Skip Navigation



Journal of Public Health Advance Access published online on January 31, 2008
Journal of Public Health, doi:10.1093/pubmed/fdm085
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
30/1/60    most recent
fdm085v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Morgan, O.
Right arrow Articles by Majeed, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Morgan, O.
Right arrow Articles by Majeed, A.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007, Published by Oxford University Press on behalf of Faculty of Public Health. All rights reserved

Antidepressant prescribing and changes in antidepressant poisoning mortality and suicide in England, 1993–2004


Oliver Morgan, Honorary Research Associate1,2
Clare Griffiths, Principal Research Officer2,
Azeem Majeed, Chair of Primary Care and Social Medicine and Head of Department1

1 Department of Primary Care and Social Medicine, Faculty of Medicine, Imperial College London, London SW7 2AZ, UK
2 Office for National Statistics, UK


Address correspondence to Oliver Morgan, E-mail: omorgan{at}bigfoot.com

Background In England, the impact of increased use of antidepressant medications is unclear. We examine associations between antidepressant use, suicide and antidepressant poisoning mortality, adjusted for important covariates.

Methods Data on suicide and antidepressant poisoning mortality were provided by the Office for National Statistics. Prescription data were provided by the Department of Health. Age- and sex-specific prescribing rates were estimated from The Health Improvement Network primary care data. We measured the association between prescribing, suicide and poisoning mortality after adjusting for age, sex, calendar year, prescribing rates and use of newer antidepressants drugs.

Results The prevalence of antidepressant treatment increased during the 1990s for all age and sex groups. Treatment prevalence remained constant from 2002 but declined among children and adolescents. Between 1993 and 2004, age-standardized rates for suicide decreased from 98.2 to 81.3 per million populations and for antidepressants from 9.2 to 7.4 per million populations. Before adjustment, increased antidepressant prescribing was associated with a decrease in suicide (rs = –0.90, P < 0.001) and antidepressant poisoning mortality rates (rs = –0.65, P = 0.023). This association disappeared after adjustment.

Conclusion In England, at a population level, there does not appear to be an association between antidepressant prescribing and antidepressant poisoning mortality or suicide.

Keywords: antidepressant, poisoning, mortality, prescribing


   Introduction
TOP
 Introduction
Methods
 Results
 Discussion
 Conclusion
 Funding
 REFERENCES
 
In response to concerns about under-diagnosis and suboptimal management of depression, the ‘Defeat Depression’ campaign was launched in 1992 by the Royal College of General Practitioners and Royal College of Psychiatrists.1,2 Since then, there has been a rapid increase in the use of antidepressant medications.3 More recently, the National Institute for Clinical Excellence (NICE) has published evidence-based guidelines that recommend a more limited use of pharmacological interventions.4

Previous studies suggest that the impact of increased antidepressant prescribing has been a reduction in suicides and antidepressant poisoning mortality. However, not all studies found such an association and few adjusted for important potential confounding factors. In this study, we use primary care prescription records to estimate age- and sex-specific prescribing trends, and measure the association between antidepressant prescribing and suicide and antidepressant poisoning mortality, adjusted for important covariates.


   Methods
TOP
 Introduction
 Methods
Results
 Discussion
 Conclusion
 Funding
 REFERENCES
 
Mortality data
Suicide deaths were identified using the International Classification of Diseases codes shown in Table 1. Suicide deaths are defined where the coroner gave a verdict of suicide or where an open verdict was reached in a death from injury or poisoning. Most open verdicts are cases where the harm was self-inflicted but there was insufficient evidence to prove that the deceased deliberately intended to kill themselves.5 Open verdicts account for ~30% of male and 40% of female suicide deaths.5


View this table:
[in this window]
[in a new window]

  		Table 1 ICD-9 and ICD-10 classification for suicides and drug poisoning deaths

 
Drug poisoning mortality data for England and Wales from 1993 onwards are stored by the Office for National Statistics (ONS) in a dedicated database.6 The database contains data on cause of death, individual characteristics (including age and sex) and textual information supplied voluntarily by coroners to ONS. The textual information provides additional details about the drugs involved, including legal and illegal drugs, prescription drugs and over-the-counter medications. Drug poisoning deaths were defined using the International Classification of Diseases codes as shown in Table 1. Antidepressant deaths were defined as those where an antidepressant drug was mentioned on the death certificate, with or without mention of alcohol or other drugs. A sensitivity analysis showed that excluding deaths with mentions of alcohol, other drugs or both, had little effect on the results. Antidepressant poisoning deaths were further categorized to their British National Formulary (BNF) categories (Table 2).


