Background
The COVID-19 pandemic prompted urgent scale-up of infection prevention practices in healthcare settings. Reference Breeher, Boon, Hainy, Murad, Wittich and Swift1 Healthcare systems rapidly implemented layered mitigation strategies to reduce SARS-CoV-2 transmission including environmental controls, enhanced personal protective equipment (PPE), symptom screening, and exposure management programs. Reference Bielicki, Duval and Gobat2–Reference Fell, Beaudoin and D’Heilly4 These strategies were deployed as healthcare institutions contended with limited data on transmission dynamics and widespread concern about system capacity.
Among these mitigation strategies, exposure management programs aimed to identify healthcare personnel (HCP) at risk of infection before symptom onset and enable proactive, mandatory work exclusion. Reference Breeher, Boon, Hainy, Murad, Wittich and Swift1 This represented a shift from prepandemic infectious illness absence protocols, where HCP with acute respiratory illness self-identified and voluntarily stayed home only after symptom onset.
Pandemic-era programs generally operated through two pathways. First, workplace contact tracing systematically identified occupational exposures to confirmed cases, building on established epidemiologic principles, Reference Klinkenberg, Fraser and Heesterbeek5,Reference Eames and Keeling6 and supported by early modeling Reference Hellewell, Abbott and Gimma7,Reference Grantz, Lee and McGowan8 and observational data from community settings. Reference Wong, Kwong and Wu3,Reference Keeling, Hollingsworth and Read9,Reference Buchtele, Rabitsch, Knaus and Wohlfarth10 Contact tracing also fulfilled healthcare institutions’ duty-of-care obligation to inform HCP of occupational exposure risk amid extreme uncertainty. Secondly, many institutions integrated protocols for HCP to self-report non-occupational exposures, in alignment with public health guidance regarding quarantine for the general public following community exposures and HCP work exclusion. 11,12 Together, these strategies aimed to prevent outbreaks, reduce occupational transmission, and provide support to affected HCP.
As the pandemic evolved to widespread transmission, high volumes of community exposures and cases strained program capacity for many healthcare systems. Reference Murugesan, Venkatesan and Ramasamy13–Reference Vaisman, Cairns, De Graeve, Dus, Hota and Granton15 Understanding whether proactive exposure management achieved sustained reductions in COVID-19 burden is an important question, Reference Wang, Tan Kuan and Tay16,Reference Armbruster and Brandeau17 but one that is methodologically challenging to answer. Systematic reviews of contact tracing highlight that programs vary in implementation, metrics, and settings, limiting broad conclusions about effectiveness. Reference Hossain, Jarolimova, Elnaiem, Huang, Richterman and Ivers18–Reference Thomas Craig, Rizvi, Willis, Kassler and Jackson22 In healthcare specifically, near-universal adoption of tracing precluded strong control groups for comparative analysis. Reference Mei, Guo, Chen and Chen23 Published studies have primarily described program operations Reference Chong, Jayaraj and Rampal24–Reference Villanueva, Lazaro and Sayo27 reported process-based metrics such as risk stratification and timeliness, Reference Hijano, Dennis and Hoffman28–Reference Seidelman, Akinboyo and Rinehart32 or documented short-term outbreak containment. Reference Buchtele, Rabitsch, Knaus and Wohlfarth10,Reference Wee, Sim and Conceicao33–Reference Lim, Tan and Htun36 It remains unclear whether contact tracing and exposure management programs achieved measurable long-term reductions in HCP COVID-19 burden compared to symptom-based work exclusion.
To address this question, we evaluated an exposure management program at a large medical center during the first 15 months of the COVID-19 pandemic. Specifically, we compared COVID-positive HCP managed through proactive exposure protocols versus standard symptom-based exclusion, across four transmission-related outcomes: infectious time worked, presenteeism, workplace exposures, and secondary cases.
Methods
COVID-19 management system
UNCMC is a large academic medical center in North Carolina with a workforce of ∼11,500 HCP. In February 2020, the Occupational Health Services (OHS) department initiated a COVID-19 management system to track HCP exposures, symptoms, and cases. Other institutional mitigation measures included universal masking, airborne/contact precautions for infected patients, enhanced environmental cleaning, limited visiting hours, daily HCP wellness checks, and, starting in December 2020, mandatory COVID-19 vaccination.
