The daily PM10 concentrations at
Chulalongkorn were highly correlated with
those at Odean Circle (r = 0.95). This correlation
suggests the curbside measurements are
reflecting the general day-to-day fluctuations
in PM10 concentrations over a reasonably
wide area in the city, because these locations
were a few miles apart. PM10 concentrations
at both sites were moderately inversely correlated
with daily temperature (r = –0.38 and
r = –0.32, respectively) and humidity (r =
–0.55 and r = –0.47, respectively), at Odean
Circle and Chulalongkorn.
Table 3 summarizes the impact of alternative
lags on the PM10 variable using the
basic logistic regression model for lower and
upper respiratory symptoms for Odean
Circle adults, nurses, and children. For
Odean Circle adults, this model controls for
a subject’s age, sex, educational level, having
a chronic respiratory condition, having no air
conditioning in the home, and daily average
temperature. For the nurses, there was no
variation in sex, education, or air conditioning,
so these were not included in the model.
For children, the model includes age, sex, having
a chronic respiratory condition, having no
air conditioning in the home, daily average
temperature, and daily average humidity. Lags
of up to 3 days and moving averages of up to
4 days (i.e., the average of the current day’s
PM10 concentration and the concentrations
on the three previous days) were examined in
these basic models. For all three panels and
both outcomes, a 4-day moving average generated
the strongest associations with PM10.
However, positive associations were indicated
for all of the lags examined, and statistically
significant results were obtained for all three
moving average measures. Based on these
results, the 4-day moving average was selected
as the basic measure of PM10 for subsequent
sensitivity analyses.
All the individual characteristics shown in
Table 1 were included in preliminary analyses,
but only those with statistically significant
relationships with symptoms were retained in
the basic model. Having a household member
who smokes (none of the subjects smoked) or
using charcoal for cooking were not significant
for the adults or for children, except for
upper respiratory symptoms in children,
which showed a higher frequency for those
who had a smoker in the house. However, the
PM10 coefficient for upper respiratory symptoms
in children was not changed when the
household smoker variable was added to the
model. Having no air conditioning had an
unexpected negative sign on symptom frequencies
in the children, but had the
expected positive sign for adults. The result
for children may have been due to correlation
with socioeconomic status rather than an
actual beneficial respiratory effect of having
no air conditioning.
Those with a chronic respiratory condition
were much more likely to have symptoms, but
interactions with the PM10 variable were not
statistically significant, suggesting those with a
chronic respiratory condition were no more
likely to be affected by daily fluctuations in
PM10 than those without a chronic condition.
Interactions between PM10 and other variables,
including no air conditioning and presence
of household smoker, were also tested
and none were found to be statistically significant.
It is important to note these are simple
binary variables for each subject and do not
reflect the potential impact of day-to-day fluctuations
in such exposures or differences in
the amount of exposure for subjects who are
exposed. These findings, therefore, suggest
only those exposed to environmental tobacco
smoke or to charcoal smoke in the home show
no evidence of a different reaction to fluctuations
in daily concentrations of outdoor
PM10. They should not be interpreted as
showing no effect of these indoor exposures
on daily symptoms, because they were not
measured as daily exposures.
The PM10 effects in the basic model and
sensitivity analyses are summarized in Table
4. First, the results for the basic model are
reported with and without a variable controlling
for the impact of daily average temperature
(unlagged). Adding temperature to the
model attenuated the effect of PM10 somewhat
for the adult panels, but caused a slight
increase in the estimated PM10 effect for children.
Temperature was negatively associated
with symptoms (i.e., fewer symptoms were
reported on hotter days). For Odean Circle
Table 3. Logistic regression PM10 coefficients (standard errors) × 100 for alternative lags and moving averages.
