Copyright Blackwell Munksgaard 2003

Indoor Air 2003; 13: 344352

www.blackwellpublishing.com/ina
Printed in Denmark. All rights reserved

INDOOR AIR

Signicance of humidity and temperature on skin and upper airway

symptoms
Abstract The objective of the present study was to assess the eect of absolute
and relative humidity, temperature and humidication on workers skin and
upper airway symptoms, and perceptions in the oce environment. Associations
between physical factors, and symptoms and perceptions were assessed in logistic
regression models. At temperatures between 18 and 26C, relative humidity of
1740%, and absolute humidity of 3.35.6 g H2O/kg air, skin symptoms and
nasal dryness and congestion were alleviated by both kinds of humidity. Pharyngeal dryness increased when temperatures rose and was alleviated with a rise
in relative humidity. Eye symptoms showed no dependence on humidity. Any
kind of humidity increased odor sensation. Stuness increased when the air was
humidied. In non-humidied conditions (21.322.7C, 20.031.7%, 3.35.6 g
H2O/kg air), skin and nasal symptoms showed no association with humidity or
temperature. Pharyngeal dryness diminished when humidity rose. In addition,
the association between humidity and odor disappeared. In humidied conditions (21.523.7C, 26.641.2%, 4.27.0 g H2O/kg air), nasal dryness and congestion were alleviated by both absolute and relative humidity, and odor
perception increased. Skin dryness and rash, pharyngeal dryness, and nasal
dryness and congestion are alleviated in higher humidity. Steam humidication
results in a risk for increased perception of odor and stuness.

L. M. Reinikainen1,2,
J. J. K. Jaakkola1,2,3
1
Department of Public Health, P.O. Box 41, University of
Helsinki, Helsinki FIN-00014, Finland, 2Laboratory of
Heating, Ventilating and Air-Conditioning, Faculty of
Mechanical Engineering, Helsinki University of
Technology, P.O. Box 4100, Helsinki FIN-02015, Finland,
3
The Institute of Occupational Health, The University of
Birmingham, Edgbaston, Birmingham B15 2TT, United
Kingdom

Key words: Indoor air; Office environment; Sick building

syndrome; Humidity; Humidification; Temperature.
Leena M. Reinikainen
Department of Public Health, P.O. Box 41, University of
Helsinki, Helsinki FIN-00014, Finland
Fax: 358 0 27540
e-mail: leena.reinikainen@helsinginenergia.fi
Received for review 20 September 2000.
Accepted for publication 25 May 2001
Indoor Air (2003)

Practical Implications

In cold climates, dry air seems to be related to skin symptoms and nasal dryness. Skin symptoms may be alleviated by
lowering room temperature which increases relative humidity. Nasal dryness is more dependent on absolute humidity.
Steam humidication includes a risk of feeling the air more stuy.

Introduction

With air-conditioning, temperature and humidity are,

despite air change rate and airow velocity, basic
physical elements aecting human comfort inside a
building. Absolute humidity is the content of water
vapor in the air (expressed, e.g. in vapor pressure, hPa,
g H2O/kg air). Relative humidity (%) is the ratio of the
actual vapor pressure and the saturation vapor pressure, which is an exponential function of air temperature. Temperature and humidity aect the thermal
balance of the human body via respiratory organs and
skin (Hoppe, 1983). They also regulate human comfort
both directly via thermal sensations and indirectly by
changing the perceived indoor air quality parameters
such as stuness (Berglund and Cain, 1989). We have
studied the eect of humidication on skin and airway
344

symptoms, and on perceived indoor air quality. We

found that during the heating season in Finland,
humidication alleviates dryness symptoms of skin and
upper airways in oce environments and excessive
indoor air temperature increases unpleasant symptoms
of the upper respiratory tract and skin (Palonen et al.,
1993; Reinikainen et al., 1991, 1992).
According to the research of Hoppe, air humidity is
the most important physical factor aecting energy
balance of the upper respiratory tract. Inhaling
requires energy for humidifying the air, which under
most conditions including the oce environment is far
from the saturation humidity, to relative humidity of
100% in the core temperature of 37C. Some energy is
gained in exhaling, when a portion of the humidity
condenses on surfaces of the mucosa. Energy loss from
upper airways depends on both absolute and relative

