Bipolar II disorder (BD-II) is associated with an excess risk of premature death from both natural and unnatural causes, according to a retrospective study from Taiwan.
Compared with matched population controls, patients with BD-II had a 62-percent higher risk of all-cause mortality (adjusted hazard ratio [aHR], 1.62, 95 percent CI, 1.47–1.78), 37-percent increased risk of mortality from natural causes (aHR, 1.37, 95 percent CI, 1.23–1.52), and more than a fourfold greater risk of mortality from unnatural causes (aHR, 4.46, 95 percent CI, 3.53–5.64). [JAMA Netw Open 2026;9:e265535]
Natural-cause deaths among BD-II patients included mental and behavioural disorders; circulatory, respiratory, digestive, and skin or subcutaneous diseases; and symptoms, signs, and abnormal clinical and laboratory findings not elsewhere classified. Unnatural-cause deaths were mostly unintentional injuries, suicide, and assault or homicide.
In within-family analyses comparing BD-II patients with their unaffected siblings, the risks of all-cause (aHR, 1.31, 95 percent CI, 1.00–1.72) and unnatural-cause (aHR, 2.05, 95 percent CI, 1.43–2.95) mortality remained elevated but not that of natural-cause mortality.
Meanwhile, in cross-subtype analyses, BD-II patients had higher risks of all-cause (aHR, 1.24, 95 percent CI, 1.01–1.53) and natural-cause (aHR, 1.45, 95 percent CI, 1.14–1.86) mortality compared with patients who had bipolar I disease (BD-I), with no difference in the risk of unnatural-cause mortality.
Findings were consistent across sex, age, and psychiatric comorbidities.
“Our study addresses a key evidence gap by focusing specifically on BD-II and quantifying its mortality burden,” said first study author Dr Chih-Wei Hsu from Kaohsiung Chang Gung Memorial Hospital, Kaohsiung, Taiwan, and colleagues.
“Clinically, BD-II is distinguished by more frequent and longer depressive episodes than BD-I, whereas hypomanic phases are relatively mild,” noted Hsu and colleagues.
They emphasized that BD-II “should not be regarded as simply a milder form of BD,” as patients do experience substantial depressive burden and functional impairment. This burden may explain why BD-II was associated with higher natural-cause mortality compared with BD-I, despite similar baseline medical comorbidity between the two subtypes.
“Longer depressive burden and more frequent delays in recognition and treatment of BD-II could reduce engagement with physical healthcare and allow greater accumulation of medical risk. The longstanding characterization of BD-II as milder may further contribute to lower clinical vigilance for physical health monitoring in these patients,” Hsu and colleagues pointed out.
The authors believed that if BD-II patients could receive prompt treatment, survival would improve. “[What’s needed is] strengthened surveillance, targeted prevention, and proactive comprehensive psychiatric care.”
For the study, Hsu and colleagues used data from Taiwan’s National Health Insurance Database. They identified 11,427 BD-II patients (mean age 39.6 years, 61.9 percent female) and 45,708 individuals without BD-II matched based on sex and birth date.
The within-family analyses included 5,063 BD-II patients and their 8,002 unaffected siblings, while the cross-subtype analyses involved all BD-II participants and 179,674 BD-I patients.
Over a mean follow-up of 7.3 years, 1,089 patients with BD-II and 1,879 controls died.