[Skip to Navigation]
February 2017

Activation in Bipolar Disorders: A Systematic Review

Author Affiliations
  • 1Department of Academic Psychiatry, Institute of Neuroscience, Newcastle University, Newcastle, England
  • 2Department of Psychology, Swinburne University of Technology, Melbourne, Australia
  • 3Pole de Psychiatrie Universitaire, Hopital A. Chenevier, Paris, France
  • 4Department of Neuroscience, Norwegian University of Science and Technology, Trondheim, Norway
  • 5Department of Medicine, University of Notre Dame, Sydney, Australia
  • 6Department of Psychiatry, Psychotherapy, and Psychosomatic, Psychiatric Hospital, University of Zurich, Zurich, Switzerland
  • 7Genetic Epidemiology Research Branch, National Institute of Mental Health, Bethesda, Maryland
  • 8Brain and Mind Centre, The University of Sydney, Sydney, Australia
JAMA Psychiatry. 2017;74(2):189-196. doi:10.1001/jamapsychiatry.2016.3459
Key Points

Questions  Is activation a distinct dimension of bipolar disorder, and what are the dynamics of objectively measured activation?

Findings  Among 56 eligible studies, few of high quality, mean activity levels were low in patients with bipolar depression and euthymia compared with patients with other disorders or healthy controls. Activation in mania may be better characterized by its variability, predictability, or complexity rather than mean levels.

Meaning  Within the limitations of the data, available evidence suggests that activation is not an epiphenomenon of mood change and broadly supports the elevation of activation as a criterion A symptom for bipolar disorder.


Importance  Increased activity and energy alongside mood change are identified in the DSM-5 as cardinal symptoms of mania and hypomania. A wide range of existing research suggests that this revision may be valid, but systematic integration of the evidence has not been reported. The term activation is understood as emerging from underlying physiological change and having objective (observable motor activity) and related subjective (energy) levels.

Objectives  To systematically review studies of the clinical phenomenon of activation in bipolar disorder, to determine whether activation is statistically abnormal in bipolar disorder and demonstrably distinct from mood, and to identify any significant between- and within-individual differences in the dynamics of activation.

Evidence Review  This systematic review of MEDLINE, PsycINFO, EMBASE, CINAHL, and PubMed databases from January 1, 1970, until September 30, 2016, identified 56 of a possible 3284 citations for (1) data-driven analyses of the dimensions and factor structure of mania and bipolar depression and (2) longitudinal studies reporting real-time objective monitoring or momentary assessment of daytime activity in individuals with bipolar disorder compared with other clinical or healthy control samples. Hand search of reference lists, specialty journals, websites, published conference proceedings, and dissertation abstracts and contact with other researchers ensured inclusion of gray literature and additional analyses as well as raw data if appropriate. Quality assessment was perfomed using the National Institutes of Health quality assessment tool.

Findings  A total of 56 studies met eligibility criteria for inclusion in the review including 29 analyses of the factor structure of bipolar disorder, 3 of activity data from experimental sampling or ecological momentary assessment, and 20 actigraphy and 4 laboratory-based studies. Synthesizing findings across the studies revealed that the most robust finding was that mean levels of activity are lower during euthymia and depression in patients with bipolar disorder compared with healthy controls and other comparison groups (11 studies). The 7 ecological and laboratory studies show less organized or predictable patterns of behavior and a relative lack of habituation among patients with bipolar disorders compared with others. Factor analytic studies provide fairly consistent evidence that mood and activation represent distinct dimensions of bipolar disorder. Ten studies that examined interindividual and intraindividual patterns of activity suggest that mania may be better characterized by differences in robustness, variability, predictability, or complexity of activation rather than mean levels of activity.

Conclusions and Relevance  Within the limitations of the data, this synthesis of available evidence broadly supports the elevation of activation as a criterion A symptom for bipolar disorder in DSM-5. Although the importance of activation in bipolar disorders has been acknowledged for more than a century, this review suggests that this critical construct is understudied and should be the topic of more systematic high-quality research.

Add or change institution