A. Estimating Effective Prevalence ()

First, the calculator estimates the chance that a person in the community is currently infectious. If you know that the people nearby differ from the community average, the people-risk profile adjusts that estimate. The result is .

Formula:

• Base Risk Probability (): This is the estimated number of infectious people per 1,000,000 residents.
  The calculator converts it to a probability using above. The value comes from one of two sources:
  • Manual Levels: Predefined fallback values (Low: 500, Moderate: 2,000, High: 4,500, Very High: 8,000 per million ≈ 0.05%, 0.20%, 0.45%, 0.80% point prevalence) used when specific location data is unavailable. These are heuristics intended to provide reasonable bands when no location-specific data is available, not a single literature-derived threshold.
  • Wastewater Data: When available for the location, LocalRisk first uses a recent WastewaterSCAN sewershed that serves the user's county. If there is no usable matching sewershed, it can use compatible raw measurements aggregated across the county from CDC's National Wastewater Surveillance System. Both paths use the relationship published by Mohring et al. in 2024 to estimate the proportion of people currently infectious.^{[1, 3, 2]} The CDC fallback uses the linear pathogen-to-PMMoV ratio, not CDC's standardized Wastewater Viral Activity Level. WVAL is relative to a site-and-method baseline and therefore does not retain the absolute concentration ratio required by the Mohring equation.^{[3, 4]} The fallback accepts only 24-hour composite raw or post-grit wastewater samples with compatible N-gene targets, PMMoV normalization, and no reported control amplification or inhibition. Numeric estimates additionally require a positive detection and a plausible ratio. Compatible all-non-detect county-weeks are preserved as non-detects rather than treating substituted values as prevalence.
  • County transfer and uncertainty: Compatible sites are calibrated to LocalRisk's WastewaterSCAN county-week series before being combined with a population-weighted median. Four source, matrix, collection, gene-target, PMMoV-control, and major-lab-method combinations passed both a whole-county holdout and a later-time holdout. Their scale factors range from 0.455 to 0.607. Other method groups are not used. The model range includes each group's validation residual, Mohring parameter uncertainty, and disagreement among detected sites in the county. These are LocalRisk calibration choices, not constants reported by CDC or Mohring. Wastewater signals remain noisy and time-shifted, and the model transfer has not been independently validated against representative U.S. clinical prevalence.^{[5, 6]} Values older than 21 days, current non-detects, and otherwise unusable measurements do not set prevalence; the calculator uses the selected manual band instead.
• People Risk Multiplier (): Use this multiplier when the people nearby have substantially more or less exposure than the community average.
  • The selected profile sets the following factor:
    • Average: 1.0
    • Standard Caution: ~0.059 (1/17)
    • High Caution: ~0.006 (1/165)
    • Went To Bar Recently: 20.0

    Note: the current UI exposes Average, Standard Caution, High Caution, and Went to a bar.

  • These are LocalRisk heuristics, but the three values exposed in the form have a traceable origin. microCOVID represented its standard- and high-caution budgets as 200 and 20 microCOVIDs per week. LocalRisk's exact ratios can be reconstructed from its first version's default prevalence of 3,300 per million: 200 / 3,300 is about 1/16.5, consistent with the rounded 1/17, and 20 / 3,300 is 1/165. microCOVID's bar scenario evaluates to 6 × 0.14 × (2 + 10 × 0.5 × 5) = 22.68; this model rounds it to 20.^{[11, 14]}
  • The conversion changes the meaning of the original caution budgets. LocalRisk keeps these as fixed ratios, so they rise and fall with current community prevalence. They do not preserve microCOVID's original fixed annual risk budget, and no epidemiological study establishes them as timeless constants.
• Universal Testing Multiplier ():
  • This input records recent negative COVID-19 test results for people involved in the activity.
  • The calculator adjusts the conditional activity risk with an attenuation factor (AF) based on test type and time since the test. In the implementation, this AF is applied as inside . Conceptually, a negative test reduces the probability a person is infectious; we implement it here because the UI treats testing as an activity condition rather than a separate prevalence input.
    A weighted average AF is calculated over the activity duration (assuming the test occurred 1 hour before the activity) using the following LocalRisk time brackets (heuristic):
    • Antigen Test AFs (heuristic): 0-6h (0.3), 6-12h (0.5), 12-24h (0.75), >24h (1.0)
    • Molecular (NAAT/PCR-like) Test AFs (heuristic): 0-8h (0.1), 8-16h (0.2), 16-24h (0.4), >24h (1.0)
    These brackets are a model choice. The cited sources inform the qualitative shape (test sensitivity, viral kinetics, and the relationship between viral load and infectiousness), but they do not provide these exact hour-by-hour AF values.^{[15, 16, 18, 19, 20, 21]}
  • A negative test reduces, but doesn't eliminate, the probability of being infectious.
    Molecular tests can detect lower viral loads than antigen tests, but how long a negative result stays reassuring is limited by within-host viral kinetics. Viral culture studies suggest infectious virus is uncommon at low viral loads and becomes more likely as viral load rises; early in infection, viral load can increase rapidly (hours-scale doubling). Together, this implies a negative high-sensitivity molecular test is most informative close to the activity and loses predictive power over the next day. Our time brackets are a simplified way to capture that effect.^{[18, 19, 20, 21]} RT-PCR cannot distinguish actively replicating virus from non-replicating viral fragments, so a positive result does not by itself establish current infectiousness.^[16] Early in infection, a nasal antigen test can be negative while saliva PCR is positive.^[17] Antigen results correlate more closely with viral culture than PCR positivity does, but antigen tests have higher false-negative rates early in infection.^{[15, 16, 19]} The protection decays over time.^{[18, 20, 21]}

