1. Shared Air – Ventilation for Airborne Infection Control
Describe the mechanism of infection control through ventilation
ventilation and its limitations in controlling transmission of airborne diseases.
Tobias van Reenen

2. CONTENTS WHAT IS THE PROBLEM Probability of Transmission
WHAT CONTROL METHODS
HOW MUCH VENTILATION IS ENOUGH
(How much transmission is too much)
APPLICATION AND LIMITS OF VENTILATION
CONCLUSION
Difficult for an engineer to talk to a group of healthcare professionals. You are more knowledgeable about clinical matters so I need to hide my limited depth of knowledge of clinical issues while not bewildering you with engineering jargon.

3. AN ISSUE OF INDOOR AIR QUALITY
TB is spread by the airborne transmission route
Consider the healthcare occupational risks7,8
Work location
Incidence
(worker relative to general population)
Outpatient facilities
4.2 – 11.6
General medical wards
3.9 – 36.6
Inpatient facilities
14.6 – 99.0
Emergency rooms
26.6 – 31.9
Laboratories
42.5 – 135.3
Joshi, R., Reingold A. L., Menzies, D., Pai, M. (2006). Tuberculosis among health-care workers in low- and middle-income countries: a systematic review. PlosMed 3(12);e494 (after Dr Paul Jensen)
Menzies, D., Joshi, R., Pai, M. (2006). Risk of tuberculosis infection and disease associated with work in healthcare settings. Int J Tuberc Lung Dis 11(6): (after Dr Paul Jensen)

4. STANDARDS AND REGULATIONS
National Building Regulations and Building Standards Act (Act 103 of 1977)
SANS XA
SANS O
SANS 204 (Invoked in part)
R1390 – HBA Regulations

5. STANDARDS AND REGULATIONS
Ventilation SANS
Table 2 all but prohibits recirculation in healthcare buildings.
Table 2 Excludes l/s/person calculations for healthcare buildings
Ignores the effects of occupancy rates on IAQ

6. 𝑃 𝑖𝑛𝑓 = 𝑛𝑒𝑤 𝑐𝑎𝑠𝑒𝑠 𝑠𝑢𝑠𝑐𝑒𝑝𝑡𝑖𝑏𝑙𝑒𝑠 =1− 𝑒 −𝐼𝑞𝑝𝑡 𝑄 𝑜𝑎
STANDARDS AND REGULATIONS
WHY NOT JUST USE THE PRESCRIBED 12 AIR CHANGES PER HOUR?
Equation for probability of airborne infection:
Where:
Pinfection = the probability of infection
Cases = the number of infection cases
Susceptibles = number of susceptible individuals
I = number of infector individuals
p = pulmonary ventilation rate of a person (m³/hour)
q = quanta generation rate (1/hr)
t = exposure time (hr)
Qoa = room ventilation rate with clean air (m³/hour)
𝑃 𝑖𝑛𝑓 = 𝑛𝑒𝑤 𝑐𝑎𝑠𝑒𝑠 𝑠𝑢𝑠𝑐𝑒𝑝𝑡𝑖𝑏𝑙𝑒𝑠
=1− 𝑒 −𝐼𝑞𝑝𝑡 𝑄 𝑜𝑎
Source: Riley et al., 1978

7. Poisson-Wells-Riley-Soper-Reed-Frost Equations - (Wells-Riley)
Probability of infection :
From:
P = D/S
D = number of infections arising from exposure
S = number of susceptible exposure cases
And:
𝑃= 1−𝑒 −µ
Poison’s probability of rare discrete events
∴𝐷=𝑆∙(1− 𝑒 −𝐼𝑞𝑝𝑡 𝑄 𝑜𝑎 )
Where:
P = the probability of infection
D = the number of new infections
S = number of susceptible individuals
I = number of infector individuals
n = number of room occupants
p = pulmonary ventilation rate of a person (m³/hour)
q = quanta generation rate (1/hr)
t = exposure time (hr)
Qoa = room ventilation rate with clean air (m³/hour)
Roe = Environmental reproductive number
𝜺 𝒗 = Ventilation Effectivness
𝑸= 𝒏𝒑𝒒𝒕 𝜺 𝒗 ∙𝒍𝒏 𝒏−𝟏 𝒏− 𝟏+ 𝑹 𝟎𝑬 𝒏

8. THE RISK OF PROXIMITY Waiting Room
100 Person Waiting Room
1 Infectious Person
J-H Grobler (CSIR, 2017)
What percentage of transmissions happen in which zones?

9. THE RISK OF PROXIMITY Relative Risk Index
2.4
0.5 m² per person
3.1
0.58
What percentage of transmissions happen in which zones?
1.0 m² per person
5.0 m² per person

10. THE RISK OF PROXIMITY (% Transmissions)
37%
0.5 m² per person
25%
What percentage of transmissions happen in which zones?
1.0 m² per person
38%
5.0 m² per person

11. WHICH ENVIRONMENTAL CONTROLS?
Reduce Occupancy Levels:
Outdoor waiting
Appointment Systems
Segregate
Screening
Isolation
Fast Tracking
VENTILATION
Natural ventilation
Mechanical ventilation
Air cleaners

12. Somerset Hospital. Circa 1890
NATURAL VS MECHANICAL VENTILATION
A 2012 study by University of Oregon proved that while clinics with open windows had more microorganisms, closed window environments were more pathogenic2
Ventilation method accounts for a greater variance in airborne bacterial pathogenicity than ventilation rates alone2
Somerset Hospital. Circa 1890
Photo: Etienne du Plessis
2Kembel, S. W., Jones, … Green, J. L. (2012). Architectural design influences the diversity and structure of the built environment microbiome. The ISME Journal, 6(8), 1469–79.

