# Clinical Symptom Patterns as Predictors of SARS-CoV-2 Infection in Healthcare Workers in Puerto Rico

## TL;DR

This study identifies key symptoms like muscle pain and loss of taste or smell that predict SARS-CoV-2 infection in healthcare workers in Puerto Rico, helping improve early detection.

## Contribution

The study introduces a symptom-based model with strong predictive power (AUC = 0.87) for identifying SARS-CoV-2 infections in healthcare workers.

## Key findings

- Symptoms like muscle pain, fever, and loss of taste or smell strongly predict SARS-CoV-2 infection.
- Symptom count is a significant predictor, with higher positivity rates among those reporting three to four symptoms.
- The model demonstrated strong discriminative ability with an AUC of 0.87.

## Abstract

Public health relevance—How does this work relate to a public health issue?
SARS-CoV-2 infection among healthcare workers represents a critical occupational and public health concern due to their dual role as a high-risk group and potential sources of onward transmission within healthcare facilities and communities.This study addresses the need for evidence-based, symptom-driven screening strategies during periods of limited testing capacity by identifying clinical symptom patterns associated with SARS-CoV-2 positivity among healthcare workers in Puerto Rico.

SARS-CoV-2 infection among healthcare workers represents a critical occupational and public health concern due to their dual role as a high-risk group and potential sources of onward transmission within healthcare facilities and communities.

This study addresses the need for evidence-based, symptom-driven screening strategies during periods of limited testing capacity by identifying clinical symptom patterns associated with SARS-CoV-2 positivity among healthcare workers in Puerto Rico.

Public health significance—Why is this work of significance to public health?
The findings demonstrate that symptom burden and specific symptoms-particularly muscle pain, fever, and loss of taste or smell-are strong predictors of SARS-CoV-2 infection, highlighting practical indicators for early case identification in frontline workers.The model showed strong discriminative ability (AUC = 0.87).

The findings demonstrate that symptom burden and specific symptoms-particularly muscle pain, fever, and loss of taste or smell-are strong predictors of SARS-CoV-2 infection, highlighting practical indicators for early case identification in frontline workers.

The model showed strong discriminative ability (AUC = 0.87).

Public health implications—What are the key implications or messages for practitioners, policy makers and/or researchers in public health?
Symptom-based screening tools that prioritize key predictive symptoms and overall symptom count can guide testing prioritization and strengthen infection control measures in healthcare environments.Public health practitioners can use findings to refine occupational health protocols while researchers can build on this work to validate symptom-based models in larger and more diverse populations and during future outbreaks.

Symptom-based screening tools that prioritize key predictive symptoms and overall symptom count can guide testing prioritization and strengthen infection control measures in healthcare environments.

Public health practitioners can use findings to refine occupational health protocols while researchers can build on this work to validate symptom-based models in larger and more diverse populations and during future outbreaks.

Coronavirus disease 2019 (COVID-19), caused by SARS-CoV-2, has posed major risks for healthcare workers (HCWs) worldwide. This study assessed the prevalence of infection and its relationship with demographic and clinical characteristics among HCWs at the University of Puerto Rico Adult Hospital. A total of 132 individuals were enrolled, of whom six tested positive (4.55%). The study population was predominantly female (78.8%) with a mean age of 41 years, and although men showed higher odds of infection (OR = 3.98), the difference was not significant. Symptom presence was strongly associated with infection: 7.4% of symptomatic participants tested positive compared to none of the asymptomatic (p < 0.001). Symptom count was also predictive, with those reporting three to four symptoms showing the highest positivity rate (14.8%) and those with five to ten symptoms at 6.7%. Specific symptoms including muscle pain (OR = 21.04, p = 0.002), taste loss (OR = 24.20, p = 0.002), smell loss (OR = 15.25, p = 0.024), and fever (OR = 20.50, p = 0.016) were significantly linked to infection, while others such as headache and congestion were not. These findings underscore the utility of symptom-based monitoring in occupational health, though the single-site design, modest sample size, reliance on self-report, and early pandemic diagnostic limitations may have led to underestimation of true cases.