Available online at www.sciencedirect.

com

ScienceDirect
Procedia Engineering 121 (2015) 52 – 58

9th International Symposium on Heating, Ventilation and Air Conditioning (ISHVAC) and the 3rd
International Conference on Building Energy and Environment (COBEE)

Field-Measurement of CO2 Level in General Hospital Wards in

Nanjing
Qi Zhoua, Zhengfei Lyub, Hua Qiana,*, Jinwei Songa, Viola C. Möbsc
a
School of Energy and Environment, Southeast University, Nanjing, 210096, China
b
Jiangsu Province Hospital, Nanjing, 210029, China
c
Department of Environmental Engineering, RWTH Aachen University, Aachen, 52056, Germany

Abstract

Hospital indoor air quality (IAQ) has a significant impact on patients’ and health care workers’ health and the indoor
carbon dioxide (CO2) level is considered to be an indicator to evaluate IAQ in some cases. This article presents a long-
term field-measurement of indoor CO2 level in a general hospital ward in Nanjing. Four months’ data are collected
and analyzed to reveal variation rule of CO2 levels in wards in this periods. The results indicate that the variation rule
of indoor CO2 level is associated to patients’ living habit. The use of natural ventilation is capable to keep the indoor
CO2 level below 1000 ppm in transition season while it is much higher in heating season due to closing openings to
maintain thermal comfort. The results also demonstrate a “ω” shape of changing of CO2 levels within a day in heating
season.
©2015
© 2015TheThe Authors.
Authors. Published
Published by Elsevier
by Elsevier Ltd.
Ltd. This is an open access article under the CC BY-NC-ND license
Peer-review under responsibility of the organizing committee of ISHVACCOBEE 2015.
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of ISHVAC-COBEE 2015
Keywords: Field-measurement; CO2 level; IAQ; Hospital wards

1. Introduction

Indoor air quality (IAQ) in hospital environment, which affects both patients’ and health care workers’ (HCWs)
health, has drawn more and more our attention. It is because of the susceptible weak immune system and long exposure

* Corresponding author. Tel.:+86-136-4518-6001

E-mail address: keenwa@gmail.com

1877-7058 © 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of ISHVAC-COBEE 2015
doi:10.1016/j.proeng.2015.08.1018
 Qi Zhou et al. / Procedia Engineering 121 (2015) 52 – 58 53

period and unexpected infection from an index inpatient [1], especially during an epidemic period. For the purpose of
evaluating IAQ, the indoor CO2 level is considered to be an indicator in some cases. For instance, since occupants are
the dominant indoor source of CO2, the indoor CO2 level is used to estimate the sufficiency of ventilation rates, levels
of contaminants related occupant activity [2] and airborne infection risk [3]. It is also a good surrogate for indoor
concentrations of bioeffluents [4], e.g. body odor. However, the indoor CO2 level is influenced by many factors such
as human behavior, occupant density and performance of ventilation systems. Moreover, the indoor CO2 level is highly
variable in different seasons and even within a day. Therefore, a long-term field-measurement is a useful approach to
identify changing characteristics of the CO2 level in hospital wards.
In the present study, the field-measurement with duration of one year has been conducted in a hospital ward in
Nanjing since October 2014. Data from October 2014 to January 2015 is analyzed and the changing characteristics of
the indoor CO2 level in autumn and winter is then revealed. The effect of different factors is also discussed in this
article. The findings of this paper should have implications for future work that investigates a change rule in spring
and summer.

2. Methods

The field-measurement was conducted in a general hospital ward in Hospital R in Nanjing, China. The ward is on
the third floor of a hospital building and it consists of several cubicles. Among them, two cubicles (namely cubicle A
and cubicle B) were selected in this study. Cubicle A contains three beds and its dimensions are 3.5(m) x 7.4(m) x
2.6(m), while cubicle B contains six beds and its dimensions are 5.8(m) x 7.4(m) x 2.6(m). These two cubicles are
adjacent to each other and each cubicle holds two openings which are connected to the outside and the corridor
respectively. There is an air-conditioning installed in each cubicle whereas ventilation system is not. Thus, it is natural
ventilation in each cubicle that exchanges indoor and outdoor air through openings. The whole ward was operated as
usual during the measurement.
Indoor CO2 level, temperature, relative humidity and outdoor wind speed, wind direction, temperature and relative
humidity were continuously measured and recorded. Indoor parameters were measured by a TES CO2 monitor (1370)
(TES Corporation, Taiwan). Outdoor parameters were measured by a Vantage Pro2 weather station (DAVIS Inc.,
Hayward, CA, USA) which was located on the roof of the building. The distribution of measurement points in cubicles
is shown in Fig.1. Monitors were put 2.4m away from the floor in order not to disturb the daily operation of the ward
and be affected by patients’ respiratory activities.
It is a long-term measurement that will last for one year. It started on Oct 11, 2014. The data from Oct 11, 2014 to
Jan 20, 2015 is analyzed in this paper, which represents a period from a transition season (autumn) to a heating season
(winter).
&XELFOH%

