Escolar Documentos
Profissional Documentos
Cultura Documentos
Research paper
art ic l e i nf o
a b s t r a c t
Article history:
Received 21 April 2016
Received in revised form
11 June 2016
Accepted 10 July 2016
Available online 16 July 2016
Background: Researchers have examined the relationship between mental health and weather/pollution
with mixed results. The current study aimed to examine a range of weather and atmospheric phenomena
and their association with time-bound mental health data.
Methods: Nineteen different weather/pollution variables were examined in connection with an archive
of self-reported mental health data for university students participating in mental health treatment
(n 16,452) using the Outcome Questionnaire 45.2 (OQ-45). Statistical approach involved randomly
selecting 500 subjects from the sample 1000 different times and testing each variable of interest using
mixed models analyses.
Results: Seasonal changes in sun time were found to best account for relationships between weather
variables and variability in mental health distress. Increased mental health distress was found during
periods of reduced sun time hours. A separate analysis examining subjects endorsement of a suicidality
item, though not statistically signicant, demonstrated a similar pattern. Initial results showed a relationship between pollution and changes in mental health distress; however, this was mediated by sun
time.
Limitations: This study examined a relatively homogenous, predominantly European American, and religious sample of college counseling clients from an area that is subject to inversions and is at a high
altitude and a latitude where sun time vacillates signicantly more than locations closer to the equator.
Conclusions: Seasonal increases in sun time were associated with decreased mental health distress. This
suggests the need for institutions and public health entities to plan for intervention and prevention
resources and strategies during periods of reduced sun time.
& 2016 Elsevier B.V. All rights reserved.
Keywords:
Weather
Pollution
Emotion
Mood
Season
1. Introduction
There is a common perception that weather can play a critical
part in the way a person feels at a given moment (Watson, 2000).
Common metaphors such as under the weather reect evidence
of the pervasiveness of this belief. The belief that weather inuences mood has also resulted in multiple research hypotheses
which have been well researched with varied results. For example,
Harmatz et al. (2000) found strong seasonal effects on levels of
depression, hostility, anger, irritability, and anxiety. Kmpfer and
n
Mutz (2013) found that respondents surveyed on sunny days reported higher life satisfaction compared with those surveyed on
days with mixed or rather bad weather (p. 587). One study in
Helsinki, Finland found signicant correlations between atmospheric pressure and suicide attempts, especially for men (Hiltunen, 2012).
Conversely, Keller et al. (2005) found that neither temperature
nor pressure was directly related to mood, but rather moods improved with time spent outside on warm, high-pressure days.
Denissen et al. (2008) found no signicant main effects of temperature, wind power, sunlight, precipitation, air pressure, or
photoperiod on positive affect. They also found that increases in
temperature resulted in increased negative affect, but increases in
sunlight and wind power decreased negative affect.
235
2. Methods
2.1. Participants
Mental health distress data for this study came from a 6-year
period (from October 2008 through September 2014) of a deidentied, archival dataset continually collected by the large university counseling center at Brigham Young University. This university is located in Provo (Utah County), Utah. The latitude of the
university is 40.2497, the longitude is 111.6489, and the altitude
is 4630 feet. Demographically, the 16,452 university students
were, 55.4% female, 44.6% male, aged 1668 (M 23.14), and
identied as 84% Caucasian/White, 6.5% Hispanic, 4.2% Asian, 1%
Black, and 4.3% Other.
2.2. Measure
The OQ-45 (Lambert et al., 2004) is a 45-item, self-report
measure of psychological functioning. Items are measured on a
5 point Likert scale: 0 never, 1 rarely, 2 sometimes,
3 frequently, 4 almost always. The range of scores possible on
the OQ-45 is 0-180, with higher scores reecting greater distress.
The OQ-45 assesses three broad domains of client functioning:
(a) subjective discomfort, (b) interpersonal relationships, and
(c) social role performance. The OQ-45, which was normed on
local and national populations, clearly discriminates between
clinical and nonclinical samples (Lambert et al., 2004; Umphress
et al., 1997), and is currently available in 17 languages. Using formulas developed by Jacobson and Truax (1991), clinical and normative data for the OQ-45 were analyzed by Lambert et al. (2004)
to provide cutoff scores for the reliable change index (RCI; 14
points) and clinically signicant change (dysfunctional/functional
cutoff: 64/63). These formulas for cutoff scores and classications
for change are supported by multiple studies (Bauer et al., 2004;
Lunnen and Ogles, 1998) and provide consensus with other selfreport measures like the Beck Depression Inventory (Beck et al.,
1996) and the Symptom Checklist-90 (Derogatis, 1996).
2.3. Procedure
Mental health treatment sessions at the university Counseling
and Psychological Services center are offered free of charge and
without session limits to full-time students of the university. Clients are referred or self-referred for a wide range of presenting
concerns, the majority of which are adjustment, anxiety, or depression related, and are given the OQ-45 at each session to track
change in symptom distress. Although outcome measurement is a
part of routine practice at this center, clients are asked for permission for their anonymized data to be included in research as
part of the intake process; only consenting clients data were included and each client was given an ID number unique to the
dataset. The study was conducted in compliance with the university's Internal Review Board.
2.4. Environmental data
Weather data were gathered from the university Physics and
Astronomy Weather Station. Pollution data were downloaded from
the US Environmental Protection Agency (EPA) online database
236
(2016). Weather and pollution data were selected that corresponded with the same six year period as the mental health data
mentioned above. Time intervals for the weather data were minute-by-minute. Minute-by-minute pollution data were not available, as the EPA provides hourly (O3, CO, NO2, & PM10) and 24-h
averages (PM 2.5 and PM10). For PM2.5 and PM10 the 24-hour
averages were used throughout the analysis.
2.5. Data Analysis
Because of the large sample, it was likely that relationships that
were practically unimportant would be identied as statistically
signicant. To ameliorate this problem, the researchers created
multiple small samples from the entire database. Each of these
samples consisted of 500 randomly chosen clients, with every
available OQ-45 score for the selected clients. The researchers
created 1000 samples of 500 clients each for every variable considered, creating distributions of p values for each variable.
All sampling and analyses were conducted in SAS, version 9.4.
Mixed models analyses were run for each of the samples using
PROC MIXED, and blocking on client ID. The distribution of p values for each variable was then used to determine the practical
signicance of that variable. The researchers used overall symptom distress (as measured by the OQ-45) and responses to item #8
on the OQ-45 (suicidal ideation) as dependent variables. An extensive list of independent variables included sun time (the hours
between sunrise and sunset), sunrise, sunset, lunar phase, rainfall,
wind speed, wind direction, wind chill, temperature, humidity,
atmospheric pressure, absolute pressure, ozone, irradiance (i.e.,
power per unit area received from the sun), dew point, carbon
monoxide, coarse particulate matter (PM10), ne particulate
matter (PM2.5), and nitrogen dioxide. Sunshine hours were not
available in the current dataset. Because values in the dataset for
ozone concentration all fell between 0.0 and 1.0 (therefore confounding the practical meaning of a one unit change in our analyses), the researchers used the standard deviation for all ozone
measurements over the time period (0.017) as the reference unit
for ozone in our mixed models.
In order to account for immediate and cumulative effects of
each independent variable, averages over one hour, four hours,
one day, and seven days were examined in each analysis (with the
exception of PM2.5 and PM10, for which only 24-h averages were
available, as well as sunrise, sunset, and sun time; these variables
were only calculated at one day and averaged over seven days).
After determining the most relevant variable related to OQ-45
scores using the method above, a second round of simulations was
performed. This second round was conducted to determine if any
variables were signicant when controlling for the most relevant
variable identied; in this round, therefore, two variables were
included in each analysis, the most relevant variable and each of
the other variables in turn.
When comparing the distribution of p values for each variable,
two indicators of statistical signicance were examined: rst, the
mean p value over the entire distribution; and second, the p value
that fell at the 50th percentile of the distribution of the p values
for that variable. If the mean p value or the 50th percentile p value
fell below 0.05, it is likely that the variable of interest was a signicant predictor of the dependent variable.
3. Results
In our rst round of analyses, 5 variables were identied that
had signicant mean p values and 12 variables that were signicant at the 50th percentile. Sun time, sunset, irradiance, and
ozone had statistically signicant, negative relationships with OQ-
Table 1
Mixed model analyses results for OQ-45 (Overall Score) and meteorological
variables.
