ASTHMA AND WEATHER: RELATIONSHIP OF WEATHER TO ASTHMA ATTACKS IN WELLINGTON

Page created by Duane Mccormick
 
CONTINUE READING
Weather and Climate (1989) 9: 15-22                                                                 15

    ASTHMA AND WEATHER: RELATIONSHIP OF WEATHER TO
             ASTHMA ATTACKS IN WELLINGTON

                                           Paul Bruce

                              New Zealand Meteorological Service
                             P.O. Box 722, Wellington, New Zealand
                                             ABSTRACT

        A number of well marked seasonal peaks of asthma attacks are found in a one year
     study of 30 patients (Beasley et al., 1988) in Wellington. The seasonal impact of various
     meteorological indices and their causal relationships with asthma are investigated.
        It is found that damp, light wind conditions in winter are significantly related to the
     frequency of asthma attacks. This is consistent with such attacks being caused by damp
     air spora and other allergens present i n confined environments such as houses.
     Overland trajectories are also significantly correlated with asthma attacks. In autumn
     this may be attributed to damp air spora produced in rural areas to the north and
     northeast, and in spring to airborne pollens. No clear relationships were found linking
     cold changes to asthma.

                INTRODUCTION
   Asthma is a condition which causes consid-       explain asthma (Tromp, 1980). The allergen
erable concern in New Zealand. For example,         theory attributes complaints to allergic reac-
an Auckland study (Mitchell, 1984) indicated        tions; the 'cold chill' theory is said to be due to
that hospital admission rates for asthma had        a deficient hypothalamic thermoregulatory
increased greatly since the mid 1960's, and a       mechanism (deficiency in rewarming capabil-
recent national (NZ) asthma mortality study         ity); the infectious theory assumes early bron-
(Sears et al., 1985) showed a mortality rate        chial i n f e c t i o n s o f a c h i l d ; t h e
three times that of the United Kingdom.             endocringological theory attributes the com-
   In this study, possible relationships are in-    plaints to endocrinological disturbances (part-
vestigated between attacks suffered by asth-        ly true in children during puberty and women
ma patients undergoing treatment a t t h e          around the age of 50); and the psycho-somatic
Wellington Clinical School f o r the 11 month       theory explains a l l complaints i n terms o f
period 12/1/84 to 17/12/84, and the weather         psycho-somatic disturbances (once a subject
over that period.                                   becomes an asthmatic, psychological factors
   The American Thoracic Society has defined        may become important due to continous respi-
asthma as " a 'disease' characterised b y an        ratory problems). Clearly some o f these
increased responsiveness of the trachea and         causes will be related to the weather.
bronchi to various stimuli, and manifested by          In this paper the weather patterns affecting
a widespread narrowing of the airways, that         Wellington are briefly described. Some of the
changes in severity either spontaneously or as      factors likely to cause asthma attacks, and
a result o f therapy". Asthma can also be           weather conditions under which such factors
described as an inflammatory disorder. Nar-         are likely to occur, are identified. The asthma
rowing of the airways includes bronchospasm,        and weather data a r e then examined and
mucous production and mucosal oedema.               conclusions drawn about likely relationships
   Various theories have been put forward to        in Wellington.
16                                                                                   Asthma and Weather
NEW ZEALAND A N D WELLINGTON'S WEATHER                     moist north or northeast airstreams to flow
              AND CLIMATE
                                                           over the country, followed b y a southerly
   New Zealand lies in the mid-latitude zone of            outbreak.
westerly winds, i n the path o f an irregular                 Wellington's proximity to Cook Strait (see
succession o f anticyclones which migrate                  Fig. 1) leads to a high frequency of strong
eastwards every week or so ( N I . Met. Ser-               winds channelled b y t h e topography t o
vice, 1986). The centres of these anticyclones             north/south direction. Westerly airstreams
generally track across the North Island with               are deviated to the northwest, and easterlies
more northerly paths being followed in spring              are normally deflected to the south. I n very
and summer, and southerly paths in autumn                  unstable situations, these airstreams are less
and winter. Anticylones are areas of descend-              affected by orography.
ing air and settled weather, with little or no                For the year of the study, 1984, Wellington
rain, which m a y bring clear skies, o r l o w             had a mean temperature of 13.1 ' C (0.6 de-
cloud and fog.                                             grees warmer than the average from 1928 to
   Between the anticyclones are troughs of low             1980), 1054 m m of rain (20% less than the
pressure. Within these troughs there are often             average) and gales on 19 days (15% fewer
cold fronts, orientated north-west t o south-              days than average).
east, which produce one o f the commonest
type of weather sequence over the country: as               FACTORS ASSOCIATED WITH ASTHMA ATTACKS
the front approaches from the west, north-                 Cold Chill Effect
westerly winds become stronger and cloud
increases, followed by a period o f rain f o r                Tromp (1980) found a highly significant sta-
several hours as the front passes over, and                tistical relationship between asthma frequen-
then a change to cold showery southerlies.                 cy and atmospheric cooling. Another more
   The frequency of the above systems shows                recent study i n Japan (Suzuki e t al., 1988),
little variation with season except for an in-             showed that low temperature and decrease in
crease in anticyclonic activity in autumn, al-             temperature was singificantly related with
lied t o a corresponding decrease i n t h e                the asthmatic symptom. However it was notes
frequency of troughs. Another feature of au-               that further studies were necessary to know
tumn is the occasional appearance of decay-                whether 'the state o f the atmosphere and
ing tropical cyclones in the area to the north             contents of the air collaborate on the airway
                                                           to evoke the asthmatic condition'.
of the North Island, which can cause very
                                                              This present study investigates asthma data
                                                           for evidence o f this cooling effect. A s the
                                                           exact time o f the asthma attack was not
                                                           available, a relationship between daily asth-
                                                           ma frequencies and a change from the warm-
                                                           er northwesterley airstreams t o t h e cold
                                                           southerly wind was examined.
                                                              The data are also examined f o r seasonal
                                                           variations, a s pollen releases occurring i n
                                                           spring with westerley winds may have a con-
                                                           founding influence.
                                                           Allergen theory
                                                              The allergic reactions of asthmatics have
                                                           been r e l a t e d t o seasonal p o l l e n peaks
                                                           (Tonkelaar, 1986), to seasonal pattern of air-
                                                           borne damp air fungal spores (Hasnain et al.,
                                                           1985; Packe e t al., 1985), t o the dust-mite,
                                                           animal proteins and fungal spores that occur
                                                           in confined environments, in addition to tobac-
                                                           co smoke and other aero-irritants (Mathison et
                                                           al., 1982).
Fig I : Location of Wellington and relative frequency of
Westerly, Southerly, Northerly, and Light and Variable     Damp air fungal spores (open environment)
sector winds.                                                Studies show that pollens are only weakly
Asthma and Weather                                                                              17