UNCMC’s COVID-19 management system had both active and passive components, including (1) contact tracing of COVID-positive patients and HCP, and (2) operation of a telephone/e-mail hotline to receive and follow up on reports of exposures, symptoms, and positive tests. Exposures were classified according to risk levels established by the Centers for Disease Control and Prevention (CDC), based on close contact (≥15 cumulative minutes at <6 feet) and some gap in usage of recommended PPE (Tables A1 and A2). 12,37 HCP with low-risk exposures to patients and/or other HCP were advised to self-monitor for symptoms. OHS arranged testing for all medium- and high-risk exposures at serial intervals according to CDC guidance. Exclusion from work was mandatory for symptomatic and positive HCP, and for 14 days post medium- or high-risk exposure. Exposure-related work restrictions for asymptomatic HCP were eliminated in March 2021 for vaccinated individuals. 37 OHS issued work exemptions via a “COVID-19-Related Absence” mechanism granting HCP special use of accrued paid leave.
Data source and definitions
We conducted a retrospective cohort study of COVID-19 encounter records and case interviews among UNCMC HCP from March 2020 through June 2021. An “encounter” was a single record, initiated by an exposure, symptoms, or a positive test, which tracked all OHS activities relevant to that event. An exposure that progressed to symptoms and/or a positive test was recorded as a single encounter. Symptoms leading to a positive test were likewise recorded as a single encounter.
Analysis
Our primary aim was to compare the standard method of HCP infectious illness management (ie, initiating absence after illness onset) to the UNCMC system of tracing contacts and actively managing self-reported exposures. We constructed three mutually exclusive study groups of COVID-positive HCP: (1) “traced contacts”, first identified via OHS case investigations as contacts of a positive HCP/patient; (2) “self-reported exposures”, cases first identified by reporting an exposure that occurred in the community, the household, or a non-traced occupational exposure; and (3) “standard illness management”, those self-reporting illness at the time of symptoms/positive test, regardless of known exposure;.
From detailed case interviews conducted by OHS, we abstracted demographics, infection time line, work history, and details of other HCP contacts. We measured four primary outcomes: hours worked while infectious, hours worked while symptomatic (“presenteeism”), number of other HCP exposed, and epidemiologically linked secondary cases. Infectious work time included all self-reported hours worked during the 48 hours prior to symptom onset and while symptomatic. OHS classified “likely” secondary cases as those with a clear epidemiologic link to the source case, with little or no evidence of reported non-occupational infection sources. “All” secondary cases included both likely and possible occupational transmission events—those with weaker epidemiologic links and/or plausible non-occupational infection sources.
Using negative binomial regression, we estimated mean count differences and ratios for each outcome comparing contact-traced HCP and self-reported exposure HCP to standard illness management HCP. We adjusted for age and job category, modeled as frontline (advanced practice providers, clinical nurses, patient-facing support) vs non-frontline (all other roles). We used mean count difference results to estimate the efficiency of exposure management and tracing via the “number of encounters needed” (NEN) to prevent one additional outcome (eg, hour of infectious work time, single onward exposure, secondary case).
Although conceptually similar to the traditional number needed to treat for a given intervention, Reference Westreich and Mooney38 the NEN incorporates the likelihood of identifying an individual to whom the intervention applies (ie, a COVID case). This reflects that any potential benefits of exposure management can only occur when an encounter involves an HCP who is truly infected with COVID-19. Accordingly, the NEN combines the difference in expected outcome counts produced by exposure management with the frequency of COVID-19 cases among encounters. Specifically, the expected count difference for each outcome was multiplied by the proportion of COVID-19 cases in each study group, yielding the expected number of outcomes saved per encounter (trace or self-report) (Figure 1). The inverse of this product yielded the number of encounters needed to prevent one unit of outcome.
Calculating efficiency of exposure management: the average “number of encounters* needed” with Occupational Health Services to prevent one “unit” of an outcome *Initiated due to an exposure, reported symptom onset, or a positive test, an “encounter” is one full record of interaction between Occupational Health Services and a healthcare staff member related to the progression of the initiating event (eg, symptoms developing following reported exposure, or a positive test following reported symptoms).
Our primary analysis included only cases with medium- and high-risk exposures in the contact tracing group, to align with UNCMC exposure exclusion policy. To quantify the effects of the full scope of UNCMC tracing, we repeated these analyses across all encounters including low-risk exposures.
For epidemiological context, we also compared weekly UNCMC system COVID-19 activity with community COVID-19 activity using cross-correlation analysis. Reference Bloom, Buckeridge and Cheng39 We used hospitalization data from the North Carolina Department of Health and Human Services 40 restricted to counties in the surrounding area. This study was approved by the UNC IRB (study #23-0304). Figures 1 and 3 were created using BioRender.com. Statistical analysis was conducted in R version 4.3.2. 41
University of North Carolina Medical Center COVID-19 hotline encounters, cases and scaled COVID-19 hospitalizations in central North Carolina (NC).
