Cognitive and behavioral predictors of light therapy use.
ABSTRACT Although light therapy is effective in the treatment of seasonal affective disorder (SAD) and other mood disorders, only 53-79% of individuals with SAD meet remission criteria after light therapy. Perhaps more importantly, only 12-41% of individuals with SAD continue to use the treatment even after a previous winter of successful treatment.
Participants completed surveys regarding (1) social, cognitive, and behavioral variables used to evaluate treatment adherence for other health-related issues, expectations and credibility of light therapy, (2) a depression symptoms scale, and (3) self-reported light therapy use.
Individuals age 18 or older responded (n = 40), all reporting having been diagnosed with a mood disorder for which light therapy is indicated. Social support and self-efficacy scores were predictive of light therapy use (p's<.05).
The findings suggest that testing social support and self-efficacy in a diagnosed patient population may identify factors related to the decision to use light therapy. Treatments that impact social support and self-efficacy may improve treatment response to light therapy in SAD.
- SourceAvailable from: Manan Pareek[Show abstract] [Hide abstract]
ABSTRACT: Low serum 25-hydroxyvitamin D levels (25(OH)D) have been associated with a higher likelihood of seasonal affective disorder (SAD) and poor mental well-being, yet firm evidence for either remains lacking. Thus, vitamin D supplementation may alleviate symptoms associated with SAD.BMC Research Notes 08/2014; 7(1):528.
Cognitive and Behavioral Predictors of Light Therapy Use
Kathryn A. Roecklein1*, Julie A. Schumacher2, Megan A. Miller1, Natalie C. Ernecoff1
1Department of Psychology, University of Pittsburgh, Pittsburgh, Pennsylvania, United States of America, 2Department of Psychiatry and Human Behavior, University of
Mississippi Medical Center, Jackson, Mississippi, United States of America
Objective: Although light therapy is effective in the treatment of seasonal affective disorder (SAD) and other mood
disorders, only 53–79% of individuals with SAD meet remission criteria after light therapy. Perhaps more importantly, only
12–41% of individuals with SAD continue to use the treatment even after a previous winter of successful treatment.
Method: Participants completed surveys regarding (1) social, cognitive, and behavioral variables used to evaluate treatment
adherence for other health-related issues, expectations and credibility of light therapy, (2) a depression symptoms scale, and
(3) self-reported light therapy use.
Results: Individuals age 18 or older responded (n=40), all reporting having been diagnosed with a mood disorder for which
light therapy is indicated. Social support and self-efficacy scores were predictive of light therapy use (p’s,.05).
Conclusion: The findings suggest that testing social support and self-efficacy in a diagnosed patient population may
identify factors related to the decision to use light therapy. Treatments that impact social support and self-efficacy may
improve treatment response to light therapy in SAD.
Citation: Roecklein KA, Schumacher JA, Miller MA, Ernecoff NC (2012) Cognitive and Behavioral Predictors of Light Therapy Use. PLoS ONE 7(6): e39275.
Editor: Antonio Verdejo Garcı ´a, University of Granada, Spain
Received March 29, 2012; Accepted May 22, 2012; Published June 13, 2012
Copyright: ? 2012 Roecklein et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was not supported by an outside funding agency. The funders had a role in study design, data collection and analysis, decision to publish, or
preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: firstname.lastname@example.org
Although light therapy is effective in the treatment of seasonal
affective disorder (SAD) , or Major Depressive Disorder with a
Seasonal Pattern, as well as for non-seasonal depression , data
suggest that only about 41% (24 out of 59) of SAD patients treated
successfully with light therapy (LT) report regularly using the
treatment in follow-up studies . In another study combining LT
and cognitive behavioral therapy (CBT), only 12% (4 out of 34) of
individuals with SAD originally treated with LT alone (n=19) or
LT plus CBT (n=15) used LT during the following winter .
These data suggest that the choice to use light therapy is itself a
target for intervention. In the present study, we aimed to test
whether certain motivational and social cognitive processes may
help explain the decision to utilize LT.
Two separate issues exist, 1) LT use during the acute treatment
phase, and 2) LT use in subsequent winters. However, similar
factors may be associated with LT use in both instances. Among
those using LT, only 53% of all those with SAD and 43% of
moderate to severe cases meet remission criteria after LT . A
more recent light therapy trial found higher remission rates,
ranging from 46% to 79%, depending on the stringency of
remission criteria . These remission rates compare favorably to
remission rates for antidepressant medications, which ranged from
43% to 62% in recent meta-analyses and pooled analyses [7–9].
