 |
| General transportation | 14 (3.0) | 12 (2.9) | >.99 |
| Health insurance | |||
| Private and public | 10 (2.1) | 18 (4.3) | .002 |
| Private only | 380 (80.7) | 302 (72.8) | |
| Public only | 56 (11.9) | 80 (19.3) | |
| None | 25 (5.3) | 15 (3.6) | |
| Tobacco, alcohol, and drug use in past 12 mo | |||
| Tobacco use | |||
| Daily or near daily | 25 (5.3) | 29 (7.0) | <.001 |
| Weekly | 3 (0.6) | 16 (3.9) | |
| Monthly | 3 (0.6) | 11 (2.7) | |
| Less than monthly | 11 (2.3) | 25 (6.0) | |
| Not at all | 429 (91.1) | 334 (80.5) | |
| Binge drinking | |||
| Daily or near daily | 4 (0.8) | 6 (1.4) | .04 |
| Weekly | 38 (8.1) | 49 (11.8) | |
| Monthly | 49 (10.4) | 63 (15.2) | |
| Less than monthly | 123 (26.1) | 94 (22.7) | |
| Not at all | 257 (54.6) | 203 (48.9) | |
| Marijuana use | |||
| Daily or near daily | 17 (3.6) | 25 (6.0) | .02 |
| Weekly | 19 (4.0) | 16 (3.9) | |
| Monthly | 11 (2.3) | 22 (5.3) | |
| Less than monthly | 45 (9.6) | 52 (12.5) | |
| Not at all | 379 (80.5) | 300 (72.3) | |
| Other drug use | |||
| Weekly or more | 3 (0.6) | 6 (1.4) | .04 |
| Monthly or less | 25 (5.3) | 38 (9.2) | |
| Not at all | 443 (94.1) | 371 (89.4) | |
| Clinical details, conditions, and vaccination | |||
| Hospitalized during initial infection | |||
| No | 447 (94.9) | 343 (82.7) | .08 |
| Yes | 14 (3.0) | 21 (5.1) | |
| Missing | 10 (2.1) | 51 (12.3) | |
| Preexisting conditions | |||
| Asthma (moderate or severe) | 54 (11.5) | 52 (12.7) | .68 |
| Kidney disease | 3 (0.6) | 6 (1.5) | .38 |
| Emphysema or COPD | 3 (0.6) | 3 (0.7) | >.99 |
| Heart conditions (eg, coronary artery disease, heart failure, or cardiomyopathy) | 5 (1.1) | 9 (2.2) | .29 |
| Diabetes mellitus | 17 (3.6) | 16 (3.9) | .98 |
| Hypertension or high blood pressure | 55 (11.7) | 47 (11.4) | .98 |
| Liver disease | 2 (0.4) | 3 (0.7) | .88 |
| Overweight or obesity | 134 (28.6) | 98 (23.8) | .13 |
| Smoking (currently smoking any type of tobacco, including smokeless tobacco) | 17 (3.6) | 17 (4.1) | .83 |
| Other | 73 (15.6) | 50 (12.2) | .18 |
| Missing (nonresponders to 3-mo follow-up survey)d | 2 (0.4) | 4 (1.0) | .57 |
| Location of COVID-19 test for first infection | |||
| At-home testing kit | 103 (21.9) | 20 (4.8) | <.001 |
| Tent/drive-up testing site | 255 (54.1) | 252 (60.7) | |
| Clinic (including urgent care) | 54 (11.5) | 71 (17.1) | |
| Hospital | 25 (5.3) | 25 (6.0) | |
| Emergency department | 4 (0.8) | 19 (4.6) | |
| Other | 30 (6.4) | 28 (6.7) | |
| Variant period at time of first infection (based on index test date; 50% cutoff) | |||
| Pre-Delta | 16 (3.4) | 85 (20.5) | <.001 |
| Delta | 88 (18.7) | 249 (60.0) | |
| Omicron | 367 (77.9) | 81 (19.5) | |
| Vaccination status before initial SARS-CoV-2 infection | |||
| Unvaccinated | 31 (6.6) | 111 (26.7) | <.001 |
| Vaccinated | 317 (67.3) | 236 (56.9) | |
| Missing | 123 (26.1) | 68 (16.4) | |
Abbreviations: COPD, chronic obstructive pulmonary disease; COVID-19, coronavirus disease 2019; GED, general equivalency diploma; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2.