B. Reference Case and Activity Conditions

The activity model starts with a reference case, then adjusts it for the conditions you enter. The result, , is the estimated chance that one infectious outsider transmits to one susceptible household attendee during the activity.

The one-hour reference case is:

This means one infectious outsider, one susceptible household attendee, one hour indoors, no masks, about one metre (three feet) apart, normal conversation, and no added air cleaning. It is a reference case, not an average for every encounter.

The main calibration comes from an analysis of seven million NHS COVID-19 app contacts. The app's risk score equals 1 for 15 minutes at two metres from an index case with standard infectiousness. The score is proportional to duration, while its proximity component follows inverse-square distance beyond one metre and is capped at one metre. A 30-minute exposure at about one metre therefore has a score near 8.^[9]

In Supplementary Table 9, a mean score of 7.86 corresponded to a 1.61% probability of reported transmission in one 30-minute window. Combining two such windows gives about 3.19% reported transmission over an hour:

Reported transmission is lower than actual infection because the study could observe only contacts who reported a positive test through the app. A separate UK analysis estimated that diagnostic testing found roughly 20–40% of infections during this period.^[10] Dividing 3.19% by that broad ascertainment range gives a rough actual-risk range of 8–16%, with 12% near the midpoint. This is a calibration inference, not a confidence interval or an exact under-reporting correction. Notified contacts may have tested more often than the general population, and not every positive test was necessarily reported through the app.

LocalRisk therefore uses 0.12 as its central calibration. For model review, 0.08–0.16 is a reasonable sensitivity range around this constant; the calculator currently reports only the central estimate. The NHS data did not record masks, ventilation, vocal activity, or exact app-reporting ascertainment, and its contacts had mixed vaccination and prior-infection histories. LocalRisk does not try to remove that background immunity from the calibration. The household vaccine setting later applies only the model's relative updated-dose adjustment.

The earlier microCOVID estimate—6%, with a stated 2–13% range—is a useful historical cross-check, not the derivation of 0.12.^[11] Likewise, household secondary attack rates and reproduction-number ratios describe different settings and are not used as multipliers in this calibration.^{[12, 13]} The absolute scale remains a major uncertainty, so comparisons between otherwise similar scenarios are more defensible than the exact point estimate.

The calculator turns that reference into a relative exposure dose:

This is a Wells-Riley-inspired exponential response, not a full Wells-Riley calculation. LocalRisk does not ask for infectious quanta, breathing rate, room volume, or clean-air flow, so is a dimensionless relative dose assembled from the inputs below.^[22]