Lower respiratory symptoms Upper respiratory symptoms
PM10 lag or Odean Circle Odean Circle
moving average adultsa Nursesb Childrenc adultsa Nursesb Childrenc
Same day 0.59*** 0.20* 0.41*** 0.78*** 0.23* 0.37**
(0.08) (0.09) (0.10) (0.08) (0.09) (0.10)
Lag 1 day 0.46*** 0.16 0.35*** 0.61*** 0.26** 0.36***
(0.08) (0.09) (0.09) (0.08) (0.09) (0.09)
Lag 2 days 0.48*** 0.14 0.20* 0.61*** 0.27** 0.31***
(0.08) (0.09) (0.08) (0.08) (0.09) (0.08)
Lag 3 days 0.41*** 0.11 0.11 0.52*** 0.26** 0.19*
(0.08) (0.09) (0.08) (0.08) (0.09) (0.08)
2-Day moving average 0.65*** 0.20* 0.48*** 0.85*** 0.29** 0.46***
(0.09) (0.10) (0.11) (0.09) (0.10) (0.11)
3-Day moving average 0.79*** 0.23* 0.52*** 1.05*** 0.37*** 0.55***
(0.10) (0.11) (0.11) (0.10) (0.11) (0.11)
4-Day moving average 0.89*** 0.27* 0.56*** 1.14*** 0.45*** 0.61***
(0.11) (0.12) (0.12) (0.11) (0.12) (0.12)
aOdean Circle adults model includes daily average temperature (same day), age, sex, educational level, having a chronic respiratory
condition, and having no air conditioning in the home. bNurses model includes daily average temperature (same day), age, and having
a chronic respiratory condition. cSchoolchildren model includes daily average temperature (same day), daily average humidity (same
day), age, sex, having a chronic respiratory condition, and having no air conditioning in the home.
* p < 0.05; ** p < 0.01; *** p < 0.001.
Table 4. Basic model and sensitivity analysis results (odds ratios and 95% CIs for 45 μg/m3 change in PM10).
Lower respiratory Upper respiratory
Modela Odean Circle adults Nurses Children Odean Circle adults Nurses Children
Basic model without 1.66 1.22 1.22 1.94 1.35 1.31
weather variables (1.52–1.82) (1.10–1.36) (1.11–1.35) (1.77–2.12) (1.22–1.48) (1.19–1.44)
Basic model with 1.49 1.13 1.29 1.67 1.22 1.32
weather variables (1.35–1.64) (1.02–1.26) (1.16–1.43) (1.52–1.84) (1.10–1.36) (1.18–1.46)
Add symptom 1.51 1.12 1.26 1.38 1.11 1.14
yesterday (1.37–1.67) (1.01–1.25) (1.14–1.40) (1.24–1.53) (0.98–1.27) (0.97–1.34)
Omit high (25%) 1.56 1.22 1.66 1.56 1.26 1.53
temperature days (1.38–1.77) (1.08–1.38) (1.40–1.96) (1.38–1.77) (1.12–1.43) (1.30–1.81)
Omit low (25%) 1.38 1.12 1.15 1.51 1.17 1.21
temperature days (1.23–1.54) (0.98–1.28) (1.03–1.29) (1.33–1.70) (1.03–1.32) (1.08–1.36)
Limit to new 1.43 1.27 1.08 1.56 1.31 0.99
symptom days (1.24–1.65) (1.05–1.54) (0.87–1.33) (1.36–1.80) (1.09–1.58) (0.78–1.25)
Fixed-effects modelb 1.53 1.10 1.13 1.69 1.19 1.10
(1.36–1.71) (0.98–1.24) (1.05–1.23) (1.49–1.91) (1.05–1.34) (1.02–1.18)
aThe PM10 measure in all models is the 4-day moving average. Other independent variables include daily temperature (same day), age,
sex, educational level, having a current chronic respiratory condition, and having no air conditioning in the home, as appropriate (see
Table 3). The models in the sensitivity analyses include daily temperature. The children’s model also includes same day humidity. The
45 μg/m3 increment in PM10 approximates the interquartile range. bThe fixed-effects models include the daily weather variables.
Air pollution and respiratory health in Bangkok
adults, for an interquartile change (75th–25th
percentile) in PM10 of approximately 45
μg/m3, the odds ratio is 1.66 [95% confidence
interval (CI) = 1.52–1.82] for lower
respiratory symptoms and 1.94 (95% CI =
1.77–2.12) for upper respiratory symptoms.
