The social costs of anxiety disorders Experience and Evidence in Psychiatry: Symposium on Somatics and Anxiety Disorders,, Madrid, 19th April 2006 Frank Jacobi & Hans-Ulrich Wittchen Technische Universität t Dresden Institute of Clinical Psychology and Psychotherapy Overview 1. The spectrum of anxiety disorders 2. What is special in anxiety disorders? high prevalence early onset persistence comorbidity low treatment rates 3. Indicators of social cost 4. The social costs of anxiety disorders in the context of all disorders of the brain 1
The spectrum of anxiety disorders (DSM-IV) panic disorder (with and without agoraphobia) agoraphobia (without panic disorder) social phobia specific phobias generalized anxiety disorder (GAD) obsessive-compulsive disorder (OCD) posttraumatic stress disorder (PTSD) anxiety disorder NOS Special features: partly different risk factors and correlates and consequences, various degree of within anxiety disorders overlap To understand the burden we need epidemiological studies: Research and reference populations Total general population Epidemiology is able to provide a more complete picture of patterns of morbidity and supplement findings from clinical research Subjects with a diagnosis in lifetime (lifetime prevalence/risk) High risk subjects (current subsyndromal, partial remission) Subjects with current disorders but not in treatment (undiagnosed, untreated) Patients in treatment services (treated prevalence) research 2
What is special in anxiety disorders? prevalence onset persistence Increasingly higher prevalence estimates for anxiety disorders in 3 decades (due to broader and more specific coverage no evidence for real increase ) 12month prevalence (%) 20 S6 13-18% S9 Studies: S1 Marks review S2 Wittchen review 15 8-10% S7 S8 S10 S3 ECA S4 MFS S5 Edmonton 10 5 5-7% S1 S2 S3 S4 S5 S6 NCS S7 NEMESIS S8 OHS S9 EDSP S10 GHS-MHS 0 pre 1980 studies in the 80ies studies in the 90ies Wittchen & Jacobi, ECNP 2003 3
First onset of anxiety disorders is predominantly before age of 20 1.00 0.75 cumulative % 0.50 0.25 "specific phobias" social phobia OCD panic disorder GAD Within anxiety disorders: PD, GAD later than phobias and OCD 0.00 0 5 10 15 20 25 30 Age of onset (years) EDSP, 2005 Persistence: 12-month / lifetime prevalence Any anxiety disorder PTSD OCD Specific phobia Unlike to depression if you have a lifetime anxiety disorder you are very likely to also have an anxiety disorder currently! Social phobia GAD Agoraphobia 12-month lifetime Panic disorder NCS-R, 2005 0 5 10 15 20 25 30 35 % 4
Where is the position of anxiety disorder within all disorders of the brain? Size and Burden of Mental Disorders in Europe : Material and methods Standardized search for EU-publications (N=212 studies all languages) Iterative data collection process (114 country-specific experts) Inclusion of unpublished material (additional 19 studies) Agreement on definition and conventions (DSM-III-R/IV-diagnoses & criteria, 12-month, etc.) Original data for standardized reanalyses (7 EU-countries, N = 28.000+, mean, 95% CI) Data compilation by country, age, gender and diagnoses for experts review Preparation of peer review state of the art papers by diagnostic domains Circulation to all country- and topic-specific experts (over 100 experts) Linkage with Health-Economic panel (collaboration with European Brain Council, EBC) Reanalyses and statistical modelling of data The collaborative EBC-ECNP ECNP network: Contributing core experts Panel members and review authors (mental disorders): Carlo Altamura, IT Jules Angst, CH Eni Becker, NL Claudine Berr, FR Terry Brugha, UK Ron de Graaf, NL Carlo Faravelli, IT Lydia Fehm, DE Tom Fryers, UK Tomas Furmark, SE Renee Goodwin, US Frank Jacobi, DE Ludwig Kraus, DE Roselind Lieb, DE Eugene Paykel, UK Antoine Pelissolo, FR Lukas Pezawas, US Stefano Pini, IT Jürgen Rehm, CH, CA Anita Riecher-Rössler, CH Karen Ritchie, FR Wulf Rössler, CH Robin Room, SE Hans Joachim Salize, DE Wim van den Brink, NL Jim van Os, NL Johannes Wancata, AT Hans-Ulrich Wittchen, DE Panel members and review authors (COI-reviews): Patrik Andlin-Sobocki, SE Jenny Berg, SE Mattias Ekman, SE Lars Forsgren, SE Bengt Jönsson, SE Linus Jönsson, SE Gisela Kobelt, FR Peter Lindgren, SE Mickael