Activity Inputs

1. Duration Multiplier ()
   • Duration is measured in hours: . Duration enters an exponential dose-response curve, which keeps the resulting probability below 1.0.
   • Saturation behavior: The exponential model saturates smoothly toward 1.0 as duration increases, rather than overshooting. For short durations (≤2 hr) it is nearly identical to the old linear approximation. For indoor activities, a conservative cap of 0.9 is also applied, informed by household secondary attack rate studies (Madewell et al. 2022: Omicron SAR 42.7%; individual studies up to 80.9%).^{[12, 51]}
   • The model assumes relative inhaled dose grows in proportion to time. It does not claim that every additional hour has the same incremental probability; the exponential curve makes each increment smaller as total probability rises.
2. Mask Multiplier ()
   • Mask type, fit, and use set two protection factors: . The multiplier combines the effective wearer protection factor for you () and the effective source control factor from others (). Higher factors indicate better protection/source control, leading to a lower overall multiplier.
   • and are effective protection factors - they are not raw filter efficiencies. In practice, real-world performance is often dominated by fit and leakage (often described as "total inward leakage"), which is why improving fit can meaningfully change protection even for the same nominal mask material.^[23]
   • Derived factors: Many studies report mask effectiveness as odds ratios or relative risks; the model converts those quantities into a compact set of values (and rounds them) so they can be combined multiplicatively. For example, Andrejko et al. report adjusted odds ratios of ~0.34 for surgical masks and ~0.17 for N95/KN95 respirators versus no mask (roughly ~3× and ~6× reductions), which we then simplify into small integer factors to reflect real-world fit variability.^[25]
   • Masks reduce inhaled particles (wearer protection) and exhaled particles (source control).^{[26, 25, 27]} Universal masking provides the greatest benefit.^[25] In healthcare settings, a randomised trial reported medical masks non-inferior to N95 respirators for preventing COVID-19 among healthcare workers under the trial's pre-specified margin.^[28] Effectiveness depends heavily on mask type and fit.^{[23, 26]}
   • Protection & Source Control Factors:

     Mask Type	Effective Wearer Protection Factor ()	Effective Source Control Factor ()	Supporting Evidence
     None / Poor Cloth	1	1	-
     Good Cloth / Surgical	2	2	[25]
     KN95 / KF94	5	3	[25]
     N95 (Fit-Tested/Sealed)	Default 6 (or FitFactor/10) [25]	30 [29]	[25, 29]
     P100 half-mask*	Default 10, or reported Fit Factor (up to 10,000)	1 (no assumed credit)	[24, 30, 31, 32, 33]

     Note: OSHA Assigned Protection Factors (APFs) assume a respiratory protection program and address wearer protection, not source control.^[24] Source-control factors and the N95 FitFactor/10 mapping are model heuristics. *"P100" identifies a filter class, not a complete respirator's protection factor or exhalation path. For the P100 half-mask option, LocalRisk uses a passing quantitative fit factor directly, up to 10,000. OSHA defines fit factor as the outside-to-inside concentration ratio and requires at least 100 for a half-mask QNFT. A PortaCount 8048 measures Series-100 fit factors through a range greater than 10,000; enter 10,000 when a report only says ">10,000." Without a passing result, LocalRisk falls back to OSHA's half-mask APF of 10. It gives no source-control credit because this input does not say whether exhaled air is filtered.^{[24, 32, 33]} The NIOSH 2021-107 report tested valved filtering facepiece respirators, not generic P100 elastomeric respirators, so it does not establish a 4× P100 factor. NIOSH separately documents approved non-valved and filtered-exhalation elastomeric designs that can provide source control; those configurations require product-specific information that this input does not collect.^{[30, 31]}

