- Email: [email protected]
Antidepressant drug use in Lombardy, Italy: a population-based study
Journal of Affective Disorders 83 (2004) 169 – 175 www.elsevier.com/locate/jad
Research report
Antidepressant drug use in Lombardy, Italy: a population-based study Mauro Percudania,*, Corrado Barbuib, Ida Fortinoa, Lorenzo Petrovicha a Health Care Directorate, Region of Lombardy, Milan, Italy Department of Medicine and Public Health, Section of Psychiatry, University of Verona, Italy
b
Received 11 May 2004; received in revised form 23 July 2004; accepted 23 July 2004
Abstract Background: The patterns of antidepressant drug prescribing have rarely been studied in large and geographically defined catchment areas. In the present study, we examined the prevalence and distribution of antidepressant prescribing in Lombardy, a northern Italy region of nine million inhabitants. Methods: This study used the Regional Administrative Database of Lombardy. This database includes all prescriptions reimbursed by the National Health System in the population living in this region. All antidepressant prescriptions dispensed from the 1st January to the 31st December 2001 were extracted and prevalence data calculated by dividing antidepressant users by the total number of male and female residents in each age group. Results: During the study period, 404,238 individuals were dispensed antidepressants, yielding a prevalence of use of 2.85 (95% confidence interval 2.84, 2.87) per 100 males and 5.92 (95% confidence interval 5.90, 5.94) per 100 females. The prevalence of use progressively rose with age in both sexes, with the highest rates in old and very old individuals. The majority of individuals received a pharmacological treatment with selective-serotonin reuptake inhibitors only, slightly more than 12% received a treatment with tricyclic antidepressants. General practitioners issued the majority of antidepressant prescriptions. Conclusions: The very high rates of antidepressant drug prescribing detected in late life suggest the need of characterising these individuals in terms of medical and psychiatric characteristics, needs and quality of life. It also suggests the need for pragmatic clinical trials, carried out in the general practice, with the aim of assessing whether antidepressants are effective in these conditions. D 2004 Elsevier B.V. All rights reserved. Keywords: Epidemiology; Antidepressants; General practice
* Corresponding author. Regione Lombardia, Via Pola 9/11, 20123 Milano, Italy. Tel.: +39 267653098; fax: +39 267653306. E-mail address: [email protected] (M. Percudani). 0165-0327/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.jad.2004.07.005
1. Introduction Since the introduction in the late 1980s of the selective serotonin re-uptake inhibitors (SSRIs) and
170
M. Percudani et al. / Journal of Affective Disorders 83 (2004) 169–175
newer antidepressants (ADs), in most countries the number of AD prescriptions has progressively increased (Barbui et al., 1999; Isacsson et al., 1999; Lawrenson et al., 2000; Middleton et al., 2001; Rosholm et al., 2001). AD agents are indicated in the pharmacological treatment of major depression but, in recent years, new labels have been increasing the number of psychiatric disorders where these agents are indicated, and new patient populations, such as the elderly, children and adolescents, have been suggested as possible target of AD therapy (Piccinelli et al., 1995; Dimmock et al., 2000; Otto et al., 2001). A relevant proportion of these prescriptions are issued by general practitioners (GPs), who have frequently been the object of educational interventions aiming at increasing the recognition and pharmacological treatment of affective disorders (Gilbody et al., 2003). In this evolving context of care, AD prescribing patterns have rarely been examined in large and geographically defined catchment areas. In the present study, we examined the prevalence and distribution of AD prescribing in Lombardy, a northern Italy region of nine million inhabitants.
2. Materials and methods 2.1. Study area Lombardy is the largest and most affluent region in Italy; the territory extends over a total surface area of 23,861 sq. km. The region represents 15.6% of the overall national population with approximately 9 million inhabitants (24% of the population is under the age of 25 and 17% over 65) (Regione Lombardia, 2002–2004). In Lombardy, in 2001, there were 7310 general practitioners (GPs) providing primary care. On average, in 2001, each GP was in charge of 1124 inhabitants. Since the first Regional Mental Health Plan approved in the early 1980s, Lombardy has adopted an organisational model for community psychiatric care focused on large multi-disciplinary teams. In 2001, there were 66 community psychiatric services (CPSs) operating in the region. Each CPS consists of various mental health centres, including psychiatric wards in the general hospital (n=54),
