Background: The "spectrum" model has advantages for the conceptualization of mental disorders, representing a complementary approach to the currently available categorical systems. We carried out a study in order to assess lifetime mood symptoms among patients with bipolar disorder (BD) and healthy controls from a dimensional perspective. Methods: The Mood Spectrum Self-Report instrument (MOODS-SR) was administered to 101 bipolar patients (52 BD I, 32 BD II, and 17 BD NOS, 36 males/65 females, mean age + SD=36.10 +/- 13.34 years) and 38 healthy controls (16 males/22females, mean age + SD=35.18 +/- 13.70 years). The scores of the different MOOD-SR scales and subscales among patients and controls were compared using non-parametric tests (Mann-Whitney and Kruskal-Wallis). Results: Bipolar patients scored significantly higher than healthy controls on the total MOOD-SR scores (BD: mean +/- SD=98.65 +/- 22.17; HC: mean +/- SD=12.92 +/- 10.72; p < 0.01) and all subdomains. Multiple comparisons revealed lower scores among controls when compared to each one of the subtypes of BD, also regarding the total scores and all subdomains (p < 0.01). Comparisons across the different subtypes of BD revealed statistically significant higher scores among BD I patients when compared to BD II and BD NOS patients, only in regard to the total MOOD-SR scores (BD I: mean +/- SD=102.94 +/- 22.79; BD II: mean +/- SD=93.53 +/- 21.97; BD NOS: mean +/- SD= 94.88 +/- 18.68; p=0.03) and two subdomains: mood mania and energy mania. Conclusions: These results, although preliminary, suggest that even though the MOODS-SR seems effective in distinguishing BD patients from HC, it is not as good in discriminating different subtypes of BD, especially in respect to lifetime depressive symptoms. Limitations: Our sample size was small, and comprised by outpatients. The MOOD-SR measures only lifetime symptoms and does not take into account the progression of mood symptoms or the current mood state of patients. (C) 2014 Elsevier B.V. All rights reserved.