Depression is a disorder of high socio-economic impact, due to its high prevalence and substantial disability, with US estimates ranking it among most expensive diseases in terms of total associated costs (Hall and Wise, 1995). In both the USA and Europe, indirect costs largely overcome direct costs (Kind and Sorensen, 1993; Hall and Wise, 1995; Ansseau, 1998). This is consistent with the fact that the missed diagnosis of depression has been shown to generate costs higher than those required to improve its detection and treatment (Rupp, 1995). The issue of increasing the recognition and diagnosis of depression in primary care and improving treatment standards has received strong attention in recent years (Perez-Stable et al., 1990; Paykel et al., 1992; Depression Guideline Panel, US Department of Health and Human Services, 1993; Katon et al., 1995; Lecrubier et al., 1996). Educational campaigns have been launched to increase public awareness about the illness and its management (e.g. the Depression: Awareness, Recognition and Treatment in the USA; and the Defeat Depression Campaign in the UK).
To address the problem of unrecognized depression, a few epidemiological surveys have been carried out in the community, with application of structured interviews for DSM criteria by lay interviewers in within-nation samples (Bland et al., 1988; Weissmann et al., 1988; Hwu et al., 1989; Blazer et al., 1994; Kessler et al., 1994), or cross-national samples (Weissmann et al., 1996). Recently, the results obtained in the Depression Research in European Society (DEPRES) study, the first large cross-national European survey, with representative national samples from Belgium, France, Germany, Netherlands, Spain and the UK, have been published (Lépine et al., 1997). In this survey, the Mini-International Neuropsychiatric Interview (Sheehan et al., 1994), which has been validated in a clinicianrated (Lecrubier et al., 1997) as well as in a patient-rated version (Sheehan et al., 1997), was used as a screening instrument and administered, following the positive results of a feasibility study, by lay interviewers as a part of an omnibus survey. The present study was carried out to evaluate the prevalence of depression in Italy as assessed by means of the modified-Mini-International-Neuropsychiatric Interview (MINI) (Lépine et al., 1997), administered to a representative national sample included in a telematic panel. This interview modality was selected in view of its cost-effectiveness, and on the basis of available evidence from a comparative evaluation, supporting its substantial equivalence to the personal interview modality in demoscopic surveys (Sani, 1998).
The modified-MINI (Lépine et al., 1997) was translated in local language by two of the authors, A.D. and P.P., on the basis of the Italian translation of the DSM-IV (Andreoli et al., 1996). The questionnaire was submitted to the telematic panel as the last module of a market research with consumers in an omnibus survey. The panel is a representative sample of the Italian population, including 3550 panel members from families located in 329 municipalities, with family head aged up to 60 years. The sampling method for panel selection is stratified on the basis of family size, residence town size and region. Families are provided with personal computers, connected with a central unit via modem, with user friendly applications allowing interactive questionnaire administration. On each Friday, questionnaires are sent via modem to each family, with the request of activating specific panel members, on the basis of pre-defined selection criteria. There is no instrumental control of the identity of the answering person, e.g. via a web camera. However, at regular intervals, ad hoc questions are included in the questionnaires, to check consistency between answers concerning behaviour, and verify responder accuracy, on the basis of which non-reliable people are excluded from the panel. During the weekend, answers are sent back to the central unit, to be checked and processed on the following Tuesday. Non responders or partial responders are contacted by telephone.
The sample socio-structural composition is shown in Table 1. The stratified random sampling method achieved a distribution matching for each parameter the distribution of the Italian population (1991 ISTAT nationwide census). The study was carried out by ISPO (Istituto per gli Studi sulla Pubblica Opinione), with fieldwork by CRA-Nielsen, on 12-14 June 1998.
Classification of subjects in diagnostic categories was performed according to the algorithm published by Lépine et al. (1997). Depression was identified by positive response to question 1 and/or 2, and with at least one positive response to question 3. In case of substantial perceived disability in association with depressive symptoms (modified-MINI questionnaire, question 4), subjects were classified in the depression category, major depression including subjects with five or more positive response to questions 1-3, and minor depression including subjects with two to four positive response to questions 1-3. In the absence of substantial perceived disability, subjects were classified in the depressive symptoms category, irrespective of the number of positive responses to the first three questions.
The survey was carried out for descriptive rather than inferential purpose, similar to the original study (Lépine et al., 1997), and no statistical hypotheses were set. Inferential statistics were not considered of interest in view of the exploratory nature of this first Italian general population survey and because of the obvious impact of the size of the sample on statistical findings. Frequency distributions by stratification factor and socio-demographic characteristics were calculated. In the indicated sample size range, it is worth noting that the standard error decreases from ± 5.0% (range for extreme-middle prevalence values 4.4-10%) for a subsample size of 100 observations to ± 1.6% (range for extreme-middle prevalence values 1.4-3.2%) for a subsample size of 1000 observations and to ± 0.9% for a sample size of 3000 observations.