Maximizing remission rates may require higher rates of adherence
to LT prescriptions. Adherence to LT in SAD is lower than hoped,
about 41%–60% [3,10–11] when treatment dropouts are includ-
ed, and this incomplete adherence may explain the incomplete
remission rates. The general literature on medication and medical
treatment compliance suggests dose-taking compliance declines as
the number of daily doses increases , and depression
symptoms are associated with noncompliance . Light therapy
requires approximately 30–45 minutes a day, which represents a
significant time commitment. Factors associated with light therapy
use and adherence could be clinically relevant if such factors could
be manipulated to help improve treatment outcomes.
In the one study to date assessing possible explanations for not
using light therapy, 59 patients originally treated with light therapy
were reassessed approximately 9 years later . Among the 56%
(n=33 out of 59) who had discontinued using light therapy, 14%
(n=8) reported they had not had sufficient symptoms, leaving
44% of individuals who did have sufficient symptoms to warrant
treatment but chose not to use light therapy. Those individuals
reported inconvenience and/or perceived ineffectiveness as
reasons for not using light therapy despite having experienced
improvement during initial light therapy treatment. To examine
further which factors predict light therapy use, we reviewed
research on treatment adherence in depression and other similar
health conditions in which treatment requires a time commitment.
The Transtheoretical Model [14–15], and Social Cognitive
Theory  both describe variables that have been identified as
important predictors of compliance with treatments such as
continuous positive airway pressure (CPAP) treatment for
obstructive sleep apnea, a treatment requiring nightly use of a
PLoS ONE | www.plosone.org1 June 2012 | Volume 7 | Issue 6 | e39275
CPAP mask and machine somewhat similar to the daily use
required for light therapy . Factors such as self-efficacy and
depression symptoms have also been associated with adherence to
treatments for depression. The Theory of Planned Behavior
hypothesizes that tangible barriers (e.g., access to care), illness
barriers (e.g., depression symptoms, concerns about treatment),
and psychological barriers (e.g., self-efficacy, stigma) are modifi-
able and, therefore, targets of interventions to improve adherence
to medication treatment of depression . An intervention based
on this model that incorporates motivational interviewing,
problem solving, and psychoeducation has shown a significant
increase in adherence to medications in older adults with
depression from baseline across 24 weeks compared to treatment
as usual .
Although different in their mechanisms of efficacy, homework
within the context of psychotherapy has similar duration and daily
frequency demands as light therapy. A recent meta-analysis of 23
studies on homework found that homework compliance had a
small to medium effect (r=.26) on treatment outcome .
Individuals with more severe or longer-lasting symptoms comply
less with homework . Variables including motivation, readi-
ness to change, and necessary skills are correlated with homework
compliance and treatment outcome in cognitive behavioral
therapy for depression [21–22].
The above data on factors associated with adherence to either
homework in psychotherapy or medication treatment for depres-
sion utilize constructs from multiple theories of behavior change.
Fishbein et al. (2001)  proposed an integration of social
cognition, health belief, and other models and focused on eight
shared variables including intention to change behavior, environ-
mental barriers, skills, outcome expectations or attitudes, norms,
self-standards, emotion, and self-efficacy. The Fishbein et al.
(2001)  report focused on AIDS-related health behaviors that
may have a similar burden of daily frequency and duration as light
therapy. Neimeyer and colleagues (2008)  recently reviewed
theoretical models appropriate for predicting homework compli-
ance in psychotherapy and found that willingness, as well as
motivation and stage of change predicted homework compliance.
Aims of the Study
We hypothesize that the aforementioned variables such as social
support, self-efficacy, and treatment credibility will predict use of
LT in individuals with SAD or non-seasonal depression. There-
fore, the present study sought to collect preliminary data using an
anonymous web survey of individuals self-identifying as having
been previously diagnosed with a disorder for which light therapy
is indicated as a treatment. Specifically, we hypothesize that the
decision to use LT in the previous winter would be associated with
cognitive, behavioral, and social variables including self-efficacy,
outcome expectations, social support, processes of change,
knowledge and the degree of perceived treatment credibility. We
further hypothesized that LT non-use would be associated with
higher frequency of self-reported depression symptoms in the
Participants were recruited through Internet websites such as
online SAD-focused groups and websites for national organiza-
tions focused on support for individuals with depression. Partic-
ipants completed the survey anonymously, and IP addresses were
not recorded. Participants were informed that they would not be
compensated for participation in order to preserve anonymity, and
the survey was estimated to take about 20 minutes to complete.
The study was approved and determined to be exempt from
Human Subjects Research by the University of Pittsburgh
Institutional Review Board, because it involved the use of survey
data recorded in such a manner that human subjects cannot be
identified, as no identifiers were collected. Written documentation
of informed consent was waived to preserve anonymity, although
information was given to participants prior to beginning the
surveys regarding the duration, content, and focus of the study.
Only individuals reporting a diagnosis for which LT is a treatment
and those who reported having heard of LT were included in the
Questionnaires specific to light therapy were based on existing
measures described below, and were tested for psychometric
properties including internal consistency as a measure of reliability
and dimensionality as a measure of construct validity. Reliability
was assessed with Chronbach’s alpha, a measure of the
intercorrelation among items in a scale, with the traditional cut
off of alpha=0.70. Scales with Chronbach’s alphas that fell within
the acceptable range for internal consistency (range: 0.70–0.94)
were retained. Dimensionality was measured with confirmatory
factor analysis, to determine if each scale is unidimensional as
hypothesized, reflecting the degree of variance in the measure due
to a single common factor. Principal components extraction with
varimax rotation was performed, and all scales except for the
knowledge scale were identified as having a single component.
With empirical evidence supporting many constructs reviewed
above, some similar constructs were excluded in the present study
to reduce participant burden or to address specific research
hypotheses. Some measures have no clear interpretation when
administered retrospectively (i.e., stage of change, intention).
Others, such as attitudes, perceived behavioral control, and social
norms may overlap with outcome expectations, self-efficacy, and
social support scales, respectively. Therefore, the present study
involved modifying a subset of all possible scales including the
following existing scales to be appropriate for light therapy for
depression: outcome expectations, self-efficacy, social support,
knowledge, process of change, and treatment credibility. Con-
struction methods and psychometric test results for each scale are
described further below.
Outcome expectations are beliefs
about the efficacy and importance of a given behavior in
producing desired outcomes, a construct derived from the social
Table 1. Demographic variables compared between those
that did and did not use LT.
Did Use LTDid Not Use LT
Variable & Sample (n)M SDMSDFp
Age46.65 13.2437.24 13.065.19.03*
White1482.419 82.60 1.00
Non-White3 17.64 17.4
Female 1588.22191.3.10 .75
Male2 11.82 8.7
Predictors of Light Therapy Use
PLoS ONE | www.plosone.org2 June 2012 | Volume 7 | Issue 6 | e39275
cognitive theory [16,24]. This scale was modified based on one
from Stepnowski et al. (2002)  for CPAP use in sleep apnea,
and another from Gyurcsik, Brawley, Spink, Glazebrook, and
Anderson (2011)  designed to measure outcomes related to
arthritis. Three items assessed how effective participants believed
regular use of light therapy is for managing mood, fatigue, and
sleep, rated on a 5-point Likert scale from ‘‘not at all effective’’ to
‘‘extremely effective.’’ A fourth item assessed how important
respondents believed regular use of light therapy is for managing
symptoms on a 5-point scale from ‘‘not at all important’’ to
‘‘extremely important.’’ Internal consistency was acceptable
(Chronbach’s alpha=.84), and factor analysis revealed a single
dimension for this light therapy outcome expectations scale.
a given behavior . Our self-efficacy scale was modeled after
those used for CPAP therapy  and exercise . Previous
items such as ‘‘I am confident I can participate in exercise when I
am tired’’ and ‘‘I am confident I will use CPAP regularly even if I
do not feel like it’’ were revised for light therapy (e.g., ‘‘I am
confident I can use light therapy regularly even when I don’t want
to get up early’’). Each of 5 items was scored on a 5-point Likert
scale from ‘‘disagree completely’’ to ‘‘agree completely.’’ Internal
consistency was high (Chronbach’s alpha=.94), and factor
analysis revealed a single dimension for this light therapy self-
Social support in this context refers to the
utility of support from friends, family, and health care staff in
supporting a given behavior change or adherence to a given
treatment, and is also derived from the social cognitive theory
. The present scale was modeled on those for CPAP  and
in light of recommendations for the assessment of social support
for behavior change [16,24]. Items such as ‘‘I have people in my
life who will support me in using CPAP regularly’’ were revised for
light therapy (e.g., ‘‘I have people in my life who support me in
using light therapy regularly’’). Each of the eight items was scored
on a 5-point Likert scale from ‘‘disagree completely’’ to ‘‘agree
completely.’’ Internal consistency was high (Chronbach’s al-
pha=.94), and factor analysis revealed a single dimension for this
light therapy social support scale.