aAt least 1 eligible reinfection during the study period, ≥90 days after the index SARS-CoV-2 infection.
bBased on Mann-Whitney U test for continuous and χ2 test for categorical variables.
cFor variables with a “missing” category, the missing category was excluded from the statistical testing shown in this table.
dQuestions about comorbid conditions were included only in the 3-month follow-up survey, so this information is missing for nonresponders at that time point.
Compared with those who reported a single infection, a higher proportion in the reinfection group were female (67.5% in the reinfection vs 59.4% in the single-infection group; P = .02), black (8.2% vs 4.0%, respectively; P = .02), provided essential healthcare related work (28.9% vs 21.7%; P = .02), and, overall, reported less educational achievement (in the reinfection group, 50.7% had less than a 4 year college degree compared with 21.9% in the single-infection group; P < .001). A higher proportion of individuals with reinfection reported not being vaccinated (26.7% vs 6.6%; P < .001).
Symptomatic Outcomes 3–6 Months After a Single Infection Versus a Reinfection
Individuals who experienced their first infection during the pre-Delta period had similar 3–6-month prevalence rates of ongoing symptoms in the single-infection and reinfection groups (Figure 3 and Supplementary Table 4). The weighted prevalence (weighted results reported in the text; nonweighted prevalence rates demonstrated similar trends and are shown in Supplementary Figure 5) of those reporting fatigue severity scores ≥25 and those reporting ≥3 symptoms were the same in both groups. Loss of smell, which was more prevalent in those reporting a single infection, was the only symptom found to differ significantly between the groups. In the multivariable model results (Figure 4; pre-Delta vs pre-Delta → pre-Delta || Delta || Omicron) the odds of having ≥3 symptoms at 3–6 months were lower in the reinfection group along with reporting of any gastrointestinal symptom and 7 other individual symptoms. Forgetfulness or memory problems were the only symptoms with a higher odds of reporting following a reinfection (w-aOR, 4.02 [95% confidence interval (CI), 1.15–14.09]; P = .03).
Figure 3.
Weighted prevalence rates of symptomatic outcomes 3–6 months after the last severe acute respiratory syndrome coronavirus 2 infection, by variant period of first infection. The transparency level of the bar represents significance. Numbers next to bars represent the rate difference (as a percentage) between single-infection and reinfection groups, with placement side of the number indicating which group has the higher rate. Abbreviation: HEENT, head, eyes, ears, nose, and throat.
Figure 4.
Weighted adjusted odds ratios with 95% confidence intervals of reported symptoms 3–6 months after the last severe acute respiratory syndrome coronavirus 2 infection for the 4 multivariable logistic models. Abbreviation: HEENT, head, eyes, ears, nose, and throat.
Individuals in the reinfection group who experienced their first infection during the Delta period reported lower 3–6-month prevalence rates of ongoing symptoms following their reinfection than individuals who had a single infection in the Delta period (Figure 3 and Supplementary Table 5). A lower proportion in the reinfection group reported having ≥3 symptoms 3–6 months after their reinfection compared with the same time point after a single infection (14.3% vs 27.1%; P = .01).
The multivariable model including individuals who experienced their only or first infection during the Delta period (Figure 4; Delta vs Delta → Delta || Omicron) showed that the odds of reporting ≥3 symptoms 3–6 months after the last infection was lower in the reinfection group (w-aOR, 0.51 [95% CI, .32–.79]; P = .003). Those with reinfections had lower odds of reporting the composite variables of any cardiovascular or any gastrointestinal symptom. Odds of reporting 9 specific symptoms were lower in the reinfection group.
Similar trends were seen in the model comparing outcomes following a single Omicron infection with those following an Omicron reinfection after an initial Delta infection (Figure 4; Omicron vs Delta → Omicron). Reinfected individuals had lower odds of reporting a fatigue severity score ≥25 as well as lower odds of reporting 3 composite outcomes and 5 specific symptoms.
However, individuals in the reinfection group who experienced both infections during the Omicron period had higher prevalence rates of multiple symptoms 3–6 months after their reinfection infection than those who experienced a single infection during the Omicron period (Supplementary Table 6). A higher proportion of those with reinfection reported ≥3 symptoms (25.2% vs 14.1%; P = .02) as well as the composite variables of any pulmonary or cardiovascular symptoms. In addition, prevalence rates were higher in the reinfection group for chills, feeling hot, cough, wheezing, diarrhea, and dizziness or fainting.
The multivariable model including individuals who experienced their infections during the Omicron period (Figure 4; Omicron vs Omicron → Omicron) showed that the odds of reporting ≥3 symptoms 3–6 months after the last infection were higher in the reinfection group (w-aOR, 1.54 [95% CI, 1.02–2.34]; P = .04), but those with a reinfection had lower odds of reporting a fatigue severity score ≥25 (0.53 [.33–.85]; P = .01). The odds of 3 of the organ-system specific composite variables—any constitutional, any pulmonary, and any cardiovascular—were higher in the reinfection group. Five symptoms had higher odds in the reinfection group: chills, cough, wheezing, chest pain, and diarrhea. Three had lower odds in the reinfection group: loss of taste, aches, and joint pain.
DISCUSSION
In this analysis, we characterized reinfection and compared rates of symptoms 3–6 months after the last infection between individuals who had a single infection and those who had a reinfection. We found that individuals who experienced their first infection during the pre-Delta and Delta periods had the same or lower odds of having persistent symptoms 3–6 months after their last infection whether they had a reinfection or a single pre-Delta or Delta infection. Similarly, we observed lower odds of long-term symptoms among those with a Delta initial infection and an Omicron reinfection compared with a single Omicron infection. However, participants who reported both initial infection and reinfection during the Omicron period were more likely than those with a single Omicron infection to report the presence of several organ system symptoms in the 3–6-month period following reinfection .
Consistent with studies reporting an increased rate of reinfection as Omicron variants spread [17, 18], most reinfections in our analysis occurred during the Omicron period. Demographic trends and lower educational attainment in the reinfection group may suggest that health disparities, which have been well documented throughout the SARS-CoV-2 pandemic [19, 20], played a role in reinfection risk during this study. While our analysis was not designed to assess the effectiveness of vaccination, there was lower vaccination coverage in our reinfection group, and previous studies have suggested that vaccination is protective against reinfection [21, 22].
Our results further our knowledge regarding the effect of reinfection on the presence of symptoms. For individuals who experienced their first infection during the pre-Delta and Delta periods, our results suggest that reinfection did not increase the risk for the presence of long-term symptoms compared with the risk following an initial infection. This mirrors the literature on acute outcomes after reinfection, which suggests that the clinical course after repeated infection is similar or milder than that after the initial infection [5, 23] as well as another study that found decreased risk for new-onset long COVID after reinfection [24].
However, our results suggest that repeated infection with Omicron was associated with increased risk for long-term symptoms. The exceptions to this finding were loss of taste, a symptom that is less common in acute Omicron infection [25], and fatigue-related symptoms. There are several possible explanations for repeated Omicron infections leading to worse long-term outcomes. The interval between infections was shorter during the Omicron period, and the closer stacking of repeated infections and potential resulting organ-specific damage [26, 27] may increase the risk of persistent, long-term symptoms. In addition, there might be differences in immune-mediated damage or direct viral effects triggered by 2 Omicron infections. Another possible explanation is that later Omicron variants may be more likely to cause long-term symptoms than earlier Omicron variants, and later variants may have been represented more frequently in the outcomes of participants with 2 Omicron infections than in those with a single Omicron infection. However, the temporal spreads of Omicron initial infections and reinfections were similar in our data, suggesting that such an effect may have been minimal.