community mental health centres (n=107) and residential facilities (n=243). In 2001, a total of 110,486 individuals had at least one contact with one of the 66 CPSs (Regione Lombardia, 2002–2004). 2.2. Data source This study used the Regional Administrative Database of Lombardy. This database includes all community (i.e., outside hospitals) prescriptions reimbursed by the National Health System (NHS) in the population living in this region. Therefore, general practitioner (GP) prescriptions, ambulatory prescriptions delivered by specialists (psychiatrists, neurologists, others) and prescriptions delivered in private care are included in the database if reimbursed by the NHS. In Italy, AD agents are reimbursed by the NHS. The Italian system works in such a way that outpatients receive AD prescriptions from GPs, psychiatrists or other specialists and then get the medicines free of charge from retail pharmacies. Outpatients receiving prescriptions in the private sector get the medicines free of charge through GP prescriptions. Each local pharmacy provides these prescriptions to the Regional Health Authority to get reimbursed. The Regional Health Authority electronically stores these prescriptions into the Regional Administrative Database. In this system, a unique patient code prevents double counting of individuals who have been prescribed AD agents by more then one GP or medical team. 2.3. Data extraction and analysis All AD (N05B, N05C and N06A group of the anatomical therapeutic chemical classification system) prescriptions dispensed from the 1st January to the 31st December 2001 were extracted from the Regional Administrative Database. Each record included the anonymised patient code, patient’s sex and age, the number of AD prescriptions and a code identifying whether the prescription was filled by a GP or by a specialist. Using the patient code, records were attributed to patients receiving one or more prescriptions of one or more AD during the census year. Prevalence data were calculated by dividing AD users by the total number of male and female residents in each age group. Rates per 100 inhabitants were thus
M. Percudani et al. / Journal of Affective Disorders 83 (2004) 169–175
171
Table 1 Male and female prevalence of antidepressant prescribing by age (rates per 100 inhabitants) Age
0–15 16–20 21–25 26–30 31–35 36–40 41–45 46–50 51–55 56–60 61–65 66–70 71–75 76–80 81–85 86–90 91–95 N95 Total
Males
Females
Individuals
Population
Rate per 100 (95% CI)
Individuals
Population
Rate per 100 (95% CI)
950 1875 4795 7751 9316 9968 9246 9277 10,833 10,141 11,778 11,805 11,468 9250 4197 2987 773 189 126,599
656,102 219,692 285,487 374,746 417,442 387,884 329,903 307,515 316,659 282,57 273,769 221,874 167,396 113,136 42,240 28,891 6954 1028 4,433,293
0.14 (0.13, 0.15) 0.85 (0.81, 0.89) 1.67 (1.63, 1.72) 2.06 (2.02, 2.11) 2.23 (2.18, 2.27) 2.56 (2.52, 2.62) 2.80 (2.74, 2.85) 3.01 (2.95, 3.07) 3.42 (3.35, 3.48) 53.58 (3.52, 3.65) 4.30 (4.22, 4.37) 5.32 (5.22, 5.41) 6,85 (6.73, 6.97) 8.17 (8.01, 8.33) 9.93 (9.65, 10.22) 10.33 (9.99, 10.69) 11.11 (10.38, 11.87) 18.38 (16.06, 20.89) 2.85 (2.84, 2.87)
930 3196 7288 11,793 15,889 19,387 19,250 20,291 24,380 22,964 26,636 26,409 27,092 25,496 13,256 9808 2891 683 277,639
618,462 208,084 272,805 352,771 391,090 367,644 322,103 307,612 321,732 296,273 299,298 269,530 238,982 202,658 98,506 86,157 29,133 5581 4,688,421
0.15 (0.14, 0.16) 1.53 (1.48, 1.58) 2.67 (2.61. 2.63) 3.34 (3.28, 3.40) 4.06 (4.00, 4.12) 5.27 (5.20, 5.34) 5.97 (5.89, 6.05) 6.59 (6.50, 6.68) 7.57 (7.48, 7.66) 7.75 (7.65, 7.84) 8.89 (8.79, 9.00) 9.79 (9.68, 9.91) 11.33 (11.20, 11.46) 12.58 (12.43, 12.72) 13.45 (13.24, 13.67) 11.38 (11.17, 11.59) 9.92 (9.58, 10.27) 12.23 (11.38, 13.12) 5.92 (5.90, 5.94)
calculated, together with the 95% confidence interval (CI).
old and very old individuals. Overall, the prevalence of use was 2.85 (95% CI 2.84, 2.87) per 100 males and 5.92 (95% CI 5.90, 5.94) per 100 females.
3. Results
3.2. AD class and the top five agents
3.1. Prevalence of AD prescribing
The distribution of male and female individuals by AD class is presented in Table 2. During the 12-month period, the majority of individuals received a pharmacological treatment with SSRI only, slightly more than 12% received a treatment with TCA only and almost one fourth was treated with other AD agents. Two or more agents belonging to different pharmacological classes were received by 11.77% and 12.83%
During the 12 months surveyed, 404,238 individuals were dispensed one or more prescriptions of ADs, yielding a prevalence of use of 4.43 individuals per 100 inhabitants (95% CI 4.41, 4.44). Table 1 shows that the prevalence of AD use progressively rose with age in both sexes, with the highest rates in
Table 2 Distribution of male and female individuals by antidepressant class, with prevalence of prescribing per 100 inhabitants AD category
Males No.
TCA only SSRI only Other AD only* Two or more AD of different classes Total
15,368 68,101 28,227 14,903 126,599
Females % 12.14 53.79 22.30 11.77 100.0
Rate per 100 (95% CI)
No.
0.346 (0.341, 0.352) 1.53 (1.52, 1.54) 0.636 (0.629, 0.644) 0.336 (0.330, 0.341)
35,007 143,156 63,862 35,614
2.85 (2.84, 2.87)
277,639
% 12.61 51.56 23.00 12.83 100.0
* Amisulpiride+oxitriptano+mianserina+trazodone+viloxazina+mirtazapina+venlafaxina+reboxetina.
Rate per 100 (95% CI) 0.746 3.053 1.362 0.759
(0.738, (3.037, (1.351, (0.751,
0.754) 3.069) 1.372) 0.767)
5.92 (5.90, 5.94)
172
M. Percudani et al. / Journal of Affective Disorders 83 (2004) 169–175
male and female individuals, respectively. In both sexes, the prevalence of SSRI prescribing was substantially higher than the prevalence of TCA and other AD prescribing (Table 2). The distribution of individuals receiving the top five antidepressive agents by age is presented in Fig. 1. Paroxetine was first in the ranking, but its use progressively declined with the increasing age. In contrast, the use of amisulpride progressively rose with age, being with paroxetine the most prescribed agent in old and very old individuals. Nearly 60% of individuals receiving ADs was treated with one of the top five agents.
(49.08%) males received between two and six AD prescriptions, 22,428 (17.72%) received between seven and 12 AD prescriptions and 23,644 (18.68%) received more than 12 AD prescriptions. Similarly, a total of 41,306 females (14.88%) received only one AD prescription during the 12 months, 125,399 (45.17%) females received between two and six AD prescriptions, 53,745 (19.36%) received between seven and 12 AD prescriptions and 57,189 (20.60%) received more than 12 AD prescriptions.
3.3. Prescriptions filled by GPs
In males, the yearly mean cost of AD prescriptions was euro 150.8 (S.D. 156.9) per patient in those receiving SSRI only, euro 19.66 (S.D. 32.87) per patient in those receiving TCA only, euro 93.94 (SD 142.72) per patient in those receiving other AD agents and euro 272.8 (SD 249.8) per patient in those receiving two or more AD agents during the 12month period. In females, the yearly mean cost of AD prescriptions was euro 155.5 (SD 154.2) per patient in those receiving SSRI only, euro 20.12 (SD 31.30) per patient in those receiving TCA only, euro 100.70 (SD 139.27) per patient in those receiving other AD agents and euro 260.8 (SD 230.3) per patient in those receiving two or more AD agents during the 12-
GPs were responsible for most AD prescriptions. TCA prescriptions were issued by GPs in 18,476 males (89.84%) and in 44,162 (91.18%) females; SSRI prescriptions were issued by GPs in 69,415 males (84.87%) and in 152,232 females (86.82%); other AD prescriptions were issued by GPs in 35,040 males (87.09%) and in 82,591 females (89.50%). 3.4. Number of prescriptions A total of 18,393 males (14.53%) received only one AD prescription during the 12 months, 62,134
3.5. Cost of AD therapy
Fig. 1. Distribution of subjects receiving the top five antidepressant agents by age.
M. Percudani et al. / Journal of Affective Disorders 83 (2004) 169–175
month period. In 2001, the total cost of AD prescriptions was euro 53,873,695, yielding an average yearly cost of euro 5.9 per inhabitant.
173
occurred for individuals receiving AD prescriptions in the private sector; in this case, it is possible either to get the medicines free of charge through GP prescriptions or to buy them from retail pharmacies paying the full price.
4. Discussion 4.2. Clinical implications 4.1. Study limitations A first study limitation is the absence of clinical data on individuals receiving AD drugs. These agents are indicated in the treatment of psychiatric disorders other than depression and the database did not allow to make a distinction between those who received AD therapy for depression from those who received AD therapy for panic disorder, obsessive compulsive disorder, anxiety disorder or other psychiatric disorders. In addition, the database did not allow to study the patterns of AD treatment longitudinally, and no information was available on patient outcomes. Finally, the lack of data on whether patients eventually took the prescribed agents should be highlighted, since around half of the medicines prescribed for people with chronic conditions are not taken (Jones, 2003). Another limitation is the possibility of underreporting. The database collects community prescriptions only, and no information is available, for example, on AD agents prescribed to long-stay patients living in nursing homes and residential facilities. Since a relevant proportion of individuals living in residential settings receive AD drugs, prevalence rates might have underestimated the use of these agents, especially in particular age groups, such as the elderly, who are more likely to be admitted to these facilities (Mann et al., 2000). Moreover, some AD prescriptions might have been missed by the database. It is possible that retail pharmacies failed to provide all AD prescriptions to the Regional Health Authority. Although this possibility cannot be completely ruled out, it should be noted that retail pharmacies are reimbursed only for those AD prescriptions provided to the Regional Health Authority. In this system, therefore, underreporting would have produced negative economic consequences. Finally, even though AD agents are fully reimbursed, it is possible that some patients got the medicines paying the full price. This might have
This study examined the prevalence and distribution of AD prescribing in a large and geographically defined area comprising one sixth of the entire Italian population. The overall 1-year prevalence of AD prescribing was similar to that estimated in a Danish prescription database study which found a prevalence of AD prescribing of 4.1% in 1997 (Rosholm et al., 2001). Similarly, a Swedish prescription database study found a prevalence of AD prescribing of 3.8% in 1996 (Isacsson et al., 1999). In the UK the proportion of individuals receiving TCAs and SSRIs was 4.0% and 3.2% in 1995/96 (Lawrenson et al., 2000). Lower prevalence rates were calculated in prescription database studies carried out in selected areas of Piedmont, a Region in Northwest Italy, in 2000 (Pietraru et al., 2001; Barbui et al., 2003). At the time of data collection, however, SSRIs were reimbursed only when prescribed for patients who could not tolerate TCAs, and therefore a relevant proportion of AD prescriptions might have been missed. Most AD prescriptions were issued by GPs. This finding does not mean that all individuals with GP prescriptions were treated by GPs, since in Italy, it is possible that psychiatrists, neurologists or other specialists make the first AD prescription only, and then refer the patient to the treating GP for the subsequent prescriptions. In the private sector, moreover, specialists are obliged to refer patients to their GPs to get the medicines free of charge. However, considering that the number of individuals with GP prescriptions in 2001 was more than three time higher than the number of individuals seen by the 66 community psychiatric services operating in Lombardy in 2001, it seems reasonable to suppose that most of these individuals were managed in the general practice only. In 2001, the first year in which all AD prescriptions were fully reimbursed, the SSRIs and newer
174
M. Percudani et al. / Journal of Affective Disorders 83 (2004) 169–175
ADs accounted for the majority of prescriptions, with paroxetine top of the ranking. In addition, interesting prescribing behaviours emerged. Amisulpiride, for example, an antipsychotic agent registered in Italy for the treatment of dysthymia, was among the most prescribed agents, especially in the elderly. Probably, marketing claims of efficacy in patients with anxiety and multiple vague physical symptoms—nausea, tiredness, gastrointestinal disturbances—were responsible for its popularity in elderly individuals. These prescribing habits had cost implications. In 2001, the total pharmaceutical expenditure in Lombardy was 1617 million of euros; of these, AD acquisition costs accounted for more than 3%, a figure similar to the national data of 2.6% (National Report Year, 2001). SSRI, other AD and TCA costs accounted for 73%, 24.3% and 2.9% of the total expenditure for AD, respectively. The progressive rise in the prevalence of individuals receiving AD treatment with age is in contrast with data on the epidemiology of depression. In Canada (Bland et al., 1988), USA (Blazer et al., 1994; Kessler et al., 1994) and Europe (Tylee et al., 1999), prevalence rates of major depression were lower in the elderly than in young and adult people. In Italy, a survey conducted in the community found no increase in the prevalence of major and minor depression with age (Dubini et al., 2001). It is therefore likely that, in later life, ADs are widely prescribed for conditions which do not fully correspond to the diagnostic labels for major depression. This has implications for research because it underlines the relevance of studying these conditions, which have to be precisely described in terms of medical and psychiatric characteristics, psychopathology, needs and quality of life. It also suggests the need for pragmatic clinical trials, carried out in the general practice, with the aim of assessing whether AD agents have a favourable impact on these conditions.
