Outcomes One and Two Winters Following Cognitive-Behavioral Therapy or Light Therapy for Seasonal Affective Disorder

Abstract

Objective

The central public health challenge for winter seasonal affective disorder (SAD) is recurrence prevention. Preliminary studies suggest better long-term outcomes following cognitive-behavioral therapy (CBT-SAD) than light therapy. This study is a large randomized head-to-head comparison of these treatments on outcomes one and two winters after acute treatment.

Method

Community adults with Major Depression, Recurrent with Seasonal Pattern (N=177) were followed one and two winters after a randomized trial of 6-weeks of CBT-SAD (n=88) or light therapy (n=89). Prospective followup visits occurred in January or February of each year, and major depression status was assessed by phone in October and December of the first year. The primary outcome was winter depression recurrence status on the Structured Interview Guide for the Hamilton Rating Scale for Depression-SAD Version (SIGH-SAD). Other outcomes were depression severity on the SIGH-SAD and the Beck Depression Inventory-Second Edition (BDI-II), remission status based on severity cutpoints, and major depression status from tracking calls.

Results

The treatments did not differ on any outcome during the first year of followup. The second winter, CBT-SAD was associated with a smaller proportion of SIGH-SAD recurrences (27.3% vs. 45.6%), less severe symptoms on both measures, and a larger proportion of remissions defined as BDI-II≤8 (63.3% vs. 43.9%) than light therapy. Non-recurrence at next winter was more highly associated with non-recurrence the second winter among CBT-SAD (RR=5.12) than light therapy (RR=1.92) subjects.

Conclusions

CBT-SAD was superior to light therapy two winters following acute treatment, suggesting greater durability for CBT-SAD.

The central public health challenge in the management of winter seasonal affective disorder (SAD) (1) is prevention of depressive episode recurrence over subsequent winters. Light therapy, the most studied treatment, is highly efficacious for acute SAD (2). However, long-term compliance with clinical practice guidelines recommending daily light therapy during the symptomatic months each year is poor. Most patients fail to re-initiate light therapy in subsequent winters (3), leaving them vulnerable to recurrence without other treatment. Cognitive-behavioral therapy tailored for SAD (CBT-SAD) (4) is an emerging, time-limited, alternative treatment. Whereas light therapy targets a chronobiological vulnerability, CBT-SAD targets a psychological vulnerability, specifically maladaptive thoughts through cognitive restructuring and avoidance behaviors through behavioral activation, to alleviate current symptoms and prevent future recurrences. Attenuated risk for relapse and recurrence of nonseasonal major depression following cognitive therapy is well-documented (5). If CBT-SAD's effects endure after treatment to prevent recurrences, it may offer a more practical method of managing long-term SAD symptoms than re-initiating daily light therapy each year.

Pilot studies found that CBT-SAD and light therapy showed comparable improvements during treatment (6), but that CBT-SAD was associated with fewer recurrences and less severe symptoms at naturalistic followup the next winter (7). The first wave of our new, largest randomized trial found large and comparable improvements in CBT-SAD and light therapy over the acute treatment phase (8). At treatment endpoint, the treatments did not differ on patient- or rater-assessed depression severity or the proportion of patients in remission (47.6% in CBT-SAD vs. 47.2% in light therapy). This study focuses on the primary aim of that project: to compare the long-term efficacy of CBT-SAD vs. light therapy one and two winters following treatment. We hypothesized that CBT-SAD would be associated with a smaller proportion of depression recurrences, less severe symptoms, and a larger proportion of remissions than light therapy over followup.

Methods

Design Overview

This trial was conducted at the Mood and Seasonality Laboratory at the University of Vermont and was approved by the University's institutional review board. The enrolled sample of patients (N = 177) was randomized to 6 weeks of CBT-SAD or light therapy and prospectively tracked through two new winters following treatment endpoint. Prior publications detail our full protocol (9) and report baseline characteristics, treatment integrity, and acute treatment outcomes (8). Participants were aged 18 or older and met DSM-IV-TR criteria for Major Depression, Recurrent, with Seasonal Pattern on the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID) and a current SAD episode on the Structured Interview Guide for the Hamilton Rating Scale for Depression–Seasonal Affective Disorder Version (same criteria as for recurrence, below). Exclusion criteria were kept to a minimum to maximize external validity. Potential participants were screened out for current light therapy or psychotherapy for depression, prior light therapy or CBT for SAD, a comorbid Axis I disorder primary to SAD requiring immediate treatment, acute and serious suicidal intent, initiation of a new antidepressant medication in the past month or plans to change the dose of a current antidepressant, or positive laboratory findings for hypothyroidism at medical workup.

