An unpublished trial in night owls with bipolar depression
STUDY: Chan JWY et al, unpublished paper presented at ISBD conference, Vancouver Canada 2026.
STUDY TYPE: Randomized sham-controlled trial
FUNDING: Unknown
Background
Light therapy treated bipolar depression in a handful of randomized trials. In theory, it may work better in night owls (phase delay), and this is the first trial to test it in bipolar depression with that sleep type.
The Study
- 66 adults in Hong Kong with bipolar depression and sleep phase delay (night owls). The did not have seaonality (which is rare in Hong Kong).
- Randomized to lightbox 10,000 lux for 30 minutes daily at midday (12:00 to 2:00 pm range) for 6 weeks. The placebo was a sham (low level negative air ion generator).
Results
- Greater reduction in depressive symptoms with light therapy, with a large effect size (1.0) on the SIGH-ADS score.
- Although light therapy improves sleep in other studies, insomnia (ISI) scores did not differ between the groups here.

Side Effects
No difference in hypomania rates (one on placebo, one with lightbox).
Limitations
Small size, possible unblinding, presented at conference (not peer reviewed).
Practice Implications
- Light therapy is effective for depression in winter and summer. Learn how to use it and find reliable devices. Other trials used 60 minutes in bipolar depression, either in morning (5:00 to 8:00 am) or midday. This one used 30 minutes at midday.
- In bipolar, there is a theoretical risk of mania, but that has not shown up in the trials. It may be minimized by delivering at midday and titrating slowly (by 15 minutes every week). This trial did not titrate. I would avoid delivering light after 4:00 pm, which can worsen mood and insomnia.
—Chris Aiken, MD
Director, Psych Partners
Editor in Chief, Carlat Psychiatry Report