Our study is subject to several limitations. To identify reinfections, we relied on individuals being motivated to test for and then report SARS-CoV-2 reinfections. We may be underdetecting reinfection cases if individuals developed a true repeated infection <90 days from their initial infection, failed to self-identify reinfection, or had a false-negative COVID-19 result in the context of self-administered testing. Study designs with less reliance on self-motivated testing and self-report might decrease bias. Our analysis was limited to INSPIRE participants who had high survey compliance and completed all symptom and testing questions, resulting in many participants who were excluded due to missing data. This may have introduced spectrum bias and limited our power to find significant differences when they truly exist.
Because the timing of data collection for INSPIRE is based on initial infection and quarterly surveys, data collection was not optimally timed to understand the course following a reinfection. For this reason, we used a 3–6-month postinfection time frame for outcomes. In addition, due to falling survey compliance over time, evaluation of outcomes ≥6 months after infection was not possible. In addition, because the follow-up time was limited to 18 months, we had limited data on multiple reinfections. Finally, we did not have the data to account for symptom severity following the first infection or pattern of ongoing symptoms from the first infection. Future studies with more frequent and detailed symptom monitoring and more granular data on the duration and temporal relationship of symptoms to repeated infections would help clarify the relationship between long COVID and reinfection.
Our reinfection group was enrolled earlier in the pandemic than the single-infection group. While this would have less effect on our models as we stratified by the variant of initial infection, it does limit our ability to characterize differences between single-infection and reinfection groups, as our findings might reflect the risk factors for early infection rather than reinfection.
Because our survey design did not allow us to temporally place administration of vaccine doses after the initial illness in the reinfection timeline, our analysis does not account for vaccines received after the initial infection or the time between infection and vaccination. Although vaccination may offer some protection, the risk of long COVID remains even after vaccination [28], so it remains unclear how much receiving vaccines between infection episodes affects outcomes.
Despite these limitations, our results suggest that the timing of reinfection and the variants involved play important roles in the longer-term clinical outcomes. Individuals who experienced their initial and reinfection episodes during the Omicron period exhibit both shorter intervals between disease episodes and increased risk of long-term symptoms following reinfection.
Supplementary Material
Contributor Information
John J Openshaw, Division of Infectious Diseases and Geographic Medicine, Stanford University, Stanford, California, USA; Covid Control Branch, Division of Communicable Disease Control, California Department of Public Health, Richmond, California, USA.
Ji Chen, Center for Outcomes Research and Evaluation, Section of Cardiovascular Medicine, Yale School of Medicine, New Haven, Connecticut, USA.
Robert Rodriguez, Department of Medicine, School of Medicine, University of California, Riverside, Riverside, California, USA.
Michael Gottlieb, Department of Emergency Medicine, Rush University Medical Center, Chicago, Illinois, USA.
Kalyani McCullough, Covid Control Branch, Division of Communicable Disease Control, California Department of Public Health, Richmond, California, USA.
Michelle Santangelo, Department of Emergency Medicine, Rush University Medical Center, Chicago, Illinois, USA.
Mandy J Hill, Department of Population Health and Health Disparities, University of Texas Medical Branch, Galveston, Texas, USA.
Kristyn Gatling, Department of Medicine, Division of Infectious Diseases, Rush University Medical Center, Chicago, Illinois, USA.
Ahamed H Idris, Department of Emergency Medicine, University of Texas Southwestern Medical Center, Dallas, Texas, USA.
Samuel McDonald, Department of Emergency Medicine, University of Texas Southwestern Medical Center, Dallas, Texas, USA.
Lauren E Wisk, Division of General Internal Medicine & Health Services Research, University of California, Los Angeles, Los Angeles, California, USA.
Jonathan Dyal, Department of Emergency Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA.
Ralph C Wang, Department of Emergency Medicine, University of California, San Francisco, San Francisco, California, USA.
Kristin L Rising, Department of Emergency Medicine, Thomas Jefferson University, Philadelphia, Pennsylvania, USA.