Löthgren, UK Jes Olesen, DK Country specific epidemiol. experts (mental disorders): Christer Allgulander, SE Jordi Alonso, ES Jules Angst, CH Terry Brugha, UK Ron de Graaf, NL Eva Dragomirecka, CZ Carlo Faravelli, IT Erkki Isometsä, FI Heinz Katschnig, AT Jean-Pierre Lèpine, FR Jouko Lönnqvist, FI Julien Mendlewicz, BE Povl Munk-Jörgensen, DK Bozena Pietrzykowska, PL Zoltan Rihmer, HU Inger Sandanger, NO Jon G. Stefánsson, IS Miguel Xavier, PT Panel members (neurological): Ettore Beghi, IT Karin Berger, DE Gudrun Boysen, DK Sonja v. Campenhausen, DE Richard Dodel, DE Lars Forsgren, SE W.H. Oertel, DE Jes Olesen, DK Maura Pugliatti, IT Franco Servadei, IT Uwe Siebert, DE Lars Stovner, NO Thomas Truelsen, SE Manfred Westphal, DE Coordinator of data collection: Frank Jacobi DE Steering committee members are underlined 5
Coverage and Definitions Mental disorders (DSM-IIR-DSM-IV) Affective disorders: Bipolar disorders, major depression, dysthymia Anxiety disorders: panic disorder, agoraphobia, GAD, social phobia, specific phobia, OCD, PTSD Dementia Psychotic disorders (focus on schizophrenia) Somatoform disorders: hypochondriasis, pain disorders, Somatisation disorder Substance use disorders: Alcohol abuse and dependence, Illegal drug abuse and dependence, nicotine dependence Eating disorders: anorexia nervosa, bulimia Other disorders of the brain: Parkinson s disease, Migraine and other headaches Stroke, Epilepsy, Brain trauma, Brain tumour, Multiple Sclerosis Geographical Scope EU member countries (EU-25) and Iceland, Norway and Switzerland Latest findings: Anxiety accounts for a large proportion of all mental disorders! 12-month prevalence (%, 95% CI) and estimated number of subjects affected in the EU eating disorders ill. subst. dep. OCD psychotic disorders bipolar disorder agoraphobia GAD panic disorder 1,1 Mio (0,9-1,7) 2,0 Mio (1,4-2,1) 2,6 Mio (2,4-3,0) 3,6 Mio (2,8-5,3 2,4 Mio (1,7-2,4) 3,9 Mio (3,3-4,7) 5,8 Mio (5,2-6,1) 5,2 Mio (4,3-5,3) social phobia alcohol dependence 6,6 Mio (5,4-9,2) 7,1 Mio (5,8-8,6) somatof. disorders specific phobias 18.9 Mio. (12.6-21.1) 18.4 Mio. (17.2-19.0) major depression 18.5 Mio. (14.3-18.6) 0 1 2 3 4 5 6 7 8 9 Note: Numbers add up to more than 27% and 82 million subjects because subjects can have more than one disorder (comorbidity) Wittchen & Jacobi (2005), Neuropsychopharmacology 6
Are these 12-month prevalence EU estimates suprisingly high? Yes if you consider that some of the previous epidemiological studies revealed somewhat lower estimates, because of + A restricted range of disorders covered + Narrower time window (e.g. restricted the prevalence period to 2 weeks) + Additional so-called clinical significance criteria Are those 12-month prevalence EU estimates surprisingly high? Yes if you consider that some of the previous epidemiological studies revealed somewhat lower estimates, because of + A restricted range of disorders covered + Narrower time window (e.g. restricted the prevalence period to 2 weeks) + Additional so-called clinical significance criteria Not however if you account in previous studies for the above mentioned methodological differences Not in comparison to somatic disorders: In this age range, over 70% of the general population has at least one somatic disorder ( Why should the brain less frequently affected? ) 7
What is special in anxiety disorders? (cont.) comorbidity Comorbidity is a fundamental characteristic of mental disorders (and the way they are defined in current classification systems) and increases by age OR Anxiety with: Suds: 2.6 Depression: 6.9 Somatoform: 3.4 OR Depression: Anxiety: 7.0 Suds: 2.7 Somatoform: 3.5 80 54.3% of all anxiety disorders are comorbid 60,2% of the mood disorders 70 60 proportion comorbid 50 40 30 41,2% of substance use disorders 20 10 OR Substance with: Anxiety: 2.5 Depression: 2.7 Somatoform: 1.9 49,2% of the somatoform disorders OR Somatoform: Anxiety: 3.5 Suds: 2.1 Depression: 3.5 0 18-29 30-39 40-49 50-59 60-65 Age group 8