The social cognitive theory proposes that
accurate information provides the basis upon which behavior
change will take place, although it is not expected to explain
behavior change alone . Knowledge measures the degree of
accuracy of information a person has, as this may provide
motivation to use a particular treatment. The first knowledge scale
Self-efficacy is the belief that one can engage in
Figure 1. Comparison of motivation, credibility, and depression symptom scale scores between those reporting use vs. non-use of
light therapy (M, SD). *p,.05. Each measure has a different scale and minimum/maximum values, and is only compared here between groups
defined by self-reported use vs. non-use of light therapy. Use: Individuals who reported using LT. Non Use: Individuals reporting no use of LT in the
Table 2. MANOVA comparing individuals that did and did
not use LT.
Did Use LT
Did Not Use
Variable & Sample (n)MSDMSDFp
Outcome Expectations 14.47 3.89 12.944.19.01.94 .00
Self-Efficacy 17.006.93 12.605.958.04 .02*.42
Social Support 25.13 10.7511.602.976.14.03*.36
Process of Change 66.5017.4953.4014.363.70.08 .25
CES-D38.29 10.90 45.82 6.9220.127.116.11
Treatment Credibility 40.817.24 41.868.78.30 .60 .03
Predictors of Light Therapy Use
PLoS ONE | www.plosone.org3 June 2012 | Volume 7 | Issue 6 | e39275
for this study was developed based on previously published scales
and included twelve true/false questions such as ‘‘Light therapy is
effective even if the user’s eyes are closed’’ and ‘‘Normal household
lights are just as effective as light boxes designed for light therapy.’’
This knowledge scale had a Chronbach’s alpha of only. 44 and a
multi-factor solution, so it was excluded from analyses. For future
studies, this scale will undergo revision to reduce the level of
knowledge assessed to a more basic level.
Processes of Change.
The Transtheoretical Model com-
bines theories of behavior change with learning theory to describe
stages of motivational readiness to make a change in a given
behavior . Because this was a retrospective survey, the
processes used to engage in change was measured, instead of the
actual stage of behavioral change a given individual is in at the
time of assessment. The hypothesis is that progress through the
stages occurs on the basis of processes of change, or behaviors and
cognitions that promote change such as reading about a particular
treatment. Processes of change describe the reasons that motivate
change, and the means by which it is achieved, and are associated
with actual behavioral change . The current processes of
change questionnaire was based on previously published scales for
exercise behavior , smoking cessation , and CPAP for
sleep apnea . Each of 20 items was rated on a 5-point Likert
scale (i.e., ‘‘never,’’ ‘‘seldom,’’ ‘‘occasionally,’’ ‘‘often,’’ and
‘‘repeatedly’’). Items include cognitive processes including con-
sciousness raising, dramatic relief, environmental reevaluation,
self-reevaluation, and social liberation. The scale also includes
behavioral processes of change including counter-conditioning,
helping relationships, reinforcement management, self-liberation,
and stimulus control. Items such as this stimulus control item were
revised from the original examples; ‘‘I put things around my home
to remind me of exercising’’ , ‘‘I remove things from my home
that remind me of smoking’’  and ‘‘I put things around my
home to remind me to use CPAP’’ , to be appropriate for light
therapy (i.e., ‘‘I put my light therapy device in a place so that I’ll be
easily reminded to use it’’). Internal consistency was high
(Chronbach’s alpha=.92), and factor analysis revealed a single
dimension for this light therapy processes of change questionnaire.
The Treatment Expectations and Credibility Survey.
standardized measure of treatment credibility  was revised to
evaluate expectations, preferences and credibility of LT as a
treatment for depression. In the original Borkovec and Nau (1972)
 measure, participants rated a new therapy intended to reduce
public speaking anxiety in five questions on a 10-point Likert scale.