Efrat Kean, Department of Emergency Medicine, Thomas Jefferson University, Philadelphia, Pennsylvania, USA.
Kelli N O’Laughlin, Departments of Emergency Medicine and Global Health, University of Washington, Seattle, Washington, USA.
Kari A Stephens, Department of Family Medicine, University of Washington, Seattle, Washington, USA.
Caitlin Malicki, Department of Emergency Medicine, Yale School of Medicine, New Haven, Connecticut, USA.
Zhenqiu Lin, Center for Outcomes Research and Evaluation, Section of Cardiovascular Medicine, Yale School of Medicine, New Haven, Connecticut, USA.
Erica S Spatz, Center for Outcomes Research and Evaluation, Section of Cardiovascular Medicine, Yale School of Medicine, New Haven, Connecticut, USA.
Huihui Yu, Center for Outcomes Research and Evaluation, Section of Cardiovascular Medicine, Yale School of Medicine, New Haven, Connecticut, USA.
Robert A Weinstein, Department of Medicine, Division of Infectious Diseases, Rush University Medical Center, Chicago, Illinois, USA.
Joann Elmore, Division of General Internal Medicine & Health Services Research, University of California, Los Angeles, Los Angeles, California, USA.
Supplementary Data
Supplementary materials are available at Clinical Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.
Notes
Acknowledgments . The authors thank Sharon Saydah, Claire M. Midgley, Ian D. Plumb, Aron J. Hall, and Melissa Briggs-Hagen for their assistance, feedback, and insight.
Disclaimer . The findings and conclusions in this article are those of the authors and do not necessarily represent the views or opinions of the California Department of Public Health or the California Health and Human Services Agency.
Financial support . The Innovative Support for Patients with SARS-COV-2 Infections Registry (INSPIRE) is funded by the Centers for Disease Control and Prevention, National Center of Immunization and Respiratory Diseases (NCIRD) (contract 75D30120C08008; principial investigator, R. A. W.).
References
- 1. Murchu E, Byrne P, Carty PG, et al. Quantifying the risk of SARS-CoV-2 reinfection over time. Rev Med Virol 2022; 32:e2260. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Nguyen NN, Houhamdi L, Hoang VT, et al. High rate of reinfection with the SARS-CoV-2 Omicron variant. J Infect 2022; 85:174–211. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Mensah AA, Lacy J, Stowe J, et al. Disease severity during SARS-COV-2 reinfection: a nationwide study. J Infect 2022; 84:542–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Abu-Raddad LJ, Chemaitelly H, Bertollini R. Severity of SARS-CoV-2 reinfections as compared with primary infections. N Engl J Med 2021; 385:2487–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Nguyen NN, Nguyen YN, Hoang VT, Million M, Gautret P. SARS-CoV-2 reinfection and severity of the disease: a systematic review and meta-analysis. Viruses 2023; 15:967. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Centers for Disease Control and Prevention . Long COVID basics. Available at: https://www.cdc.gov/covid/long-term-effects/. Accessed 6 March 2025.
- 7. National Academies of Sciences, Engineering, and Medicine . A Long COVID definition: a chronic, systemic disease state with profound consequences. Available at: https://nap.nationalacademies.org/download/27768. Accessed 6 March 2025. [PubMed]
- 8. Jones JM, Manrique IM, Stone MS, et al. Estimates of SARS-CoV-2 seroprevalence and incidence of primary SARS-CoV-2 infections among blood donors, by COVID-19 vaccination Status—United States, April 2021-September 2022. MMWR Morb Mortal Wkly Rep 2023; 72:601–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. World Health Organization . Tracking SARS-CoV-2 variants. Available at: https://www.who.int/activities/tracking-SARS-CoV-2-variants. Accessed 12 March 2024.
- 10. O’Laughlin KN, Thompson M, Hota B, et al. Study protocol for the Innovative Support for Patients with SARS-COV-2 Infections Registry (INSPIRE): a longitudinal study of the medium and long-term sequelae of SARS-CoV-2 infection. PLoS One 2022; 17:e0264260. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Gottlieb M, Wang RC, Yu H, et al. Severe fatigue and persistent symptoms at 3 months following severe acute respiratory syndrome coronavirus 2 infections during the pre-Delta, Delta, and Omicron time periods: a multicenter prospective cohort study. Clin Infect Dis 2023; 76:1930–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Puhach O, Meyer B, Eckerle I. SARS-CoV-2 viral load and shedding kinetics. Nat Rev Microbiol 2023; 21:147–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Bowe B, Xie Y, Al-Aly Z. Acute and postacute sequelae associated with SARS-CoV-2 reinfection. Nat Med 2022; 28:2398–405. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Spatz ES, Gottlieb M, Wisk LE, et al. Three-month symptom profiles among symptomatic adults with positive and negative severe acute respiratory syndrome coronavirus 2 tests: a prospective cohort study from the INSPIRE Group. Clin Infect Dis 2023; 76:1559–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Stürmer T, Rothman KJ, Avorn J, Glynn RJ. Treatment effects in the presence of unmeasured confounding: dealing with observations in the tails of the propensity score distribution–a simulation study. Am J Epidemiol 2010; 172:843–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. O’Laughlin KN, Klabbers RE, Ebna Mannan I, et al. Ethnic and racial differences in self-reported symptoms, health status, activity level, and missed work at 3 and 6 months following SARS-CoV-2 infection. Front Public Health 2023; 11:1324636. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Cohen D, Izak M, Stoyanov E, et al. Predictors of reinfection with pre-Omicron and Omicron variants of concern among individuals who recovered from COVID-19 in the first year of the pandemic. Int J Infect Dis 2023; 132:72–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Chen Y, Zhu W, Han X, et al. How does the SARS-CoV-2 reinfection rate change over time? The global evidence from systematic review and meta-analysis. BMC Infect Dis 2024; 24:339. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Hollis ND, Li W, Van Dyke ME, et al. Racial and ethnic disparities in incidence of SARS-CoV-2 infection, 22 US states and DC, January 1-October 1, 2020. Emerg Infect Dis 2021; 27:1477–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Benoit TJ, Kim Y, Deng Y, et al. Association between social vulnerability and SARS-CoV-2 seroprevalence in specimens collected from commercial laboratories, United States, September 2021-February 2022. Public Health Rep 2024; 139:501–11 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Ellingson KD, Hollister J, Porter CJ, et al. Risk factors for reinfection with SARS-CoV-2 Omicron variant among previously infected frontline workers. Emerg Infect Dis J 2023; 29:599–604. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Gómez-Gonzales W, Chihuantito-Abal LA, Gamarra-Bustillos C, et al. Risk factors contributing to reinfection by SARS-CoV-2: a systematic review. Adv Respir Med 2023; 91:560–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Wei J, Stoesser N, Matthews PC, et al. Risk of SARS-CoV-2 reinfection during multiple Omicron variant waves in the UK general population. Nat Commun 2024; 15:1008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Bosworth ML, Shenhuy B, Walker AS, et al. Risk of new-onset long COVID following reinfection with severe acute respiratory syndrome coronavirus 2: a community-based cohort study. Open Forum Infect Dis 2023; 10:ofad493. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Reiter ER, Coelho DH, French E, Costanzo RM; N3C Consortium . COVID-19-associated chemosensory loss continues to decline. Otolaryngol Neck Surg 2023; 169:1386–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Han X, Chen L, Fan Y, et al. Longitudinal assessment of chest CT findings and pulmonary function after COVID-19 infection. Radiology 2023; 307:e222888. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Watanabe A, So M, Iwagami M, et al. One-year follow-up CT findings in COVID-19 patients: a systematic review and meta-analysis. Respirol Carlton Vic 2022; 27:605–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Xie Y, Choi T, Al-Aly Z. Postacute sequelae of SARS-CoV-2 infection in the pre-Delta, Delta, and Omicron eras. N Engl J Med 2024; 391:515–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.