All the 3550 individuals involved in the survey provided complete responses to the questionnaire. Among the 3550 responding individuals, 1350 were positive for depression, corresponding to a 6-month prevalence of depression in the community of 38%. The majority of cases (27.1%) were classified in the depressive symptoms category, since symptoms were judged not to have caused substantial disability. Major depression and minor depression accounted for 8.0% and 2.9% of the cases, respectively. In 10.9% of the cases, therefore, depression was judged to have substantially interfered with work or social functioning. Among the 3550 responding individuals, 81.2% answered by Sunday evening, while 18.8% answered by the following Tuesday, after being contacted by telephone. Socio-demographic characteristics and depression rates of these two subject sub-sets were not different.
The 6-month prevalence of each depressive category by gender and age is given in Table 2.
The most evident between-gender differences were seen for major depression, which was more prevalent in women, and minor depression, which was more prevalent in men. Depressive symptoms and the total depression cases were only slightly more prevalent in women. Prevalence of major depression increased from 7.8% in the lowest age range to a maximum of 11.9% in the age ranges from 30-39 years and from 40-49 years, and decreased, subsequently, up to a minimum of 4.1% in subjects aged 60 years or older. A similar pattern was seen for other categories, with minimum frequencies in the lowest and highest age ranges.
The 6-month prevalence of each depressive category by education level and occupation is given in Table 3. Prevalence of both major and minor depression is similar across education levels, with the exception of the decreased values at the lowest education level, and similar across occupations, with the exception of the decreased values among unemployed and, particularly, retired people, which is consistent with and confounded by the above commented age-related distribution.
The 6-month prevalence of depressive categories is similar in the four Italian geographical areas, with no relationship with the size of residence towns.
The 6-month prevalence of depression in the Italian community was evaluated by means of the modified-MINI (Lépine et al., 1997) administered to a panel of 3550 individuals representative of the Italian population through a telematic computer-prompted interview. All involved individuals provided complete responses to the questionnaire. This full response rate was not unexpected since the telematic panel interview modality is generally characterized by a very high response rate: the omnibus surveys carried out by ISPO and CRA-Nielsen in the 6-month period preceding the present survey had a 96% average response rate (excluding computer or modem failure, which was always lower than 5%). In fact, with respect to personal interviews, this interview modality allows maximization of response rate with the advantage of requiring limited resources. In addition, it can be expected to provide more truthful answers with respect to delicate matters, in that there is no direct relationship between the interviewer and the interviewee. Its weaknesses are expected to be related to the possible sample selection bias resulting from the acceptance of the computer-prompted interview procedure per se and to the absence of a direct control of the identity of the answering person, although measures were taken to check the reliability of the panel individuals. In any case, the telematic data acquisition modality has been shown to provide similar results to other interview modalities in the frame of omnibus surveys (Sani, 1998).
This was also the case for our survey with respect to the prevalence of the modified-MINI major depression (8.0%) and minor depression categories (2.9%), which were well within the range reported by Lépine et al. (1997) for other European countries (3.8-9.9% for major depression, 1.5-3.0% for minor depression). On the other hand, the prevalence of the modified-MINI depressive symptoms category (27.1%) was higher in our survey than that reported by Lépine et al. (1997) for other European countries (5.6-11.6%). This finding is possibly connected to the different interview modality. This was carried out door-to-door by lay interviewers in the study of Lépine et al. (1997) but was telematic in our study, with the impersonal computer-prompted interview favouring positive answers on isolated symptoms, while not impacting on more robust classification scales, such as those behind the modified MINI major and minor depression categories. It is worth noting, however, that a given scale may be unidimensional (i.e. uniquely measuring the dimension for which it was constructed) for one group of individuals, but not for another (Miller, 1991). In our case, the MINI could have been unidimensional for our target, i.e. major and minor depression.
Our results on the gender-related distribution of the modified MINI major and minor depression categories are in accordance with the published data from community surveys in various countries and cultures on the higher prevalence of major depression in women (Bland et al., 1988; Weissmann et al., 1988; Hwu et al., 1989; Blazer et al., 1994; Kessler et al., 1994; Weissmann et al., 1996; Lépine et al., 1997). With respect to the evidence obtained with the same classification instrument in other European countries (Lépine et al., 1997), the between-gender difference for major depression observed in our sample was limited (0.3 times higher prevalence in women than in men) and similar to that reported for the UK (0.4 times). Consistent with the UK findings, there was also a higher prevalence of minor depression in man (0.6 and 0.4 times higher in Italy and the UK, respectively).