represented in autumn, but other allergy-pro-      live mite populations diminishing rapidly for
ducing micro-organisms, especially fungal          humidities below 4 5%
spores from the basidiospore group (damp air          Fungal spores, the house mite, and other
spore component), a r e especially favoured       contaminants known to cause asthma, such as
(Hasnain et al., 1985; Pennycook, 1980).           formaldehyde in treated plywood or particle
   Factors contributing t o higher damp a i r      board and glues (Matthews, 1985; Yodaiken,
spore concentrations will be: high night-time      1981; Robertson et al., 1986), have been identi-
humidity (more likely in Autumn with length-       fied in the indoor environment. Their impact
ening nights and a seasonal lack o f wind);        would b e exacerbated i n badly ventilated,
rising air motion at ground level; and winds       damp buildings, a set of conditions most likely
blowing overland towards the observer. Ris-        to occur with cold, wet periods or low winds
ing air motions occur with day-time convec-        and shorter sunlight hours, in April through to
tion on sunny days, and are especially strong      July (late autumn and winter). New Zealand
in airstreams preceding active fronts or dur-      houses may be especially susceptible to the
ing thunderstorms when the atmosphere i s          influence of prolonged damp periods, due to
'unstable' (defined in Data and Methods).          the predominance of wood, which will absorb
   Acccording to Hasnain et al., (1985) urban      moisture when badly protected.
areas with restricted vegetation provide less         In addition, there may be a tendency in the
opportunity for fungal growth than forest or      early winter period to keep windows shut to
farmlands. Hasnain also documented a close        conserve heat in inclement weather. A behav-
association between the basidiospore and hos-     ioural response to cold changes of restricting
pital admissions for acute asthma. In the case    air movement and closing windows (following
of Wellington, the north to east sector has the   a retreat indoors to a confined environment),
largest overland and rural trajectory and thus    will result in a higher humidity air mass and
airstreams from those directions would be         higher concentrations of pollutants and aller-
expected to contain higher concentrations of      gens. The relative humidity will be higher in
spores.                                           unheated rooms, typically bedrooms, w i t h
                                                  condensation on cooler surfaces of humid air
   If allergens are present when the wind is      from warmer parts of the house. Localised
blowing from the north to east sector, and i f    warmth in beds (from body heat), in combina-
the asthma is related to the contents of the
                                                  tion with this dampness, may then produce the
airstream, then a wind change t o a shorter       ideal environment for the house mite.
overland trajectory (westerlies or southerlies)       To ensure that there i s no compounding
should decrease spore concentrations and thus     factor. of damp air spores orginating from the
the number of asthma attacks. To avoid the
                                                  external environment, t h e contribution o f
possible confounding influence o f the 'cold      moist long overland trajectory northerly air-
chill' effect associated with the change to the   streams was also investigated. In spring, high
south, only the change t o the westerly flow      wind levels, increasing insolation and warm-
was investigated.                                 ing temperatures will result in a decrease in
House mite and damp a i r spores in confined      dampness i n the house, and asthma should
environments                                      then decrease.
   The Wellington study on aero-allergeds in          Cold changes that arrive after midnight
the homes of asthmatics (Sheridan et al., 1985)   when potential sufferers are probably in bed
indicated a range of fungi known to be signifi-   (and most likely to be subject to allergens
cant in asthma. These showed seasonal varia-      such as produced b y the house mite), m a y
tions, with the number of colonies of two of      result in the behavioural response of shutting
the m o s t frequent, basidiomycetes a n d        any windows that remain open. This would
cladosporium, showing a decrease i n t h e        lead to higher concentrations o f allergenic
months July, August and September.                gases or spores with a corresponding increase
   The mite of the genus 'dermatophagoides',      in the r i t e of asthma attacks.
one of the most significant causes of asthma,     Pollen, dry air spora
occurs in house dust and is temperature and          Pollen releases occur in spring and continue
humidity dependent (Mathison e t al., 1982;       into summer, when other dry a i r spora re-
Korsgarrd, 1982). Optimal growth occurs at 25     leases become important. T h e triggering
degrees C and 80% relative humidity, with         mechanism is some initial dampness, rising
18                                                                                Asthma and Weather
temperatures, and the absence of rain during               Asthma was defined with reference t o a
the critical period of release. Each plant has          peak flow meter, and confirmed b y other