Power

The study was powered to detect clinically meaningful differences between treatments on SAD recurrences (primary outcome) following treatment of the index episode. With these sample sizes (88 CBT-SAD, 89 light therapy), there was 80% power to detect differences between treatments of .21 in recurrence proportions and ≥ 3.6 points in SIGH-SAD scores at followup.

Treatments

Light Therapy

We used the 23 × 15 ½ × 3 ¼ in. SunRay® (SunBox Company, Gaithersburg, MD), which emits 10,000-lux of cool-white fluorescent light through a UV filter, initiated at 30 minutes immediately upon awakening. Weekly clinical adjustments were made per a treatment algorithm, in consultation with our study psychiatrist and light therapy expert (TTP), to maximize treatment response and reduce side effects. Final doses of light therapy were reported in Rohan et al. (8). Participants were advised to continue with daily light therapy until their typical time of spontaneous remission, then to return the light boxes in May.

Cognitive-Behavioral Therapy for SAD (CBT-SAD)

CBT-SAD (4) uses psychoeducation, behavioral activation, and cognitive restructuring to specifically target winter depression. The format is 90-minute closed-group therapy sessions twice per week for 6 weeks (12 sessions). Each group was facilitated by the Principal Investigator (KJR) or one of two community Ph.D.-level psychologists. Session attendance descriptive statistics and analyses for therapist and group membership effects (all ns) were reported in Rohan et al. (8).

Standardized Instructions for Continued Study Treatment the Next Winter

The first week of September, letters were mailed to participants treated the winter before prompting resumption of study treatment. For light therapy-treated participants, the letter encouraged re-initiating daily light therapy upon onset of first depressive symptom and provided two options: borrowing a study light box for the duration of winter or purchasing a unit. The letter provided contact information for manufactures with a list of specifications to match our devices (i.e., full-sized units emitting 10,000-lux cool-white light through UV filter). For CBT-SAD-treated participants, the letter encouraged use of the skills learned in CBT-SAD on their own (without a therapist). Both letters stated that if the recommended strategy proved insufficient, participants should pursue formal treatment and provided contact information for local mental health centers and treatment providers. These letters were intended to promote fidelity with study treatment over followup, while addressing ethical concerns about proscribing additional treatment, if needed.

Outcome Measures

The 29-item Structured Interview Guide for the Hamilton Rating Scale for Depression—Seasonal Affective Disorder Version SIGH-SAD (10) includes the 21-item Hamilton Rating Scale for Depression (HAM-D) and the 8-item atypical symptoms subscale. The primary outcome was SIGH-SAD recurrence status as assessed at next and second winter followup, indicated by total SIGH-SAD ≥ 20, HAM-D ≥ 10, and atypical score ≥ 5. Other SIGH-SAD-derived outcomes at next and second winter followup included continuous depression scores (total as well as HAM-D and atypical) and remission status. Remission status was classified as either: ≥ 50% improvement in SIGH-SAD score from pretreatment to followup + followup HAM-D score ≤ 7 + followup atypical score ≤ 7 OR followup HAM-D score ≤ 2 + followup atypical score ≤ 10. A second blind rater rated audio-recordings of the SIGH-SADs. Intra-class correlations for inter-rater reliability were 0.965 at next winter and 0.967 at second winter.

The Beck Depression Inventory—Second Edition (BDI-II) (11), a 21-item self-report measure of depressive symptom severity, was also administered at next winter and second winter. BDI-II outcomes at followup included total scores and a cutoff score of ≤ 8 as a secondary marker of remission, consistent with our prior trials.

Followup Assessment Procedures

Phone Tracking of Recurrences and Retreatment

Participants were contacted twice by phone (in October and December) in the interim between treatment completion and the in-person next winter followup to track recurrences and new treatments initiated. These tracking procedures were implemented starting with the second enrolled cohort (n = 153; 76 CBT-SAD, 77 light therapy). These calls were conducted by a trained, blinded clinical psychology graduate student and involved: (2) assessing DSM-IV-TR criteria for a major depressive episode on the SCID since date of last contact (i.e., formal assessment or last tracking call), and (3) documenting any treatments initiated since last contact using scripted questions about light therapy, psychotherapy, and medications.

In-Person Next Winter and Second Winter Followup Visits

In-person visits were conducted in January or February of the next winter and the second winter. Consistent with the intent-to-treat principle, all randomized participants were invited to attend followups. A trained clinical psychology graduate student, blind to treatment assignment, administered the SIGH-SAD interview; the BDI-II; and a questionnaire assessing use of light therapy, psychotherapy, and medications since initial study treatment (at next winter) and since the next winter followup (at second winter).

Statistical Analyses

The primary analysis was an intent-to-treat (ITT) analysis based on multiple imputation (MI) of missing next winter SIGH-SAD scores, which were then used to classify depression recurrence status for individuals who dropped out during the treatment phase, withdrew from protocol, or were subsequently lost to followup. The fully conditional specification (fcs) regression method was used to obtain imputed values based on age, sex, baseline comorbid diagnosis status, and depression scores at other timepoints. Separate regressions were used to impute values for the CBT-SAD and light therapy conditions because of differing effects of the predictors in the two treatments. Imputed values included a random component reflecting the residual distribution for the dependent variable and ten data sets with differing imputed values were generated. The difference between the CBT-SAD and light therapy groups in the proportions of subjects with a recurrence and in remission the next winter was estimated for each of the ten imputed data sets and the estimates were combined using the inference methods for MI described by Little and Rubin (12). SAS PROC MI and PROC MIANALYZE were used to carry out the imputation and analysis. Sensitivity analysis was conducted to examine robustness of the results under alternative imputation methods, including using the best- and worst-case scenarios for each treatment, and using a logistic regression in the MI analysis to directly impute recurrence (rather than linear regression to impute SIGH-SAD scores). The BDI-II outcomes were analyzed in the same manner, as were outcomes for the second winter followup. Analyses based on available data, without imputation, were also performed using logistic and linear regression analyses for dichotomous and continuous outcomes, respectively. In addition to prospectively assessed SIGH-SAD recurrence status, analyses using available data were performed for fulfilling DSM-IV-TR major depression criteria for the interims assessed by the October and December tracking telephone calls. For dichotomous outcomes that differed by treatment at next and second winter followup, logistic regressions were performed to assess the effect of treatment group after adjustment for ongoing treatment(s) reported at that time point using data without imputation. We considered any treatment(s), in general, and any new treatment, psychotherapy, light therapy, and antidepressant medications, specifically. When coded for analysis, light therapy-treated patients reporting light therapy were counted as light therapy but not as any new treatment, and psychotherapy with a therapist among CBT-SAD patients was considered as both psychotherapy and any new treatment. For continuous outcomes that differed by treatment, linear regressions were used to assess the effect of treatment on depression scores after adjustment for any, new, and each ongoing treatment.

Results

Figure 1 displays the CONSORT flow diagram from the point of randomization through the fall tracking phone calls and the next and second winter followups. See Rohan et al. (8) for prior stages of participant flow, including screening. Missing data was minimal. At the in-person followups, 170/177 (96%) provided data the next winter and 169/177 (95%) provided data the second winter. Of those enrolled after the initial year, 144/153 (94%) and 132/153 (86%) completed the October December tracking calls, respectively.

Figure 1.

Note: The October and December tracking phone call procedures were in place for all but the first cohort of 24 participants that was recruited in the initial fall/winter of study (2006-2007).

Primary and Secondary Outcomes

Comparisons between the treatments on next and second winter outcomes are shown in Tables 1 and 2, separately for the multiple imputation analyses using the intent-to-treat sample and the secondary analyses using all available data. Table 1 displays the dichotomous outcomes of recurrence and remission, and Table 2 presents the continuous outcomes of SIGH-SAD and BDI-II scores. Figure 2 graphically displays the main findings. There were no statistically significant differences between CBT-SAD and light therapy on any of the outcomes at the next winter followup. There was also no significant difference between treatments in recurrence status based on the October and December tracking calls: 9/73 (12.3%) CBT-SAD participants and 16/76 (21.1%) light therapy participants with tracking call data met major depression criteria, p = .154.

