> Most studies agree that bipolar disorder is not evenly distributed across the sexes.

I don’t think that actually impacts the statistics here the way you think it would.

Suppose 2% of the population had X. If 1% of women had X, ~3% of men would have X and therefore ~4% of couples would.

Age of onset, differences in rates of relationships, etc are all worth considering. But my point was simply it’s significant not that I had calculated some exact number. With divorced people having the possibility of remarrying simple calculations aren’t enough.