Columbia University Medical Center

Individualized Timing Key To Success Of Light Therapy For Winter Depression

New York, NY – Synchronizing light therapy with a person’s biological clock doubles its effectiveness as a treatment for winter depression, Columbia University researchers have demonstrated.
In a study published in the Jan. 15 issue of the Archives of General Psychiatry, the team of Drs. Michael and Jiuan Su Terman measured the plasma melatonin pattern in 42 patients with seasonal affective disorder (SAD) before and after they received bright light therapy. Treatment was administered for 30 minutes a day using a specialized high-intensity light box made for this purpose. The precise treatment time for each patient, in the morning or evening, was measured relative to each individual’s evening onset of melatonin production by the pineal gland, which lies deep within the brain.
Previously, researchers had hypothesized that people with SAD suffered from a delay in their circadian rhythms during the winter. But the Termans found that the melatonin cycle of patients can occur early, late, or in between.
“You cannot conclude that these patients as a group suffer from a circadian phase delay that causes their depression,” says Michael Terman, professor of clinical psychology (in psychiatry) at Columbia’s College of Physicians & Surgeons. Dr. Terman directs the Winter Depression Program, established in 1984, at the New York State Psychiatric Institute (NYSPI). Instead, he explained, “People are falling asleep too soon after their melatonin onset, and their biological clock needs to be shifted earlier to make the correction.”
While morning light treatment pushes internal clocks forward, evening treatment pushed them back. Patients whose clocks were pushed forward the farthest in the study experienced the strongest response to therapy. The researchers found that treatment timed 8.5 hours after melatonin onset was by far the most effective at pushing the clock forward and relieving depression.
“Melatonin onset varies by up to four hours between individuals and serves as an anchor point to specify the optimum time of light administration,” Dr. Terman says. “If treatment is appropriately early in circadian time, it is twice as effective as later in the morning or in the evening. The contrast in remission rates is dramatic — approximately 80 percent vs. 35 percent — and the lower rate can be suspected to be nothing more than a placebo effect.”
Currently, melatonin onset is measured by testing saliva samples taken throughout the evening. Although this assay is still only available as a research tool, doctors and patients can estimate the optimum time for light therapy by closely examining the depressed patient’s sleeping pattern. The Termans found that regardless of the amount of sleep, its midpoint falls about six hours after melatonin onset.
“We can rewrite the prescription for timing light therapy in relation to habitual sleep: The best response occurs when light is taken promptly upon waking 2.5 to 3.5 hours after the midpoint of the depressed patient’s sleep cycle,” Dr. Terman says. “Short sleepers will use the lights around the time of normal waking, while longer sleepers will need to wake up earlier.”
Dr. Terman notes that the implications of these findings go beyond SAD treatment. The effectiveness of many drugs — for example, antihypertensives, antiasthmatics, and chemotherapy agents — vary with the time of day. “The importance for medicine as a whole is that you have to anchor your manipulations to the individual patient’s circadian phase, not to a standard external clock time,” he says. This “inside out” approach stands in stark contrast to dosing schedules that neglect the patient’s oscillating state.
The research by the Termans is supported by the National Institute of Mental Health. Information about clinical trials is available at http://www.light-and-ion-therapy.org.

Source: Terman JS, Terman M, Lo ES, Cooper TB. Circadian time of morning light administration and therapeutic response in winter depression. Arch Gen Psychiatry 2001;58:69-75.

# # #

Also on CUMC News: