Table 2.
Study characteristics and findings at follow-up
| Article | FU n | % lost at FU | FU duration (years) | Diagnostic Tool | Diagnostic assessor | Diagnosed with BD n (%) | Rates of BD type | Affective lability/BD association |
|---|---|---|---|---|---|---|---|---|
| Akiskal et al. 1995 | 559 | NR | 11 | SADS (DSM /RDC) | Clinician | 70 (13) | 22 BD-I, 48 BD-II |
BD-II X2 = 19.92 + + + (specificity = 86%, sensitivity = 42%) |
| Angst et al. 2003 | 591 | NR | 15 | DSM-IV criteria | Clinician | 86 (15) | 41 BD-II, 45 BSD |
BD-II + + BSD OR = 3.4, 95% CI [1.7, 6.6] + + |
| DeGeorge et al. 2014* | 112 | 23% | 3.1 (SD = 0.5, range 1.7—4.8) | SCID | Advanced grad (81%), psychologist + undergrad (19%) | 13 (14) |
BSD and BD BSD |
BSD OR = 2.99 + BD OR = .532, CI [.08–3.45] |
| Sperry et al. 2020* | 108 | 22% | 4 (4) |
PA: OR = 1.91, 95% CI [1.14, 3.18] + NA: OR = 1.54, 95% CI [1.00, 2.38] + |
||||
| Egeland et al. 2012 | 221 | NR | 16 | Adapted K-SADS, clinical records | Clinician | 9 (4) | All BD-I | Mood lability more present in BD at risk sample than controls (p = 0.063) |
| Gan et al. 2011 | 268 | 22% | 1 | SCID-I | Psychiatrist | 27 (24) | 2 BD-I, 25 BD-II | OR = 0.487 + |
| Hafeman et al. 2017 | 412 | 14% | Mean = 8.34 | K-SADS (DSM-IV) | Trained interviewers + psychiatrist review | 44/299 at-risk (15) | 15 BD-I/II, 29 BSD | X2 = 4.00 + |
| Kochman et al. 2005 | 80 | 27% | 2–4 (27 months, SD = 9 months) | K-SADS | Investigator | 35 (43) | All BSD | Prior instability in 64% of BSD; BSD vs non-BSD difference + + + |
| Ratheesh et al. 2015 | 52 | 26% | 1 | LIFE | Unspecified | 4 (8) | 3 BD-II, 1 BD-NOS | SES = 0.27 (p = 0.13), 95% CI [0.00,0.59] |
| Salvatore et al. 2013 | 107 | 79% | Mean = 4 | SCID | Blinded investigator | 20 (19) | 10 BD-I, 10 BD-NOS | RR = 1.45 + |
| Tohen et al. 2012 | 49 | 13% | 4 | SCID | Blinded experienced raters | 14 (33) | BD-I or BD-NOS | X2 = 4.85 + |
FU follow-up, n number of, BD bipolar disorder, BPSD bipolar spectrum disorder, DSM Diagnostic and Statistical Manual of Mental Disorders, NR not reported, SADS The Schedule for Affective Disorders and Schizophrenia, RDC the Research Diagnostic Criteria, BPI Bipolar Disorder Type 1, BPII bipolar disorder type 2, DSM-IV the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders, BSD bipolar spectrum disorders, OR Odds ratio, CI Confidence intervals, SD standard deviation, SCID Structured Clinical Interviews for DSM, PA positive affect, NA negative affect, K-SADS Kiddie Schedule for Affective Disorders and Schizophrenia, SCID-I Structured Clinical Interview for the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders for Axis I Disorders, LIFE The Longitudinal Interval Follow-up Evaluation for DSM IV, BD-NOS Bipolar disorder not otherwise specified, SES standardised effect size. RR Risk Ratio
*Sperry et al. 2020 and DeGeorge et al. 2014 are two papers from the same study. In both papers ‘% lost at FU’ and ‘FU duration’ are reported for the participant pool including some participants with initial BD diagnoses (we report averages from the total sample, which are expected to be comparable)
+ p < 0.05, + + p < 0.01, + + + p < 0.001