   • These are effective protection factors, not raw filter efficiencies. Fit and leakage often dominate real-world performance, so high-filtration material can still provide modest protection when the seal leaks.^{[23, 25]} For "sealed N95", the implementation optionally uses a quantitative fit factor (FitFactor/10) as a proxy for this fit/leakage effect.
   • In the exponential calculation, a better mask always lowers the dose and therefore lowers risk. The indoor 0.9 cap is applied afterward.
3. Location Multiplier ()
   • Indoor = 1.0; outdoor = 0.05, or one-twentieth of the indoor reference.
   • Studies report much less transmission outdoors than indoors.^[34] Outdoor risk is much lower but not zero, especially in crowded or close-contact situations.
4. Distance Multiplier ()
   • Relative to the roughly three-foot reference (1.0), the multipliers are 2.0 at less than one foot, 0.5 at six feet or more, and 0.25 at ten feet or more.
   • Aerosol concentration generally decreases with distance from the source, but the relationship depends strongly on ventilation, mixing, and whether you are in the emitter's near-field plume versus the room-average far-field.^{[35, 36]} Our distance multipliers are a simple stepwise proxy and should not be interpreted as a precise aerosol-physics law or as fully capturing far-field accumulation in poorly ventilated spaces. The cited studies discuss near-field vs far-field differences and airflow effects; they do not imply fixed universal multipliers by distance.
5. Vocal Volume Multiplier ()
   • Relative to normal conversation (1.0), silence is 0.2 and loud speech or singing is 5.0.
   • This multiplier represents vocal emission, not room size. The calculator does not currently model room volume explicitly (room-size effects are usually handled via ventilation / dilution in mechanistic models).
   • Loud speech, shouting, and singing emit more respiratory aerosol than normal speech or silence.^[37]
6. Ventilation/Filtration Multiplier ()
   • The current UI offers one simplified choice: no added air cleaning = 1.0; HEPA filtration = 0.25.
   • Outdoor or filtered air dilutes and removes infectious aerosols.^{[38, 39, 40]} HEPA filters can remove virus-containing particles.^{[40, 41]} In this calculator, "HEPA" is a simplified toggle rather than a full clean-air delivery (ACH/CADR) calculation. The multiplier is a heuristic meant to represent meaningful additional clean air; real reductions vary with CADR, room volume, placement, mixing, and duration.
7. Far-UVC Multiplier ()
   • This input is for a filtered 222 nm system whose equivalent clean-air delivery rate has been documented or professionally designed. The old labels remain in saved data, but the choices now mean about 5, 15, or at least 30 added equivalent air changes per hour (eACH).
   • LocalRisk assumes 3 background eACH and well-mixed, steady-state air. Existing ventilation or filtration is converted to an implied removal rate, then the Far-UVC eACH is added: . With no other added air cleaning, this gives 0.375 at 5 eACH, 0.167 at 15 eACH, and 0.091 at 30 eACH. When HEPA and Far-UVC are both selected, their implied clean-air rates add rather than their standalone reductions being multiplied. Because the HEPA option itself is a heuristic, its implied eACH is also only a model interpretation. The Far-UVC factor applies while the system is operating; there is no eight-hour cutoff.
   • Laboratory work shows that 222 nm Far-UVC can inactivate airborne coronaviruses.^[42] In a room-sized chamber with 3 ACH, Eadie et al. measured a 98.4% steady-state reduction with five sources, equivalent to 184 additional ACH. The paper's eight-hour figure describes the exposure-limit condition used for safety assessment, not a point when disinfection stops working.^[44] General UVGI performance still depends on installation, dose, air mixing, maintenance, and applicable safety standards.^[43]

C. Aggregating Risk from Multiple People

The calculator combines risk from nearby people who are not part of your household ("outsiders") across everyone from your household who attends.

For each activity:

1. Conditional Risk (): Calculated as described in Section B, incorporating all activity-specific multipliers (Duration, Mask, Location, Distance, Volume, Ventilation/Filtration, Far-UVC, and Universal Testing) applied to the `BASELINE_RISK_PER_HOUR`.
2. Effective Prevalence (): Calculated as described in Section A, incorporating the base risk probability and the people risk multiplier for the "Outsiders".
3. Multi-Target Attack Rate (): Before aggregating, the calculator applies the shared household vaccine multiplier to the pairwise risk. This is the probability that one infectious outsider infects at least one of the household members present.
   Where = Number of household members present.
4. Probability of Infection (): This is the final probability of at least one household member getting infected by any of the outsiders. It is 1 minus the probability that no outsider infects any household member.
   Where = Number of outsiders present. If , then .

This calculation covers transmission from outsiders to household members during the shared activity. When (just the user), is the user's infection probability for that activity.

D. Household Vaccination and microCOVIDs ()

The calculator assumes that every attending household member has the vaccination status and booster timing selected in the form. It applies that shared factor to each pairwise transmission risk before calculating the chance that anyone in the household group is infected. If attendees have different statuses, run separate estimates.

• Attending Household Vaccine Multiplier ()
  • The shared vaccination status and booster timing set a multiplier on each pairwise conditional risk before the household risks are combined.
    • Unvaccinated / Outdated Dose: Multiplier = 1.0
    • Updated Booster: Multiplier calculated using a simple exponential waning model based on months since booster (): ^{[46, 49]}. The model's month zero is shorthand for early post-vaccination protection; it does not model the immune response building during the first few weeks. The multiplier starts at 0.6 and increases toward 1.0 over time. For example, at m = 6 months the multiplier is ~0.83.
  • The two formula constants come from different evidence. Du et al. estimated effectiveness against infection of 44.7% at week 4, 35.5% at week 10, and 16.7% at week 20 for 2024-2025 vaccines; LocalRisk rounds the early estimate down to 40%.^[49] Menegale et al. estimated a 143-day half-life for effectiveness against laboratory-confirmed Omicron infection; that becomes 4.7 months in the formula.^[46]
  • This smooth curve is a comparison heuristic, not a fit to every point in either study. It gives about 20% effectiveness at week 20, compared with Du et al.'s 16.7% estimate. Effectiveness also varies with population, variant, product, and immune history.^{[48, 45]} The calculator does not model prior infection or mixed vaccination statuses within one attending household group.
  • FDA selected a monovalent JN.1-lineage XFG vaccine for the 2026-2027 U.S. formula. Because that formulation is only now being introduced, LocalRisk does not claim that this curve is validated specifically for it.^[50] Nor does LocalRisk substitute evidence about medically attended outcomes (ED/urgent-care visits or hospitalization) for effectiveness against infection.^[47]
• Final Risk in microCOVIDs ()
  • Formula:
  • Definition: One microCOVID (1 ) represents a one-in-a-million unit of infection risk. In this model, per-activity values are computed from , the probability that at least one attending household member becomes infected during the activity. The vaccine multiplier has already been applied to each pairwise risk.
  • Purpose: A standardized unit for comparing relative risks and managing a personal risk budget.
  • Total across activities: The calculator combines per-activity probabilities as so the final result remains a valid probability.
  • Context Examples: 1 µCoV = 1 in 1M chance; 10 µCoV = 1 in 100k; 100 µCoV = 1 in 10k; 1,000 µCoV = 1 in 1k (0.1%); 10,000 µCoV = 1 in 100 (1%).