Acknowledgments The authors wish to thank Mr. Adriano Martinelli (Santer SpA) for data collection and Professor Giovanni Fattore for invaluable comments on the manuscript.
References Barbui, C., Campomori, A., D’Avanzo, B., 1999. Antidepressant drug use in Italy since the introduction of SSRIs: national trends, regional differences and impact on suicide rates. Soc. Psychiatry Psychiatr. Epidemiol. 34, 152 – 156. Barbui, C., Broglio, E., Costa Laia, A., 2003. Cross-sectional database analysis of antidepressant prescribing in Italy. J. Clin. Psychopharmacol. 23, 31 – 34. Bland, R.C., Orn, H., Newman, S.C., 1988. Lifetime prevalence of psychiatric disorders in Edmonton. Acta Psychiatr. Scand. (338 Suppl.), 24 – 32. Blazer, D.G., Kessler, R.C., McGonagle, K.A., 1994. The prevalence and distribution of major depression in a national community sample: the National Comorbidity Survey. Am. J. Psychiatry 151, 979 – 986. Dimmock, P.W., Wyatt, K.M., Jones, P.W., 2000. Efficacy of selective serotonin-reuptake inhibitors in premenstrual syndrome: a systematic review. Lancet 356, 1131 – 1136. Dubini, A., Mannheimer, R., Pancheri, P., 2001. Depression in the community: results of the first Italian survey. Int. Clin. Psychopharmacol. 16, 49 – 53. Gilbody, S., Whitty, P., Grimshaw, J., 2003. Educational and organisational interventions to improve the management of depression in primary care. A systematic review. JAMA 289, 3145 – 3151. Isacsson, G., Boethius, G., Henriksson, S., 1999. Selective serotonin reuptake inhibitors have broadened the utilisation of antidepressant treatment in accordance with recommendations. Findings from a Swedish prescription database. J. Affect. Disord. 53, 15 – 22. Jones, G., 2003. Prescribing and taking medicines. Concordance is a fine theory but it is mostly not being practised. BMJ 327, 819. Kessler, R.C., McGonagle, K.A., Zhao, S., 1994. Lifetime and 12month prevalence of DSM-III-R psychiatric disorders in the United States. Results from the National Comorbidity Survey. Arch. Gen. Psychiatry 51, 8 – 19. Lawrenson, R.A., Tyrer, F., Newson, R.B., 2000. The treatment of depression in UK general practice: selective serotonin reuptake inhibitors and tricyclic antidepressants compared. J. Affect. Disord. 59, 149 – 157. Mann, A.H., Schneider, J., Mozley, C.G., 2000. Depression and the response of residential homes to physical health needs. Int. J. Geriatr. Psychiatry 15, 1105 – 1112. Middleton, N., Gunnell, D., Whitley, E., 2001. Secular trends in antidepressant prescribing in the UK, 1975–1998. J. Public Health Med. 23, 262 – 267. National Report Year, 2001. Drug utilization in Italy. Ministero della Salute 2002 (www.ministerosalute.it). Otto, M.W., Tuby, K.S., Gould, R.A., 2001. An effect-size analysis of the relative efficacy and tolerability of serotonin selective reuptake inhibitors for panic disorder. Am. J. Psychiatry 158, 1989 – 1992. Piccinelli, M., Pini, S., Bellantuono, C., 1995. Efficacy of drug treatment in obsessive–compulsive disorder: a meta-analytic review. Br. J. Psychiatry 166, 424 – 452.
M. Percudani et al. / Journal of Affective Disorders 83 (2004) 169–175 Pietraru, C., Barbui, C., Poggio, L., 2001. Antidepressant drug prescribing in Italy, 2000: analysis of a general practice database. Eur. J. Clin. Pharmacol. 57, 605 – 609. Regione Lombardia, 2002–2004. Regional Community Health Plan. Regione Lombardia. www.sanita.regione.lombardia.it. Rosholm, J.U., Andersen, M., Gram, L.F., 2001. Are there differences in the use of selective serotonin reuptake inhibitors and
175
tricyclic antidepressants? A prescription database study. Eur. J. Clin. Pharmacol. 56, 923 – 929. Tylee, A., Gastpar, M., Le´pine, J.P., 1999. Identification of depressed patients in the community and their treatment needs: findings from the DEPRES II (Depression Research in European Society II) survey. Int. Clin. Psychopharmacol. 14, 153 – 165.