&XELFOH$

+:

+:
+:

Fig. 1. Schematic plan view of cubicles and distribution of measurement points (red dot)
54 Qi Zhou et al. / Procedia Engineering 121 (2015) 52 – 58

35

30

25

Temperature (qC) 20

15

10
cubicle A
5
cubicle B
spare room
outside
0
2014/10/11 2014/11/11 2014/12/11 2015/1/11
Date

Fig. 2. Daily averaged indoor/outdoor temperature

3. Results

Fig.2 presents daily averaged indoor/outdoor temperature. In order to judge whether an air-conditioner was
switched on or not, the temperature of a spare room without an air-conditioner was also measured to make comparisons.
It can be observed that the outdoor temperature gradually decreases from a value above 20ćin October to that below
5ć in December 2014 and January 2015. It is a typical temperature variation process from autumn to winter in
Nanjing. The indoor temperature is remarkably higher than the outdoor temperature due to building thermal insulation
and indoor heat source. The indoor temperature of cubicle A and cubicle B and a spare room are nearly the same
before the middle of November, which indicates that air-conditioners were not operated during this period and patients
adjusted thermal comfort via opening or closing windows and doors. However, after the middle of November, it is
obvious that the temperatures of the two cubicles are higher than the one of a spare room, indicating that air-
conditioners were turned on. It should be noted that the critical outdoor and indoor temperature is 15ć and 20ć
below which patients are tend to close openings of cubicles and turn on air-conditioners. The temperature difference
between cubicle A and cubicle B may result from patients’ living habits and cubicle volumes.

2000 2000
OCT 2014 NOV 2014
1800 1800

1600 1600
CO2 concentration (ppm)

CO2 concentration (ppm)

1400 1400

1200 1200

1000 1000

800 800

600 600

400 400
00

00

0
00

00
0

0
:0

:0
:0

:0
:3

:3

:3

:0

:0

:3

:3

:0

:3

:0
:

:
1:

1:
~5

~8

~5

~8
12

14

17

18

21

24

12

14

17

18

21

24
~1

~1

A B
0~

0~

0~

0~

0~

0~

0~

0~

0~

0~

0~

0~
00

00

00

00
00

00
0:

5:

:0

:3

:3

:0

:3

:0

0:

5:

:0

:3

:3

:0

:3

:0
8:

8:
11

12

14

17

18

21

11

12

14

17

18

21
 Qi Zhou et al. / Procedia Engineering 121 (2015) 52 – 58 55

2000 2000
DEC 2014 JAN 2015
1800 1800

1600 1600
CO2 concentration (ppm)

CO2 concentration (ppm)

1400 1400

1200 1200

1000 1000

800 800

600 600

400 400
00

00

0
00

00
0

0
:0

:0
:0

:0

:0
:3

:3

:0

:3

:0

:3

:3

:0

:3
:

:
1:

1:
~5

~8

~5

~8
12

14

17

18

21

24

12

14

17

18

21

24
~1

~1
C D
0~

0~

0~

0~

0~

0~

0~

0~

0~

0~

0~

0~
00

00

00

00
00

00
0:

5:

:0

:3

:3

:0

:3

:0

0:

5:

:0

:3

:3

:0

:3

:0
8:

8:
11

12

14

17

18

21

11

12

14

17

18

21
2000 2000
OCT 2014 NOV 2014
1800 1800

1600 1600
CO2 concentration (ppm)

CO2 concentration (ppm)

1400 1400

1200 1200

1000 1000

800 800

600 600

400 400
00

00

0
00

00
0

0
:0

:0
:3

:3

:0

:3

:0
:3

:0

:3

:0

:0

:3

:0
:

:
1:

1:
~5

~8

~5

~8
14
12

17

18

21

24

12

14

17

18

21

24
~1

~1

E F
0~

0~

0~

0~

0~

0~

0~

0~

0~

0~

0~

0~
00

00

00

00
00

00
0:

5:

:0

:3

:3

:0

:3

:0

0:

5:

:0

:3

:3

:0

:3

:0
8:

8:
11

12

14

17

18

21

11

12

14

17

18

21
2000 2000
DEC 2014 JAN 2015
1800 1800

1600 1600
CO2 concentration (ppm)

CO2 concentration (ppm)

1400 1400

1200 1200

1000 1000

800 800

600 600

400 400
00

00

0
00

00

0
:0

:0
:3

:3

:0

:3

:0

:0

:3

:3

:0

:0
:3

:0
:

:
1:

1:
~5

~8

~5

~8
12

14

17

18

21

24

12

21

24
14

17

18
~1

~1

G H
0~

0~

0~

0~

0~

0~

0~

0~

0~

0~

0~

0~
00

00

00

00
00

00
0:

5:

:0

:3

:3

:0

:3

:0

0:

5:

:0

:3

:3

:0

:3

:0
8:

8:
11

12

14

17

18

21

11

12

14

17

18

21

Fig. 3. CO2 level of every periods (A, B, C, D-cubicle A, E, F, G, H-cubicle B)

In the present study, for analyzing CO2 change rule within a day, every day is divided into nine periods on the basis
of a patients’ daily routine: 00:00~5:00, 5:00~8:00, 8:00~11:00, 11:00~12:30, 12:30~14:30, 14:30~17:00,
17:00~18:30, 18:30~21:00, 21:00~24:00. Every period stands for specific daily activities, e.g. 8:00~11:00 as ward
round, 11:00~12:30 as lunchtime, 18:30~21:00 as evening leisure. The average CO2 levels of every period are shown
in Fig.3.
56 Qi Zhou et al. / Procedia Engineering 121 (2015) 52 – 58

CO2 levels of the two cubicles show similar tendency during the measurement. In October (Fig. 3A and 3E), the
CO2 level of every period fluctuates within a narrow range. Then, in December and January (Fig 3C, 3D and 3G, 3H),
it is obvious that CO2 levels of three periods (0:00~5:00, 12:30~14:30, 21:00~24:00) are higher than those of other
periods, just like a “ω”. This phenomenon may be related to patients’ daily routine and living habits. These three
periods represent sleeping time and noon break. During these periods, each cubicle is likely to be fully occupied and
more CO2 is released due to more “sources”. Besides, as previous analyzed, after the middle of November, patients
preferred using air-conditioners to maintain thermal comfort and thus openings of a cubicle should be closed,
especially when they were sleeping. Under these circumstances extremely low ventilation rate is resulted in.
Consequently, indoor CO2 cannot be effectively diluted by fresh air from outside and the indoor level rises higher.
Fig.3 also demonstrates that CO2 levels of the same period rise per month. Focusing on the period of 14:30~17:00 for
example, in October, CO2 levels of this period are 766 ppm and 785 ppm in cubicle A and cubicle B, respectively,
while in November CO2 levels increase to 823 ppm and 920 ppm and finally reach 987 ppm and 960 ppm in January
2015.
Fig.4 shows the daily averaged CO2 levels of cubicle A and cubicle B. It can be seen from Fig.4 that a rising process
of CO2 level in both cubicle A and cubicle B occurs in November and the CO 2 level of cubicle B is higher in this
period, compared with that of cubicle A. There are likely two reasons included. Firstly, it is due to the difference
between patients’ living habits. With the decrease of outdoor temperature, patients who are sensitive to temperature
variation are likely to take actions earlier in order to maintain thermal comfort, e.g. closing doors and windows,
switching on the air-conditioner. Consequently, as described before, air exchange is hindered and the CO2 level rises.
Secondly, it is because the occupant density is different in the two cubicles. Under normal conditions, cubicle A holds
3 patients while cubicle B holds 6. Considering the volumes of the two cubicles, if each cubicle is fully occupied, the
occupant density of cubicle A and cubicle B should be almost equal (approximate 22 m3/p of cubicle A and 19 m3/p
of cubicle B). However, in case that patients left or visitors came, the occupant density of the two cubicles would be
definitely different and as a consequence, CO2 level of each cubicle should correspondingly be different.
Table 1 summarizes the monthly averaged CO2 level of the two cubicles, which also indicates a rising tendency.
The CO2 levels are below 1000 ppm in October and November while from December the levels are above 1000 ppm,
reaching as high as 1095 ppm and 1120 ppm respectively in cubicle A and cubicle B.