Eect of humidity and temperature on skin and upper airway symptoms

humidity because it is dependent also on ambient
temperature (Hoppe, 1983). Hoppe and Martinac have
postulated that evaporation from dierent parts of skin
is dependent on whether the area is covered with
clothing. Evaporation from clothed skin is more
dependent on absolute than relative humidity. Bare
skin, such as the head, has a temperature closer to the
temperature outside the body, and thus, the evaporation is more dependent on relative humidity (Hoppe
and Martinac, 1998). Sommer and coworkers have
studied the stability of the tear lm in oce conditions.
They found that in dry climates the breakup time of the
tear lm is reduced, causing a phenomenon known as
the oce eye syndrome (Sommer et al., 1994). An
association also appears to exist in the oce environment between eye and throat irritations, and an
increase in reported eye irritations at temperatures
between 22 and 26C (Backman and Haghighat, 1999).
The objective of our study was to investigate the
relationship of symptoms and perceptions reported by
workers to temperature, and to the absolute and
relative humidity in oce environment. We presumed
that some of the symptoms, such as sensation of
dryness of mucosa and eyes, and skin symptoms, are
associated with a loss of energy from the corresponding part of the body. Some others such as nasal
dripping would reect the condensation of water
vapor on the mucosa. We also wanted to analyze the
role of humidication itself on the symptoms and
perceptions.
Methods

A 6-week cross-over study was conducted in 1989 in

the Pasila Oce Center by changing humidication
and studying corresponding changes in the symptoms
and perceptions of the occupants. The building consists
of six symmetrical wings joined by a central part
(Figure 1). This study was conducted in the three
northern wings of the building (A, B, and C).The wings
comprise small oces on oors 38 which were
included in the study. Two of the wings (A and B)
were humidied with steam humidication, and the
third wing (C) on the same side of the building was
used as a non-humidied reference. The humidication
took place for one week in each of the wings of
experiment. Changes in humidication were conducted
during the weekends representing the washout period.
In the third wing (C), personal data and air quality
measurements were similarly collected, but the wing
was not humidied (Reinikainen et al., 1992).
The data concerning symptoms and perceived indoor
air quality were received from structured diaries which
the participants lled in every afternoon. The symptoms were coded from 0 for none to 3 for strong
symptoms. Odor and stuness were coded from 0 to 5.
The diary also comprised questions concerning symp-

Fig. 1 Pasila Oce Center, experiment wings A and B, and

control C

toms of upper respiratory infections. Answers from

days of any symptoms of infection were rejected from
the analysis, and the workers had to have spent at least
2 h in their oce to be accepted in the analysis.
In connection with the humidication trial, relative
humidity and temperature were measured continuously
in all three wings in two or three oces giving
information on relative humidity and temperature in
the corresponding wing. In addition, each participant
received a dry bulb thermometer whose reading was
registered in connection of lling in the symptom diary.
Thus, we received a detailed picture of the participants
exposure to both humidity and temperature. Based on
the diagrams constructed for the calculation of air
humidity (Molliers diagram), we rst calculated the
absolute humidity of each day of the study from the
registrations of the continuous measurements. Next,
we calculated the relative humidity in each oce based
on the daily temperature data registered by the
participants.
Indoor air quality

The air change was measured in each room in every

wing before the beginning of the study. Chemical and
biological impurities measured were formaldehyde,
fungal spores and growth, bacterial concentration,
and the total amount of suspended particles, as these
were suspected to be aected by the humidication. Air
samples for formaldehyde, fungal spores, and bacteria
were gathered from a sample of oces both during and
without humidication. In addition, surface samples for
detecting possible fungal growth were gathered from
the ventilation pipelines. The total amount of dust was
collected from the intake and outlet conducting pipe345

Reinikainen & Jaakkola

lines in the wings of the humidication experiment and
the outlet pipeline in the reference wing.
Statistical methods

Answers for symptoms and perceptions were dichotomized (no symptoms, no odor or stuness 0, any
degree of symptoms, any odor or stuness 1) and
associations between them and humidity were assessed
in logistic regression models using a repeated measurements technique to take into account the dierent
number of answers received from individuals.
Associations between symptoms and perceptions,
and the exposure to either absolute or relative
humidity, temperature and humidication were
assessed in logistic regression models including an
indicator variable in order to take into account the
uneven number of answers received from the individuals (max 30, min 1).
The associations were assessed in four types of
models. The rst model contained one of the two
humidity types from each individuals oce, the
corresponding temperature in the oce, and humidication. The second model assessed absolute or relative
humidity stratied with temperature. The third model
assessed absolute or relative humidity, taking into
account the presence or absence of humidication, and
the fourth model contained only the absolute or
relative humidity. Odor and stuness were connected
with how long the workers had their windows open,
and this time was taken into these models as a
confounding factor. In addition, two sets of models
were calculated using either humidied conditions
(wings A and B) or non-humidied conditions (wings
A, B, and C). All calculations were made with the SPSS
statistical package release 8.0.