The original included questions such as ‘‘1. How logical does this
type of therapy seem to you?’’ and ‘‘2. How confident would you
be that this treatment would be successful in eliminating fear of
speaking before a group?’’ For the present study, we rephrased
questions to reflect light therapy for depression, (e.g., 2. ‘‘How
confident would you be that light therapy would be successful in
eliminating depression?’’). This is consistent with the treatment
expectation evaluation in Michalak et al. (2007)  which
assessed 4 items: 1) how logical the treatment seems, 2) how
confident they are it would be successful, 3) how useful it might be,
and 4) how confident they would be in recommending light
therapy to a friend. Each of the 5 items on the scale was rated on a
10-point Likert scale from ‘‘not at all logical’’ to ‘‘very logical’’ or
‘‘not at all confident’’ to ‘‘very confident.’’ The resulting scale had
acceptable internal consistency (alpha=.74) and a yielded single
factor in factor analysis.
Depression Symptom Frequency.
miologic Studies Depression Scale (CES-D)  is a 20 item self-
report depression symptom scale developed by the National
Institutes of Mental Health Center for Epidemiologic Studies.
The Center for Epide-
Studies have validated the CES-D as a screening tool to detect
depression symptoms and to measure change in symptom severity
over time . This scale was not re-tested for psychometric
properties in the present study, and was used instead of other
measures because it does not assess suicidality, which would be
difficult to respond to in an anonymous survey conducted on-line.
Participants were asked to ‘‘Think about the week you felt the
most depressed during the past fall-winter season. Below is a list of
the ways you may have felt or behaved during that week. Please
rate how often you felt the following ways during that particular
week, even if you were not depressed last fall or winter.’’
Light Therapy Usage.
A series of items assessing the
quantity and quality of light therapy usage by self-report were
developed for the current study. A meta-analysis by Golden and
colleagues (2005)  reported that the effective starting ‘‘dose’’ of
light therapy ranged from 30 minutes per day to 1–2 hours,
depending on the intensity of the light . Participants were asked
to estimate the amount of time they used light therapy in the
previous winter on both weekday and weekend days, which were
combined for a weighted daily average (i.e., weekday min.65, plus
weekend min. 62, quantity divided by 7). Because 59.5% of the
sample reported not using LT in the previous winter, this variable
was dichotomized into two groups: those that did use LT and
those that did not use LT.
Multivariate analysis of variance was used to compare those that
did use light therapy to those that did not use light therapy on
dependent variables including total scores for processes of change,
treatment credibility, outcome expectations, self-efficacy, social
support, and depression symptoms, while controlling for age. The
significance level was set at 0.05 for the study. Data analyses were
conducted using SPSS 17.0 (SPSS Inc., Chicago, IL).
responded to the survey and met study inclusion criteria. Inclusion
criteria were that individuals had to self-report that they have been
diagnosed with a disorder for which light therapy is a treatment
(i.e., SAD or MDD), and had to report that they had at least heard
of light therapy as a treatment for their disorder. Participants
reported diagnoses of Major Depressive Disorder With Seasonal
Pattern (MDD-SP; n=16, 38.1%), Bipolar I Disorder With
Seasonal Pattern (n=12, 28.6%), Bipolar II Disorder With
Seasonal Pattern (n=9, 21.4%), or Major Depressive Disorder
without a seasonal pattern (MDD; n=5, 11.9%). The number of
individuals with a Bipolar Disorder With Seasonal Pattern did not
differ across the groups using or not using light therapy, X2(1,
42)=0.48, p=.49. Similarly, the number of individuals reporting a
seasonal pattern diagnosis (i.e., MDD-SP, and Bipolar I or II With
Seasonal Pattern) did not differ between groups reporting use or
non-use, X2(1, 42)=.26, p=.67. Groups that did and did not use
light therapy were also compared by race, age, and gender
(Table 1). As the group reporting LT use was older, age was
included as a covariate in the analysis. Overall, 17 out of 40
(42.5%) respondents reported using LT the previous winter.
Predictors of light therapy use.
OVA, the self-efficacy and social support scales were significantly
associated with whether or not individuals reported using light
therapy in the previous winter (Table 2 & Figure 1). Some other
group differences were in the expected direction, albeit not
significantly different across groups statistically (see Table 2).
A total of 40 individuals age 18 or older
In the omnibus MAN-
Predictors of Light Therapy Use
PLoS ONE | www.plosone.org4 June 2012 | Volume 7 | Issue 6 | e39275
Light therapy self-efficacy and social support were associated
with self-report of light therapy use. These data suggest that self-
efficacy and social support predict use of light therapy, just as they
predict use of treatments or behavior change in other health
conditions or other depression treatments. Other variables such as
outcome expectations, processes of change, depression scores, and
treatment credibility were not significantly different in those
reporting light therapy use. Michalak et al. (2007)  found that
treatment credibility and expectations for light therapy were not
associated with adherence to light therapy in SAD, consistent with
our findings. In that study, as well as ours, scores on the treatment
expectations scale were high in relation to the maximum score
possible on each scale. In our study, the group mean scores on
treatment credibility approached the maximum possible score of
10 across all 5 items (Did use: M=8.16, SD=1.45; Didn’t use:
M=8.37, SD=1.76). Therefore, it appears that treatment
credibility is high across both samples, and a ceiling effect may
explain the lack of predicted association between treatment
credibility and LT use.
The main limitation is that the sample used was a self-selected
group of individuals that self-identified as having been diagnosed
with a mood disorder for which LT is recommended, rather than a
group of clinically diagnosed participants. However, means and
standard deviations for the CES-D scores were 38.29 (10.90) for
the group reporting light therapy use, and 45.82 (6.96) for the
group reporting no use, indicating that individuals were reporting
significant depression symptom frequency for the previous winter.
Individuals in this study reported similar symptom levels to those
with a diagnosed mood disorder (M=38.1 for individuals in a
mood episode) . Because self-reported diagnosis may be
inaccurate, future studies will include in-person structured clinical
interviews for diagnosis.
Another limitation of this study is that only 42.5% of the
respondents reported using light therapy, so our measures are
predicting any amount of light therapy use, rather than degree of
adherence to a light therapy prescription. Additionally, the present
study did not measure factors associated with adherence to other
medication or psychotherapy treatments for depression, such as
intention and willingness. These constructs may overlap with those
of process of change, treatment credibility, and outcome
expectations. Intention and willingness are related but theoreti-
cally separate constructs of behavioral change that have been
defined as follows. Willingness reflects how willing an individual
would be to engage in a particular coping strategy if a friend or
treatment provider suggested it . On the other hand, the
construct of intention reflects a person’s intention to perform a
specific behavior (e.g., ‘‘I intend to do X’’) . Multiple theories
including cognitive attitude-behavior relations, models of health
behavior, and goal theory all propose that one’s intention to
complete homework determines motivation and performance [see
35–38]. In a meta-analysis of studies measuring intention and
behavior change, intention accounted for 28% of the variance in
behavior . Burns and Nolen-Hoeksema (1991)  found that
willingness and homework compliance individually predicted
clinical improvement in CBT for depression. Because the present
study is retrospective, and interpretation of intent retrospectively
would be difficult to interpret, we measured overall willingness to
engage in coping strategies for depressed mood. Future studies
could measure intention and willingness at the time of diagnosis of
a mood disorder, before a trial of light therapy, if patients who had
not previously been diagnosed could be recruited.
In some ways, it is surprising that individuals choose not to use
light therapy given that side effects are generally mild [3,41], and
that the rationale linking seasonal recurrence to treatment with
light seems credible . However, we found that other factors
besides low side effects and treatment credibility that may also be
important, namely self-efficacy and social support. Therefore,
interventions that manipulate these motivational, cognitive, and
behavioral factors may increase LT use rates, such as Motivational
Enhancement Therapy  or Motivational Interviewing .
These approaches have demonstrated efficacy for a variety of
health and mental health conditions for which successful treatment
requires complex and sustained behavior change . Under-
standing the impact of thoughts and appraisals regarding
symptoms and treatment on the decision to use LT could be used
to inform cognitive-behavioral interventions to maximize LT use
and improve treatment in mood disorders. Further measurement
of cognitive behavioral predictors of LT use should be performed
prospectively with participants who have undergone structured
diagnostic interviews, to increase confidence in the relationships
between self-efficacy and social support and the use of LT.
Conceived and designed the experiments: KAR JAS. Performed the
experiments: MAM NCE. Analyzed the data: KAR. Wrote the paper:
KAR MAM JAS.
1. Golden RN, Gaynes BN, Ekstrom RD, Hamer RM, Jacobsen FM, et al. (2005)
The efficacy of light therapy in the treatment of mood disorders: a review and
meta-analysis of the evidence. Am J Psychiatry 162: 656–662.
2. Tuunainen A, Kripke DF, Endo T. (2004) Light therapy for non-seasonal
depression (Cochrane Review). The Cochrane Library. Chichester, UK: John
Wiley & Sons, Ltd.