Research report
Antidepressant drug use in Lombardy, Italy: a population-based study Mauro Percudania,*, Corrado Barbuib, Ida Fortinoa, Lorenzo Petrovicha a Health Care Directorate, Region of Lombardy, Milan, Italy Department of Medicine and Public Health, Section of Psychiatry, University of Verona, Italy
b
Received 11 May 2004; received in revised form 23 July 2004; accepted 23 July 2004
Abstract Background: The patterns of antidepressant drug prescribing have rarely been studied in large and geographically defined catchment areas. In the present study, we examined the prevalence and distribution of antidepressant prescribing in Lombardy, a northern Italy region of nine million inhabitants. Methods: This study used the Regional Administrative Database of Lombardy. This database includes all prescriptions reimbursed by the National Health System in the population living in this region. All antidepressant prescriptions dispensed from the 1st January to the 31st December 2001 were extracted and prevalence data calculated by dividing antidepressant users by the total number of male and female residents in each age group. Results: During the study period, 404,238 individuals were dispensed antidepressants, yielding a prevalence of use of 2.85 (95% confidence interval 2.84, 2.87) per 100 males and 5.92 (95% confidence interval 5.90, 5.94) per 100 females. The prevalence of use progressively rose with age in both sexes, with the highest rates in old and very old individuals. The majority of individuals received a pharmacological treatment with selective-serotonin reuptake inhibitors only, slightly more than 12% received a treatment with tricyclic antidepressants. General practitioners issued the majority of antidepressant prescriptions. Conclusions: The very high rates of antidepressant drug prescribing detected in late life suggest the need of characterising these individuals in terms of medical and psychiatric characteristics, needs and quality of life. It also suggests the need for pragmatic clinical trials, carried out in the general practice, with the aim of assessing whether antidepressants are effective in these conditions. D 2004 Elsevier B.V. All rights reserved. Keywords: Epidemiology; Antidepressants; General practice
* Corresponding author. Regione Lombardia, Via Pola 9/11, 20123 Milano, Italy. Tel.: +39 267653098; fax: +39 267653306. E-mail address: [email protected] (M. Percudani). 0165-0327/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.jad.2004.07.005
1. Introduction Since the introduction in the late 1980s of the selective serotonin re-uptake inhibitors (SSRIs) and
170
M. Percudani et al. / Journal of Affective Disorders 83 (2004) 169–175
newer antidepressants (ADs), in most countries the number of AD prescriptions has progressively increased (Barbui et al., 1999; Isacsson et al., 1999; Lawrenson et al., 2000; Middleton et al., 2001; Rosholm et al., 2001). AD agents are indicated in the pharmacological treatment of major depression but, in recent years, new labels have been increasing the number of psychiatric disorders where these agents are indicated, and new patient populations, such as the elderly, children and adolescents, have been suggested as possible target of AD therapy (Piccinelli et al., 1995; Dimmock et al., 2000; Otto et al., 2001). A relevant proportion of these prescriptions are issued by general practitioners (GPs), who have frequently been the object of educational interventions aiming at increasing the recognition and pharmacological treatment of affective disorders (Gilbody et al., 2003). In this evolving context of care, AD prescribing patterns have rarely been examined in large and geographically defined catchment areas. In the present study, we examined the prevalence and distribution of AD prescribing in Lombardy, a northern Italy region of nine million inhabitants.
2. Materials and methods 2.1. Study area Lombardy is the largest and most affluent region in Italy; the territory extends over a total surface area of 23,861 sq. km. The region represents 15.6% of the overall national population with approximately 9 million inhabitants (24% of the population is under the age of 25 and 17% over 65) (Regione Lombardia, 2002–2004). In Lombardy, in 2001, there were 7310 general practitioners (GPs) providing primary care. On average, in 2001, each GP was in charge of 1124 inhabitants. Since the first Regional Mental Health Plan approved in the early 1980s, Lombardy has adopted an organisational model for community psychiatric care focused on large multi-disciplinary teams. In 2001, there were 66 community psychiatric services (CPSs) operating in the region. Each CPS consists of various mental health centres, including psychiatric wards in the general hospital (n=54),
