Light, Ion, and Fluoxetine Efficacy (LIFE) in Depression

Rationale Many effective treatments exist for depression, including psychotherapies and antidepressant medications. However, antidepressants are not always effective, and they can produce significant side effects and pose a risk of overdose. Medication can also be expensive, thereby limiting accessibility. Furthermore, many people with MDD prefer to use non-pharmacological treatments. Some studies have found that combination antidepressant and psychotherapy is more effective than either monotherapy in people with chronic forms of MDD. However, evidence-based psychotherapies such as cognitive-behavioural therapy are not widely available within the Canadian health care system. For these reasons, there is great interest in finding alternative treatments to antidepressants and psychotherapy. Bright light therapy, which is well established as an effective treatment for seasonal affective disorder (SAD), has many attributes that make it highly attractive to use for nonseasonal MDD: it is a safe, well-tolerated, inexpensive, easy to use, non-pharmacological treatment that can be used as monotherapy or combined with medications without the worry of drug-drug interactions. The "gold standard" method for applying light therapy is via a 10,000 lux white fluorescent light box for 30 minutes a day, usually in the early morning upon arising from bed. The mechanism of action of light therapy is still unknown, but major hypotheses involve resynchronizing circadian rhythms and/or restoring neurotransmitter dysfunction. Bright light has predictable phase-shifting effects on circadian rhythms in humans, but studies of light therapy have not consistently demonstrated correlations of phase shift with antidepressant response. Although most of these studies have been done in patients with SAD, there is considerable evidence that nonseasonal MDD is also associated with disturbances in circadian rhythms. Other studies have shown that rapidly depleting serotonin and catecholamines can reverse the antidepressant effects of light therapy in SAD, thereby demonstrating monoaminergic effects of bright light similar to those seen with antidepressants. These chronobiologic and monoaminergic effects of bright light provide a rationale for the use of light therapy in nonseasonal MDD. As well, initial studies have shown that negative ion generators may be an effective treatment for neurovegetative symptoms of SAD, specifically oversleeping, overeating, and fatigability. However, relatively few studies have been conducted of light and ion therapy in nonseasonal MDD. Several systematic reviews of light and ion therapy for nonseasonal MDD have shown some support for efficacy, but these are based on a limited number of small-sample, poorly controlled RCTs. Research Method This study is an 8-week, multi-centre, double-blind (subject and rater), randomized, parallel-design trial to assess the efficacy of light and ion therapy when combined with fluoxetine in the treatment of subjects with MDD, compared to treatment with fluoxetine alone. A total of 216 depressed patients meeting entry criteria will be enrolled over a three-year period. Eligible patients will be randomized to one of four treatment conditions for the entire treatment period (8 weeks): 1. light treatment using a fluorescent light box (30 minutes daily) plus a placebo pill every day; or 2. negative ion generator (30 minutes daily) plus 20 mg of fluoxetine per day; or 3. negative ion generator (30 minutes daily) plus placebo pill every day; or 4. light treatment using a fluorescent light box (30 minutes daily) plus 20 mg of fluoxetine per day. Half of all devices (light boxes and negative ions) will be deactivated to test placebo conditions. Outcomes (HAM-D) will be primarily assessed over the telephone by raters blind to treatment assignment. Other outcome measures will be assessed by patient-rated questionnaires administered over the Internet using a secure web site, and by ratings from the treating physician (CGI and Health Economics Assessment). Statistical Analysis All randomized subjects who have at least one follow-up visit will be included in the analysis based on intent-to-treat. Ineligible subjects who are inappropriately randomized will be excluded from the analysis. Missing data will be imputed using last observation carried forward (LOCF). For the analyses the treatment variables will remain coded and the analysts and investigators will remain blinded to variable identity during analysis and interpretation. The pre-specified primary efficacy endpoint is the adjusted mean change from baseline to endpoint (8 weeks) in the HAM-D score using LOCF. All comparisons will be analyzed using ANCOVA adjusting for baseline value and centre. The secondary outcomes will also be analyzed using a similar analysis, when appropriate. Post-hoc analyses will also examine observed case data. Categorical data (such as proportions of the sample with adverse events) will be analyzed using chi-square tests or Fisher's test where cell sizes warrant.