Results
Study population

A total of 368 workers (71.2%) returned the baseline

questionnaire and at least one of the diaries with
information on any of the symptoms or perceptions

of interest, sucient information

symptoms of respiratory infection,
least 2 h in their oce. In all,
received from non-humidied and
ied conditions.

on
and
342
233

the possible
had spent at
diaries were
from humid-

Temperature and humidity

The mean temperature according to the continuous

measurements collected during the non-humidied
period in wings A and B and the total period in C was
21.9C (21.322.7C), and during the humidied period
in A and B, 22.4C (21.523.7C). The mean relative
humidity was 25.8% (20.031.7%) and 32.7% (26.6
41.2%) for non-humidied and humidied conditions,
respectively. The corresponding absolute humidities
were 4.2 (3.35.6) and 5.6 (4.27.0). More detailed data
of temperature and humidity are given in Table 1.
Ventilation rate and air impurities

The ventilation rate was in average very high, 770 l/s

for each person (mean 24; s.d. in the trial area 9.9 l/s
for each person). The formaldehyde concentrations
were all below 0.1 mg/m3, and there was no association with humidication. Concentrations of particles,
bacteria, and fungal spores in the building were low.
The species of the fungi were the same which are
present in the outdoor air. The surface samples did
not show any fungal growth. No major dierences
occurred between the three wings in any of the
concentrations.
Associations between symptoms and physical indoor air conditions

The associations between symptoms and physical

indoor air conditions, including humidity, temperature,
and humidication, are shown in Tables 2 and 3, the
former focusing on absolute humidity and the latter on
relative humidity, both including the total population.
Tables 4 and 5 contain the associations between
absolute and relative humidity, and temperature separately in non-humidied and humidied conditions
correspondingly.

Table 1 Temperatures, and absolute and relative humidity in the Pasila Office Center; mean (range)
Continuous measurements
Wing
Non-humidified
A
B
C
Humidified
A
B

346

Conditions in individual offices

Temperature (C)

Relative humidity (%)

Absolute humidity (g H2O/kg air)

Temperature (C)

Relative humidity (%)

21.5 (21.322.5)
22.4 (22.022.7)
21.8 (21.322.2)

25.4 (21.231.7)
25.0 (20.030.2)
26.4 (21.231.4)

4.1 (3.55.3)
4.3 (3.35.6)
4.3 (3.55.2)

21.9 (19.025.0)
22.8 (18.025.0)
22.4 (18.025.0)

24.8 (17.439.9)
24.6 (16.637.3)
24.9 (16.638.0)

22.0 (21.822.3)
22.8 (21.523.7)

34.3 (31.741.2)
31.1 (26.638.3)

5.7 (5.26.9)
5.4 (4.27.0)

22.2 (19.026.0)
23.0 (19.526.0)

34.0 (27.949.1)
30.8 (22.244.2)

Eect of humidity and temperature on skin and upper airway symptoms

Table 2 Associations between absolute humidity and temperature, and skin, eye and upper airway symptoms, odor and stuffiness in the total population; logistic regression; odds ratio
(OR) per 1 g H2O/kg air and 1C
Absolute humidity

Temperature

Humidification

Outcome

Model number

OR

95% CI

OR

95% CI

OR

95%CI

Skin dryness

1
2
3
4

0.81
0.79
0.81
0.80

(0.621.06)
(0.650.95)*
(0.621.06)
(0.660.96)*

1.19
1.14

(0.911.55)
(0.881.48)

0.88

(0.521.51)

0.93

(0.551.58)

1
2
3
4

0.81
0.70
0.81
0.69

(0.551.19)
(0.510.95)*
(0.551.19)
(0.510.94)

0.78
0.78

0.60

(0.261.37)

0.59

(0.261.33)

1
2
3
4

0.96
0.96
0.96
0.89

(0.741.25)
(0.741.25)
(0.741.25)
(0.741.07)

1.00
1.01

0.80

(0.481.33)

0.80

(0.481.33)

1
2
3
4

0.83
0.87
0.84
0.88

(0.631.11)
(0.701.07)
(0.631.11)
(0.711.08)

1.09
1.11

1.09

(0.582.05)

1.12

(0.602.09)

1
2
3
4

0.74
0.71
0.74
0.72

(0.570.95)*
(0.590.85)*
(0.570.95)*
(0.610.86)*

1.28
1.26

0.86

(0.521.42)

0.92

(0.561.52)

1
2
3
4

1.02
0.84
1.03
0.86

(0.771.34)
(0.681.03)
(0.781.35)
(0.701.05)

1.34
1.30

0.51

(0.280.91)*

0.54

(0.300.96)*

1
2
3
4

1.05
0.98
1.03
0.94

(0.731.50)
(0.741.31)
(0.721.48)
(0.711.25)

0.71
0.70

0.81

(0.351.87)

0.72

(0.321.65)

1
2
3
4

1.24
1.34
1.23
1.30

(0.921.68)
(1.041.72)*
(0.911.66)
(1.011.67)*

0.72
0.73

1.38

(0.633.02)

1.27

(0.59-2.74)