Variable
Mean p
value
0.021n
0.037n
0.018n
0.033n
0.040n
0.061
0.165
0.124
0.129
0.133
0.083
0.080
o 0.001n
o 0.001n
o 0.001n
o 0.001n
o 0.001n
0.001n
0.038n
0.015n
0.035n
0.037n
0.013n
0.010n
0.78
0.72
1.53
1.40
1.36
1.23
0.94
1.43
0.70
0.71
1.14
1.47
0.27
0.29
0.54
0.57
0.54
0.55
0.59
0.88
0.35
0.35
0.48
0.64
Note: W/m2 Watts per square meter; SE standard error; A negative estimate
indicates a decrease in OQ-45 scores, and likewise a decrease in symptom distress.
The units for each variable are noted in parentheses. Ozone units were based on the
standard deviation for all ozone measurements over the time period studied
(0.017).
n
p o 0.05.
4. Discussion
While many researchers have attempted to nd correlations
between meteorological phenomena and mental health, few studies have examined overall distress as measured by a mental
health treatment outcome measure to explore this relationship.
Because this study involved collecting data from a university
where the counseling center requires every client to complete the
OQ-45 before each session and where the physics department
collects detailed meteorological data (minute by minute since
February 19, 2000), the researchers were able to examine the relationship between measured client distress (including suicidal
ideation) and numerous meteorological variables (including multiple measures of weather and pollution) that were specically
tied to the location and time of survey administration. This gave
this study a signicant methodological advantage over previous
237
were typically collected by regional weather centers, which suggests a large and potentially diverse weather area. This presents
limitations because it is difcult to know if the reported weather
represented the area in which the client data was collected. All
subjects in the current study attended the same university where
meteorological data was collected. Therefore, the researchers
made the reasonable assumption that subjects were spending
considerable time in the same general geographic area.
4.1. Limitations and future research
While the researchers were able to correct many limitations
from previous research in this study, there were still some limitations that could have impacted the ndings. The rst is that the
mountainous region where this study occurred has a history of
being subject to the inversion effect. Typically, temperatures decrease as altitude increases, but the inversion effect causes temperatures to increase with altitude which can increase pollution
levels. Because of this, pollution changes often follow seasonal
patterns, with inversions being more prevalent during winter
months. This could explain the nding that sun time best accounted for changes in mental health distress. In the future it
would be important to perform this study in an area that is not
subject to inversion effects, eliminating the possible confounds
this could create.
Previous research has found correlations between living at a
higher altitude and increased rates of suicide (Brenner et al., 2011;
Cheng, 2010). The current study only included clients living at a
high altitude, presenting the possible limitation that it cannot be
generalizable to populations living at lower altitudes.
Additionally, in the city where the current study was performed sun time vacillates more than in cities that are closer to the
equator, like Honolulu. Future studies comparing areas that are
closer to the equator to areas that are far from it could provide
more information on the effects of sun time. Since the geographic
area under study also receives roughly 17 in. less rain per year than
the US average and sits at an elevation of over three times the US
average (Sperling, 2016), these meteorological and regional differences could conceivably cause some limitations to the effectiveness of the study.
Lastly, the student population studied may be a source of potential limitation. The sample was collected from a relatively
homogeneous university that is composed of mostly European
American, religious students. The unique and homogeneous nature of the sampling group may not represent how people in other
areas may respond to the OQ-45 or to meteorological phenomena
and pollution.
It seems important to perform similar research in other locations to see if results are similar. Because it is unclear how sun
time may affect mental health distress and treatment outcome,
additional research is needed to explore sun time related variables,
such as supplemental vitamin D intake, amount of outdoor exercise, exposure to nature, tanning, etc.
References
American Psychological Association, 2016. Seasonal affective disorder, Psychology
Help Center. http://www.apa.org/helpcenter/seasonal-affective-disorder.aspx
(accessed 19.04.16).
Bauer, S., Lambert, M.J., Nielson, S.L., 2004. Clinical signicance methods: a comparison of statistical techniques. J. Pers. Assess. 82, 6070. http://dx.doi.org/
10.1207/s15327752jpa8201_11.