its own favoured period of release of pollen.           measurements when patients reported to the
Pollen grains on release are entrained b y              clinic (Beasley et al., 1988).
wind gusts and by small turbulent wind ed-                 The daily weather elements used were max-
dies. Peaks in concentration may occur during           imum and mean temperatures, wind run, and
the day or at night time when a temperature             significant rain at Kelburn (Wellington). Other
inversion occurs in the atmosphere trapping             indicators such as wind direction, frontal pas-
spores i n a relatively shallow layer o f the           sage, change of airstream, and atmospheric
atmosphere (Tonkelaar, 1986). I f spore re-             stablility indicators were derived from inves-
leases are occurring upstream, large local              tigation of synoptic charts. Results were found
concentrations may then occur. Rain clears              to be best represented for the four seasons,
the atmosphere of contaminants and suppress-            defining autumn a s February, March and
es pollen production, so some allergic symp-            April; winter as May, June and July; etc.
toms should b e a l l e v i a t e d d u r i n g a n d      Airstream winds over Wellington are cho-
immediately after rain. However, note that              sen by sectors to highlight over-land and on-
considerable amounts of damp air spores may             shore components:
then be produced, resulting in problems for a              a) Northerly - the overland sector repre-
different set of subjects.                              senting north, northeast, east winds.
   Wellington suburban gardens are f u l l o f             b) Northwest - the short overland sector
flowering species, and the unoccupied hills to          representing northwest and westerly winds.
the north, west and east, are covered in gorse,            c) Southerly - the oversea (onshore) sector
pine trees and native bush. A study in Welling-         representing southwest, south, and southeast
ton (Licitis, 1952) indicated that marked pol-          winds, which are generally associated with
len peaks occured i n spring, i n the period            lower temperatures.
August through to October. If this relationship            d) Light and variable - conditions which
holds, we would expect significantly higher             might indicate a short overland trajectory,
level o f pollen and high asthma levels i n             local allergen sources, or factors relating to a
spring during fine weather with winds arriv-            confined environment.
ing from the west through north to east sector.            Unstable conditions in the atmosphere indi-
Southerlies, representing the sector which has          cate upward, often turbulent motion of air,
the smallest surburban trajectory, should con-          and also the tendency for airstreams to flow
tain the least amount of pollen spores given            over, rather than around hills. When a trough
other ideal conditions. The predominant winds           of low pressure of cyclonically curved isobars
of spring are northwesterlies and these are             on the synoptic map is present, the airmass is
normally d r y (apart from brief periods o f            taken to be unstable. In cases where there is
showers), and turbulent.                                some uncertainty, the presence of significant
                DATA AND METHODS                        quantities of rain (greater than 0.5 mm) is
   This study uses asthma data on 30 patients           noted a s indicating atmospheric dynamic
collected by the Wellington Clinical School.            instability.
Patients came from a range of occupations                                  RESULTS
and ages. All resided in the Wellington region,            Over the 336 day period of the study, indi-
except for one who commuted daily from the              vidual attacks persisted for an average of 3.4
Wairarapa. The group were asked to record,              days but could extend up to 19 days. The mean
on forms, when asthma attacks began and                 daily asthma 'onset' rate totalled for the 30
finished in the period from 17th January to the         patients is 0.5 attacks per day, with a maxi-
17th December 1984 (336 days).                          mum occurrence o f 3 onsets per day. The
   The asthma data were categorised in two              mean asthma occurrence is 1.6, with a maxi-
ways:                                                   mum value of 6. Major peaks in asthma occur-
   a) Asthma Onsets. The numbers of patients            rence (Fig. 2 ) w e r e observed i n autumn
suffering an onset of an attack on any day.             (centred around late March) and i n winter
   b)Asthma Occurrence. The number of pa-               (end of July), with smaller ones in spring (late
tients suffering from asthma on any one day.            October) and summer (early December).
This includes onsets and characterises the                 A three factor analysis o f variance was
persistence of asthma.                                  used to show that variations by season, stabil-
Asthma and Weather                                                                                                                                                                     19
                   3
                                                          Fig 2: Seasonal variation o f asthma (smoothed over 31
                                                          days).