1
2
3
4

1.36
1.33
1.36
1.34

(1.041.79)*
(1.111.61)*
(1.041.79)*
(1.121.62)*

1.09
1.09

0.96

(0.541.70)

0.99

(0.561.73)

1
2
3
4

1.13
1.35
1.13
1.37

(0.911.40)
(1.161.58)*
(0.911.40)
(1.171.60)*

1.13
1.12

1.71

(1.112.65)*

1.76

(1.142.71)*

Skin rash

Eye dryness

Pharyngeal dryness

Nasal dryness

Nasal congestion

Nasal excretion

Sneezing

Odora

Stuffinessa

(0.521.16)
(0.521.15)

(0.771.31)
(0.771.31)

(0.841.43)
(0.851.44)

(0.991.67)
(0.971.63)

(1.011.79)
(0.991.72)

(0.491.04)
(0.491.02)

(0.530.98)*
(0.540.99)*

(0.791.49)
(0.811.48)

(0.891.45)
(0.931.51)

Models: 1, absolute humidity + temperature + humidification; 2, absolute humidity + temperature; 3, absolute humidity + humidification; 4, absolute humidity.
Controlled for the number of hours people had their windows open daily.
* P < 0.05.

Skin symptoms seemed to be alleviated by increasing

any type of humidity. The association with relative
humidity was slightly stronger. The eect was constant
in all conditions though the association was statistically signicant only in the total population, and in
models containing either the humidity alone or humidity and temperature. In models containing the humidication, the eect of humidity disappeared. As a rule,
high temperature increased the sensation of skin
dryness. In the non-humidied conditions the eect

was opposite. Skin rash was alleviated on temperature

increase in all but the humidied conditions. The
association between temperature and skin symptoms
was not statistically signicant.
Eye dryness was also constantly alleviated by
humidity, though the eect was not statistically signicant in any of the models. Humidication also had an
alleviating eect, though not statistically signicant.
High temperature either had no eect or alleviated eye
dryness.
347

Reinikainen & Jaakkola

Table 3 Associations between relative humidity and temperature, and skin, eye and upper airway symptoms, odor and stuffiness in the total population; logistic regression; OR per %
relative humidity (RH) and 1C
Relative humidity

Temperature

Humidification

Outcome

Model number

OR

95% CI

OR

95% CI

OR

95% CI

Skin dryness

1
2
3
4

1.00
0.96
0.97
0.96

(0.921.08)
(0.930.99)*
(0.921.02)
(0.931.00)*

1.20
1.13

(0.911.54)
(0.881.47)

0.84

(0.501.43)

0.89

(0.531.50)

1
2
3
4

0.96
0.94
0.96
0.94

(0.901.04)
(0.881.00)*
(0.901.04)
(0.880.99)*

0.77
0.77

0.59

(0.261.35)

0.57

(0.251.30)

1
2
3
4

0.99
0.98
0.99
0.98

(0.951.04)
(0.951.01)
(0.951.04)
(0.951.01)

1.00
1.00

0.80

(0.481.32)

0.80

(0.481.32)

1
2
3
4

0.96
0.97
0.96
0.97

(0.911.01)
(0.931.01)
(0.921.02)
(0.931.01)

1.09
1.11

1.12

(0.602.08)

1.15

(0.622.12)

1
2
3
4

0.95
0.94
0.95
0.95

(0.911.00)*
(0.910.97)*
(0.911.00)*
(0.910.98)*

1.28
1.24

0.80

(0.481.33)

0.87

(0.531.42)

1
2
3
4

1.01
0.97
1.01
0.97

(0.961.06)
(0.931.01)
(0.961.06)
(0.941.01)

1.34
1.29

0.50

(0.280.89)*

0.53

(0.300.95)

1
2
3
4

0.99
0.99
0.99
0.98

(0.931.06)
(0.941.04)
(0.931.06)
(0.931.03)

0.71
0.71

0.90

(0.392.06)

0.99

(0.931.06)

1
2
3
4

1.05
1.06
1.05
1.06

(1.001.11)
(1.021.11)*
(0.991.11)
(1.011.11)*

0.72
0.73

1.28

(0.5892.78)

1.17

(0.542.51)

1
2
3
4

1.07
1.06
1.07
1.06

(1.021.13)*
(1.021.10)*
(1.021.13)*
(1.031.10)*

1.10
1.09

0.87

(0.491.55)

0.90

(0.511.58)

1
2
3
4

1.03
1.06
1.03
1.06

(0.991.07)
(1.031.09)*
(0.991.07)
(1.031.10)*

1.13
1.19

1.65

(1.062.55)*

1.70

(1.102.62)*

Skin rash

Eye dryness

Pharyngeal dryness

Nasal dryness

Nasal congestion

Nasal excretion

Sneezing

Odora

Stuffinessa

(0.521.15)
(0.521.14)

(0.771.31)
(0.771.31)

(0.841.42)
(0.851.44)

(0.981.66)
(0.961.61)

(1.011.78)*
(0.981.71)

(0.491.04)
(0.491.03)

(0.530.98)*
(0.540.99)*

(0.801.51)
(0.811.48)

(0.891.45)
(0.941.51)

Models: 1, relative humidity + temperature + humidification; 2, relative humidity + temperature; 3, relative humidity + humidification; 4, relative humidity.
Controlled for the number of hours people had their windows open daily.
* P < 0.05.