Figure 3. Long description
A line graph depicting weekly COVID-19 encounter volume and related metrics from March 2020 to June 2021. The horizontal axis represents the weeks, while the vertical axis on the left shows the weekly COVID-19 encounter volume, and the vertical axis on the right shows the percent of encounters positive for COVID-19. The graph includes four data lines: cases, encounters non-positive, percent of encounters positive for COVID-19, and Central North Carolina COVID-19 hospitalizations scaled. The cases and encounters non-positive are represented by bars, with cases in light teal and encounters non-positive in gray. The percent of encounters positive for COVID-19 is shown by a solid teal line, and Central North Carolina COVID-19 hospitalizations scaled is depicted by a dashed purple line. The graph shows an increase in cases and hospitalizations starting around November 2020, peaking in January 2021, and then declining through June 2021. The percent of encounters positive for COVID-19 also peaks around January 2021 and then decreases.
Results
From March 2020 through June 2021, OHS recorded 30,102 encounters with HCP at UNCMC regarding COVID-19 exposure concerns, symptoms, and/or positive tests (Figure 3). Most encounters (62.8%, n = 18,899) were initiated by COVID-19 exposures. These included self-reported medium/high-risk exposures (n = 2,371) and occupational traces (n = 14,149) for contact with a positive patient (n = 13,995) or HCP (n = 920). Approximately 5% (n = 766) of occupational traces qualified as medium- or high-risk exposures eligible for work exclusion. Of all encounters, 37.2% (n = 11,203) were reports of symptoms and/or positive test results, adhering to the standard illness management process (ie, calling out sick).
During the study period, 1,114 HCP tested positive for COVID-19, including 12 cases associated with low-risk exposures which were excluded from the primary outcome analysis. The proportion of encounters testing positive was 2.9% (n = 22) for traced contacts, 7.0% (n = 166) for self-reported exposures, and 8.2% (n = 914) for standard illness management.
Age and vaccination status were similar across all COVID-positive HCP (Table 1). Frontline HCP were more likely to self-report exposure or have been contact-traced than non-frontline HCP. Contact-traced cases were most likely to work in inpatient units, whereas self-reported exposures occurred more evenly across departments. Support and inpatient staff made up the majority of the standard illness management group. These demographic trends were consistent in HCP who did not test positive for COVID-19 (Table A3).
Positive COVID-19 cases among healthcare personnel by identification mechanism (contact tracing, self-reported exposure, or standard illness management process)
Table 1. Long description
The table compares COVID-19 positive cases among healthcare personnel by identification mechanism, including traced contacts, self-reported exposure, and standard illness management. It has 22 rows and 11 columns. Column headers are Case characteristics, Traced contacts (Medium- or high-risk exposure to HCP or patient) n = 22, Self-reported exposure (Community, household or occupational exposure) n = 166, Standard illness management (Symptoms and/or positive test) n = 914, and P-value. Row labels include Age, median (interquartile range), Job category, Frontline role, Department, Primary vaccination status, Asymptomatic at the time of first encounter, Asymptomatic at follow-up interview, Known exposure to COVID + person before symptoms or asymptomatic test, Presenteeism (worked while symptomatic), Caused any occupational exposure, and Non-occupational contact with other employees. Each row provides specific data points for each category.
a Kruskal–Wallis rank sum test for age; Other categorical variables used Fisher’s exact test with simulated P value (2000 replicates).
b Patient-facing support includes paraprofessionals, technicians and assistants; frontline roles include job categories Advanced Practice Provider, clinical nurse, patient-facing support, and trainee.
c Departmental staff includes those in medicine, pediatrics, nursing and residents assigned to specific units; Inpatient includes HCP in Medicine, Surgery, OBGYN, Oncology, Psychiatry and Pediatrics; Intensive Care Units across Medicine, Surgery and Pediatrics.
d Likely secondary cases include occupational transmission events with strong epidemiological link and little evidence of other non-occupational infection sources as adjudicated by case investigation team.
e Total secondary cases include both likely transmission and “possible” occupational transmission as adjudicated by case investigation team, based on strength of epidemiologic link as well as other possible non-occupational infection sources.