Asthmaatacksp
            e
            d
            ray.
                                                                                                                                             Ons

                           /

                                Autumn          Winter

                       0       1         oo               200                                                    300                                                               400
                           Julian day
ity and wind direction are significant at the                    3
5% level (F-test). The null hypothesis, 'that
there is no difference i n asthma levels be-                     2
tween environmental factors', was tested us-
ing a one tailed t-test on paired differences.                                      GO           G O                                 0                           0 El

The analysis was also carried out on the
square root of asthma levels to 'normalise' the
data.
Cold chill effect:                                                                                                                                   G

  The change from west to southerly sectors                     •-2
occurred on 21 days with a mean decrease in                            Autumn       W    i       n   t       e   r   S   p   r   i       n   g   S       u   m       m       e     r
temperature of 2.3 degrees C. The increase in             Fig. 3 : Impact o f change from westerly t o southerly
asthma by .04 onsets per day is significant at            sector (cold change) on asthma levels.
only the 10% level (p < 0.07). There is a hint of
a "U" shape response, of reduced asthma on-
sets in winter and spring, and increased asth-                   2-

ma onsets at other times of the year. Fig. 3
shows that the onset response t o southerly
change is very scattered, but with a tendency
to increased onset. There were ten cases of
increase, seven cases of no change, and four
cases of decrease.                                                                      03                                                                                   133

 Fig 4 : Impact o f change from northerly t o westerly          -2 n
 sector (cutt-off o f long trajectory spores) on asthma
 levels.
                                                                -3              0

                                                                       Autumn       W        i   n       t   e   r   S   p   r   i       n   g   S       u   m   m       e    r
20               A                s                 t             h             m                 a                       and Weather
Damp a i r f u n g a l s p o r e s ( e x t e r n a l                      1.5

environment)
   The change from northerly to westerly sec-
tors occurred on 20 occasions over the year
with a mean decrease of 0.9 asthma onsets (p                              1.0