View this table:
[in this window]
[in a new window]

  		Table 2 BNF categories for antidepressant drugs

 
Prescription data
The Department of Health provided data on prescriptions for antidepressant drugs in England between 1993 and 2004. Prescription statistics include all drugs dispensed in the community in England, even if the prescription was written in Scotland, Wales or Northern Ireland. This includes prescriptions written by community pharmacists, dispensing doctors and prescriptions submitted by doctors for items personally administered (i.e. given by the doctor during a consultation). The data does not include prescriptions dispensed in hospitals. Department of Health prescription data is likely to provide a relatively complete picture of antidepressant prescribing in the UK as 90–98% of patients with depression are treated in primary care.7 We converted prescription data for each antidepressant drug to Average Daily Quantities (ADQs). ADQ is the assumed average maintenance dose per day for a drug used for its main indication in adults.8 It is analogous to the Defined Daily Dose9 and is based on prescribing patterns in England, and therefore better reflects prescribing practice in this country. ADQs are calculated as:


Formula

Department of Health prescription statistics are only available as aggregated data. To estimate the proportion of all prescriptions within different age and sex groups, we used data from English general practices in The Health Improvement Network (THIN) primary care dataset. For each year between 1993 and 2004, we identified patients who were diagnosed with depression and who received an antidepressant drug. Depression was defined using Read codes as specified by the ONS (Internet supplementary document).10

Analysis
For each year, we calculated the prevalence of treated depression for individuals in each age- (0–14, 15–24, 25–34, 35–44, 45–54, 55–64, 65–74, 75+ ) and sex-specific group. For all antidepressants and for tricyclic antidepressants (TCAs, BNF 4.3.1), selective serotonin re-uptake inhibitors (SSRIs, BNF 4.3.3) and antidepressant drugs in the Other category (BNF 4.3.4) separately, we summed the total quantity of drugs prescribed to individuals with a diagnosis of depression in each year and calculated the proportion of the ADQ total in the age- and sex-specific groups for each year. By multiplying national-level data on ADQs by the proportion of ADQs in each age- and sex-specific group within the THIN practices, we estimated the national annual ADQ within each sub-group and between 1993 and 2004.

We calculated Spearman's rank correlation coefficient to assess whether suicides or antidepressant poisoning mortality were associated with antidepressant prescribing, with no adjustment for age and sex.11 Both suicide deaths and antidepressant poisoning deaths were expressed as directly age-standardized rates per million population (standardized to the European population). Antidepressant prescribing was expressed as ADQs per person per year. In addition to considering the association between suicides rates and prescribing rates for all antidepressant drugs combined, we considered all antidepressant prescriptions and poisonings, as well as poisoning and prescription rates for TCAs and SSRIs. As the number of deaths involving other antidepressants was small, we combined this with SSRIs to make an additional category ‘SSRI & Other’ which we assessed in a similar way. Because drugs other than antidepressants are mentioned more commonly for deaths involving SSRIs than TCAs, we also conducted a sensitivity analysis whereby we calculated Spearman's rank correlation coefficient for each drug poisoning mortality and prescribing category, excluding deaths where multiple drugs were mentioned.

To consider other covariates in addition to antidepressant prescribing, we used a Poisson regression model for suicide and antidepressant poisoning mortality.12 The Poisson regression model had counts of deaths as the dependent variable and the main independent variable was antidepressant prescribing rate expressed as deciles of ADQ rate (ADQs per person). Covariates included year (1993–2004), sex (male, female), age group (0–14, 15–24, 25–34, 35–44, 45–54, 55–64, 65–74, 75 + ) and the proportion of prescribed antidepressants that were SSRIs, expressed as deciles. A key assumption underlying Poisson regression is that at each level of the covariates, the variance of the dependent variable is equal to its mean. We found that this was not the case for either suicides or poisoning deaths. The mean was consistently smaller than the variance. This is referred to as overdispersion that results in P-values that are too small. One solution to this problem is to use a negative binomial regression model, which is similar to the Poisson regression model, except that it allows for the effect of overdispersion.13 Although the point estimates of the model coefficients remain similar, the P-values will tend to be larger. All statistical analysis was done using STATA 8.2.