community mental health centres (n=107) and residential facilities (n=243). In 2001, a total of 110,486 individuals had at least one contact with one of the 66 CPSs (Regione Lombardia, 2002–2004). 2.2. Data source This study used the Regional Administrative Database of Lombardy. This database includes all community (i.e., outside hospitals) prescriptions reimbursed by the National Health System (NHS) in the population living in this region. Therefore, general practitioner (GP) prescriptions, ambulatory prescriptions delivered by specialists (psychiatrists, neurologists, others) and prescriptions delivered in private care are included in the database if reimbursed by the NHS. In Italy, AD agents are reimbursed by the NHS. The Italian system works in such a way that outpatients receive AD prescriptions from GPs, psychiatrists or other specialists and then get the medicines free of charge from retail pharmacies. Outpatients receiving prescriptions in the private sector get the medicines free of charge through GP prescriptions. Each local pharmacy provides these prescriptions to the Regional Health Authority to get reimbursed. The Regional Health Authority electronically stores these prescriptions into the Regional Administrative Database. In this system, a unique patient code prevents double counting of individuals who have been prescribed AD agents by more then one GP or medical team. 2.3. Data extraction and analysis All AD (N05B, N05C and N06A group of the anatomical therapeutic chemical classification system) prescriptions dispensed from the 1st January to the 31st December 2001 were extracted from the Regional Administrative Database. Each record included the anonymised patient code, patient’s sex and age, the number of AD prescriptions and a code identifying whether the prescription was filled by a GP or by a specialist. Using the patient code, records were attributed to patients receiving one or more prescriptions of one or more AD during the census year. Prevalence data were calculated by dividing AD users by the total number of male and female residents in each age group. Rates per 100 inhabitants were thus
M. Percudani et al. / Journal of Affective Disorders 83 (2004) 169–175
171
Table 1 Male and female prevalence of antidepressant prescribing by age (rates per 100 inhabitants) Age
0–15 16–20 21–25 26–30 31–35 36–40 41–45 46–50 51–55 56–60 61–65 66–70 71–75 76–80 81–85 86–90 91–95 N95 Total
Males
Females
Individuals
Population
Rate per 100 (95% CI)
Individuals
Population
Rate per 100 (95% CI)
950 1875 4795 7751 9316 9968 9246 9277 10,833 10,141 11,778 11,805 11,468 9250 4197 2987 773 189 126,599
656,102 219,692 285,487 374,746 417,442 387,884 329,903 307,515 316,659 282,57 273,769 221,874 167,396 113,136 42,240 28,891 6954 1028 4,433,293
0.14 (0.13, 0.15) 0.85 (0.81, 0.89) 1.67 (1.63, 1.72) 2.06 (2.02, 2.11) 2.23 (2.18, 2.27) 2.56 (2.52, 2.62) 2.80 (2.74, 2.85) 3.01 (2.95, 3.07) 3.42 (3.35, 3.48) 53.58 (3.52, 3.65) 4.30 (4.22, 4.37) 5.32 (5.22, 5.41) 6,85 (6.73, 6.97) 8.17 (8.01, 8.33) 9.93 (9.65, 10.22) 10.33 (9.99, 10.69) 11.11 (10.38, 11.87) 18.38 (16.06, 20.89) 2.85 (2.84, 2.87)
930 3196 7288 11,793 15,889 19,387 19,250 20,291 24,380 22,964 26,636 26,409 27,092 25,496 13,256 9808 2891 683 277,639
618,462 208,084 272,805 352,771 391,090 367,644 322,103 307,612 321,732 296,273 299,298 269,530 238,982 202,658 98,506 86,157 29,133 5581 4,688,421
0.15 (0.14, 0.16) 1.53 (1.48, 1.58) 2.67 (2.61. 2.63) 3.34 (3.28, 3.40) 4.06 (4.00, 4.12) 5.27 (5.20, 5.34) 5.97 (5.89, 6.05) 6.59 (6.50, 6.68) 7.57 (7.48, 7.66) 7.75 (7.65, 7.84) 8.89 (8.79, 9.00) 9.79 (9.68, 9.91) 11.33 (11.20, 11.46) 12.58 (12.43, 12.72) 13.45 (13.24, 13.67) 11.38 (11.17, 11.59) 9.92 (9.58, 10.27) 12.23 (11.38, 13.12) 5.92 (5.90, 5.94)
calculated, together with the 95% confidence interval (CI).
old and very old individuals. Overall, the prevalence of use was 2.85 (95% CI 2.84, 2.87) per 100 males and 5.92 (95% CI 5.90, 5.94) per 100 females.
3. Results
3.2. AD class and the top five agents
3.1. Prevalence of AD prescribing
The distribution of male and female individuals by AD class is presented in Table 2. During the 12-month period, the majority of individuals received a pharmacological treatment with SSRI only, slightly more than 12% received a treatment with TCA only and almost one fourth was treated with other AD agents. Two or more agents belonging to different pharmacological classes were received by 11.77% and 12.83%
During the 12 months surveyed, 404,238 individuals were dispensed one or more prescriptions of ADs, yielding a prevalence of use of 4.43 individuals per 100 inhabitants (95% CI 4.41, 4.44). Table 1 shows that the prevalence of AD use progressively rose with age in both sexes, with the highest rates in
Table 2 Distribution of male and female individuals by antidepressant class, with prevalence of prescribing per 100 inhabitants AD category
Males No.
TCA only SSRI only Other AD only* Two or more AD of different classes Total
15,368 68,101 28,227 14,903 126,599
Females % 12.14 53.79 22.30 11.77 100.0
Rate per 100 (95% CI)
No.
0.346 (0.341, 0.352) 1.53 (1.52, 1.54) 0.636 (0.629, 0.644) 0.336 (0.330, 0.341)
35,007 143,156 63,862 35,614
2.85 (2.84, 2.87)
277,639
% 12.61 51.56 23.00 12.83 100.0
* Amisulpiride+oxitriptano+mianserina+trazodone+viloxazina+mirtazapina+venlafaxina+reboxetina.
Rate per 100 (95% CI) 0.746 3.053 1.362 0.759
(0.738, (3.037, (1.351, (0.751,
0.754) 3.069) 1.372) 0.767)
5.92 (5.90, 5.94)
172
M. Percudani et al. / Journal of Affective Disorders 83 (2004) 169–175
male and female individuals, respectively. In both sexes, the prevalence of SSRI prescribing was substantially higher than the prevalence of TCA and other AD prescribing (Table 2). The distribution of individuals receiving the top five antidepressive agents by age is presented in Fig. 1. Paroxetine was first in the ranking, but its use progressively declined with the increasing age. In contrast, the use of amisulpride progressively rose with age, being with paroxetine the most prescribed agent in old and very old individuals. Nearly 60% of individuals receiving ADs was treated with one of the top five agents.