In addition, our data confirm the published findings on a lower prevalence of major depression in the elderly (Weissmann et al., 1988; Lépine et al., 1997). However, similar to the DEPRES study (Lépine et al., 1997), the modified-MINI criteria, being based on the degree of perceived impairment of work or social activities, might underestimate the prevalence of depression for social groups that were less likely to be engaged in a paid occupation. Consistent and confounded with the lower prevalence in the elderly was the evidence obtained in the study on the decreased prevalence of depression among retired people. In fact, when the association between occupation and depression categories was evaluated within age classes, a trend towards decreasing prevalence of depression in the older age classes was seen for all occupations.
In conclusion, this is the first survey on the prevalence of depression in the Italian community, conducted with a telematic, computer-prompted modified-MINI interview. We obtained prevalence figures for major and minor depression comparable to those reported in the DEPRES study for other European countries, thus confirming a common depression burden in post-industrial Western societies, and supporting the use of telematic data acquisition for health-related surveys in the community.
Andreoli V, Cassano GB, Rossi R, eds (1996) DSM IV Manuale Diagnostico c Statistico dei Disturbi Mentali.
Ansseau M (1998). The socio-economics of depression
. Revue Médicale de Liège53:
Bland RC, Orn H, Newman SC (1988). Lifetime prevalence
of psychiatric disorders in Edmonton. Acta Psychiatrica Scandinavica77:
Blazer DG, Kessler RC, McGonagle KA, Swartz MS (1994). The prevalence
and distribution of major depression
in a national community
sample: the national comorbidity survey. Am J Psychiatry151:
Guideline Panel (1993) Depression in Primary Care, Vol. 1. Detection and Diagnosis.
Rockville, MD: US Department of Health and Human Services.
Hall RCW, Wise MG (1995). The clinical and financial burden of mood disorders: costs and outcome. Psychosomatics36:
Hwu HG, Yeh EK, Chang LY (1989). Prevalence
of psychiatric disorders in Taiwan defined by the Chinese Diagnostic Interview Schedule. Acta Psychiatr Scand79:
Katon W, Von Korff M, Lin E, Walker E, Simon GE, Bush T et al.
(1995). Collaborative management to achieve treatment guidelines. Impact on depression
in primary care. J Am Med Assoc273:
Kessler RC, McGonagle KA, Zhao S, Nelson CB, Hughes M, Eshleman S et al.
(1994). Lifetime and 12-month prevalence
of DSM-III-R psychiatric disorders in the United States. Arch Gen Psychiatry51:
Kind P, Sorensen J (1993). The costs of depression
. Int Clin Psychopharmacol7:
Lecrubier Y, Boyer P, Lépine JP, Weiller E (1996). The identification of psychiatric disorders in primary care. Eur Psychiatry11(Suppl 4):
Lecrubier Y, Sheehan DV, Weiller E, Amorim P, Bonora I, Sheehan KH et al.
(1997). The Mini-International-Neuropsychiatric Interview (MINI). A short diagnostic structured interview: reliability and validity according to the CIDI. Eur Psychiatry12:
Lépine JP, Gastpar M, Mendlewicz J, Tylee A, on behalf of the DEPRES Steering Committee (1997). Depression
in the community
: the first pan-European study DEPRES (Depression
Research in European Society). Int Clin Psychopharmacol12:
Miller DC (1991). Handbook of Research Design and Social Measurement.
London: SAGE Publications, p. 325.
Paykel ES, Priest RG, on behalf of conference participants (1992). Recognition and management of depression
in general practice: consensus statement. BMJ305:
Perez-Stable EJ, Miranda J, Munoz RF, Ying YW (1990). Depression
in medical outpatients. Underrecognition and misdiagnosis. Arch Gen Med150:
Rupp A (1995). The economic consequences of not treating depression
. Br J Psychiatry166(Suppl 27):
Sani G (1998). Otto domande a cinque campioni: una comparazione tra indagini demoscopiche. Polis12:
Sheehan DV, Lecrubier Y, Janavs. J, Knapp E, Weiller E, Bonora LI, et al.
(1994) Mini-International-Neuropsychiatric Interview (MINI).
Tampa, FL: University of South Florida Institute for Research in Psychiatry; Paris: INSERM-Hôpital de la Salpêtrière.
Sheehan DV, Lecrubier Y, Sheehan KH, Janavs. J, Weiller E, Bonora LI et al.
(1997). The validity of the Mini-International-Neuropsychiatric Interview (MINI) according to the SCI-P and its reliability. Eur Psychiatry12:
Weissmann MM, Leaf PJ, Tischler GL, Blazer DG, Karno M, Bruce ML, Florio LP (1988). Affective disorders in five United States communities. Psychol Med18:
Weissmann MM, Bland RC, Canino GJ, Faravelli C, Greenwald S, Hwu HG et al.
(1996). Cross-national epidemiology of major depression
and bipolar disorder. JAMA276: