Excess mortality in bipolar disorder and risk of major physical complications in patients on lithium and other mood-stabilizing medications

Abstract

Bipolar disorder is a severe mental illness that is associated with increased risk of mortality and physical morbidities, compared to the general population. Earlier literatures suggested that lithium, the first-line mood-stabilizing medication treatment for bipolar disorder, was associated with elevated risk of hypothyroidism, hyperthyroidism and chronic kidney diseases (CKD), than non-lithium counterparts. However, findings of previous research regarding excess mortality in bipolar disorder and major physical complications in relation to mood-stabilizers were hampered by important methodological limitations. Despite a substantial rise on use of non-lithium mood stabilizers, risk of mortality and physical complications related to these mood-stabilizing alternatives is rarely investigated. This work aimed to study the magnitude of excess mortality and reduced lifespan associated with bipolar disorder, as well as the risk of premature death and major physical complications in relation to lithium and other common mood stabilizers. First, a systematic review and meta-analysis of 11 publications about life expectancy of bipolar disorder was conducted. It is observed that bipolar disorder patients had a life expectancy of 67 years and lost 13 more years of life than the general population. Second, utilizing electronic medical records retrieved from Hospital Authority, our study showed that bipolar disorder was associated with a 2- to 3-fold higher mortality rate than the general population. Bipolar disorder was related to 7 years of reduced lifespan when the longevity gap was calculated using life-years lost (LYL), a measure taking into consideration different ages of onset of bipolar disorder. The overall mortality gap remained similar in the recent decade. Third, rate of physical morbidities prior to diagnosis of bipolar disorder as well as the association of physical (multi)morbidity with all-cause mortality were examined. It was noted that gastrointestinal diseases represented the most frequently-diagnosed pre-existing physical morbidity. An increase in physical multimorbidity level was related to increasing mortality rate. Fourth, the risk of premature mortality in bipolar disorder patients treated with other mood-stabilizing medications relative to lithium was investigated. Our data revealed that olanzapine and risperidone were associated with higher rate of mortality than lithium. Fifth, analysis of laboratory data with comprehensive adjustment for covariates identified a priori showed that lithium was related to an increased rate of hypothyroidism and CKD, but not hyperthyroidism or more severe stages of CKD (i.e., ≥ stages 4, 5 and end-stage renal diseases), compared to the non-lithium group. Lithium was associated with a higher rate of hypothyroidism, hyperthyroidism and CKD than valproate, olanzapine, quetiapine and risperidone. Finally, a risk prediction model for CKD was developed to provide prognostic information of lithium-treated individuals and hence patients who are at high risk of developing CKD would be identified. The risk prediction model was developed in a derivation dataset and validated in a separate validation dataset. Important clinical implications of the study findings and future research directions for physical health disparity in bipolar disorder were also discussed in the thesis. Overall, this thesis indicates that people with bipolar disorder are experiencing physical health disparity, which represent a serious public health challenge that warrants urgent attention.