Beck, A.T., Steer, R.A., Ball, R., Ranieri, W., 1996. Comparison of Beck depression
inventories -IA and -II in psychiatric outpatients. J. Pers. Assess. . http://dx.doi.
org/10.1207/s15327752jpa6703_13
Brenner, B., Cheng, D., Clark, S., Camargo, C.A., 2011. Positive association between
altitude and suicide in 2584 U.S. counties. High Alt. Med. Biol. 12, 3135. http:
238
//dx.doi.org/10.1089/ham.2010.1058.
Cheng, D., 2010. Higher suicide death rate in Rocky Mountain States and a correlation to altitude. Wilderness Environ. Med. 21, 177178. http://dx.doi.org/
10.1016/j.wem.2010.01.004.
Chew, K.S., McCleary, R., 1994. A life course theory of suicide risk. Suicide LifeThreat. Behav. 24, 234244.
Clark, L.A., Watson, D., 1988. Mood and the mundane: relations between daily life
events and self-reported mood. J. Pers. Soc. Psychol. 54, 296308. http://dx.doi.
org/10.1037/00223514.54.2.296.
Dalgleish, T., Rosen, K., Marks, M., 1996. Rhythm and blues: the assessment and
treatment of seasonal affective disorder. Br. J. Clin. Psychol. 35, 163182.
Deisenhammer, E.A., 2003. Weather and suicide: the present state of knowledge on
the association of meteorological factors with suicidal behaviour. Acta Psychiatr.
Scand. 108, 402409. http://dx.doi.org/10.1046/j.0001-690X.2003.00209.x.
Demeneix, B., 2014. Losing Our Minds: How Environmental Pollution Impairs Human Intelligence and Mental Health. Oxford University Press, New York.
Denissen, J.J.A., Butalid, L., Penke, L., van Aken, M.A.G., 2008. The effects of weather
on daily mood: a multilevel approach. Emotion 8, 662667. http://dx.doi.org/
10.1037/a0013497.
Derogatis, L.R., 1996. SCL-90-R: Symptom Checklist-90-R: Administration, Scoring,
and Procedures Manual. NCS Pearson, Minneapolis, MN.
Fritze, J.G., Blashki, G.A., Burke, S., Wiseman, J., 2008. Hope, despair and transformation: climate change and the promotion of mental health and wellbeing. Int.
J. Ment. Health Syst. 2, 13. http://dx.doi.org/10.1186/17524458-2-13.
Gordijn, M.C.M., t Mannetje, D., Meesters, Y., 2012. The effects of blue-enriched
light treatment compared to standard light treatment in seasonal affective
disorder. J. Affect. Disord. 136, 7280. http://dx.doi.org/10.1016/j.
jad.2011.08.016.
Harmatz, M.G., Well, A.D., Overtree, C.E., Kawamura, K.Y., Rosal, M., Ockene, I.S.,
2000. Seasonal variation of depression and other moods: a longitudinal approach. J. Biol. Rhythm. 15, 344350. http://dx.doi.org/10.1177/
074873000129001350.
Hiltunen, L., Ruuhela, R., Ostamo, A., Lnnqvist, J., Suominen, K., Partonen, T., 2012.
Atmospheric pressure and suicide attempts in Helsinki, Finland. Int. J. Biometeorol. 56, 10451053. http://dx.doi.org/10.1007/s00484-011-0518-2.
Huibers, M.J.H., de Graaf, L.E., Peeters, F.P.M.L., Arntz, A., 2010. Does the weather
make us sad? Meteorological determinants of mood and depression in the
general population. Psychiatry Res. 180, 143146. http://dx.doi.org/10.1016/j.
psychres.2009.09.016.
Jacobson, N.S., Truax, P., 1991. Clinical signicance: a statistical approach to dening
meaningful change in psychotherapy research. J. Consult. Clin. Psychol. 59,
1219. http://dx.doi.org/10.1037/0022-006X.59.1.12.
Kadotani, H., Nagai, Y., Sozu, T., 2014. Railway suicide attempts are associated with
amount of sunlight in recent days. J. Affect. Disord. 152154, 162168. http://dx.
doi.org/10.1016/j.jad.2013.08.040.