Pharyngeal dryness was slightly increased by higher

humidity, and increased by high temperature. In the
non-humidied conditions the eect of temperature
was opposite. Humidication increased pharyngeal
dryness. None of the eects was statistically signicant.
Nasal dryness was alleviated by an increase in
humidity of any kind in all but the non-humidied
conditions, and the alleviating eect was statistically
signicant. High temperature increased nasal dryness,
and humidication decreased it. These eects were not
statistically signicant.
348

Nasal congestion was constantly decreased by

humidication, and the eect was statistically signicant in all models containing the humidication. In
humidied conditions, both types of humidity seemed
to have a statistically signicant alleviating eect if
they were in the model alone. High temperature
increased nasal congestion, and the eect was statistically signicant in the models containing either
absolute or relative humidity, temperature, and
humidication. In humidied conditions the association was statistically signicant in the model

Eect of humidity and temperature on skin and upper airway symptoms

Table 4 Associations between absolute and relative humidity and temperature, and skin, eye and upper airway symptoms, odor and stuffiness in the non-humidified conditions; logistic
regression; OR per 1 g H2O/kg air, %RH, and 1C
Absolute humidity

Temperature

Relative humidity

Temperature

Outcome

Model

OR

95% CI

OR

95% CI

Model

OR

95% CI

OR

95% CI

Skin dryness

nHat
nHa
nHat
nHa
nHat
nHa
nHat
nHa
nHat
nHa
nHat
nHa
nHat
nHa
nHat
nHa
nHat
nHa
nHat
nHa

0.76
0.76
0.74
0.72
0.97
0.97
0.78
0.77
1.02
1.04
1.24
1.26
0.90
0.88
1.15
1.13
0.90
0.90
1.06
1.06

(0.531.07)
(0.531.07)
(0.461.20)
(0.451.16)
(0.691.37)
(0.691.36)
(0.551.11)
(0.551.10)
(0.731.43)
(0.741.45)
(0.891.71)
(0.911.74)
(0.591.38)
(0.581.34)
(0.811.62)
(0.801.60)
(0.591.36)
(0.591.37)
(0.781.45)
(0.771.44)

0.99

(0.721.35)

0.60

(0.370.98)

0.96

(0.711.31)

0.92

(0.671.26)

0.93

(0.681.26)

0.83

(0.601.15)

1.22

(0.891.69)

1.26

(0.901.76)

1.22

(0.891.69)

1.23

(0.891.72)

0.68

(0.451.03)

0.66

(0.431.01)

0.70

(0.500.98)

0.74

(0.521.05)

1.40

(0.932.09)

1.31

(0.862.00)

0.90

(0.661.21)

(0.931.03)
(0.931.04)
(0.891.04)
(0.911.06)
(0.921.03)
(0.931.03)
(0.901.00)
(0.911.01)
(0.961.07)
(0.961.06)
(0.951.05)
(0.941.04)
(0.921.05)
(0.941.07)
(0.971.08)
(0.991.09)
(0.901.04)
(0.891.02)
(0.951.05)
(0.961.05)

(0.681.30)

(0.411.05)

0.98
0.98
0.96
0.98
0.97
0.98
0.95
0.95
1.02
1.01
1.00
0.99
0.98
1.00
1.02
1.04
0.97
0.95
1.00
1.00

0.94

0.66

nHrt
nHr
nHrt
nHr
nHrt
nHr
nHrt
nHr
nHrt
nHr
nHrt
nHr
nHrt
nHr
nHrt
nHr
nHrt
nHr
nHrt
nHr

0.89

(0.651.23)

Skin rash
Eye dryness
Pharyngeal dryness
Nasal dryness
Nasal congestion
Nasal excretion
Sneezing
Odora
Stuffinessa

Models: nHat, non-humidified, absolute humidity + temperature; nHa, non-humidified, absolute humidity; nHrt, non-humidified, relative humidity + temperature; nHr, non-humidified,
relative humidity.
a
Controlled for the number of hours people had their windows open daily.
* P < 0.05.
Table 5 Associations between absolute and relative humidity and temperature, and skin, eye and upper airway symptoms, odor and stuffiness in the humidified conditions; logistic
regression; OR per 1 g H2O/kg air, %RH, and 1C
Absolute humidity