3. Schwartz PJ, Brown C, Wehr TA, Rosenthal NE (1996) Winter seasonal
affective disorder: a follow-up study of the first 59 patients of the National
Institute of Mental Health Seasonal Studies Program. Am J Psychiatry 153:
4. Rohan KJ, Roecklein KA, Lacy TJ, Vacek PM (2009) Winter depression
recurrence one year after cognitive-behavioral therapy, light therapy, or
combination treatment. Behav Ther 40: 225–38.
5. Terman M, Terman JS, Quitkin FM, McGrath PJ, Stewart JW, Rafferty B
(1989) Light therapy for seasonal affective disorder: a review of efficacy.
Neuropsychopharmacology 2: 1–22.
6. Flory R, Ametepe J, Bowers B (2010) A randomized, placebo-controlled trial of
bright light and high-density negative air ions for treatment of Seasonal Affective
Disorder. Psychiatry Res 177: 101–108.
7. Thase ME, Nierenberg AA, Vrijland P, van Oers HJ, Schutte AJ, Simmons JH
(2010) Remission with mirtazapine and selective serotonin reuptake inhibitors: a
meta-analysis of individual patient data from 15 controlled trials of acute phase
treatment of major depression. Int Clin Psychopharmacol 25: 189–198.
8. Lam RW, Lonn SL, Despiegel N (2010) Escitalopram versus serotonin
noradrenaline reuptake inhibitors as second step treatment for patients with
major depressive disorder: a pooled analysis. Int Clin Psychopharmacol 25: 199–
9. Watanabe N, Omori IM, Nakagawa A, Cipriani A, Barbui C, et al. (2008)
Mirtazapine versus other antidepressants in the acute-phase treatment of adults
with major depression: systematic review and meta-analysis. J Clin Psychiatry
10. Michalak EE, Hayes S, Wilkinson C, Hood K, Dowrick C (2002) Treatment
compliance in light therapy: Do patients do as they say they do? J Affect Disord
11. Michalak EE, Murray G, Wilkinson C, Dowrick C, Lam RW (2007) A pilot
study of adherence with light treatment for seasonal affective disorder.
Psychiatry Res 149: 315–320.
12. Claxton AJ, Cramer J, Pierce C (2001) A systematic review of the associations
between dose regimens and medication compliance. Clin Ther 23: 1296–1310.
13. DiMatteo MR, Lepper HS, Croghan TW (2000) Depression is a risk factor for
noncompliance with medical treatment: meta-analysis of the effects of anxiety
and depression on patient adherence. Arch Intern Med 160: 2101–2107.
Predictors of Light Therapy Use
PLoS ONE | www.plosone.org5 June 2012 | Volume 7 | Issue 6 | e39275
14. Prochaska JO, Redding CA, Evers KE (1997) The Transtheoretical Model and
Stages of Change. In: Glanz K, Lewis FM, Rimer BK, eds. Health Behavior and
Health Education. San Francisco: Jossey-Bass Publishers. 60–84.
15. Wilson GT, Schlam TR (2004) The transtheoretical model and motivational
interviewing in the treatment of eating and weight disorders. Clin Psychol Rev
16. Baundra A (1986) Social foundations of thought and action: a social cognitive
theory. Englewood Cliffs, NJ: Prentice Hall.
17. Stepnowsky CJ Jr., Marler MR, Ancoli-Israel S (2002) Determinants of nasal
CPAP compliance. Sleep Med 3: 239–247.
18. Sirey JA, Bruce ML, Kales HC (2010) Improving antidepressant adherence and
depression outcomes in primary care: The treatment initiation and participation
(TIP) program. Am J Geriatr Psychiatry 18: 554–562.
19. Mausbach BT, Moore R, Roesch S, Cardenas V, Patterson TL (2010) The
relationship between homework compliance and therapy outcomes: an updated
meta-analysis. Cognit Ther Res 34: 429–438.
20. Worthington EL, (1986) Client compliance with homework directives during
counseling. J Couns Psychol 33: 124–130
21. Neimeyer RA, Kazantzis N, Kassler DM, Baker KD, Fletcher R (2008) Group
cognitive behavioural therapy for depression outcomes predicted by willingness
to engage in homework, compliance with homework, and cognitive restructuring
skill acquisition. Cogn Behav Ther 37: 199–215.
22. Yovel I, Safren SA (2007) Measuring homework relevance in psychotherapy:
CBT for adult ADHD as an example. Cognit Ther Res 31: 385–399.