Contact tracing and self-reporting exposures identified a higher proportion of both presymptomatic and asymptomatic cases compared to standard illness management (Table 1, P < .001). Nearly all COVID-19 cases in the traced and self-reported exposure groups were a-/presymptomatic at the time of identification in contrast to just 5% of standard illness management. Moreover, the proportion of cases remaining asymptomatic at follow-up was significantly greater among contact-traced (14%) and self-reported exposure groups (18%) compared to standard illness management (5%, P < .001).
Presenteeism was significantly higher among the standard illness management HCP (33.6%) compared with contact-traced HCP (4.5%) and self-reported exposure (0%, P < .001) (Table 1). Notably, among the 310 cases reporting presenteeism, 20.1% (n = 62) had a known, unreported exposure to a confirmed COVID-positive individual prior to their own symptom onset.
Overall, HCP COVID-19 cases exposed <1 other HCP on average (mean 0.96, 95%CI [0.81,1.11]). Fewer than 8% exposed >3 other HCP. Large occupational exposure clusters were investigated (15 individual HCP each exposed >10 others, including 2 cases who exposed >30 others), however no occupational superspreading events were detected.
Of COVID-positive traced contacts, one (4.5%) caused likely transmission to one other HCP, and positive HCP with self-reported exposure caused no likely transmission (0%). In contrast, among positive HCP following standard illness management, 2.5% (n = 23) caused likely occupational transmission to 26 secondary cases. In contrast, Including those with possible epidemiologic links, the standard illness management group generated 42 secondary cases compared to 2 from contact-traced HCP and one from HCP with self-reported exposure.
We found significant differences in outcome measures for contact tracing and self-reporting exposure compared to standard illness management (Table 2). Both traced contacts and self-reported exposure groups worked fewer hours while infectious (count difference (CD): −7.8, 95%CI [−10.3,−4.6] traced contacts, −7.8 [−9.0,−6.6] self-reported exposure) and fewer hours while symptomatic (−3.6 [−4.3,−3.1] for both groups) than the standard illness management group. The number of encounters needed (NEN) to prevent one hour of infectious work time was 4.3 for traced contacts and 1.8 for self-reported exposure. For one hour of presenteeism, NENs were 9.4 traces and 3.9 self-reported exposures.
Efficiency of COVID-19 exposure management to minimize presenteeism, onward exposures and secondary cases among healthcare personnel (HCP), University of North Carolina Medical Center, March 2020 to July 2021
Table 2. Long description
The table compares traced contacts, self-reported exposure, and standard illness management for COVID-19 exposure among healthcare personnel. It includes data on total encounters, COVID-19 positive cases identified, COVID-19 cases identified per encounter, number of encounters per case identified, hours worked while infectious, expected count ratio, expected count difference, infectious hours prevented per encounter, number of encounters needed to prevent one infectious hour worked, hours worked with symptoms, presenteeism hours prevented per encounter, number of encounters needed to prevent one hour of COVID-19 presenteeism, number of colleagues exposed, exposures prevented per encounter, number of encounters needed to prevent one exposure, likely secondary cases detected, secondary cases prevented per encounter, number of encounters needed to prevent one likely secondary case, and all secondary cases detected. The table has 9 rows and 11 columns, with specific values for each category.
a vs standard illness management as reference, generated from negative binomial regression model.
b Calculated as: COVID-19 cases per encounter x expected count difference.
c Number of encounters needed (NEN), calculated as: number self-reported or traced per case / expected count difference.
d Likely secondary cases include occupational transmission events with strong epidemiological link and low evidence of other non-occupational infection sources as adjudicated by case investigation team.
e All secondary cases include both likely transmission and “possible” occupational transmission as adjudicated by case investigation team, based on strength of epidemiologic link as well as other possible non-occupational infection sources.
The average number of coworkers exposed by traced contacts (0.4 [0.1, 0.8]) and HCP self-reporting exposure (0.4 [0.2,0.6]) was lower than HCP following standard illness management (CD [95%CI]: 1.1 [0.94,1.25]). This translated to small reductions on the absolute scale, however the relative reduction in onward exposures was 62% for traced contacts and 67% for self-reported exposure. The NEN to prevent one onward exposure was 48 for traced contacts and 20 for self-reporting exposure.
Secondary case rates were low overall. Compared with standard illness management, contact tracing and exposure self-reporting reduced secondary cases by 2%–4%, depending on transmission definition. This translated to low efficiency for both strategies: the NEN to prevent one likely secondary case was 1,111 for contact tracing and 714 for self-reporting exposure. When including possible secondary cases (with weaker or multiple epidemiologic links), there was no significant difference in secondary cases associated with contact tracing. For self-reported exposure, the NEN to prevent one possible or likely secondary case was 357.