Lower effect magnitudes were observed for
the panels of nurses and children, but PM10
was associated with statistically significant
increases in frequencies for both symptom
categories for all three panels, with and
without daily weather variables included in
the models.
The inclusion of a variable indicating the
presence of a symptom on the prior day
caused virtually no change in the estimated
PM10 effects for lower respiratory symptoms
for all three panels relative to the basic model
with daily weather variables, but attenuated
the estimated effect of PM10 on upper respiratory
symptoms in all three panels. Omitting
the hottest 25% of the days tended to increase
the estimated association, whereas omitting
the coldest 25% of the days lowered the estimate,
but the PM10 effect remained statistically
significant in nearly all cases. The effects
of PM10 on the likelihood of a new symptom
episode were examined in an analysis that
included only those days for which there were
no symptoms reported on the previous day.
The results indicate an association exists for
both adult panels and for both outcomes. The
magnitude and statistical significance of the
PM10 effect remained comparable to that
found in the original model for all days.
However, the results for children showed no
statistically significant effect of PM10 on new
symptoms. We also ran a model in which
cough was not included as a lower respiratory
symptom. The results were similar to those
obtained for lower respiratory symptoms
when cough was included.
The last row in Table 4 shows the PM10
results estimated with a fixed-effects model
that included the daily weather variables. The
fixed-effects model allows the baseline symptom
incidence to vary for each individual and
corrects for the correlations among repeated
responses from the same individuals. The
PM10 results for the fixed-effects models for
the two adult panels were little changed from
the results using the basic model with daily
weather variables. However, the PM10 results
for the children were about 50
The daily PM10 concentrations atChulalongkorn were highly correlated withthose at Odean Circle (r = 0.95). This correlationsuggests the curbside measurements arereflecting the general day-to-day fluctuationsin PM10 concentrations over a reasonablywide area in the city, because these locationswere a few miles apart. PM10 concentrationsat both sites were moderately inversely correlatedwith daily temperature (r = –0.38 andr = –0.32, respectively) and humidity (r =–0.55 and r = –0.47, respectively), at OdeanCircle and Chulalongkorn.Table 3 summarizes the impact of alternativelags on the PM10 variable using thebasic logistic regression model for lower andupper respiratory symptoms for OdeanCircle adults, nurses, and children. ForOdean Circle adults, this model controls fora subject’s age, sex, educational level, havinga chronic respiratory condition, having no airconditioning in the home, and daily averagetemperature. For the nurses, there was novariation in sex, education, or air conditioning,so these were not included in the model.For children, the model includes age, sex, havinga chronic respiratory condition, having noair conditioning in the home, daily averagetemperature, and daily average humidity. Lagsof up to 3 days and moving averages of up to4 days (i.e., the average of the current day’sPM10 concentration and the concentrationson the three previous days) were examined inthese basic models. For all three panels andboth outcomes, a 4-day moving average generatedthe strongest associations with PM10.However, positive associations were indicatedfor all of the lags examined, and statisticallysignificant results were obtained for all threemoving average measures. Based on theseresults, the 4-day moving average was selectedas the basic measure of PM10 for subsequentsensitivity analyses.All the individual characteristics shown inTable 1 were