1400
cubicle A
cubicle B
1200
CO2 concentration (ppm)

1000

800

600

400
2014/10/11 2014/11/11 2014/12/11 2015/1/11
Date

Fig. 4. Daily averaged CO2 level

 Qi Zhou et al. / Procedia Engineering 121 (2015) 52 – 58 57

Table 1. Summary of monthly averaged CO2 level

Month Cubicle A (ppm) Cubicle B (ppm)
Oct 2014 699 722
Nov 2014 794 952
Dec 2014 1059 1030
Jan 2015 1095 1120

A 0

90 270

180
2.5 360
B Wind speed Wind direction
315
2.0
270
Wind speed (m/s)

225 Wind direction (q)

1.5

180

1.0
135

90
0.5
45

0.0 0
2014/10/11 2014/10/18 2014/10/25 2014/11/1 2014/11/8
Date

Fig. 5. (A) A satellite image of the hospital building (B) Wind speed and wind direction from Oct 11 to Nov 11

A satellite image of the hospital building and wind speed and wind direction measured by the weather station from
Oct 11 to Nov 11 are shown in Fig.5. As previously discussed, air-conditioners were not in operation during this
period and patients maintained thermal comfort by opening/closing windows and doors. Therefore, natural ventilation
is formed and indoor/outdoor air exchange via openings is driven by wind force or buoyancy force. The daily averaged
CO2 levels (see Fig.4) during this period are mostly below 900 ppm which is remarkably lower than those in heating
58 Qi Zhou et al. / Procedia Engineering 121 (2015) 52 – 58

period. Although the measured wind speed is relatively low (0.5m/s in average, and wind speed should be even lower
at the height of window due to the gradient wind) and wind direction varies significantly, the type of natural ventilation
is capable to maintain the indoor CO2 level at a low level. The results demonstrate the potential of natural ventilation
to create an acceptable IAQ in general hospital wards.

4. Discussion

The change rule of CO2 level revealed in this article may be meaningful when analyzing the data of other two
seasons, spring and summer. It can be imagined that CO2 levels in those two seasons would present a similar process
to the one described in this paper. The change rule of the whole year might be like a sine curve, with peaks in
heating/cooling season and rising or decreasing process in transition seasons. Further, the IAQ of a hospital ward
through a year can thus be evaluated and it helps patients or HCWs to take adequate measures to create higher IAQ in
hospital wards.
A limiting value of 1000 ppm is widely accepted and this value is set as the daily maximum limit in an indoor air
quality standard of China. In this study, most of the measured daily averaged CO2 levels in December 2014 and
January 2015 exceed 1000 ppm, which implicates a potential threat to patients’ and HCWs’ health.
With the purpose of reducing daily CO2 levels and keeping them below the standard value of 1000 ppm, some
measures could be taken. The simplest way is to keep windows or doors partly open during patients’ stay. The openings
could be small and it should not enormously influence the indoor thermal comfort. Besides, installing a ventilation
system or personal ventilation (PV) is also a choice. However, the operating cost and energy consume will increase
and thus a further investigation is needed.

5. Conclusions

The field-measurement of indoor CO2 level in general hospital wards in Nanjing is reported in this paper and the
influencing factors are also discussed. The following conclusions could be drawn.
1. Patients’ living habit is a main influencing factor of the indoor CO 2 variation rule.
2. Patients tend to close openings as the outdoor temperature is below 15 ć to maintain thermal comfort, which
causes a rising process of indoor CO2 level.
3. As the outdoor temperature is above 15 ć, openings are kept open and thus natural ventilation is formed. The
use of natural ventilation is capable to maintain the indoor CO2 level at a low level, e.g. below 1000 ppm in the present
study.
4. In transition season, CO2 level in hospital wards fluctuates within a narrow range during the day while in heating
season, CO2 level varies significantly and three specific period of a day presents higher CO2 levels than others, which
illustrates the change rule as a “ω”.

Acknowledgements

The work described in this paper was funded by the Natural Science Foundation of China under the Project
no.51378103.

References

[1] Y. Li, X. Huang, I. T. S. Yu, T. W. Wong, H. Qian, Role of air distribution in SARS transmission during the largest nosocomial outbreak in
Hong Kong, Indoor Air. 15 (2005) 83-95.
[2] ASTM, ASTM Standard D6245-12, Standard guide for using indoor carbon dioxide concentrations to evaluate indoor air quality and ventilation.
American Society for Testing and Materials. West Conshohocken, PA, USA, 2012.
[3] S.N. Rudnick and Milton D.K., Risk of indoor airborne infection transmission estimated from carbon dioxide concentration, Indoor Air. 13
(2003) 237-245.
[4] O.A. Seppänen, W.J. Fisk, M.J. Mendell, Association of ventilation rates and CO2 concentrations with health and other responses in commercial
and institutional buildings, Indoor Air. 9 (1999) 226-252.