Temperature

Relative humidity

Temperature

Outcome

Model

OR

95% CI

OR

95% CI

Model

OR

95% CI

OR

95% CI

Skin dryness

Hat
Ha
Hat
Ha
Hat
Ha
Hat
Ha
Hat
Ha
Hat
Ha
Hat
Ha
Hat
Ha
Hat
Ha
Hat
Ha

0.82
0.76
0.86
0.80
0.73
0.74
0.85
0.83
0.42
0.41
0.50
0.44
1.32
1.33
1.62
1.69
3.23
3.27
0.95
0.94

(0.501.34)
(0.471.23)
(0.372.00)
(0.361.79)
(0.441.23)
(0.441.24)
(0.471.54)
(0.461.48)
(0.250.69)*
(0.250.66)*
(0.241.05)
(0.220.92)*
(0.612.90)
(0.612.93)
(0.813.22)
(0.853.34)
(1.915.44)*
(1.945.52)*
(0.661.36)
(0.651.35)

1.77

(1.003.13)

2.17

(0.677.04)

0.88

(0.431.78)

0.81

(0.391.70)

1.47

(0.772.79)

1.32

(0.662.63)

1.15

(0.622.13)

0.85

(0.441.67)

2.43

(1.045.68)*

1.96

(0.794.88)

1.32

(0.612.90)

0.47

(0.121.87)

0.66

(0.261.65)

0.80

(0.302.13)

0.79

(0.371.68)

1.19

(0.542.63)

1.27

(0.702.30)

(0.891.07)
(0.861.02)
(0.811.15)
(0.791.09)
(0.871.05)
(0.881.06)
(0.851.06)
(0.841.04)
(0.780.93)*
(0.790.93)*
(0.781.02)
(0.750.97)*
(0.901.21)
(0.921.23)
(0.981.28)*
(1.001.28)*
(1.945.52)*
(1.131.36)*
(0.921.06)
(0.921.05)

(0.923.16)

(0.766.95)

0.97
0.94
0.97
0.93
0.95
0.96
0.95
0.93
0.85
0.86
0.89
0.85
1.04
1.06
1.12
1.13
3.27
1.24
0.99
0.98

1.70

2.29

Hrt
Hr
Hrt
Hr
Hrt
Hr
Hrt
Hr
Hrt
Hr
Hrt
Hr
Hrt
Hr
Hrt
Hr
Hrt
Hr
Hrt
Hr

1.24

(0.672.30)

Skin rash
Eye dryness
Pharyngeal dryness
Nasal dryness
Nasal congestion
Nasal excretion
Sneezing
Odora
Stuffinessa

Models: Hat, humidified, absolute humidity + temperature; Ha, humidified, absolute humidity; Hrt, humidified, relative humidity + temperature; Hr, humidified, relative humidity.
Controlled for the number of hours people had their windows open daily.
* P < 0.05.

containing absolute humidity and temperature or

relative humidity alone.
Nasal excretion showed no constant patterns other
than the non-signicant alleviating eect of humidi-

cation. Sneezing was alleviated in higher temperatures,

and increased in higher humidity. The eect of
temperature was statistically signicant in the total
population, and the eect of any kind of humidity also
349

Reinikainen & Jaakkola

in the total population in the models not containing
humidication.
The sensation of odor was increased by humidity.
The eect was statistically signicant in the total
population in all models. The association with absolute
humidity was slightly stronger compared to the relative
humidity. In the total population, association between
odor and humidication was non-signicant and
showed a decrease in the sensation of odor during
humidication. There was a non-signicant direct
association with temperature. In the non-humidied
conditions the associations with humidity or temperature were not signicant.
Stuness seemed to be associated with humidication in the total population. Humidication increased
the sensation of stuness. The eects of high
temperature and an increase in both absolute and
relative humidity had the same direction in the total
population. The association to temperature was not
signicant, and the association to humidity was
statistically signicant only if humidication was not
in the model. In non-humidied conditions, the
association with temperature was opposite to the
total population and humidied conditions. In humidied conditions, stuness and humidity had a reverse
association. These associations were not statistically
signicant.
As a summary, it seemed that in the non-humidied
conditions (temperature 1825C, absolute humidity
3.55.6 g water/kg air, and relative humidity 2032%),
no statistically signicant associations were found
between humidity and temperature, and the symptoms
of upper airways or odor or stuness. In the humidied conditions (temperature 1926C, absolute
humidity 4.27.0 g water/kg air, and relative humidity
2741%), statistically signicant increase was found
between odor and sneezing, and decrease between nasal
dryness and congestion and humidity. In humidied
conditions, nasal congestion was increased by high
temperature, and the association was statistically
signicant in the model containing absolute humidity
and temperature. Having the total population in the
models assessing the associations between absolute or
relative humidity, and temperature and humidication,
skin symptoms showed a decrease when humidity was
higher, but only the models not containing humidication were statistically signicant. The association
with absolute humidity was stronger. Nasal dryness
seemed to be constantly alleviated by humidity, absolute humidity having a stronger eect. Nasal congestion was alleviated by humidication, and high
temperature increased it. The sensation of odor was
consistently increased by humidity, and the association
was stronger with absolute humidity. Stuness showed
an association with humidication, and there was
association with humidity only in the models where
humidication was not present.
350