References

1. [1] WastewaterSCAN. (n.d.). About the WastewaterSCAN Dashboard. https://data.wastewaterscan.org/about/
2. [2] Mohring, J., Leithäuser, N., Wlazło, J., Schulte, M., Pilz, M., Münch, J., & Küfer, K.-H. (2024). Estimating the COVID-19 prevalence from wastewater. Scientific Reports, 14, 14384. https://doi.org/10.1038/s41598-024-64864-1
3. [3] Centers for Disease Control and Prevention. (n.d.). CDC Wastewater Data for SARS-CoV-2. National Wastewater Surveillance System. https://data.cdc.gov/Public-Health-Surveillance/CDC-Wastewater-Data-for-SARS-CoV-2/j9g8-acpt/about_data
4. [4] Centers for Disease Control and Prevention. (2026, August 25). Using the Wastewater Viral Activity Level. https://www.cdc.gov/wastewater/about/wval.html
5. [5] WastewaterSCAN. (n.d.). About WastewaterSCAN. https://www.wastewaterscan.org/en/about
6. [6] Network of Wastewater-Based Epidemiology (NWBE). (n.d.). Frequently Asked Questions About Public Health Wastewater Monitoring. https://nwbe.org/?page_id=77
7. [7] Schenk, H., et al. (2024). SARS-CoV-2 surveillance in US wastewater: Leading indicators and data variability analysis in 2023-2024. PLOS ONE, 19(11), e0313927. https://doi.org/10.1371/journal.pone.0313927
8. [8] Feng, S., et al. (2021). Evaluation of Sampling, Analysis, and Normalization Methods for SARS-CoV-2 Concentrations in Wastewater to Assess COVID-19 Burdens in Wisconsin Communities. ACS ES&T Water, 1(8), 1955-1965. https://doi.org/10.1021/acsestwater.1c00160
9. [9] Ferretti, L., Wymant, C., Petrie, J., et al. (2024). Digital measurement of SARS-CoV-2 transmission risk from 7 million contacts. Nature, 626, 145-150. https://doi.org/10.1038/s41586-023-06952-2
10. [10] Colman, E., Puspitarani, G. A., Enright, J., & Kao, R. R. (2023). Ascertainment rate of SARS-CoV-2 infections from healthcare and community testing in the UK. Journal of Theoretical Biology, 558, 111333. https://doi.org/10.1016/j.jtbi.2022.111333
11. [11] microCOVID Project. (n.d.). microCOVID Project — White Paper. https://www.microcovid.org/paper/all
12. [12] Madewell, Z. J., Yang, Y., Longini, I. M., Halloran, M. E., & Dean, N. E. (2022). Household Secondary Attack Rates of SARS-CoV-2 by Variant and Vaccination Status: An Updated Systematic Review and Meta-analysis. JAMA Network Open, 5(4), e229317. https://doi.org/10.1001/jamanetworkopen.2022.9317
13. [13] Ito, K., Piantham, C., & Nishiura, H. (2022). Relative instantaneous reproduction number of Omicron SARS-CoV-2 variant with respect to the Delta variant in Denmark. Journal of Medical Virology, 94(5), 2265-2268. https://doi.org/10.1002/jmv.27560
14. [14] microCOVID Project. (2024). Calculator risk-profile implementation (commit 8ecaf11; src/data/calculate.ts and src/data/data.ts). GitHub. https://github.com/microCOVID/microCOVID/tree/8ecaf1129576f889016ff1ad8df6bb2fda0dff7e/src/data
15. [15] Soni, A., et al. (2023). Performance of Rapid Antigen Tests to Detect Symptomatic and Asymptomatic SARS-CoV-2 Infection: A Prospective Cohort Study. Annals of Internal Medicine, 176(7), 975-982. https://doi.org/10.7326/M23-0385
16. [16] Smith-Jeffcoat, S. E., Mellis, A. M., Grijalva, C. G., et al. (2024). SARS-CoV-2 Viral Shedding and Rapid Antigen Test Performance - Respiratory Virus Transmission Network, November 2022-May 2023. MMWR Morbidity and Mortality Weekly Report, 73(16), 365-371. https://doi.org/10.15585/mmwr.mm7316a2
17. [17] Adamson, B., Sikka, R., Wyllie, A. L., & Premsrirut, P. (2022). Discordant SARS-CoV-2 PCR and Rapid Antigen Test Results When Infectious: A December 2021 Occupational Case Series (preprint). medRxiv. https://doi.org/10.1101/2022.01.04.22268770
18. [18] Iyaniwura, S. A., Ribeiro, R. M., Zitzmann, C., Phan, T., Ke, R., & Perelson, A. S. (2024). The kinetics of SARS-CoV-2 infection based on a human challenge study. Proceedings of the National Academy of Sciences, 121(46), e2406303121. https://doi.org/10.1073/pnas.2406303121
19. [19] Kirby, J. E., Riedel, S., Dutta, S., Arnaout, R., Cheng, A., Ditelberg, S., Hamel, D. J., Chang, C. A., & Kanki, P. J. (2023). Sars-Cov-2 antigen tests predict infectivity based on viral culture: comparison of antigen, PCR viral load, and viral culture testing on a large sample cohort. Clinical Microbiology and Infection, 29(1), 94-100. https://doi.org/10.1016/j.cmi.2022.07.010