Table 1. Depression Recurrence and Remission Status at the Next Winter and Second Winter Followups.

Multiple Imputation Analysis (N = 177)
	CBT-SAD		Light Therapy
	%	SE	%	SE	t Statistic	df	p value
	Next Winter Followup
SIGH-SAD recurrencea	28.9	4.7	24.9	5.0	0.57	61	.571
SIGH-SAD remissionb	37.0	5.2	34.2	5.1	0.40	65	.694
BDI-II remissionc	63.5	5.3	65.3	5.7	0.23	63	.819
	Second Winter Followup
SIGH-SAD recurrence	27.3	5.1	45.6	5.1	2.52	63	.014
SIGH-SAD remission	34.1	4.9	22.9	4.8	1.62	63	.111
BDI-II remission	68.3	5.5	44.5	5.4	3.06	62	.003
Analysis using All Available Data
	CBT-SAD		Light Therapy
	n/N	%	n/N	%	X2 Statistic	df	p value
	Next Winter Followup
SIGH-SAD recurrence	25/85	29.4	20/84	23.8	0.68	169	.410
SIGH-SAD remission	32/85	37.6	30/84	35.7	0.07	169	.794
BDI-II remission	52/85	61.2	51/85	60.0	0.02	170	.875
	Second Winter Followup
SIGH-SAD recurrence	23/82	28.0	40/86	46.5	6.11	168	.013
SIGH-SAD remission	28/82	34.1	20/86	23.3	2.44	168	.118
BDI-II remission	52/83	62.7	38/86	44.2	5.78	169	.016

Note. CBT-SAD = SAD-tailored cognitive-behavioral therapy. SIGH-SAD = Structured Interview Guide for the Hamilton Rating Scale for Depression-Seasonal Affective Disorder Version; BDI-II = Beck Depression Inventory—2^nd Edition; HAM-D = 21-item Hamilton Rating Scale for Depression; Atypical = 8-item, atypical subscale score of the SIGH-SAD. One light therapy participant completed the BDI-II, but not the SIGH-SAD at next winter followup. One CBT-SAD participant completed the BDI-II, but not the SIGH-SAD at second winter followup.

^atotal SIGH-SAD score ≥ 20 + HAM-D score ≥ 10 + Atypical score ≥ 5.

^b≥ 50% improvement in SIGH-SAD + HAM-D ≤ 7 + Atypical ≤ 7 OR HAM-D ≤ 2 + Atypical ≤10.

^cBDI-II ≤ 8.

Table 2. Continuous Outcomes at the Next Winter and Second Winter Followups.

Multiple Imputation Analysis (N = 177)
	CBT-SAD		Light Therapy
	Mean	SE	Mean	SE	t Statistic	df	p value
	Next Winter Followup
SIGH-SAD	15.0	0.9	15.5	0.9	0.33	64	.743
HAM-D	9.4	0.6	9.1	0.6	0.30	66	.764
Atypical	5.7	0.4	6.4	0.5	1.02	47	.313
BDI-II	8.2	0.8	7.8	0.8	0.28	65	.779
	Second Winter Followup
SIGH-SAD	15.0	1.0	18.7	0.9	2.77	65	.007
HAM-D	9.4	0.6	11.9	0.6	2.89	65	<.001
Atypical	5.6	0.5	6.9	0.5	1.86	65	.067
BDI-II	7.7	0.9	11.3	0.9	2.98	65	.004
Analysis using All Available Data
	CBT-SAD		Light Therapy
	Mean	SD	Mean	SD	t Statistic	df	p value
	Next Winter Followup
SIGH-SAD	15.0	9.1	15.1	8.0	0.08	167	.935
HAM-D	9.4	6.0	9.0	5.3	0.51	167	.609
Atypical	5.6	4.1	6.2	4.1	0.88	167	.381
BDI-II	8.2	7.7	7.9	7.0	0.31	168	.755
	Second Winter Followup
SIGH-SAD	15.1	8.6	18.7	9.3	2.65	166	.009
HAM-D	9.5	5.2	11.9	6.2	2.73	166	.007
Atypical	5.6	4.4	6.8	4.5	1.83	166	.069
BDI-II	7.8	7.0	11.1	8.7	2.73	167	.007

Note. See Table 1 for treatment and measure abbreviations.