Discussion

We studied the inuence of temperature and humidity

under real-life conditions in the oce environment. In
addition to our previous report (Reinikainen et al.,
1992) we included in the analysis, the population in the
C wing, who were only exposed to the natural variation
in humidity. Although the control of environmental
conditions was poorer than in a laboratory setting, the
ndings are expectedly more relevant and describe the
conditions in a cold climate area.
Skin symptoms of dryness and rash were alleviated
by both absolute and relative humidity. Alleviation
was slightly more dependent on absolute humidity. The
skin symptoms did not include a denition of the body
part where the symptoms were felt. Taking into
consideration that during winter the majority of the
skin is covered with clothing, the nding is in agreement with the model of Hoppe and Martinac
(1998).Our previous results showed that humidication
alleviates skin dryness (Reinikainen et al., 1992).
Surprisingly, eye symptoms were not signicantly
associated with either kind of humidity. Our previous
results had shown a decrease in the amount of eye
symptoms when the air was humidied (Reinikainen
et al., 1992). According to Carsten and Boge (1993) the
oce eye syndrome shows no seasonal variation,
which may partly explain the virtual independence of
eye symptoms from humidity in this study population,
comprising the workers in the wings of the experiment
and a third part of the population whose conditions
were only dependent on natural conditions. On the
other hand, neither was there a dependence on
temperature, contrary to the result of Backman and
Haghighat (1999) revealing associations between eye
symptoms and temperature rise, as well as pharyngeal
dryness and eye symptoms.
Our previous result showed that humidication does
not signicantly alleviate pharyngeal dryness, and the
present result, showing a slight decrease when relative
humidity was increased, is in line with this nding
(Reinikainen et al., 1992). The result concurs with the
model of Hoppe and Martinac (1998) in that both
humidity and temperature aect energy loss from
upper airway mucosa, taken that the feeling of dryness
is dependent on energy loss. The contradictory (though
non-signicant) eects of temperature in humidied
and non-humidied conditions may mirror the energy
content of the humidity.
The alleviation of nasal dryness and congestion by
both absolute and relative humidity concurs with
Hoppe and Martinac (1998), who speculate that these
symptoms are associated with energy loss from
mucosa. Temperature rise seems to increase nasal
congestion. Humidication alone caused no dierence
in these symptoms in our previous study, possibly
because of the higher temperature in the humidied

Eect of humidity and temperature on skin and upper airway symptoms

conditions (Reinikainen et al., 1992). Sneezing
increased in higher humidity. Thus, while nasal dryness
and congestion appeared to act as a single entity,
increasing when temperature rose and diminishing
when humidity rose, sneezing behaved in the opposite
manner.
Both absolute and relative humidity increased the
sensation of odor in humidied conditions. By contrast, in non-humidied conditions, no association was
present. Higher humidity is not likely to increase
emissions from the building and furnishing materials
other than highly water soluble compounds (Fang
et al., 1999). According to the results, humidication is
not associated with odor, as shown in the models tted
with the humidication variable. Our previous results
show an increase in odor sensation when air is
humidied which may be seen to be concordant with
the present result as humidity alone was not assessed in
the previous model (Reinikainen et al., 1992). The
nding that the sensation of odor is associated with
humidity is opposite to the results of two chamber
studies where the intensity of odor was found to be
independent of humidity in temperature range 1828C
and 3070% relative humidity. (Fang et al., 1998a,b).
We cannot totally exclude the possibility in our study
that some compounds of the humidication water, or
emissions from the building or furnishing materials
would have been the cause of an increase in the odor
perception in higher humidity. Factors striking against
this possibility are the nature of the water which
originates from a very pure raw water source and is not
chlorinated. The ventilation rate, in average 2030 l/s
per person is exceptionally high, thus diluting the
possible emissions eectively. We suspect that very low
humidity conditions may partly cause the insuciency
in odor perception, and a rise in humidity enhances the
odor perception ability of the olfactory mucosa.
Stuness showed possibly the most interesting
features in this assessment. It was associated with
humidity only if humidication was not included in the
model. We suspect that this phenomenon mirrors the
nature of stuness as a sensation of more than only
odor or sensation in the airways mucosa. The researchers who visited the building described the feeling during
humidication as moisture on the skin and a congestive
feeling in the airways. Some of the workers reported
that during the humidication there was something in
the air resembling a Finnish sauna. The surprising
nding that during non-humidied conditions higher
temperature was associated with lower sensation of
stuness may be a consequence of the people opening