   Results
TOP
 Introduction
 Methods
 Results
Discussion
 Conclusion
 Funding
 REFERENCES
 
The prevalence of antidepressant treatment in general practices included in the THIN dataset is shown in Fig. 1. In all age groups (except 0–14 years), treatment with antidepressants was more common among females than males. For both sexes and all age groups, prevalence of antidepressant treatment increased between 1993 and 2004. Among males, the highest treatment prevalence was among ages 35–44, 45–54 and 75+ . Males age 25–34 and 65–74 had similar, and slightly lower treatment prevalence. Among adult patients, the 15–24-year olds had the lowest treatment prevalence. Unlike other age groups, there was a decline in treatment prevalence in 2004. For females, the general pattern was similar, with the highest treatment prevalence in 35–44- and 45–54-year olds. Women 25–34, 65–74 and 75+ years had a slightly lower treatment prevalence. Also, women 15–24 had the lowest prevalence of antidepressant treatment. There was a notable decline in treatment prevalence in the 15–24 and 25–34-year olds from 2002, which was more pronounced than the decline in males.


Figure 1
View larger version (26K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 Prevalence of antidepressant treatment per 1000 patients registered with a General Practitioner by age group and sex, England, 1993–2004.

 
Prevalence by drug class for males and females is shown in Fig. 2. For TCAs, the prevalence of treatment declined for both males and females; the decline was similar for all age groups. SSRI prescribing increased for all age groups, but the increase was more pronounced among females, who were far more likely to receive an SSRI than males. Similarly, the increase in treatment prevalence for other antidepressants was greater among females. There has been a fall in SSRI treatment prevalence among 15–24- and 25–34-year olds since 2002, which was greatest for females but which also occurred to a lesser degree among males in the same age groups. For those receiving other antidepressants, there was a smaller decrease for males and females aged 15–24 only after 2003. Prevalence among 0–14-year olds was much lower than other ages. The prevalence of children receiving TCAs decreased between 1993 and 2004 from 0.64 to 0.15 per 1000. However, for SSRIs, prevalence increased between 1993 and 2002 from 0.13 to 1.13 per 1000, after which it declined to 0.43 per 1000 in 2004. Similarly, for other antidepressants, prevalence in this age group increased from 0 to 0.13 in 2002, declining to 0.04 in 2004.


Figure 2
View larger version (44K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 Prevalence of antidepressant treatment per 1000 patients registered with a General Practitioner by age group, sex and drug class, England, 1993–2004.

 
Table 3 shows age-standardized suicide and antidepressant poisoning rates per million population and the rate of antidepressant prescribing expressed as ADQs per person. Suicide rates by all methods decreased from 98.2 to 81.3 per million population between 1993 and 2004. There was a similar downward trend in non-drug poisoning suicide. For all antidepressant poisoning deaths there was a slight increase in 1996 and 1997, followed by a decline until 2004. This was similar for TCAs, but not for SSRIs and SSRIs & Other combined, which increased steadily throughout the study period. Prescribing rates for all antidepressants increased from 6.7 to 18.6 ADQs per person. For TCAs there was a slight decrease in prescribing rates, whereas SSRIs and SSRIs & Others combined increased sharply.


View this table:
[in this window]
[in a new window]

  		Table 3 Age-standardized mortality rates for suicide and antidepressant-related poisoning deaths and prescriptions of antidepressant drugs, England, 1993–2004

 
Spearman's rank correlation coefficient between all suicides and prescriptions for all antidepressants was rs = –0.90 (first row of Table 4). This indicates that as antidepressant prescribing increased, the suicide rate decreased. The 95% CI indicates that the correlation could be as strong as –0.97 (almost perfect agreement), or as low as –0.68, which would still be considered a substantial correlation. The P-value (P = 0.0001) indicates that this finding is not likely to be due to chance. The final column of the table indicates that over 80% of the variation in suicide rates over the period 1993–2004 can be explained by changes in prescribing rates. A similar correlation was also observed between non-drug poisoning suicides and all antidepressant prescriptions (second row). Antidepressant prescribing (all drugs combined) was less well correlated with antidepressant poisoning deaths (third row, rs = –0.65). The wide 95% CI indicates considerable uncertainty around the estimate of correlation and prescribing trends only explained 42% of the variation in poisoning mortality.