Although absolute reductions in exposures and secondary cases were low, exposure management strategies generated large relative reductions across all outcomes: approximately 80% fewer infectious work hours, 60%–70% fewer exposures, and 80%–100% fewer secondary cases compared to standard illness management.
A sensitivity analysis found similar effectiveness results when including low-risk exposures (Table 3). However, efficiency was greatly reduced due to the high volume of low-risk traces, with higher NEN values of 81 traces to prevent one infectious work hour, 139 for one symptomatic work hour, 704 for one exposure, and >16,000 for one secondary case.
Efficiency of COVID-19 contact tracing exposures of all risk levels to minimize presenteeism, onward exposures and secondary cases among UNCMC healthcare personnel, March 2020 to July 2021
Table 3. Long description
The table compares the efficiency of traced contacts, including low-risk exposures, and standard illness management for COVID-19 among healthcare personnel. It has 15 rows and 3 columns. The columns are labeled ʼTraced contacts, including low-risk exposures (Occupational exposure to HCP or patient)ʼ, ʼStandard illness management (Symptoms and/or positive test)ʼ, and various metrics. Row 1: Total encounters (n), 14915, 11203. Row 2: COVID-19 positive cases (n), 34, 914. Row 3: COVID-19 cases identified per encounter, 0.002, 0.082. Row 4: Number of encounters per case identified, 500, 12.3. Row 5: Hours worked while infectious (95%CI), 4.0 (1.8, 6.6), 10.2 (9.4, 11.1). Row 6: Expected count ratio, 0.4 (0.2, 0.6), -. Row 7: Expected count difference, -6.2 (-8.6, -3.5), -. Row 8: Infectious hours prevented per encounter, 0.1 hours, -. Row 9: Number of encounters needed (NEN) to prevent one infectious hour worked, 81 traces, -. Row 10: Hours worked with symptoms (95%CI), 0.1 (0, 0.2), 3.6 (3.0, 4.3). Row 11: Expected count ratio, 0.02 (0, 0.05), -. Row 12: Expected count difference, -3.6 (-4.2, -3.0), -. Row 13: Presenteeism hours prevented per encounter, 0.007 hours, -. Row 14: NEN to prevent one hour of COVID-19 presenteeism, 139 traces, -. Row 15: Number of colleagues exposed (95% CI), 0.4 (0.2, 0.6), 1.1 (0.9,1.3).
a vs standard illness management as reference.
b Calculated as: COVID-19 cases per encounter x expected count difference.
c NEN calculated as: number reported or traced per case / expected count difference.
d Likely secondary transmission includes occupational transmission events with strong epidemiological link and low evidence of other non-occupational infection sources as adjudicated by case investigation team.
e All secondary transmission includes both likely transmission and “possible” occupational transmission as adjudicated by case investigation team, based on strength of epidemiologic link as well as other possible non-occupational infection sources.
Across the study period, UNCMC consistently fielded a high volume of COVID-19 exposure- and infection-related encounters with HCP each week. Initial waves in spring and summer 2020 peaked at approximately 600–700 HCP encounters, and a winter surge of community transmission aligned with a weekly peak of 1,241 HCP encounters in January 2021 (Figure 2). Once quarantine was eliminated for vaccinated HCP in February 2021, encounter volume decreased but still exceeded 100 most weeks. Across the study period, the proportion of HCP encounters identified a positive COVID case averaged 3.5%, peaking at 11.2% in January 2021. Encounter positivity averaged 2.5% during the first wave of community transmission (March–September 2020), 5.3% during the second wave (October 2020–March 2021), and 3.5% thereafter. Cross-correlation analysis showed consistent temporal alignment of HCP encounters and cases compared with hospitalizations in local counties (Figure A1, Table A4).
Discussion
Both contact tracing and self-reporting of exposures performed as intended, detecting pre and asymptomatic infections more often than standard illness management. Cases identified from active exposure management also spent less time working while infectious and/or symptomatic, and exposed fewer HCP. Although exposure management was associated with statistically significant reductions in secondary case rates, the low volume of these events translated to a modest absolute effect.
Exposure management efficiency was mixed. Tracing was most efficient at reducing infectious work time, requiring <5 encounters to prevent a single hour. Preventing one onward occupational exposure required 20 self-reported exposures or 48 occupational traces. Ultimately we found that hundreds or thousands of encounters were required to prevent one secondary COVID-19 case, no matter how strictly transmission was defined.