23. Fishbein M, Triandis HC, Kanfer FH, Becker M, Middlestadt SE, Eichler A
(2001) Factors influencing behavior and behavior change. In: Baum A,
Revenson TA, Singer JE, eds. Handbook of Health Psychology. Mahwah, NJ:
Lawrence Erlbaum. 3–17.
24. Bandura A (1997) The anatomy of stages of change. Am J Health Promot 12: 8–
25. Gyurcsik NC, Brawley LR, Spink KS, Glazebrook KE, Anderson TJ (2011) Is
level of pain acceptance differentially related to social cognitions and behavior?
The case of active women with arthritis. J Health Psychol 16: 530–539.
26. Marcus BH, Selby VC, Niaura RS, Rossi JS (1992) Self-efficacy and the stages of
exercise behavior change. Res Q Exerc Sport 63: 60–66.
27. Prochaska JO (1984) Systems of psychotherapy: a transtheoretical analysis. 2nd
ed. Pacific Grove, CA: Brooks-Cole.
28. Marcus BH, Rossi JS, Selby VC, Niaura RS, Abrams DB (1992) The stages and
processes of exercise adoption and maintenance in a worksite sample. Health
Psychol 11: 386–395.
29. Prochaska JO, Velicer WF, DiClemente CC, Fava J (1988) Measuring processes
of change: applications to the cessation of smoking. J Consult Clin Psychol 56:
30. Borkovec TD, Nau SD (1972) Credibility of analogue therapy rationales. Journal
of Behavior Therapy 3: 257–260.
31. Radloff L (1977) The CES-D Scale: a self-report depression scale for research in
the general population. Applied Psychological Measurement 1: 385–401.
32. Weissman MM, Sholomskas D, Pottenger M, Prusoff BA, Locke BZ (1977)
Assessing depressive symptoms in five psychiatric populations: a validation study.
Am J Epidemiol 106: 203–214.
33. Burns DD, Shaw BF, Croker W (1987) Thinking styles and coping strategies of
depressed women: an empirical investigation. Behav Res Ther 25: 223–225.
34. Sheeran P, Abraham C (2003) Mediator of moderators: temporal stability of
intention and the intention-behavior relation. Pers Soc Psychol Bull 29: 205–
35. Abraham C, Sheeran P, Johnston M (1989) From health beliefs to self-
regulation: theoretical advances in the psychology of action control. Psychol
Health 13: 569–591.
36. Austin JT, Vancouver JB (1996) Goal constructs in psychology: structure,
process, and content. Psychol Bull 120: 338–375.
37. Conner M, Norman P (1996) Body weight and shape control: examining
component behaviours. Appetite 27: 135–150.
38. Maddux JE (1996) Expectancies and the social-cognitive perspective: basic
principles, processes, and variables. In: Kirsch I, Ed. How expectancies shape
behavior. Washington, DC: American Psychological Association. 17–40.
39. Sheeran P (2002) Intention-behavior relations: a conceptual and empirical
review. Eur Rev Soc Psychol 12: 1–36.
40. Burns DD, Nolen-Hoeksema S (1991) Coping styles, homework compliance, and
the effectiveness of cognitive-behavioral therapy. J Consult Clin Psychol 59:
41. Pail G, Huf W, Pjrek E, Winkler D, Willeit M, et al. (2011) Bright-light therapy
in the treatment of mood disorders. Neuropsychobiology 64: 152–162.
42. Murray G. Seasonality: the importance of longitudinal measurement. (2001)
In:Magnusson A, Partonen T, editors. Seasonal Affective Disorder: Practice and
Research. New York: Oxford University Press. 55–62.
43. Aloia MS, Smith K, Arnedt JT, Millman RP, Stanchina M, et al. (2007) Brief
behavioral therapies reduce early positive airway pressure discontinuation rates
in sleep apnea syndrome: preliminary findings. Behav Sleep Med 5: 89–104.
44. Miller WR, Rollnick S (2002) Motivational interviewing: preparing people for
change. 2nded. New York: Guilford Press.
45. Lundahl BW, Kunz C, Brownell C, Tollefson D, Burke BL (2010) A meta-
analysis of motivational interviewing: twenty-five years of empirical studies. Res
Soc Work Pract 20: 137–160.
Predictors of Light Therapy Use
PLoS ONE | www.plosone.org6June 2012 | Volume 7 | Issue 6 | e39275