their windows because of too high temperature, excessive odor, or stuness. That the temperature dependence did not appear under humidied conditions may
mirror the rather strong stuness-causing eect of
steam humidication. Our previous results also showed
that the greatest dierence in the individual symptoms
and perceptions between humidied and non-humidied conditions was in the perception of stuness
(Reinikainen et al., 1992). In a report by Berglund and
Cain (1989), indoor air was perceived as less stuy at a
lower temperature and humidity. In their study, the
increase in humidity was achieved articially, and thus,
it is impossible to evaluate the role of humidication
and humidity separately. On the other hand, a Swedish
4-month follow-up study in hospitals, also using steam
humidication, did not show an increase in stuness
experienced in humidied buildings (Nordstrom et al.,
1994). Possibly, the contrast between humidied and
non-humidied conditions disappears over a longer
time period.
Conclusions

If no humidication is used during the heating season

which in Finland means very low humidity, skin and
upper airway symptoms, odor perception, and the
sensation of air stuness show no associations with a
slight rise in humidity or temperature.
Our results suggest that increasing indoor air
humidity during the heating season, to a reasonable
level, will alleviate skin symptoms and nasal dryness
and congestion. The goals can partly be achieved by
lowering room temperature alone, which increases the
relative humidity, and simultaneously decreases the
eects of temperature, which often are opposite to the
eects of an increase in humidity. Humidication
emphasizes the alleviating eect of humidity on nasal
dryness and congestion, as well as the perception of
odor.
Diminishing nasal dryness and skin symptoms may
further be alleviated by increasing indoor air humidity.
However, the trade-o of having a less dry and
congested nose, is an increase in sneezing and odor
perception. Eye and pharyngeal dryness show no
association with humidity or temperature. Stuness
seems to be increased by articial steam humidication.
Acknowledgements

This study was supported by the Finnish Work

Environment Fund.

351

Reinikainen & Jaakkola

References
Backman, H. and Haghighat, F. (1999)
Indoor-air quality and ocular discomfort,
J. Am. Optometr. Assoc., 70, 309316.
Berglund, L.G. and Cain, W.S. (1989) Perceived air quality and the thermal environment. In: Proceedings of the
ASHRAE/SOEH Conference, IAQ 89,
San Diego, American Society of Heating,
Refrigerating, and Air-Conditioning
Engineers, Inc., 9399.
Carsten, F. and Boge, I. (1993) Break-up
time and lissamine green epithelial damage in office eye syndrome, Six-month
and one-year follow-up investigations,
Acta Ophthalmol., 71, 6264.
Fang, L., Clausen, G. and Fanger, P.O.
(1998a) Impact of temperature and
humidity on perception of indoor air
quality, Indoor Air, 8, 8090.
Fang, L., Clausen, G. and Fanger, P.O.
(1998b) Impact of temperature and
humidity on perception of indoor air
quality during immediate and longer

352

whole-body exposures, Indoor Air, 8, 276

284.
Fang, L., Clausen, G. and Fanger, P.O.
(1999) Impact of temperature and
humidity on chemical and sensory emissions from building materials, Indoor Air,
9, 193201.
Hoppe, P. (1983) Die Energiebilanz des
Menschen, Dissertation der Fakultat fur
Physik der Ludwig-Maximilians-Universitat in Munchen, Muchen.
Hoppe, P.R. and Martinac, I. (1998) Indoor
climate and air quality, Int. J. Biometeorol., 42, 17.
Nordstrom, K., Norback, D. and Akselsson,
R. (1994) Effect of air humidification on
the sick building syndrome and perceived
indoor air quality in hospitals, a four
month longitudinal study, Occup. Environ. Med., 51, 683688.
Palonen, J., Reinikainen, L.M. and Jaakkola, J.J.K. (1993) The effects of air temperature and relative humidity on thermal

comfort in the office environment. In:

Proceedings of Indoor Air 93, International Conference on Indoor Air Quality
and Climate, Helsinki, Vol. 6, 4348.
Reinikainen, L.M., Jaakkola, J.J.K. and
Heinonen, O.P. (1991) The effect of air
humidification on different symptoms in
office workers an epidemiologic study,
Environ. Int., 17, 243250.
Reinikainen, L.M., Jaakkola, J.J.K. and
Seppanen, O. (1992) The effect of air
humidification on symptoms and perception of indoor air quality in office
workers, a six-period cross-over trial,
Arch. Environ. Health, 47, 815.
Sommer, H.J., Johnen, J., Schongen, P. and
Stolze, H.H. (1994) Adaptation of the
tear film to work in air-conditioned
rooms (office-eye syndrome), German J.
Ophtalmol., 3, 406408.