Characterizing and Predicting Post-Acute Sequelae of SARS CoV-2 Infection (PASC) in a Large Academic Medical Center in the US

Abstract

Background: A growing number of Coronavirus Disease-2019 (COVID-19) survivors are affected by post-acute sequelae of SARS CoV-2 infection (PACS). Using electronic health record data, we aimed to characterize PASC-associated diagnoses and develop risk prediction models.

Methods: In our cohort of 63,675 patients with a history of COVID-19, 1724 (2.7%) had a recorded PASC diagnosis. We used a case-control study design and phenome-wide scans to characterize PASC-associated phenotypes of the pre-, acute-, and post-COVID-19 periods. We also integrated PASC-associated phenotypes into phenotype risk scores (PheRSs) and evaluated their predictive performance.

Results: In the post-COVID-19 period, known PASC symptoms (e.g., shortness of breath, malaise/fatigue) and musculoskeletal, infectious, and digestive disorders were enriched among PASC cases. We found seven phenotypes in the pre-COVID-19 period (e.g., irritable bowel syndrome, concussion, nausea/vomiting) and sixty-nine phenotypes in the acute-COVID-19 period (predominantly respiratory, circulatory, neurological) associated with PASC. The derived pre- and acute-COVID-19 PheRSs stratified risk well, e.g., the combined PheRSs identified a quarter of the cohort with a history of COVID-19 with a 3.5-fold increased risk (95% CI: 2.19, 5.55) for PASC compared to the bottom 50%.

Conclusions: The uncovered PASC-associated diagnoses across categories highlighted a complex arrangement of presenting and likely predisposing features, some with potential for risk stratification approaches.

Keywords: Coronavirus Disease-2019 (COVID-19); electronic health records; phenome-wide association study; phenotype risk score; post-acute sequelae of SARS CoV-2 (PASC, long COVID, post-COVID conditions).

Figure 1

Schematic on study design. Three time periods were defined relative to the 1. positive COVID-19 test or diagnosis (index date): pre-COVID-19 until −14 days, acute-COVID-19 from −14 to +28 days, and post-COVID-19 from +28 days onwards. The post-COVID-19 PheWAS is used to validate features of PASC cases compared to COVID-19 cases without PASC diagnoses. The pre-COVID-19 and acute-COVID-19 PheWAS on the training data (index date in 2020–2021) inform on phenotype risk scores (PheRS) that will be used to predict PASC in the testing data (index date in 2022).

Figure 2

PheWAS on symptoms that occurred between 28 days and 6 months after the first COVID-19 test (outcome: post-COVID-19 symptoms/phecodes; predictor: PASC diagnosis yes/no). Among PheCodes that reached phenome-wide significance (red dashed line, p ≤ 0.05/960 = 5.2 × 10⁻⁵), only the strongest association per PheCode category was labeled. The analysis was adjusted using the following covariates: age at index date, gender, race/ethnicity, Elixhauser Score AHRQ, population density (quartiles), NDI (quartiles), health care worker status, vaccination status, post-test years in EHR, and severity. Summary statistics can be found in File S1.

Figure 3

PheWAS on symptoms that occurred at least 14 days before the first positive COVID-19 test (outcome: PASC diagnosis yes/no; predictors: PheCodes). Among PheCodes that reached phenome-wide significance (red dashed line, p ≤ 0.05/1404 = 3.56 × 10⁻⁵), only the strongest association per PheCode category was labeled. The analysis was adjusted using the following covariates: age at index date, gender, race/ethnicity, Elixhauser Score, population density (quartiles), NDI (quartiles), health care worker status, vaccination status, pre-test years in EHR, and severity. Summary statistics can be found in File S1.

Figure 4

Acute-COVID-19 PheWAS on symptoms that occurred between −14 and +28 days relative to testing positive for COVID-19 (outcome: acute-COVID-19 symptoms/PheCodes; predictor: PASC diagnosis yes/no). Among PheCodes that reached phenome-wide significance (red dashed line, p ≤ 0.05/663 = 7.5 × 10⁻⁵), only the strongest association per PheCode category was labeled. The analysis was adjusted using the following covariates: age at index date, gender, race/ethnicity, Elixhauser Score AHRQ, population density (quartiles), NDI (quartiles), health care worker status, vaccination status, post-test years in EHR, and severity. Summary statistics can be found in File S1.

Figure 5

PheRS-based risk stratification in the testing data. The proportion of PASC cases among different PheRS bins is shown for (A) the pre-COVID-19 PheRS (PheRS1) and (B) the acute-COVID-19 PheRS (PheRS2). The analysis is based on patients with history of COVID-19 in 2022 with at least 28 days between the first COVID-19 and first PASC diagnosis; 123 cases and 1154 controls. Risk bins correspond to selected ranges of the PheRS distributions. Vertical lines represent confidence intervals for binomial proportions [46].