(49.08%) males received between two and six AD prescriptions, 22,428 (17.72%) received between seven and 12 AD prescriptions and 23,644 (18.68%) received more than 12 AD prescriptions. Similarly, a total of 41,306 females (14.88%) received only one AD prescription during the 12 months, 125,399 (45.17%) females received between two and six AD prescriptions, 53,745 (19.36%) received between seven and 12 AD prescriptions and 57,189 (20.60%) received more than 12 AD prescriptions.
3.3. Prescriptions filled by GPs
In males, the yearly mean cost of AD prescriptions was euro 150.8 (S.D. 156.9) per patient in those receiving SSRI only, euro 19.66 (S.D. 32.87) per patient in those receiving TCA only, euro 93.94 (SD 142.72) per patient in those receiving other AD agents and euro 272.8 (SD 249.8) per patient in those receiving two or more AD agents during the 12month period. In females, the yearly mean cost of AD prescriptions was euro 155.5 (SD 154.2) per patient in those receiving SSRI only, euro 20.12 (SD 31.30) per patient in those receiving TCA only, euro 100.70 (SD 139.27) per patient in those receiving other AD agents and euro 260.8 (SD 230.3) per patient in those receiving two or more AD agents during the 12-
GPs were responsible for most AD prescriptions. TCA prescriptions were issued by GPs in 18,476 males (89.84%) and in 44,162 (91.18%) females; SSRI prescriptions were issued by GPs in 69,415 males (84.87%) and in 152,232 females (86.82%); other AD prescriptions were issued by GPs in 35,040 males (87.09%) and in 82,591 females (89.50%). 3.4. Number of prescriptions A total of 18,393 males (14.53%) received only one AD prescription during the 12 months, 62,134
3.5. Cost of AD therapy
Fig. 1. Distribution of subjects receiving the top five antidepressant agents by age.
M. Percudani et al. / Journal of Affective Disorders 83 (2004) 169–175
month period. In 2001, the total cost of AD prescriptions was euro 53,873,695, yielding an average yearly cost of euro 5.9 per inhabitant.
173
occurred for individuals receiving AD prescriptions in the private sector; in this case, it is possible either to get the medicines free of charge through GP prescriptions or to buy them from retail pharmacies paying the full price.
4. Discussion 4.2. Clinical implications 4.1. Study limitations A first study limitation is the absence of clinical data on individuals receiving AD drugs. These agents are indicated in the treatment of psychiatric disorders other than depression and the database did not allow to make a distinction between those who received AD therapy for depression from those who received AD therapy for panic disorder, obsessive compulsive disorder, anxiety disorder or other psychiatric disorders. In addition, the database did not allow to study the patterns of AD treatment longitudinally, and no information was available on patient outcomes. Finally, the lack of data on whether patients eventually took the prescribed agents should be highlighted, since around half of the medicines prescribed for people with chronic conditions are not taken (Jones, 2003). Another limitation is the possibility of underreporting. The database collects community prescriptions only, and no information is available, for example, on AD agents prescribed to long-stay patients living in nursing homes and residential facilities. Since a relevant proportion of individuals living in residential settings receive AD drugs, prevalence rates might have underestimated the use of these agents, especially in particular age groups, such as the elderly, who are more likely to be admitted to these facilities (Mann et al., 2000). Moreover, some AD prescriptions might have been missed by the database. It is possible that retail pharmacies failed to provide all AD prescriptions to the Regional Health Authority. Although this possibility cannot be completely ruled out, it should be noted that retail pharmacies are reimbursed only for those AD prescriptions provided to the Regional Health Authority. In this system, therefore, underreporting would have produced negative economic consequences. Finally, even though AD agents are fully reimbursed, it is possible that some patients got the medicines paying the full price. This might have
This study examined the prevalence and distribution of AD prescribing in a large and geographically defined area comprising one sixth of the entire Italian population. The overall 1-year prevalence of AD prescribing was similar to that estimated in a Danish prescription database study which found a prevalence of AD prescribing of 4.1% in 1997 (Rosholm et al., 2001). Similarly, a Swedish prescription database study found a prevalence of AD prescribing of 3.8% in 1996 (Isacsson et al., 1999). In the UK the proportion of individuals receiving TCAs and SSRIs was 4.0% and 3.2% in 1995/96 (Lawrenson et al., 2000). Lower prevalence rates were calculated in prescription database studies carried out in selected areas of Piedmont, a Region in Northwest Italy, in 2000 (Pietraru et al., 2001; Barbui et al., 2003). At the time of data collection, however, SSRIs were reimbursed only when prescribed for patients who could not tolerate TCAs, and therefore a relevant proportion of AD prescriptions might have been missed. Most AD prescriptions were issued by GPs. This finding does not mean that all individuals with GP prescriptions were treated by GPs, since in Italy, it is possible that psychiatrists, neurologists or other specialists make the first AD prescription only, and then refer the patient to the treating GP for the subsequent prescriptions. In the private sector, moreover, specialists are obliged to refer patients to their GPs to get the medicines free of charge. However, considering that the number of individuals with GP prescriptions in 2001 was more than three time higher than the number of individuals seen by the 66 community psychiatric services operating in Lombardy in 2001, it seems reasonable to suppose that most of these individuals were managed in the general practice only. In 2001, the first year in which all AD prescriptions were fully reimbursed, the SSRIs and newer
174
M. Percudani et al. / Journal of Affective Disorders 83 (2004) 169–175
ADs accounted for the majority of prescriptions, with paroxetine top of the ranking. In addition, interesting prescribing behaviours emerged. Amisulpiride, for example, an antipsychotic agent registered in Italy for the treatment of dysthymia, was among the most prescribed agents, especially in the elderly. Probably, marketing claims of efficacy in patients with anxiety and multiple vague physical symptoms—nausea, tiredness, gastrointestinal disturbances—were responsible for its popularity in elderly individuals. These prescribing habits had cost implications. In 2001, the total pharmaceutical expenditure in Lombardy was 1617 million of euros; of these, AD acquisition costs accounted for more than 3%, a figure similar to the national data of 2.6% (National Report Year, 2001). SSRI, other AD and TCA costs accounted for 73%, 24.3% and 2.9% of the total expenditure for AD, respectively. The progressive rise in the prevalence of individuals receiving AD treatment with age is in contrast with data on the epidemiology of depression. In Canada (Bland et al., 1988), USA (Blazer et al., 1994; Kessler et al., 1994) and Europe (Tylee et al., 1999), prevalence rates of major depression were lower in the elderly than in young and adult people. In Italy, a survey conducted in the community found no increase in the prevalence of major and minor depression with age (Dubini et al., 2001). It is therefore likely that, in later life, ADs are widely prescribed for conditions which do not fully correspond to the diagnostic labels for major depression. This has implications for research because it underlines the relevance of studying these conditions, which have to be precisely described in terms of medical and psychiatric characteristics, psychopathology, needs and quality of life. It also suggests the need for pragmatic clinical trials, carried out in the general practice, with the aim of assessing whether AD agents have a favourable impact on these conditions.