Our findings demonstrate limited measurable long-term impact of exposure management when conducted alongside other sustained mitigation strategies, including mandatory masking (with high reported adherence), air filtration, and enhanced cleaning protocols. Nevertheless, we found that nearly 30% of positive HCP worked while symptomatic, including many with known exposures that went unreported. HCP presenteeism remains a concern not only for COVID-19 but also for other respiratory viruses posing high risk to vulnerable patients such as influenza and respiratory syncytial virus. As healthcare institutions consider policies on sick leave, masking, and wellness checks, they should do so with careful attention to the risk posed by unmitigated presenteeism.
Beyond epidemiological effectiveness, exposure management programs served an important ethical and operational function. In the first pandemic year, with vaccines unavailable and scientific understanding of COVID-19 evolving, timely notification of workplace exposures was one of few definitive ways for HCP to manage personal risk. Exposure management programs also aligned with public health guidance that healthcare institutions were expected to follow in the absence of new evidence-based interventions. Our findings suggest that while similar programs may be necessary in early phases of future pandemics when uncertainty is high, adaptability is key as evidence changes. For example, as our sensitivity analysis demonstrated, tracing low-risk exposures may not be a useful pursuit in high-demand periods. Future work should explore definable metric-based inflection points that can indicate safe scale-back, particularly with time-series data over multiple surge periods.
Strengths
This analysis utilized a large sample of encounters including all COVID-19 cases identified at UNCMC through June 2021. Exposure-related calls were thorough to verify that work exclusion was warranted. Comprehensive, detailed case interviews supported in-depth assessment of infection timelines to accurately determine epidemiologic links between HCP. Although recall errors are possible, 97% of interviews were conducted within one day of a positive test result. Additionally, contact ascertainment was supplemented by input from supervisors and patient chart abstraction.
During the study period, public access to SARS-CoV-2 testing was challenging. Availability of free, fast, onsite testing at UNCMC optimized case ascertainment. Although some cases were inevitably missed, the combination of convenient testing and heightened COVID-19 concern provided strong incentives for HCP to seek testing through UNCMC when symptomatic or exposed.
Consistent with public health quarantine guidelines, CDC’s guidance for HCP work exclusion applied to exposures from any source, which expanded UNCMC’s program scope beyond typical occupational exposure management. Including community and household exposures provided an opportunity to evaluate the effectiveness of voluntarily reporting off-duty exposures posing workplace transmission risk.
Limitations
Although the CDC definition of close contact is operationally practical, its fixed time and distance thresholds may have misclassified or excluded some meaningful exposures. Even as understanding of transmission modes has evolved, the CDC 6-feet/15-minute rule remains a widely used standard, therefore the results here remain relevant to other hospital-based exposure management programs. Similarly, the widely adopted but static presymptomatic infectious window of 48 hours does not account for variability in infectiousness during that period.
Our assessment of exposures and infectious periods depended on self-reported information from case interviews conducted in service of their employer. HCP may have underreported behaviors that could reflect poorly on them professionally, particularly working while symptomatic or lapses in infection control protocols. Further, if present, differential underreporting could have influenced the magnitude of observed differences.
Summary
Over the first 15 months of the pandemic, UNCMC’s exposure management system processed >30,000 encounters related to exposures, symptoms, and infections, resulting in substantial operational impact. The system identified 1,114 HCP COVID-19 cases, with 27 secondary cases attributed to likely occupational transmission. Temporal patterns in HCP data aligned with community transmission trends, indicating that the HCP experience paralleled broader epidemiological conditions.
At UNCMC, contact tracing and exposure management identified cases earlier, reduced infectious and symptomatic work time, and prevented onward exposures. However, given infrequent secondary case detection overall, these efforts had low efficiency in reducing secondary cases. In a period of widespread community transmission and comprehensive institutional mitigation strategies, exposure management provided limited additional case prevention benefits. These findings suggest that while exposure management programs serve important duty-of-care roles, their greatest epidemiological impact may occur during localized outbreaks or early containment. Our data also demonstrate that presenteeism among HCP remains a challenge to safe healthcare environments, even in high-concern pandemic periods.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/ice.2026.10553
Acknowledgements
The authors would like to thank Tracy Watson and Dr Erica Pettigrew for their assistance in understanding exposure management program operations and the data collection process. The authors would also like to thank Dana Giandomenico for proofreading.
Financial support
None reported.
Competing interests
None reported.
Open access