13. HOW MUCH VENTILATION? (100 person waiting room)
It can be seen that the 12 ACH criteria (in the regulations) I intended for low occupancy rooms. It becomes ineffective for high occupancy spaces with overcrowding

14. HOW MUCH TRANSMISSION IS TOO MUCH?
Environmental Reproductive Number (R0E)
the number of secondary infections that arise from each index case in a space *
When R0E > 1 the space conditions amplify the disease and may increase healthcare burden
* Rudnick, S. N., & Milton, D. K. (2003). Risk of indoor airborne infection transmission estimated from carbon dioxide concentration. Indoor Air, 13(3), 237–245.
* LIAO, C. M., CHEN, S. C., & CHANG, C. F. (2008). Modelling respiratory infection control measure effects. Epidemiology and Infection, 136(3), 299–308.
Image:

15. Probability of Airborne Transmission
From:
We can calculate a ventilation rate 𝑄 𝑜𝑎 equivalent to the transmission rate limit (R0E) for any:
Limiting transmission rate (R0E)
Disease (q)
Number of occupants (n)
exposure time (t)
𝑸= 𝒏𝒑𝒒𝒕 𝜺 𝒗 ∙𝒍𝒏 𝒏−𝟏 𝒏− 𝟏+ 𝑹 𝟎𝑬 𝒏
Rudnick, S. N., & Milton, D. K. (2003).

16. HOW MUCH VENTILATION? 1. Limiting transmission rate: R0E = 1
Assumptions (For a 100 person waiting room)
1. Limiting transmission rate:
R0E = 1
2. Disease:
q = 60 /h
3. Number of occupants:
n = 100
4. Exposure time:
t = 4h
Assumptions are 2.7m ceiling heights and 0.8m²/person. Assuming 6m ceiling heights and 1.5m² per person would yield parity

17. HOW MUCH VENTILATION? (for a 100 person waiting room)
Assumptions are 2.7m ceiling heights and 0.8m²/person. Assuming 6m ceiling heights and 1.5m² per person would yield parity
Ventilation Rate ≥ 3200 L/s
Mechanical Ventilation
Room Air Cleaners
Natural Ventilation
Full Fresh Air
Particle Removal
Fully Passive
Recirculation
Pollutant Destruction
Mixed Mode
VENTILATION RATE:
≥ 3200 L/S
≈21 ACH

18. MECHANICAL VENTILATION (3 600 L/s)
Q= 3200 L/s
≈ 2 x laminar flow theatres
145 kW cooling
5 kW fan power
75 kW total
Assumptions are 2.7m ceiling heights and 0.8m²/person. Assuming 6m ceiling heights and 1.5m² per person would yield parity

19. HOW MUCH VENTILATION? Roe = 1
Convergence with 12 ACH is coincidental. Changing ceiling heights, exposure times or occupancy densities would greatly affect this point.
Roe = 1
1 patient with undetected pulmonary TB (*Nardell et al., 1991)

20. HOW MUCH VENTILATION? R0 ≈ 4
1 patient with undetected laryngeal TB (*Catanzaro, 1982)

21. HOW MUCH VENTILATION?
Index case with measles (*Liao et al, 2008)

22. HOW MUCH VENTILATION? 1% risk in a work year (q=13):
Ventilation alone cannot protect against long term exposure:
1% risk in a work year (q=13):
Q = 3,300,000 L/s (≈55M air changes per hour)
So lets talk about natural ventilation
Rockstar Games - Grand Theft Auto

23. NATURAL VENTILATION Highly technical design Benefits Pitfalls
Highest potential ventilation rates
ACH typical
Greater perception of comfort
Lower comfort expectations
Potentially better air quality
Lowest running costs
Healthier and greener buildings
Lowest potential ventilation rates
Variability in performance
Comfort levels not guaranteed
Potentially poor air quality
Outdoor pollution
Noise
Ambient, wind and speech
Highly technical design
Pitfalls not drawbacks

24. NATURAL VENTILATION Windows ≥ 5% floor area
Capacity Benefits:
Windows ≥ 5% floor area
100 person waiting: 100*0.9 = 90 m² (floor area)
Window area = 4.5 m²
+ Doors = 8.3 m²
Imperceptible breeze = m/s
= – L/s (17 – 41 ACH) (using only half the openable area)
The ventilation capacity of simple open windows…

25. CONCLUSION Ventilation is important to control airborne transmission
Ventilation is not the panacea often promised
Often higher rates are required than feasible
12 ACH may not be sufficient
A combination of controls is greater than the sum of its parts
Adapted from - LIAO, C. M., CHEN, S. C., & CHANG, C. F. (2008). Modelling respiratory infection control measure effects. Epidemiology and Infection, 136(3), 299–308.

26. Tobias van Reenen (tvreenen@csir.co.za)
Thank you
Tobias van Reenen