included in preliminary analyses,but only those with statistically significantrelationships with symptoms were retained inthe basic model. Having a household memberwho smokes (none of the subjects smoked) orusing charcoal for cooking were not significantfor the adults or for children, except forupper respiratory symptoms in children,which showed a higher frequency for thosewho had a smoker in the house. However, thePM10 coefficient for upper respiratory symptomsin children was not changed when thehousehold smoker variable was added to themodel. Having no air conditioning had anunexpected negative sign on symptom frequenciesin the children, but had theexpected positive sign for adults. The resultfor children may have been due to correlationwith socioeconomic status rather than anactual beneficial respiratory effect of havingno air conditioning.Those with a chronic respiratory conditionwere much more likely to have symptoms, butinteractions with the PM10 variable were notstatistically significant, suggesting those with achronic respiratory condition were no morelikely to be affected by daily fluctuations inPM10 than those without a chronic condition.Interactions between PM10 and other variables,including no air conditioning and presenceof household smoker, were also testedand none were found to be statistically significant.It is important to note these are simplebinary variables for each subject and do notreflect the potential impact of day-to-day fluctuationsin such exposures or differences inthe amount of exposure for subjects who areexposed. These findings, therefore, suggestonly those exposed to environmental tobaccosmoke or to charcoal smoke in the home showno evidence of a different reaction to fluctuationsin daily concentrations of outdoorPM10. They should not be interpreted asshowing no effect of these indoor exposureson daily symptoms, because they were notmeasured as daily exposures.The PM10 effects in the basic model andsensitivity analyses are summarized in Table4. First, the results for the basic model arereported with and without a variable controllingfor the impact of daily average temperature(unlagged). Adding temperature to themodel attenuated the effect of PM10 somewhatfor the adult panels, but caused a slightincrease in the estimated PM10 effect for children.Temperature was negatively associatedมีอาการ (เช่น อาการน้อยลงได้รายงานในวันร้อน) สำหรับวงกลม Odeanตาราง 3 ถดถอยโลจิสติก PM10 สัมประสิทธิ์ (ข้อผิดพลาดมาตรฐาน) × 100 lags ทางเลือกและย้ายหาค่าเฉลี่ยลดอาการหายใจบนอาการทางเดินหายใจPM10 ความล่าช้าหรือ Odean กลม Odean วงย้ายเฉลี่ย adultsa Nursesb Childrenc adultsa Nursesb Childrencเดียววัน 0.59* ** 0.20* 0.41* ** 0.78* ** 0.23* 0.37* *(0.08) (0.09) (0.10) (0.08) (0.09) (0.10)ล่าช้า 1 วัน 0.46* ** 0.16 0.35* ** 0.61* ** 0.26* * 0.36* **(0.08) (0.09) (0.09) (0.08) (0.09) (0.09)ล่าช้า 2 วัน 0.48* ** 0.14 0.20* 0.61* ** 0.27* * 0.31* **(0.08) (0.09) (0.08) (0.08) (0.09) (0.08)ล่าช้า 3 วัน 0.41* ** 0.11 0.11 0.52* ** 0.26* * 0.19*(0.08) (0.09) (0.08) (0.08) (0.09) (0.08)2 วันย้ายเฉลี่ย 0.65* ** 0.20* 0.48* ** 0.85* ** 0.29* * 0.46* **(0.09) (0.10) (0.11) (0.09) (0.10) (0.11)วันที่ 3 ย้ายเฉลี่ย 0.79* ** 0.23* 0.52* ** 1.05* ** 0.37* ** 0.55* **(0.10) (0.11) (0.11) (0.10) (0.11) (0.11)วันที่ 4 ย้ายเฉลี่ย 0.89* ** 0.27* 0.56* ** 1.14* ** 0.45* ** 0.61* **(0.11) (0.12) (0.12) (0.11) (0.12) (0.12)aOdean วงผู้ใหญ่รุ่นมีอุณหภูมิเฉลี่ยประจำวัน (วันเดียว), อายุ เพศ ระดับการศึกษา มีการเรื้อรังทางเดินหายใจเงื่อนไข และมีไม่มีเครื่องปรับอากาศในบ้าน รุ่น bNurses มีอุณหภูมิเฉลี่ยประจำวัน (วันเดียว), อายุ และมีเป็นโรคทางเดินหายใจ รุ่น cSchoolchildren รวมรายวันเฉลี่ยอุณหภูมิ (วันเดียว), ความชื้นเฉลี่ยรายวัน (เหมือนกันday), age, sex, having a chronic respiratory condition, and having no air conditioning in the home.