Kmpfer, S., Mutz, M., 2013. On the sunny side of life: sunshine effects on life satisfaction. Soc. Indic. Res. 110, 579595. http://dx.doi.org/10.1007/
s11205-011-9945-z.
Keller, M.C., Fredrickson, B.L., Ybarra, O., Ct, S., Johnson, K., Mikels, J., Conway, A.,
Wager, T., 2005. A warm heart and a clear head: the contingent effects of
weather on mood and cognition. Psychol. Sci. 16, 724731. http://dx.doi.org/
10.1111/j.14679280.2005.01602.x.
Lam, R.W., Levitt, A.J., Levitan, R.D., Enns, M.W., Morehouse, R., Michalak, E.E., Tam,
E.M., 2006. The can-SAD study: a randomized controlled trial of the effectiveness of light therapy and uoxetine in patients with winter seasonal affective
disorder. Am. J. Psychiatry 163, 805812. http://dx.doi.org/10.1176/appi.
ajp.163.5.805.
Lambert, M.J., Kahler, M., Harmon, C., Shimokawa, K., Burlingame, G., 2011. Administration and scoring manual for the Outcome Questionnaire-45.2. American Professional Credentialing Services, Orem, UT.
Lambert, M.J., Morton, J.J., Hateld, D.R., Harmon, C., Hamilton, E.G., Reid, R.C., et al.,
2004. Administration and Scoring Manual for the OQ-45.2 (Outcome Questionnaire). American Professional Credentialing Services, Salt Lake City, UT.
Levitt, A.J., Lam, R.W., Levitan, R., 2002. A comparison of open treatment of seasonal
major and minor depression with light therapy. J. Affect. Disord. 71, 243248.
Lunnen, K.M., Ogles, B.M., 1998. A multiperspective, multivariable evaluation of
reliable change. J. Consult. Clin. Psychol. 66, 400410. http://dx.doi.org/10.1037/
0022-006X.66.2.400.
Martiny, K., Simonsen, C., Lunde, M., Clemmensen, L., Bech, P., 2004. Decreasing TSH
levels in patients with Seasonal Affective Disorder (SAD) responding to 1 week
of bright light therapy. J. Affect. Disord. 79, 253257. http://dx.doi.org/10.1016/
S0165-0327(02)00361-0.
Michalak, E.E., Tam, E.M., Manjunath, C., Levitt, A.J., Levitan, R.D., Lam, R.W., 2005.
Quality of life in patients with seasonal affective disorder: summer vs winter
scores. Can. J. Psychiatry 50, 292295.
Rastad, C., Ulfberg, J., Lindberg, P., 2008. Light room therapy effective in mild forms
of seasonal affective disorder a randomised controlled study. J. Affect. Disord.
108, 291296. http://dx.doi.org/10.1016/j.jad.2007.10.009.
Roecklein, K.A., Schumacher, J.A., Miller, M.A., Ernecoff, N.C., 2012. Cognitive and
behavioral predictors of light therapy use. PLoS One, 7. http://dx.doi.org/
10.1371/journal.pone.0039275.
Sperling, B., 2016. Sperling's Best Places. http://www.bestplaces.net/climate/city/
utah/provo (accessed 19.04.16).
Traffanstedt, M.K., Mehta, S., LoBello, S.G., 2016. Major depression with seasonal
variation: is it a valid construct? Clin. Psychol. Sci. . http://dx.doi.org/10.1177/
2167702615615867
Umphress, V.J., Lambert, M.J., Smart, D.W., Barlow, S.H., Clouse, G., 1997. Concurrent
and construct validity of the Outcome Questionnaire. J. Psychoeduc. Assess. 15,
4055.
US Environmental Protection Agency, 2016. Air quality system data mart [data le].
http://www.epa.gov/ttn/airs/aqsdatamart (accessed 12.01.16).
Watson, D., 2000. Mood and Temperament. Guilford Press, New York.
Wesson, V.A., Levitt, A.J., 1998. Light therapy for seasonal affective disorder. In: Lam,
R.W. (Ed.), Seasonal Affective Disorder and beyond: Light Treatment for SAD
and Non-SAD Conditions. American Psychiatric Association, Washington, DC,
pp. 4589.