[Screening for bipolar disorders: a review of the literature]. / Dépistage des troubles bipolaires: une revue de la littérature.

OBJECTIVE: The aim of this article is to review the major instruments proposed for screening for bipolar disorder among clinical or general, adult or paediatric populations. They were developed in order to improve the detection of this illness which, far too often, remains unrecognized. Several of these screening instruments are already translated into several languages and validated. METHODOLOGY: A systematic review of the literature published on this topic up to July 2007 was carried out, using the main electronic data base (Medline). The keywords employed included bipolar disorder, screening, questionnaire, diagnosis and early recognition. RESULTS: The studies reported here examine whether screening instruments perform similarly in various clinical and non-clinical samples. Different forms of the same questionnaire (like self-report or parent report used in paediatric samples) are sometimes compared, usually showing that parent reports supersede the adolescent self-report form. This is namely the case for the Mood Disorder Questionnaire (MDQ) which is a brief and widely tested tool, available both in adult and adolescent versions. The MDQ exhibits good psychometric properties in relation to sensitivity and specificity in adult psychiatric samples, but these are more limited in the general population. Moreover, it yields better sensitivity for BP type I than for other bipolar subtypes. This is also true for other screening instruments like the hypomania check list (HCL-32). In order to optimize the sensitivity for bipolar II disorders, proposals for changing the MDQ screening algorithm have been tested. DISCUSSION: Even though it does not replace a thorough clinical interview, the use of screening tools for bipolar disorder is widely advocated. We discuss the need for clinicians to rely upon instruments allowing for a rapid and economically feasible identification of this disorder. Involving family members in the evaluation process may also increase the rate of recognition. More studies are still required in order to improve diagnostic efficiency of the screening instruments.

[Validation of the QFS measuring the frequency and satisfaction in social behaviours in psychiatric adult population]. / Validation du Questionnaire de fonctionnement social (QFS), un autoquestionnaire mesurant la fréquence et la satisfaction des comportements sociaux d'une population adulte psychiatrique.

INTRODUCTION: Although everyone working in routine mental health services recognizes the scientific and ethical importance to ensure that treatments being provided are of highest quality, there is a clear lack of consensus regarding what outcome domains to include, what measure of assessment to use and, moreover, who to question when assessing. LITERATURE FINDINGS: Since the fifties, social functioning is considered as an important dimension to take into account for treatment planning and outcome measuring. But for many years, symptoms scales have been considered as sufficient outcome measures and social functioning improvement expected on the basis of symptoms alleviation. As symptoms and social adjustment sometimes appear relatively independent, no accurate conclusion concerning the patient's social functioning can so be driven on the basis of his clinical symptoms. More attention has then been directed toward the development of instruments specifically intended to measure the extent and nature of social functioning impairments observed in most psychiatric syndromes. Many of these instruments are designed to be completed by caregivers or remain time consuming and difficult to use routinely. Presently, in clinical practice, there is a need to rely on simple and brief instruments considering patients'perspective about their social adjustment as a function of time. AIM OF THE STUDY: The aim of this study is to present a new instrument, the QFS, initially developed in order to assess social functioning in patients involved in group psychotherapy programs conducted in a specialist mental health setting, as well as its psychometric characteristics. METHODOLOGY: It was designed to be completed in less than 10 minutes and the questions are phrased in a simple and redundant way, in order to limit problems inherent to illiteracy or language comprehension. The QFS is a 16 items self-report instrument that assesses both the frequency of (8 items) and the satisfaction with (8 items) various social behaviours adopted during the 2 weeks period preceding the assessment. It yields three separate indexes of social functioning, defined a priori and labelled "frequency", "satisfaction" and "global". The higher the scores, the better the social functioning. The QFS was administered to 457 subjects, aged between 18 and 65, including 176 outpatients (99 with anxious or depressive disorders, 25 with personality disorders and 52 with psychotic disorders) and 281 healthy control subjects. RESULTS: No significant difference was found between patients and controls according to age or gender distribution. Acceptance rate was high (>95%). Moreover, the QFS was generally acceptable to the clinicians who used it. Internal consistency calculated for each index ranged from 0.65 to 0.83 (Cronbach alpha). Test-retest reliability, calculated within a 15 days time interval on a sample of 49 healthy controls, ranged from 0.69 to 0.71 (intraclass correlation coefficient). Discriminant validity was calculated on healthy controls and patients divided into sub-groups according to their diagnosis. It showed to be excellent, with significantly higher scores in control subjects than in psychiatric patients and significant differences across diagnostic categories (Kruskal-Wallis ANOVA with post-hoc tests, all p<0.05). The convergent validity of the QFS with other measures of social functioning was calculated, using the Social Adaptation Self-Evaluation Scale (SASS) and the Social Adjustment Scale Self-Report (SAS-SR). With the SASS, the convergent validity was higher among patients (Spearman rS 0.71 to 0.92, p<0.01) than controls (rS from 0.49 to 0.66, p<0.001). In healthy controls, correlation with the SAS-SR was moderate but statistically significant (rS from - 0.21 to - 0.44, p<0.05). When comparing QFS scores with self-rated symptoms severity, lower levels of social functioning were significantly associated with more severe symptoms according to the Brief Symptom Inventory (BSI: rS from - 0.38 to - 0.65, p<0.001). The QFS indexes demonstrated sensitivity to change (Wilcoxon: all p<0.05) on a sample of 27 out-patients suffering from anxious-depressive disorders questioned before and after 4 months of cognitive behavioural group therapy running on a weekly basis during 16 sessions of 2 hours each.The factorial validity of the QFS was measured through 3 separate factor analysis conducted using the data of 457 subjects. The first analysis considered only Frequency items; 7 out of 8 items had loadings above 0.5 on Factor 1 accounting for 30.7% (unrotaded) of the variance. The second analysis considered only Satisfaction items; all items had loadings above 0.6 on Factor 1 explaining 43.4% (unrotaded) of the variance. And finally, in the third factor analysis, all QFS items were included; 15 out of 16 items had loadings above 0.4 on Factor 1 accounting for 30% (unrotated) of the variance. Concerning the factorial validity of the instrument, these results suggest that all QFS items belong to the same underlying dimension. DISCUSSION: Finally, provisional norms for the QFS are provided for healthy controls, in order to characterise individual patients or patient subgroups. In conclusion, the need for assessment in clinical routine, in order to estimate different aspects of patients conditions as well as the quality of the treatment provided, has contributed to the development of a large variety of instruments measuring several domains. Concerning the level of social functioning, many instruments fail to meet chief criterion of feasibility, remaining often too complex or time onsuming. Moreover, only few of them are available in French. CONCLUSION: The QFS presented here is a brief, simple and easy to administer self-rating scale that displays satisfactory psychometric properties. It seems to be a valuable instrument for the monitoring of social functioning in psychiatric patients which, from a therapeutic point of view, may have a clear impact as it sets up expectation of change and allows both to reality test patients and therapists beliefs about the presence of progress or not and to identify if therapy is working on this specific outcome domain. Though, to date, the administration of the QFS to other populations and treatment modalities requires further investigation.