Figure 2. Recurrence Rates and Depression Severity at Next and Second Winter Followup in Cognitive-Behavioral Therapy and Light Therapy.

Note: See Table 1 for measure abbreviations. These data were derived from the primary analysis, an intent-to-treat analysis based on multiple imputation of missing followup depression scores.

At the second winter followup, CBT-SAD was statistically superior to light therapy on 2 out of 3 dichotomous outcomes and on 3 out of 4 continuous outcomes, and this pattern was consistent across analyses using multiple imputation and all available data. On the primary outcome (SIGH-SAD recurrences), CBT-SAD was associated with fewer recurrences at the second winter than light therapy (27.3% vs. 45.6% using imputation; 28.0% vs. 46.5% without imputation). CBT-SAD was also associated with more remissions at second winter than light therapy using BDI-II criteria (68.3% vs. 44.5% using imputation; 62.7% vs. 44.2% without imputation), but not using SIGH-SAD criteria. Depression scores in CBT-SAD were significantly lower than in light therapy at second winter on both the SIGH-SAD and the BDI-II. Considering the component scales of the SIGH-SAD separately, HAM-D scores were lower in CBT-SAD than in light therapy at second winter, but the treatments did not differ on atypical subscale scores.

Given that the treatments did not differ at next winter, but differed on the majority of outcomes at second winter, we conducted exploratory analyses to probe this. Using SIGH-SAD recurrence status, we examined the association between recurrence status at next winter and the risk of recurrence at second winter within each treatment group. CBT-SAD subjects without recurrence at next winter were about five times more likely to not have recurrence the second winter relative to CBT-SAD subjects with recurrence at next winter (87.5% vs. 15.5%, RR = 5.12). In contrast, light therapy subjects without recurrence at next winter were only about twice as likely to not have recurrence the second winter relative to light therapy subjects with recurrence at next winter (63.5% vs. 30.0%, RR = 1.92). The ratio of RR (2.67) differed significantly from one (z = 2.38, p = .017), indicating that although next winter recurrence status was predictive of second winter recurrence in both treatments, the relationship was stronger in CBT-SAD than in light therapy. McNemar tests indicated that CBT-SAD subjects with recurrence in one, but not both, winters were just as likely to have their recurrence in either winter (p = .804), whereas light therapy subjects with recurrence in only one winter were significantly more likely to recur the second winter (p = .003). These findings suggest greater durability of a treatment effect in CBT-SAD than light therapy.

Treatment Utilization over Followup

Table 3 presents descriptive information for ongoing treatment utilization reported at each winter followup visit. All subjects who reported treatment on the October and/or December tracking phone calls also reported it at the next winter followup. Therefore, the more inclusive next winter followup reports are presented here. A larger proportion of light therapy than CBT-SAD subjects reported any treatment at next winter, but this difference was driven by 34.9% of light therapy subjects reporting continued light therapy as instructed (vs. only 7.1% of CBT-SAD subjects reporting light therapy). Consequently, the treatment groups did not differ in reports of any new treatment at next winter. At the second winter, the treatments did not differ in the proportions reporting any treatment or any new treatment, but more light therapy (30.6%) than CBT-SAD subjects (13.4%) reported using light therapy. CBT-SAD and light therapy subjects did not differ on psychotherapy or antidepressant medication use at either followup. Some cross-over was evident with generally more light therapy subjects pursuing psychotherapy than CBT-SAD subjects pursing light therapy.

Table 3. Number and Percentages of Participants with Followup Data Reporting Ongoing Treatment.

	All	CBT-SAD	Light Therapy	Statistic
Next Winter Followup
Light therapy	35/167 (21.0)	6/84 (7.1)	29/83 (34.9)	χ2 (1, N=167) =19.47, p<.001
Psychotherapy	39/167 (23.4)	22/84 (26.2)	17/83 (20.5)	χ2 (1, N=167)=.76, p=.38
Antidepressant Meds	43/166 (25.9)	20/83 (24.1)	23/83 (27.7)	χ2 (1, N=166)=.28, p =.59
Any Treatment	88/167 (52.7)	37/84 (44.0)	51/83 (61.4)	χ2 (1, N=167)=5.07, p =.02
Any New Treatment	73/167 (43.7)	37/84 (44.0)	36/83 (43.4)	χ2 (1, N=167)=0.01, p =.93
Second Winter Followup
Light therapy	37/167 (22.2)	11/82 (13.4)	26/85 (30.6)	χ2 (1, N=167)=7.14, p<.01
Psychotherapy	53/167 (31.7)	28/82 (34.1)	25/85 (29.4)	χ2 (1, N=167)=.43, p =.51
Antidepressant Meds	48/167 (28.7)	20/82 (24.4)	28/85 (32.9)	χ2 (1, N=167)=1.49, p=.22
Any Treatment	103/167 (61.7)	45/82 (54.9)	58/85 (68.2)	χ2 (1, N=167)=3.15, p =.08
Any New Treatment	88/167 (53.7)	45/82 (54.9)	43/85 (50.6)	χ2 (1, N=167)=0.31, p =.58

Note. See Table 1 for treatment abbreviations. Any new treatment excluded light therapy subjects who reported light therapy and CBT-SAD subjects who reported using CBT-SAD skills on their own without a therapist. Both of these scenarios represent fidelity to study treatment. Psychotherapy included only formal treatment with a provider. Therefore, for CBT-SAD subjects, any new treatment is equivalent to any treatment.

Logistic and linear regression analyses indicated that differences between CBT-SAD and light therapy at second winter were not attributable to concurrent treatment(s), in general, or to any new treatment or ongoing light therapy, psychotherapy, or antidepressant medication, specifically. After adjustment for concurrent treatment utilization, CBT-SAD continued to have fewer SIGH-SAD recurrences and lower SIGH-SAD, BDI-II, and HAM-D scores the second winter than light therapy. The difference between CBT-SAD and light therapy in BDI-II remissions at second winter persisted with these adjustments, except when adjusted for ongoing light therapy. (See Table 4).

Table 4. Differences between Treatment Conditions on Second Winter Outcomes, After Adjustment for Concurrent Treatments.

	Light Therapy		Psychotherapy		Medication		Any Treatment		Any New Treatment
Outcome	OR	p	OR	p	OR	p	OR	p	OR	p
SIGH-SAD Recurrence	0.49	0.032	0.44	0.013	0.46	0.018	0.46	0.017	0.44	0.013
BDI-II Remission	1.76	0.082	2.07	0.022	2.06	0.023	1.98	0.031	2.05	0.023
	Difference	p	Difference	p	Difference	p	Difference	p	Difference	p
SIGH-SAD Score	-3.2	0.026	-3.6	0.007	-3.6	0.013	-3.5	0.014	-3.8	0.008
HAM-D Score	-2.1	0.021	-2.5	0.005	-2.4	0.008	-2.3	0.011	-2.5	0.006
BDI-II Score	-2.8	0.025	-3.3	0.008	-3.2	0.010	-3.2	0.012	-3.3	0.008

Note. See Table 1 for treatment and measure abbreviations. Only outcomes with statistically significant differences between treatments in the primary analysis were examined after adjusting for treatment. OR = Odds ratio for the outcome in CBT-SAD compared to LT from logistic regression. Difference = the difference between the mean score in the CBT-SAD and LT conditions (i.e. CBT-SAD minus LT).

Discussion

Outcomes for SAD patients initially treated with CBT-SAD or light therapy were comparable at followup the next winter, but differed two winters after initial treatment. Relative to light therapy, CBT-SAD was associated with fewer depression recurrences using SIGH-SAD criteria, more remissions based on BDI-II cutpoint, less severe blind interviewer-rated depression on the SIGH-SAD, and less severe patient-rated depression on the BDI-II at the second winter. The superiority of CBT-SAD over light therapy the second winter persisted after adjustment for any concurrent treatment, in general, and for any new treatment, psychotherapy, or antidepressant medications, specifically. Differences between CBT-SAD and light therapy were somewhat attenuated after adjustment for ongoing light therapy use, which was not unexpected because it is partially confounded with treatment. Nevertheless, BDI-II remission status at second winter was the only treatment group difference that became non-significant after adjusting for ongoing light therapy. Only two of the 11 (18.2%) CBT-SAD subjects reporting light therapy the second winter met BDI-II remission criteria compared to 69.0% of those who reported no light therapy. Interestingly, BDI-II remission the second winter was also lower in light therapy subjects who reported ongoing light therapy (38.5%) than in those who did not (47.5%), but the difference was much smaller than among CBT-SAD subjects. Light therapy use may be a marker for non-remission at second winter, particularly following CBT-SAD.