Acknowledgments The authors wish to thank Mr. Adriano Martinelli (Santer SpA) for data collection and Professor Giovanni Fattore for invaluable comments on the manuscript.
References Barbui, C., Campomori, A., D’Avanzo, B., 1999. Antidepressant drug use in Italy since the introduction of SSRIs: national trends, regional differences and impact on suicide rates. Soc. Psychiatry Psychiatr. Epidemiol. 34, 152 – 156. Barbui, C., Broglio, E., Costa Laia, A., 2003. Cross-sectional database analysis of antidepressant prescribing in Italy. J. Clin. Psychopharmacol. 23, 31 – 34. Bland, R.C., Orn, H., Newman, S.C., 1988. Lifetime prevalence of psychiatric disorders in Edmonton. Acta Psychiatr. Scand. (338 Suppl.), 24 – 32. Blazer, D.G., Kessler, R.C., McGonagle, K.A., 1994. The prevalence and distribution of major depression in a national community sample: the National Comorbidity Survey. Am. J. Psychiatry 151, 979 – 986. Dimmock, P.W., Wyatt, K.M., Jones, P.W., 2000. Efficacy of selective serotonin-reuptake inhibitors in premenstrual syndrome: a systematic review. Lancet 356, 1131 – 1136. Dubini, A., Mannheimer, R., Pancheri, P., 2001. Depression in the community: results of the first Italian survey. Int. Clin. Psychopharmacol. 16, 49 – 53. Gilbody, S., Whitty, P., Grimshaw, J., 2003. Educational and organisational interventions to improve the management of depression in primary care. A systematic review. JAMA 289, 3145 – 3151. Isacsson, G., Boethius, G., Henriksson, S., 1999. Selective serotonin reuptake inhibitors have broadened the utilisation of antidepressant treatment in accordance with recommendations. Findings from a Swedish prescription database. J. Affect. Disord. 53, 15 – 22. Jones, G., 2003. Prescribing and taking medicines. Concordance is a fine theory but it is mostly not being practised. BMJ 327, 819. Kessler, R.C., McGonagle, K.A., Zhao, S., 1994. Lifetime and 12month prevalence of DSM-III-R psychiatric disorders in the United States. Results from the National Comorbidity Survey. Arch. Gen. Psychiatry 51, 8 – 19. Lawrenson, R.A., Tyrer, F., Newson, R.B., 2000. The treatment of depression in UK general practice: selective serotonin reuptake inhibitors and tricyclic antidepressants compared. J. Affect. Disord. 59, 149 – 157. Mann, A.H., Schneider, J., Mozley, C.G., 2000. Depression and the response of residential homes to physical health needs. Int. J. Geriatr. Psychiatry 15, 1105 – 1112. Middleton, N., Gunnell, D., Whitley, E., 2001. Secular trends in antidepressant prescribing in the UK, 1975–1998. J. Public Health Med. 23, 262 – 267. National Report Year, 2001. Drug utilization in Italy. Ministero della Salute 2002 (www.ministerosalute.it). Otto, M.W., Tuby, K.S., Gould, R.A., 2001. An effect-size analysis of the relative efficacy and tolerability of serotonin selective reuptake inhibitors for panic disorder. Am. J. Psychiatry 158, 1989 – 1992. Piccinelli, M., Pini, S., Bellantuono, C., 1995. Efficacy of drug treatment in obsessive–compulsive disorder: a meta-analytic review. Br. J. Psychiatry 166, 424 – 452.
M. Percudani et al. / Journal of Affective Disorders 83 (2004) 169–175 Pietraru, C., Barbui, C., Poggio, L., 2001. Antidepressant drug prescribing in Italy, 2000: analysis of a general practice database. Eur. J. Clin. Pharmacol. 57, 605 – 609. Regione Lombardia, 2002–2004. Regional Community Health Plan. Regione Lombardia. www.sanita.regione.lombardia.it. Rosholm, J.U., Andersen, M., Gram, L.F., 2001. Are there differences in the use of selective serotonin reuptake inhibitors and
175
tricyclic antidepressants? A prescription database study. Eur. J. Clin. Pharmacol. 56, 923 – 929. Tylee, A., Gastpar, M., Le´pine, J.P., 1999. Identification of depressed patients in the community and their treatment needs: findings from the DEPRES II (Depression Research in European Society II) survey. Int. Clin. Psychopharmacol. 14, 153 – 165.