Pre-COVID health-related quality of life predicts symptoms and outcomes for patients with long COVID

Abstract

Background: Post-acute sequelae SARS-CoV-2 (PASC) is a prevalent condition with variable symptom presentation. PASC occurs more often with pre-existing medical conditions, however it is unknown whether pre-COVID health-related quality of life (HRQL) is associated with PASC. Similarly, the trajectory of HRQL following PASC is unknown.

Objective: Our study sought to evaluate (1) whether pre-COVID HRQL is associated with PASC symptoms; (2) whether PASC patients have worse pre-COVID HRQL compared to matched controls; and (3) to compare HRQL trajectories from pre-COVID to 1-year follow-up between PASC patients and matched controls.

Design: Retrospective cohort study with propensity-score matched control group.

Participants: The cohort included 1,114 adult patients (mean age 53 ± 14, 75% female) seen in a PASC clinic between 2/10/21 and 3/27/24 who completed HRQL surveys prior to their initial COVID-diagnosis in a large health system. A propensity-score matched control group included patients with COVID-19 without PASC.

Main measures: HRQL was measured with PROMIS Global Health [global mental health (GMH) and global physical health (GPH) summary scores].

Key results: PASC symptoms were significantly associated with pre-COVID HRQL. Symptoms most associated with PROMIS-GMH included diarrhea/nausea [odds ratio (OR) = 1.27 (95% CI: 1.16-1.39) per five-point worsening] and brain fog [OR = 1.25 (95% CI: 1.14-1.37)], while fatigue [OR = 1.39 (95% CI: 1.15-1.68)] had the highest association with PROMIS-GPH. Pre-COVID GMH and GPH were significantly worse for PASC patients compared to controls [-2.6 (SE 0.4) and -3.4 (0.3) T-score points, respectively]. At 1-year following COVID, PASC patients worsened significantly in GMH and GPH (-2.0 ± 8.2 and -1.2 ± 7.5 T-score points, respectively), compared to controls who worsened significantly on GMH but not GPH (-0.8 ± 7.7 and 0.2 ± 7.4 T-score points, respectively).

Conclusions: In patients with PASC, worse pre-COVID HRQL was associated with more PASC-related symptoms. PASC patients had worse pre-COVID HRQL compared to matched controls and experienced a greater decline in HRQL 1-year after COVID-diagnosis; however, this decline was below the threshold for clinical significance.

Keywords: PROMIS Global Health; health-related quality of life; long-COVID symptoms; patient-reported outcomes; post-acute sequelae of COVID.

Figure 1

Proportion of patients with PASC experiencing symptoms, n = 1,114 patients.

Figure 2

Change in PROMIS Global Health pre-COVID diagnosis to 1-year after. (a) Change in PROMIS Global Mental Health pre- and post- initial COVID diagnosis in patients by PASC status (PASC vs. no PASC controls); (b) Change in PROMIS Global physical health pre- and post- initial COVID diagnosis in patients by PASC status (PASC vs. no PASC controls). Note, meaningful change in PROMIS Global Health is considered between 2.5 and 5 T-score points.