20. [20] Larremore, D. B., Wilder, B., Lester, E., Shehata, S., Burke, J. M., Hay, J. A., Tambe, M., Mina, M. J., & Parker, R. (2021). Test sensitivity is secondary to frequency and turnaround time for COVID-19 screening. Science Advances, 7(1), eabd5393. https://doi.org/10.1126/sciadv.abd5393
21. [21] Wells, C. R., Gokcebel, S., Pandey, A., Galvani, A. P., & Townsend, J. P. (2022). Testing for COVID-19 is Much More Effective When Performed Immediately Prior to Social Mixing. International Journal of Public Health, 67, 1604659. https://doi.org/10.3389/ijph.2022.1604659
22. [22] Rudnick, S. N., & Milton, D. K. (2003). Risk of indoor airborne infection transmission estimated from carbon dioxide concentration. Indoor Air, 13(3), 237-245. https://doi.org/10.1034/j.1600-0668.2003.00189.x
23. [23] Brooks, J. T., Beezhold, D. H., Noti, J. D., et al. (2021). Maximizing Fit for Cloth and Medical Procedure Masks to Improve Performance and Reduce SARS-CoV-2 Transmission and Exposure. MMWR Morbidity and Mortality Weekly Report, 70(7), 254-257. https://doi.org/10.15585/mmwr.mm7007e1
24. [24] Occupational Safety and Health Administration (OSHA). (2009). Assigned Protection Factors for the Revised Respiratory Protection Standard (Publication 3352). https://www.osha.gov/sites/default/files/publications/3352-APF-respirators.pdf
25. [25] Andrejko, K. L., Pry, J., Myers, J. F., et al. (2022). Effectiveness of Face Mask or Respirator Use in Indoor Public Settings for Prevention of SARS-CoV-2 Infection - California, February-December 2021. MMWR Morbidity and Mortality Weekly Report, 71(6), 212-216. https://doi.org/10.15585/mmwr.mm7106e1
26. [26] Lindsley, W. G., Blachere, F. M., Law, B. F., Beezhold, D. H., & Noti, J. D. (2021). Efficacy of face masks, neck gaiters and face shields for reducing the expulsion of simulated cough-generated aerosols. Aerosol Science and Technology, 55(4), 449-457. https://doi.org/10.1080/02786826.2020.1862409
27. [27] Howard, J., Huang, A., Li, Z., et al. (2021). An evidence review of face masks against COVID-19. Proceedings of the National Academy of Sciences, 118(4), e2014564118. https://doi.org/10.1073/pnas.2014564118
28. [28] Loeb, M., Bartholomew, A., Hashmi, M., Tarhuni, W., Hassany, M., Youngster, I., Somayaji, R., Larios, O., Kim, J., Missaghi, B., Vayalumkal, J. V., Mertz, D., Chagla, Z., Cividino, M., Ali, K., Mansour, S., Castellucci, L. A., Frenette, C., Parkes, L., et al. (2022). Medical Masks Versus N95 Respirators for Preventing COVID-19 Among Health Care Workers: A Randomized Trial. Annals of Internal Medicine, 175(12), 1629-1638. https://doi.org/10.7326/M22-1966
29. [29] Lai, J., Coleman, K. K., Tai, S.-H. S., German, J., Hong, F., Albert, B., … Milton, D. K. (2024). Relative efficacy of masks and respirators as source control for viral aerosol shedding from people infected with SARS-CoV-2: a controlled human exhaled breath aerosol experimental study. eBioMedicine, 104, 105157. https://doi.org/10.1016/j.ebiom.2024.105157
30. [30] National Institute for Occupational Safety and Health (NIOSH). (2020). Filtering Facepiece Respirators with an Exhalation Valve: Measurements of Filtration Efficiency to Evaluate Their Potential for Source Control. DHHS (NIOSH) Publication No. 2021-107. https://doi.org/10.26616/NIOSHPUB2021107
31. [31] Fernando, R., Peterson, J., & Portnoff, L. (2021, March 1). Advancements in Elastomeric Respirator Technology for Use as Source Control. NIOSH Science Bulletin. https://www.cdc.gov/niosh/bulletin/2021/elastomeric_source-control.html
32. [32] Occupational Safety and Health Administration (OSHA). Respiratory Protection Standard, 29 CFR 1910.134 and Appendix A: Fit Testing Procedures. https://www.osha.gov/laws-regs/regulations/standardnumber/1910/1910.134
33. [33] TSI Incorporated. (2022). PortaCount Respirator Fit Tester Model 8048 Specifications (P/N 5001949, Rev. M). https://tsi.com/getmedia/fa9402a2-eaa7-4dc5-a589-ade04ea75aaa/PortaCount_Model_8048_A4-UK_5001949_RevG_Web?ext=.pdf
34. [34] Bulfone, T. C., Malekinejad, M., Rutherford, G. W., & Razani, N. (2021). Outdoor Transmission of SARS-CoV-2 and Other Respiratory Viruses: A Systematic Review. The Journal of Infectious Diseases, 223(4), 550-561. https://doi.org/10.1093/infdis/jiaa742
35. [35] Parhizkar, H., et al. (2023). A novel VOC breath tracer method to evaluate indoor respiratory exposures in the near- and far-fields; implications for the spread of respiratory viruses. Journal of Exposure Science & Environmental Epidemiology, 33(3), 339-346. https://doi.org/10.1038/s41370-022-00499-6