These followup results are consistent with our prior study (7), except that treatment group differences emerged at the second winter here and were apparent at the next winter in our earlier study (7). In contrast to our prior work, the current protocol added the October and December phone calls to prospectively track depression recurrences and retreatment in the interim between treatment endpoint and the next winter. It is possible that these extra measures and timepoints created a testing effect in the light therapy group. Treatment differences emerged when subjects were “left to their own devices” between the next and second winter followups, as was the case at the single, naturalistic followup the next winter in our prior trial (7).

The following observations are consistent with the interpretation of a testing effect in the light therapy condition at next winter and also support the interpretation that CBT-SAD had an enduring effect that reduced risk of recurrence after acute treatment relative to light therapy. CBT-SAD subjects with a prospective SIGH-SAD recurrence at one followup were equally likely to recur in either winter. In contrast, light therapy subjects with SIGH-SAD recurrences at one followup were significantly more likely to recur at the second than the next winter. In addition, the predictive relation between next winter and second winter recurrence status was significantly stronger in CBT-SAD than in light therapy. Relative to subjects with recurrence at next winter, subjects without recurrence at next winter were 5-times more likely to go without recurrence the second winter in CBT-SAD (versus just 2-times more likely in light therapy).

Reinitiating light therapy each fall/winter season is widely regarded as the most effective means of preventing winter depression recurrence. To our knowledge, clinical trials that have prospectively followed SAD patients treated acutely with light therapy in subsequent winters are limited to our prior study (7) and the current study. Here, only about one-third of light therapy subjects reported continued light therapy at each followup, even though we explicitly prompted continued compliance and made our light boxes available to them the next winter. In our pilot study, only 2/19 (11%) light therapy subjects reported light use the next winter. These results suggest that providing CBT-SAD during acute treatment is a more effective means to reduce risk of recurrence than treating acute SAD with light therapy and instructing patients to resume light use each fall/winter, as practice guidelines recommend. Not resuming light therapy in future winters is problematic, given that light therapy is a palliative treatment that exerts its effects by suppressing symptoms as long as treatment is actively applied.

SAD research has generally focused on achieving acute remission without reference to longer-term outcomes, an approach that does not match the recurrent course of the disorder. These results underscore the importance of including longitudinal followup over subsequent winters in SAD clinical trial designs. There is currently no accepted benchmark for defining a SAD treatment failure, but we argue that recurrence is more clinically meaningful than post-treatment non-remission status. By that standard, our data suggest that slightly more than one-quarter of SAD patients initially treated with CBT-SAD and slightly less than one-half of SAD patients initially treated with light therapy would require a next-step treatment strategy within the next two years. Therefore, there is room for improvement beyond these two empirically-validated SAD treatments, suggesting several future directions. Possible next steps to test empirically include maintenance strategies (e.g., early fall booster sessions focused on using CBT-SAD skills or motivational interviewing around continuing light therapy) and “switch” (i.e., cross-over) or “combine” decision-rules. Although our prior study found that solo CBT-SAD had greater benefit the next winter than CBT-SAD combined with light therapy (7), there might be a subgroup of patients that would benefit from combination treatment. Narrowing the pool to patients who failed a first-line solo treatment should help identify those for whom the benefit of combined treatment outweighs the cost.

In conclusion, our prior report found that CBT-SAD and light therapy are comparably effective treatment modalities for acute SAD (8), but these followup data show better outcomes for CBT-SAD than light therapy two winters later. Accordingly, CBT-SAD should be considered as an efficacious SAD treatment and disseminated into practice, particularly if the focus is on recurrence prevention.