View this table:
[in this window]
[in a new window]

  		Table 4 Spearman's rank correlation coefficient (rs) for directly age-standardized suicide and antidepressant poisoning rates and prescription rates, England, 1993–2004

 
Unlike all antidepressants combined, which showed a negative relationship with antidepressant poisoning, increased TCA prescribing was (positively) associated with higher poisoning rates (rs = 0.81), explaining two-thirds of the variation. Similarly, higher prescribing rates for SSRIs and SSRI and others combined were also positively associated with higher poisoning rates. Restricting the analysis of antidepressant poisoning mortality to deaths where only an antidepressant was mentioned on the death certificate did not substantially alter the correlations (shown in the lower half of Table 4).

The result of the negative binomial model is shown in Table 5. In the model for suicide, increasing decile of ADQ was associated with a small decrease in suicides, but this was not statistically significant (IRR = 0.942, P = 0.682). Year was not associated with suicide (IRR = 1.008, P = 0.95), but the rate of suicide was lower among females (IRR = 0.323, P < 0.0001). There was a small increase in suicide rate with older age (IRR = 1.012, P = 0.02). An increasing proportion of prescribed antidepressants that were SSRIs caused a notable increase in suicide rate, although this was not statistically significant (IRR = 2.303, P = 0.331). For antidepressant poisoning deaths, ADQ rate was associated with a small increase in poisoning deaths, but was not statistically significant (IRR = 1.017, P = 0.924). None of the other covariates showed strong or statistically significant associations with poisoning mortality.


View this table:
[in this window]
[in a new window]

  		Table 5 Incident rate ratios for antidepressant prescribing and suicide and poisoning mortality, England, 1993–2004

 

   Discussion
TOP
 Introduction
 Methods
 Results
 Discussion
Conclusion
 Funding
 REFERENCES
 
Main findings
In England between 1993 and 2004, there was a rapid increase in SSRI prescribing. This is similar to trends reported from other countries in previous studies.14 A similar, but more gradual increase also occurred for drugs within the group of other antidepressants. Despite the large increase in prescriptions for antidepressants from these classes, the quantity of TCAs prescribed remained relatively constant. SSRIs and other antidepressants do not appear to have replaced TCAs but instead been used to increase the volume of prescribing of antidepressants.

Increased prescribing of antidepressants was statistically associated with reduced mortality rates due to all suicides as well as non-poisoning suicides. Antidepressant poisoning mortality also decreased, although the evidence for a statistical association with prescribing rates was weaker. The reversal in the direction of association between prescribing and drug poisoning mortality when drug class subgroups (TCAs, SSRIs and Other antidepressants) are considered, a phenomenon often referred to as Simpson's Paradox,15 is due to the influence of confounding by drug class on the association between prescriptions and deaths for all antidepressants combined. After adjusting for year, sex, age and the proportion of prescriptions for SSRIs, there was no evidence to support the hypothesis that antidepressant prescribing was associated with suicide or antidepressant poisoning rates. Only in the model for suicide was female sex and older age associated with suicide rates, which concurs with previously reported findings about the epidemiology of suicide.5

What is known already
Several previous studies reported associations between increased prescribing of antidepressant drugs and decreases in suicides and antidepressant poisoning deaths.16–31 Few of these studies were able to adjust for the potentially confounding effect of age and sex due to the lack of subgroup-specific data on antidepressant prescribing.

What this study adds
It seems reasonable that increased diagnosis and treatment of depression in the community would lead to reduced antidepressant poisoning deaths and suicides. However, even among individuals suffering depression, overdose and suicide are relatively rare32 and only a small proportion of depressed individuals may be at risk of fatal self harm.33 Although increased diagnosis and treatment may provide important gains in other dimensions such as quality of life, there may be a number of reasons why the impact on antidepressant poising deaths and suicide may be more limited. First, it is uncertain that non-recognition of depression is clinically significant.34–36 Secondly, the efficacy of antidepressant drugs, which are the mainstay of treatment for depression, has been difficult to quantify.4 Thirdly, there is an uncertainty about the efficacy of antidepressant drugs in the routine clinical setting.37 As many as 29–46% of depressed patients show partial or no response to antidepressant drugs38 and 30–60% of patients to not take their medication as prescribed,39 possibly due to side effects and misconceptions about the addictive properties of antidepressant drugs.40 Finally, increased diagnosis and treatment may not achieve reductions in antidepressant poisoning deaths and suicide as more than half the patients treated for depression in primary care have mild depression, for which efficacy of antidepressant drugs is unproven and is thought to be lower than for individuals with severe depression.41

The large increase in antidepressant prescriptions throughout the 1990s and early 2000s probably reflects increased diagnosis, access and provision of mental health care services in primary care following the Defeat Depression Campaign and government policy initiatives.1 Increased popularity of newer antidepressant drugs among GPs, mainly due to their better safety profile and fewer side effects, may have also contributed to this increase.42 Although prevalence of treated depression in primary care increased in all age and sex groups, there was a decline among younger age groups and children after 2002. This is probably due to the MHRAs recommendation not to treat children with SSRIs and Venlafaxine and to be cautious when using these drugs among young adults.43

Simple correlation analysis suggested that at a population level, increased antidepressant prescribing was associated with reduced rates of suicide and antidepressant poisoning mortality. This was similar to the majority of previous studies.14 However, after adjusting for age and sex in a multivariable regression model, we found no association. This was also reported by Dahlberg and Lundin in Sweden.44

Limitations of this study
This study reports ecological associations between antidepressant prescribing, suicide and antidepressant poisoning mortality. Although unable to infer causality, such analysis can make an important contribution to the body of evidence about the population-level impact of antidepressant prescribing.45

Few previous studies considered the effects of age and sex in their analysis.14 We were able to estimate the prevalence of treated depression and proportion of prescriptions in different age and sex groups by projecting prescribing patterns recorded in a primary care dataset. Using this information, we could include prescribing rates by age and sex in a multivariable regression model of suicide and antidepressant poisoning mortality. Although this approach was relatively simple, it assumes that all prescriptions for antidepressants in the national dataset were for the treatment of depression. However, as much as 30% of prescriptions for some antidepressant drugs may be for non-psychiatric indications, such as pain relief or insomnia.46

In an earlier paper, we showed that poisoning deaths involving SSRIs were more likely to be given a coroner's verdict of accidental death.47 Our definition of non-poisoning deaths excluded those with a verdict of accidental death. However, it is possible that some of these deaths may have actually been suicides48 and that such misclassification may have been associated with an individual's use of a particular class of antidepressant.

The regression models in this study were limited to individual-level covariates. However, early studies by Durkheim showed that societal characteristics also influence suicide.49 A number of such contextual factors have been studied and reviewed by Agerbo et al.50 These include social fragmentation and disintegration, religious beliefs, socio-economic position, unemployment and area deprivation.


   Conclusion
TOP
 Introduction
 Methods
 Results
 Discussion
 Conclusion
Funding
 REFERENCES
 
As antidepressants are both a treatment and method of suicidal behaviour, the epidemiology of antidepressant prescribing, antidepressant poising mortality and suicide is complex. There has been a large increase in prescriptions for antidepressant drugs and the prevalence of treated depression in primary care. This has not been reflected by a large fall in antidepressant poisoning deaths or suicides. However, concerns that poisoning deaths would increase have also not materialised. In England, there does not appear to be an association between antidepressant prescribing and antidepressant poisoning mortality or suicide.


   Funding
TOP
 Introduction
 Methods
 Results
 Discussion
 Conclusion
 Funding
REFERENCES
 
This work was supported by a bursary from EPIC and The Health Improvement Network (THIN). We are grateful to Mary Thompson at EPIC who helped with us define the extract of primary care data.


   REFERENCES
TOP
 Introduction
 Methods
 Results
 Discussion
 Conclusion
 Funding
 REFERENCES
 

  1. Paykel ES, Priest RG. Recognition and management of depression in general practice: consensus statement. BMJ (1992) 305(6863):1198–202.[Free Full Text]
  2. Paykel ES, Tylee A, Wright A, et al. The defeat depression campaign: psychiatry in the public arena. Am J Psychiatry (1997) 154(6 suppl):59–65.[Abstract]
  3. Gunnell D, Ashby D. Antidepressants and suicide: what is the balance of benefit and harm. BMJ (2004) 329(7456):34–8.[Free Full Text]
  4. National Institute for Clinical Excellence. Depression: Management of Depression in Primary and Secondary Care – NICE Guidance (2004) London: NICE.
  5. Brock A, Griffiths C. Trends in suicide by method in England and Wales 1979 to 2001. Health Stat Q (2001) 20:7–18.[Medline]
  6. Christophersen O, Rooney C, Kelly S. Drug-related mortality: methods and trends. Popul Trends (1998) (93):29–37.
  7. Henry JA, Alexander CA, Sener EK. Relative mortality from overdose of antidepressants. BMJ (1995) 310(6974):221–4.[Abstract/Free Full Text]
  8. Walley T, Roberts D. Average daily quantities: a tool for measuring prescribing volume in England. Pharmacoepidemiol Drug Saf (2000) 9:55–8.[CrossRef][Web of Science][Medline]
  9. WHO Collaborating Centre for Drug Statistics Methodology. The DDD – definition and principle. (2005) http://www.whocc.no/atcddd/.
  10. Office for National Statistics. Key Health Statistics from General Practice 1998. Series MB6 No.2 (2000) London: HMSO.
  11. Altman D. Practical Statistics for Medical Research (1999) London: Chapman & Hall/CRC.
  12. Woodward M. Epidemiology. Study Design and Data Analysis (1999) London: Chapman & Hall/CRC.
  13. Greenland S. Introduction to Regression Modeling. In: Modern Epidemiology—Rothman K, Greenland S, eds. (1998) Second ed. Philidelphia: Lippincott Williams & Wilkins. Chapter 21.
  14. Morgan O. Population-level trends in antidepressant prescribing and changes in fatal antidepressant poisoning and suicide: a systematic review. J Public Health (2007) submitted for publication (linked-paper).
  15. Pearl J. Causality. Models, Reasoning and Inference (2000) Cambridge: Cambridge University Press.
  16. Morgan OW, Griffiths C, Majeed A. Association between mortality from suicide in England and antidepressant prescribing: an ecological study. BMC Public Health (2004) 4:63.[CrossRef][Medline]
  17. Barak Y, Aizenberg D. Association between antidepressant prescribing and suicide in Israel. Int Clin Psychopharmacol (2006) 21(5):281–4.[CrossRef][Web of Science][Medline]
  18. Carlsten A, Waern M, Ekedahl A, et al. Antidepressant medication and suicide in Sweden. Pharmacoepidemiol Drug Saf (2001) 10(6):525–30.[CrossRef][Web of Science][Medline]
  19. Gibbons RD, Hur K, Bhaumik DK, et al. The relationship between antidepressant prescription rates and rate of early adolescent suicide. Am J Psychiatry (2006) 163(11):1898–904.[Abstract/Free Full Text]
  20. Gibbons RD, Hur K, Bhaumik DK, et al. The relationship between antidepressant medication use and rate of suicide. Arch Gen Psychiatry (2005) 62(2):165–72.[Abstract/Free Full Text]
  21. Grunebaum MF, Ellis SP, Li S, et al. Antidepressants and suicide risk in the United States, 1985–1999. J Clin Psychiatry (2004) 65(11):1456–62.[Web of Science][Medline]
  22. Guaiana G, Andretta M, Corbari L, et al. Antidepressant drug consumption and public health indicators in Italy, 1955 to 2000. J Clin Psychiatry (2005) 66(6):750–5.[Web of Science][Medline]
  23. Hall WD, Mant A, Mitchell PB, et al. Association between antidepressant prescribing and suicide in Australia, 1991–2000: trend analysis. BMJ (2003) 326(7397):1008.[Free Full Text]
  24. Isacsson G. Suicide prevention–a medical breakthrough? Acta Psychiatr Scand (2000) 102(2):113–7.[CrossRef][Web of Science][Medline]
  25. Kelly CB, Ansari T, Rafferty T, et al. Antidepressant prescribing and suicide rate in Northern Ireland. Eur Psychiatry (2003) 18(7):325–8.[CrossRef][Web of Science][Medline]
  26. Ludwig J, Marcotte DE. Anti-depressants, suicide, and drug regulation. J Policy Anal Manage (2005) 24(2):249–72.[CrossRef][Web of Science][Medline]
  27. Milane MS, Suchard MA, Wong ML, et al. Modeling of the temporal patterns of fluoxetine prescriptions and suicide rates in the United States. PLoS Med (2006) 3(6).