* p < 0.05; ** p < 0.01; *** p < 0.001.Table 4. Basic model and sensitivity analysis results (odds ratios and 95% CIs for 45 μg/m3 change in PM10).Lower respiratory Upper respiratoryModela Odean Circle adults Nurses Children Odean Circle adults Nurses ChildrenBasic model without 1.66 1.22 1.22 1.94 1.35 1.31weather variables (1.52–1.82) (1.10–1.36) (1.11–1.35) (1.77–2.12) (1.22–1.48) (1.19–1.44)Basic model with 1.49 1.13 1.29 1.67 1.22 1.32weather variables (1.35–1.64) (1.02–1.26) (1.16–1.43) (1.52–1.84) (1.10–1.36) (1.18–1.46)Add symptom 1.51 1.12 1.26 1.38 1.11 1.14yesterday (1.37–1.67) (1.01–1.25) (1.14–1.40) (1.24–1.53) (0.98–1.27) (0.97–1.34)Omit high (25%) 1.56 1.22 1.66 1.56 1.26 1.53temperature days (1.38–1.77) (1.08–1.38) (1.40–1.96) (1.38–1.77) (1.12–1.43) (1.30–1.81)Omit low (25%) 1.38 1.12 1.15 1.51 1.17 1.21temperature days (1.23–1.54) (0.98–1.28) (1.03–1.29) (1.33–1.70) (1.03–1.32) (1.08–1.36)Limit to new 1.43 1.27 1.08 1.56 1.31 0.99symptom days (1.24–1.65) (1.05–1.54) (0.87–1.33) (1.36–1.80) (1.09–1.58) (0.78–1.25)Fixed-effects modelb 1.53 1.10 1.13 1.69 1.19 1.10(1.36–1.71) (0.98–1.24) (1.05–1.23) (1.49–1.91) (1.05–1.34) (1.02–1.18)aThe PM10 measure in all models is the 4-day moving average. Other independent variables include daily temperature (same day), age,sex, educational level, having a current chronic respiratory condition, and having no air conditioning in the home, as appropriate (see
Table 3). The models in the sensitivity analyses include daily temperature. The children’s model also includes same day humidity. The
45 μg/m3 increment in PM10 approximates the interquartile range. bThe fixed-effects models include the daily weather variables.
Air pollution and respiratory health in Bangkok
adults, for an interquartile change (75th–25th
percentile) in PM10 of approximately 45
μg/m3, the odds ratio is 1.66 [95% confidence
interval (CI) = 1.52–1.82] for lower
respiratory symptoms and 1.94 (95% CI =
1.77–2.12) for upper respiratory symptoms.
Lower effect magnitudes were observed for
the panels of nurses and children, but PM10
was associated with statistically significant
increases in frequencies for both symptom
categories for all three panels, with and
without daily weather variables included in
the models.
The inclusion of a variable indicating the
presence of a symptom on the prior day
caused virtually no change in the estimated
PM10 effects for lower respiratory symptoms
for all three panels relative to the basic model
with daily weather variables, but attenuated
the estimated effect of PM10 on upper respiratory
symptoms in all three panels. Omitting
the hottest 25% of the days tended to increase
the estimated association, whereas omitting
the coldest 25% of the days lowered the estimate,
but the PM10 effect remained statistically
significant in nearly all cases. The effects
of PM10 on the likelihood of a new symptom
episode were examined in an analysis that
included only those days for which there were
no symptoms reported on the previous day.
The results indicate an association exists for
both adult panels and for both outcomes. The
magnitude and statistical significance of the
PM10 effect remained comparable to that
found in the original model for all days.
However, the results for children showed no
statistically significant effect of PM10 on new
symptoms. We also ran a model in which
cough was not included as a lower respiratory
symptom. The results were similar to those
obtained for lower respiratory symptoms
when cough was included.
The last row in Table 4 shows the PM10
results estimated with a fixed-effects model
that included the daily weather variables. The
fixed-effects model allows the baseline symptom
incidence to vary for each individual and
corrects for the correlations among repeated
responses from the same individuals. The
PM10 results for the fixed-effects models for
the two adult panels were little changed from
the results using the basic model with daily
weather variables. However, the PM10 results
for the children were about 50
การแปล กรุณารอสักครู่..