.. And might have important etiological implications, for example Symptom progression models: Sequential comorbidity in anxiety disorders Onset of cascade Precursors: Behavioral inhibition/separation anxiety, (trauma) Increased neurobiological, cognitive, behavioral sensitization Specific and social phobia panic attacks, agoraphobia, panic disorder GAD Secondary depression Increased impairment/disability Suicidality Substance use disorders 5 10 15 20 25 30 35+ age Cumulative risk of cases with primary anxiety disorder by age of onset of secondary depressive disorder Cumulative % of Cum. risk (%) depression 60 50 40 30 no anxiety dx 20 10 0 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 By age of onset EDSP, 2001 9
Cumulative risk of cases with primary anxiety disorder by age of onset of secondary depressive disorder Cumulative % of depression 60 50 40 30 20 PD GAD AG SPP SoP no anxiety dx 10 0 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 By age of onset EDSP, 2001 Anxiety disorders are also associated with increased somatic morbidity: Comorbidity with selected physical conditions Physical Conditions No Anxiety Disorder n (%) Anxiety Disorder n (%) AOR (95% CI) significant reduction of health related QoL (SF-36) Cardiac diseases 88 (2.3) 18 (3.7) 1.79 (0.85-3.79) Respiratory diseases 191 (5.4) 43 (10.5) 1.71 (1.13-2.57)** X Gastrointestinal diseases 113 (2.9) 29 (7.4) 2.10 (1.24-3.54)** X Arthritic conditions 956 (24.6) 138 (32.0) 1.66 (1.24-2.21)** X Metabolic syndromes 279 (7.6) 38 (9.9) 1.56 (1.02-2.37)* X Allergic conditions 461 (12.3) 75 (18.1) 1.39 (1.00-1.95)* X Migraine headaches 271 (6.2) 72 (17.0) 2.12 (1.51-2.98)** X Thyroid diseases 340 (8.4) 68 (15.9) 1.59 (1.13-2.24)** Any past month physical condition 2295 (59.6) 315 (74.2) 1.70 (1.27-2.27)** X AOR: Odds Ratio adjusted for sociodemographic variables and comorbid depression and SUD Sareen et al. (subm.) 10
What is special in anxiety disorders? (cont.) treatment rates Treatment rates are extremely low in almost all mental disorders increase by degree of comorbidity By type of disorder anxiety By comorbidity 100% 100% 90% 90% 80% 80% 70% 60% 50% no treatment both only psychological only drug no consultation 70% 60% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% any mood disorder any anxiety disorder any alcohol disorder psychotic disorder 0% just one disorder more than one disorder ECNP-Task Force Report 2005 : Size and burden of Mental Disorders in the EU 11
and if treatment occurs it occurs predominantly very late! Cumulative lifetime probability of treatment contact in anxiety disorders Wang et al. (2005), Archives of General Psychiatry Summary (1): Special features of anxiety disorders High prevalence Early onset and persistence High comorbidity rates (with secondary mental disorders and with somatic disorders) Low rates of treatment Wittchen & Jacobi (2005), Neuropsychopharmacology Goodwin et al. (2005), Neuropsychopharmacology Fehm et al. (2005), Neuropsychopharmacology Lieb et al. (2005), Neuropsychopharmacology ESEMeD/MHEDEA 2000 Investigators (2004), Acta Psychiatrica Scandinavica 12
Burden indicators and cost estimates social impairment high utilizers total and hidden costs Almost invariably marked social impairment and disability in anxiety disorders By diagnostic definition and criteria anxiety disorders always imply social role impairment and clinically significant suffering Unlike to episodic disorders like depression, persistent (anticipatory anxiety) and special (panic attacks) features are associated with a pronounced long-term burden in terms of subjective suffering and disruption in social roles However, research has been slow to incestigate this in greater detail! A crude indicator applied across the majority of studies is number of disability (sick leave) days Also reduced work productivity (when at work) important factor for high indirect costs (Greenberg et al., 2001; Simon et al., 2000) 13
Example: Proportion of subjects with days lost, days impaired (or both) due to mental health problems in pure and comorbid 12-month GAD % subjects having any lost/impaired days 80 70 60 50 40 30 20 10 0 Days lost Days impaired Total lost/impaired 67,8 55,4 50,5 52,5 30,7 31,5 13 14,9 6,5 6,8 5,4 0,7 No GAD/No MDD GAD/No MDE MDD/No GAD GAD + MDD Diagnostic comparison groups Wittchen et al. (2002), International Clinical Psychopharmacology Total number of disability days in the past month in the population: Population attributable fraction of 12-month mental and anxiety disorders any somatic disorder 56% mental disorder other than anxiety 16% any anxiety disorder 25% PAF controlled for the presence of other types of disorder other 3% 14
High and overutilization Number of doctor visits in the past 12-months by diagnostic status: anxiety patients are high utilizers of health care resources no mental disorder 6,9 any anxiety disorder 12,4 any mental disorder 10,0 OCD 18,8 specific phobia 12,1 social phobia GAD agoraphobia 14,6 15,4 16,1 panic disorder 19,1 0 4 8 12 16 20 But only a small proportion of excess utilization rates can be explained by mental health care visits! The total estimated cost of brain disorders in Europe by disease area ( PPP billion) Andlin-Sobocki et al 2005, modified) All brain disorders: 386.176 billion Health care costs: 135.446 Direct non-medical 72.201 Indirect costs: 178.529 Mental disorders Health care costs: 110.061 Direct non-medical 51.673 Indirect costs: 132.985 mental disorders 294.719 billion neurological 83.934 Neurological disorders Health care costs: 21.286 Direct non-medical 20.259 Indirect costs: 42.389 Addiction 57.274 Health care costs: 16.655 Direct non-medical 3.962 Indirect costs: 36.657 Affective disorders 105.666 Health care costs: 28.639 Direct non-medical - NE- Indirect costs: 77.027 Anxiety disorders 41.373 Health care costs: 22.072 Direct non-medical -NE- Indirect costs: 19.301 Psychotic disorders 35.229 Health care costs: 29.885 Direct non-medical 5.374 Indirect costs: - NE- neurosurgical 7.523 billion Neurosurgical disorders Health care costs: 4.099 Direct non-medical 269 Indirect costs: 3.155 Note: under-estimation (especially indirect costs) 15
Despite past limitations and vast variation with regard to the relative contribution of cost components good concordance across studies Annual cost estimates for anxiety Rice & Miller (1996): 46 billion $ DuPont et al. (1996): 47 billion $ Greenberg et al. (1999): 42 billion $ (in 1998 costs: 63.1 billion $) Andlin-Sobocki et al. (2005): 41 billion Annual cost estimates for depression Rice & Miller (1996): 31 billion $ DuPont et al (1996): 44 billion $ Greenberg et al (1999): 53 billion $ Andlin-Sobocki et al. (2005): 105 billion (including bipolar disorders) The total health care and societal costs of anxiety disorders are roughly the same as for depression Additional effects of illness-related life course changes with adverse financial implications that have so far never been taken into account in cost studies! under-estimation of (especially indirect) costs / Further hidden costs of untreated anxiety disorders (e.g., Candilis & Pollack, 1997) Other indices not or only partially covered in these cost estimations: Subsequent unemployment (Etner et al., 1997; Leon et al., 1995; Yayakody et al., 1998) Work in under-payed jobs (Etner et al., 1997; Kessler & Greenberg, in press) Educational under-achievement (Kessler et al. 1995) Teen childbearing, marital timing and instability (Kessler et al. 1997, 1998) 16
Summary (2): The underestimated cost and burden of anxiety disorders burden as a function of prevalence x active time within an affected individual X cost per case Summary (2): The underestimated cost and burden of anxiety disorders prevalence: 1/4 of the population will suffer an anxiety disorder at least once in their lifetime, ~15% are affected in any given year active time within an affected individual: early onset, persistence cost: risk factor status, high degree of current and lifetime comorbidity cost (cont.): extremely high indirect costs and relatively low direct costs despite burden: large degree of unmet needs of patients with anxiety disorders (low treatment rates unless complex comorbid complications occur, considerably delayed treatment, particularly low treatment rates in adolescents and young adults) Limitations: incomplete data base with regard to prevalence/incidence in the elderly and in children, incomplete costs estimates, incomplete data for many countries, lack of data on sequential comorbidity, lack of data concerning burden 17
Summary (2): The underestimated cost and burden of anxiety disorders burden as a function of prevalence x active time within an affected individual x cost per case Avoidable burden? jacobi@psychologie.tu-dresden.de 18