 < 0.005). Fig. 4 shows the influence o f this
change in sector on asthma levels through the
year.
   In autumn the most dominant factor associ-
ated with asthma levels is the presence o f
northerlies (mean temperature around 17 deg
C). Asthma levels in autumn for the northerly
sector airstreams and for light and variable                              Do
condiitons, are found to be significantly higher                                           2            6   1    0        1

than that for northwesterlies (p < 0.025). On                                                  Time of day NZ standard time (hrs)
the 3 autumn days with thunder and lightning                          Fig 6: Impact o f arrival time o f southerly changes in
events, mean asthma persistence levels were                           Wellington (with standard error bars) on daily asthma
                                                                      onset rate.
 167% above the seasonal mean, but this was
not significant due t o the l o w number o f                             From winter to spring, the decrease in asth-
events. No clear correlations were found with                         ma onsets overall i s significant (p < 0.05).
airmass instability and the passage of fronts.                        However, there is a more highly significant
                                                                      drop for both unstable and wet conditions (p<
House mite and damp air spores in the inter-                          0.005), and similarly for southerly sector wind
nal environment                                                       conditions (p < 0.005).
   In winter, the most important factors influ-                          A higher asthma onset rate i s observed
encing the number of patients suffering asth-                         when southerly wind changes arrive during
ma attacks a r e found t o b e atmospheric                            the night hours (Fig. 6). Individual standard
instability, rain, southerlies (mean tempera-                         error bars show that changes over four hour
ture 8 deg C), and light o r variable wind                            adjacent p e r i o d s a r e n o t n e c e s s a r i l y
conditions. These relationships, shown in Fig.                        significant.
5, are observed to begin in Autumn. The over-
all seasonal change in asthma rate from au-                           Pollen, dry air spora
tumn to winter is not significant. However for                           In spring, the mean rate of asthma attacks
northerly conditions there i s a significant                          per day is significantly lower than that for the
decrese in importance in winter (p < 0.025),                          other seasons (Fig. 2). Asthma occurs most
clearly indicating the lower importance of the                        readily for northeasterlies, in this period (Au-
overland sector at this time of the year.                             gust, September, October) with levels signifi-
      -                                                               cantly above those occurring for southerlies (p
                                             O Stable                 < 0.025), a reversal of the situation found in
                                             tO Unstabaa              winter. The decrease from winter to spring in
                                                                      mean asthma levels occurring with southerly
                                                                      airstreams, is also statistically significant (p <
                                                                      0.005).
                                                                         Rain inhibits asthma in spring (p 0.025) in
                                                                      contrast to winter, where rain is associated
                                                                      with high asthma occurrence. The change in
                                                                      relationship is highly significant (p < 0.001).
                                                                         The six changes from westerly to southerly
                                                                      airstreams found i n spring led t o a mean
                                                                      decrease of 0.9 asthma onsets (p
Asthma and Weather                             2                            1
November/December p e r i o d w i t h these        lution, a n d , a s a result, asthma l e v e l s
airstreams.                                        (Mathison et al., 1982). A tendency towards
                                                   higher asthma levels i n the early morning
                 CONCLUSIONS
                                                   hours may also be an indication of a stronger
   A number of different meteorological influ-     influence o f internal environments a t that
ences have been found to be associated with        time.
asthma attacks through the year. These are            Further research is needed to clarify the
highlighted at peak times:                         importance in New Zealand of these allergens
— an autumn peak associated with northerly         and their relationship to environmental and
    airstreams;                                    behavioural factors. T h e correlations also
— the winter peak associated with a wet, low       point to a possible relation between meteoro-
    wind period;                                   logical conditions and asthma which might be
— and the spring peak associated with dry          found in specific microclimates, such as damp
    weather and winds f r o m t h e overland       river valleys. The data collected in this study
    sector.                                        did not include where people actually were,
   The periods of asthma may be attributed to      when they experienced their asthma attacks -
allergens of distinct varieties:                   the importance of micro-climates in asthma
Autumn — outdoor spores, damp variety pre-         incidence i n New Zealand deserves further
              dominating and originating from      study.
              inland sector;                          The asthma attack rates found in this analy-
Winter — i n d o o r spores, house mite and oth-   sis indicate the origin o f asthma-producing
              er contaminants;                     allergens present i n spring and summer -
Spring — l o c a l pollens;                        northwesterly airstreams and light wind situ-
Summer — airborne spores (mainly dry vari-         ations representing pollens derived from local
              ety) from inland sector.             vegetation in spring, and northerlies and east-
   The close correspondence of northerlies in      erlies representing longer trajectory pollens
autumn with asthma indicates a seasonal re-        and spores f r o m r u r a l and forested areas
lationship w i t h a n airstream t h a t has the   further afield in summer.
larges overland bias. Studies such as that done       The cold chill factor is not clearly shown to
by Hasnain et al., (1985), suggest damp a i r      be present. This may be due to possible con-
spores as the allergen responsible f o r the       founding factors resulting from a behavioural
asthma attacks.                                    response to colder conditions, or to the gener-
   The present study suggests that a major         ally small temperature drops experienced in
asthma peak also occurs with wet, low-wind         Wellington's maritime climate.
or cold conditions in winter, which is likely to                  ACKNOWLEDGEMENTS
be a t least partly related t o a behavioural
response of retreating indoors to closed envi-       I thank the staff of the Wellington Clinical
ronment. This is supported by the finding that     School for advise and the use of data, and also
during winter, the northerly overland winds        Reid Basher, Steve Goulter, John Revfeim and
have the lowest correspondence with asthma.        others for discussions and helpful reviews.
A marked change in environmental conditions
associated with asthma attacks then occurs                             REFERENCES
from winter through to spring, with rain and
instability losing t h e i r importance, a n d a   Beasley, R., Coleman, E.D., Hermon, Y. , Hoist, R E . ,
change to longer and more windy days. This            O'Donnell, TN., Tobias, M., 1988: Viral respiratory
                                                      tract infection and exacerbations of asthma in adult
points t o allergens occurring i n the indoor         patients. Thorax 43, 679-683.
environment as the major cause of asthma           Hasnain, S.M., Wilson, I D . , Newhook, KJ., 1985: Fungi
during the winter period. It may indicate that        and Disease. Fungal allergy and respiratory disease.
the asthma rate could be diminished i f the           N.Z. Med. J. 98, 342-6.
                                                   Korsgarrd, J., 1982: Preventive measures in house dust
predominant indoor environments were drier             allergy. Am. Rev. Resp. Dis. 125, 80-4.
and better ventilated (as in spring when the       Licitis, R., 1952: Air-borne pollen and spores sampled at
mean wind is higher). Recent changes towards           five New Zealand Stations. N.Z. J. Sci. and Tech.B4,
                                                       289-316.
less draughty homes with the introduction of       Mathison, D.A., Stevenson, D., Simon, R.A., 1982: Asthma
draught strips, aluminium joinery etc., may            and the home environment. Annals of Internal Medi-
have exacerbated internal dampness and pol-            cine. 97, 1.
22               A              s              t               h   m   a   and Weather