  28. Ohberg A, Vuori E, Klaukka T, et al. Antidepressants and suicide mortality. J Affect Disord (1998) 50(2–3):225–33.[CrossRef][Web of Science][Medline]
  29. Olfson M, Shaffer D, Marcus SC, et al. Relationship between antidepressant medication treatment and suicide in adolescents. Arch Gen Psychiatry (2003) 60(10):978–82.[Abstract/Free Full Text]
  30. Reseland S, Bray I, Gunnell D. Relationship between antidepressant sales and secular trends in suicide rates in the Nordic countries. Br J Psychiatry (2006) 188:354–8.[Abstract/Free Full Text]
  31. Rihmer Z, Belso N, Kalmar S. Antidepressants and suicide prevention in Hungary. Acta Psychiatr Scand (2001) 103(3):238–9.[CrossRef][Web of Science][Medline]
  32. Jick SS, Dean AD, Jick H. Antidepressants and suicide. BMJ (1995) 310(6974):215–8.[Abstract/Free Full Text]
  33. Simon GE, Savarino J, Operskalski B, et al. Suicide risk during antidepressant treatment. Am J Psychiatry (2006) 163(1):41–7.[Abstract/Free Full Text]
  34. Moncrieff J, Wessely S, Hardy R. Active placebos versus antidepressants for depression. Cochrane Database Syst Rev (2004) (1). CD003012.
  35. Ormel J, Tiemens B. Recognition and treatment of mental illness in primary care. Towards a better understanding of a multifaceted problem. Gen Hosp Psychiatry (1995) 17(3):160–4.[CrossRef][Web of Science][Medline]
  36. Peveler R, Kendrick T. Treatment delivery and guidelines in primary care. Br Med Bull (2001) 57:193–206.[Abstract/Free Full Text]
  37. Anderson IM. Selective serotonin reuptake inhibitors versus tricyclic antidepressants: a meta-analysis of efficacy and tolerability. J Affect Disord (2000) 58(1):19–36.[CrossRef][Web of Science][Medline]
  38. Muller-Oerlinghausen B, Berghofer A. Antidepressants and suicidal risk. J Clin Psychiatry (1999) 60(Suppl 2):94–9. Discussion 111–6.[Web of Science][Medline]
  39. Demyttenaere K. Compliance during treatment with antidepressants. J Affect Disord (1997) 43(1):27–39.[CrossRef][Web of Science][Medline]
  40. Priest RG, Vize C, Roberts A, et al. Lay people's attitudes to treatment of depression: results of opinion poll for Defeat Depression Campaign just before its launch. BMJ (1996) 313(7061):858–9.[Abstract/Free Full Text]
  41. Anderson IM, Nutt DJ, Deakin JF. Evidence-based guidelines for treating depressive disorders with antidepressants: a revision of the 1993 British Association for Psychopharmacology guidelines. British Association for Psychopharmacology. J Psychopharmacol (2000) 14(1):3–20.[Abstract/Free Full Text]
  42. Diab T, Singh AN. The new generation of antidepressants. Hosp Med (2003) 64(1):12–5.[Web of Science][Medline]
  43. Expert Committee on the Safety of Selective Serotonin Reuptake Inhibitors (SSRIs). Report of the CSM Expert Committee on the Safety of Selective Serotonin Reuptake Inhibitor Antidepressants (2004) London: MHRA.
  44. Dahlberg M, Lundin D. Antidepressants and the Suicide Rate: Is there really a connection? (2005) Vol. 5. Department of Economics, Uppsala Universitet: Working Paper 2005.
  45. Rothman K, Greenland S. Causation and causal inference. In: Modern Epidemiology—Rothman K, Greenland S, eds. (1998) Philadelphia: Lippincott Williams & Wilkins.
  46. Lawrenson RA, Tyrer F, Newson RB, et al. The treatment of depression in UK general practice: selective serotonin reuptake inhibitors and tricyclic antidepressants compared. J Affect Disord (2000) 59(2):149–57.[CrossRef][Web of Science][Medline]
  47. Morgan O, Griffiths C, Baker A, et al. Fatal toxicity of antidepressants in England and Wales, 1993–2002. Health Stat Q (2004) (23):18–24.
  48. O'Donnell I, Farmer R. The limitations of official suicide statistics. Br J Psychiatry (1995) 166(4):458–61.[Abstract/Free Full Text]
  49. Durkheim E. Suicide (originally published 1897) (1952) London: Routledge and Kegan Paul.
  50. Agerbo E, Sterne J, Gunnell D. Combining individual and ecological data to determine compositional and contextual socio-economic risk factors for suicide. Soc Sci Med (2007) 64(2):451–61.[CrossRef][Web of Science][Medline]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
30/1/60    most recent
fdm085v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Morgan, O.
Right arrow Articles by Majeed, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Morgan, O.
Right arrow Articles by Majeed, A.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?