36. [36] Mariam, Magar, A., Joshi, M., Rajagopal, P. S., Khan, A., Rao, M. M., & Sapra, B. K. (2021). CFD Simulation of the Airborne Transmission of COVID-19 Vectors Emitted during Respiratory Mechanisms: Revisiting the Concept of Safe Distance. ACS Omega, 6(26), 16876-16889. https://doi.org/10.1021/acsomega.1c01489
37. [37] Asadi, S., Wexler, A. S., Cappa, C. D., Barreda, S., Bouvier, N. M., & Ristenpart, W. D. (2019). Aerosol emission and superemission during human speech increase with voice loudness. Scientific Reports, 9, 2348. https://doi.org/10.1038/s41598-019-38808-z
38. [38] Centers for Disease Control and Prevention (CDC). (2023, May 12). Ventilation in Buildings. https://archive.cdc.gov/www_cdc_gov/coronavirus/2019-ncov/community/ventilation.html
39. [39] Centers for Disease Control and Prevention (CDC). (2024, October 3). Ventilation Mitigation Strategies. NIOSH Workplace Safety and Health Topics. https://www.cdc.gov/niosh/ventilation/prevention/index.html
40. [40] Environmental Protection Agency (EPA). (Accessed 2024). Air Cleaners, HVAC Filters, and Coronavirus (COVID-19). https://www.epa.gov/indoor-air-quality-iaq/air-cleaners-hvac-filters-and-coronavirus-covid-19
41. [41] Das, D., Babik, K. R., Moynihan, E., & Ramachandran, G. (2023). Experimental studies of particle removal and probability of COVID-19 infection in passenger railcars. Journal of Occupational and Environmental Hygiene, 20(1), 1-13. https://doi.org/10.1080/15459624.2022.2137298
42. [42] Buonanno, M., Welch, D., Shuryak, I., & Brenner, D. J. (2020). Far-UVC light (222 nm) efficiently and safely inactivates airborne human coronaviruses. Scientific Reports, 10, 10285. https://doi.org/10.1038/s41598-020-67211-2
43. [43] American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE). (2019). Ultraviolet Air and Surface Treatment (Handbook — HVAC Applications, Chapter 62). https://www.ashrae.org/file%20library/technical%20resources/covid-19/i-p_a19_ch62_uvairandsurfacetreatment.pdf
44. [44] Eadie, E., Hiwar, W., Fletcher, L., et al. (2022). Far-UVC (222 nm) efficiently inactivates an airborne pathogen in a room-sized chamber. Scientific Reports, 12, 4373. https://doi.org/10.1038/s41598-022-08462-z
45. [45] Bobrovitz, N., Ware, H., Ma, X., Li, Z., Hosseini, R., Cao, C., … Subissi, L. (2023). Protective effectiveness of previous SARS-CoV-2 infection and hybrid immunity against the omicron variant and severe disease: a systematic review and meta-regression. The Lancet Infectious Diseases, 23(5), 556-567. https://doi.org/10.1016/S1473-3099(22)00801-5
46. [46] Menegale, F., Manica, M., Zardini, A., et al. (2023). Evaluation of Waning of SARS-CoV-2 Vaccine-Induced Immunity: A Systematic Review and Meta-analysis. JAMA Network Open, 6(5), e2310650. https://doi.org/10.1001/jamanetworkopen.2023.10650
47. [47] Link-Gelles, R., et al. (2025). Interim Estimates of 2024-2025 COVID-19 Vaccine Effectiveness Among Adults Aged ≥18 Years - VISION and IVY Networks, September 2024-January 2025. MMWR Morbidity and Mortality Weekly Report, 74(6), 73-82. https://doi.org/10.15585/mmwr.mm7406a1
48. [48] Berber, E., & Ross, T. M. (2024). Factors Predicting COVID-19 Vaccine Effectiveness and Longevity of Humoral Immune Responses. Vaccines, 12(11), 1284. https://doi.org/10.3390/vaccines12111284
49. [49] Du, Y., Paritala, S., Xu, Y., Maloney, P., & Lin, D.-Y. (2025). Durability of 2024-2025 COVID-19 Vaccines Against JN.1 Subvariants. JAMA Internal Medicine, 185(12), 1501-1504. https://doi.org/10.1001/jamainternmed.2025.5465
50. [50] U.S. Food and Drug Administration. (2026, May 28). COVID-19 Vaccines (2026-2027 Formula) for Use in the United States Beginning in Fall 2026. https://www.fda.gov/vaccines-blood-biologics/industry-biologics/covid-19-vaccines-2026-2027-formula-use-united-states-beginning-fall-2026
51. [51] López-Muñoz, I., Torrella, A., Pérez-Quílez, O., Castillo-Zuza, A., Martró, E., Bordoy, A. E., Saludes, V., Blanco, I., Soldevila, L., Estrada, O., Valerio, L., Roure, S., & Vallès, X. (2022). SARS-CoV-2 Secondary Attack Rates in Vaccinated and Unvaccinated Household Contacts during Replacement of Delta with Omicron Variant, Spain. Emerging Infectious Diseases, 28(10), 1999-2008. https://doi.org/10.3201/eid2810.220494