Matthews, R., 1985: Pollution begins at home. New Sci. 5
   Dec, 34.
Mitchell, EA., 1984: Paediatric admissions to Auckland
   Hospital for asthma from 1970-80. N.Z. Med. J. 97, 67-
   70.
N.Z. Met. Service, 1986: Climate of New Zealand. Info
   Pub, 15.
Packe, GE., Ayres, Jon G., 1985: Asthma outbreak during
   a thunderstorm. Lancet. July 27, 199-203.
Pennycook, SIC, 1980: The air spora of Auckland city,
   New Zealand. N.Z. J. Sci. 23, 27-37.
Robertson, A.S., 1986: Sherwood Burge P. Building Sick-
   ness. Patient Management. Sep, 139-48.
Sears, M R . , Rea, RH., Beaglehole, R., G i l lies, A.J.D.,
   O'Donnel, TV., Rothwell, R.P.G., Sutherland, D.C.,
   1985: Asthma mortality in New Zealand. A two year
   national study. N.Z. Med. J. 98, 271-5.
Sheridan, I E . , Coleman, E D . , Hoist, R E . , O'Donnell,
   T.V., 1985: Aero-allergens in the homes of Wellington
   asthmatics. N.Z. J. Sci. 26, 1-7.
Suzuki, S., Kamakura, T., Tadokoro, K., Takeuchi, F. ,
   Yukiyama, Y. , Miyamoto, T., 1988: Correlation be-
   tween the atmospheric conditions and the asthmatic
   symptom. Int. J. Blometeoral 32, 129-33.
Tonkelaar, den J.F., 1986: Grass pollen counts in the west
   Netherlands a s a function o f weather conditions.
   Weather Ministry of Netherlands, Sci Report. W.R.
   7.
Tromp, S.W., 1980: The impact of the weather and climate
   on humans and their environment. London: Heydon,
   148-54.
Yodaiken, ILE., 1981: The uncertain consequences o f
   formaldehyde toxicity. Jama 246, 1677-8.
You can also read