 |
 |
|
|
|
Intentional drug overdose is one of the most common methods of attempted suicide (1,2). In 1997, intentional drug or chemical overdose was the third most commonly used method of suicide in British Columbia (74 cases, 15%), next to hanging or suffocation (154 cases, 32%) and using firearms or explosives (125 cases, 26%) (3). Many studies have identified the types of medications consumed in intentional drug overdoses, varying from barbiturates in the 1960s to tricyclic antidepressants (TCAs), other psychotropic medications, and over-the-counter (OTC) medications in the 1990s (1,2,4). Although most psychotropic medications can provide positive clinical outcomes if used appropriately, clinicians are often concerned about potential toxicity and lethality of these agents. This concern can result in underdosing or in inadequate treatment that may subsequently increase the risk of suicide (5). Recent reports, however, suggest that freely accessible OTC medications, especially acetaminophen, are among the most common and lethal agents used in overdose (4,6). There are well-established relations between suicide attempts and psychiatric conditions, especially with depression and borderline personality disorders (7–9). Even so, no data are available with respect to the association between the characteristics of those attempting suicide—particularly, a history of psychiatric diagnosis or personality disorder—and the type of medications used in overdose. Identifying these associations may help in developing strategies to decrease the incidence of intentional drug overdose and to assess the role of medication accessibility in the risk of overdose. For this reason, the focus of this study was to identify associations between the characteristics of those attempting suicide and the type of medication used; specifically, prescription vs OTC medications. MethodsSt Paul’s Hospital is a 450-bed acute care teaching hospital located in downtown Vancouver, British Columbia. To identify patient characteristics associated with the use of OTC medications in intentional overdose, we performed a retrospective chart review that included all patients who presented at St Paul’s Hospital from August 1, 1997, to July 31, 1998, with a discharge diagnosis of medication overdose. Patients were excluded if they had accidental poisoning or overdose or if they exclusively used alcohol, illicit drugs, and nondrug chemicals. Information that was recorded from the charts comprised patient demographics, types of medications used in the overdose, and presence of prescription medications at the time of overdose. In addition, the latest DSM-IV diagnosis written by a psychiatrist was noted. However, if this was unavailable, we used the latest diagnosis written on the chart by other medical- service individuals. Charts with no diagnosis were excluded. To ensure the consistency and quality of the diagnosis data collected, a psychiatrist reviewed the diagnosis of one-third of the 95 usable charts. Data Analysis We chose the use of any OTC medications for overdose as the dependent variable. Univariate analysis was carried out, comparing characteristics of patients who used any OTC medications in their suicide attempt with those who did not use OTC medications. We used Student’s t-test for independent samples for interval data and the chi-square test for nominal data. Variables exhibiting a P value of < 0.25 were selected for multivariate logistic regression analysis, using the approach described by Hosmer and Lemeshow (11). A liberal threshold was employed to avoid excluding variables from the multivariate analysis that might appear important once other variables were considered. The final multivariate model included only those variables with P < 0.05.
ResultsOf 333 patient charts with medication overdose over the study period, 180 were randomly selected. The sample size resulted from a 1-year limit on the study duration and the coinvestigators’ availability for chart review. Of the 180 charts, 85 were excluded, owing to incomplete data or lack of evidence indicating the patient met the diagnostic inclusion criteria, leaving 95 cases for analysis. In the study cohort, the patient mean (SD) age was 40 (13) years, and 57% were men. Most patients had a previous suicide (74%) or overdose (71%) attempt and required treatment in the emergency room only (53%). Most overdose attempts included multiple medications (71%) and were limited to prescription drugs only (71%). The most common type of prescription medication used for overdose was a benzodiazepine, and the most common type of OTC medication used for overdose was acetaminophen (Table 1). Patients typically overdosed with their own medications (85%), and most possessed prescription medications at the time of the overdose (76%). Concurrent use of alcohol was evident in 34% of the patients. Further, most patients had a psychiatric diagnosis, with the most common being depression (52%), and most had previously overdosed on medications (71%).
Between patients using OTC medications in overdose and those using exclusively prescription medications, univariate analysis identified potentially important differences in sex, age, and possession of prescription medications (Table 2). For this reason, these variables were controlled in the multivariate model that assessed the association between psychiatric diagnosis and the use of OTC medications in overdose. Because only 1 patient in the cohort met the DSM-IV criteria for anxiety disorders, this variable was not entered into the regression model. When differences in age, sex, and possession of prescription medications were controlled, a DSM-IV diagnosis of substance abuse was the only psychiatric diagnosis that was a significant independent predictor of OTC medication use in overdose (Table 3). Patients with a diagnosis of substance abuse were 11% as likely (89% less likely) to use OTC medications in overdose, compared with those without this diagnosis. Not surprisingly, possession of prescription medications also significantly decreased the likelihood of using OTC medications. Most patients in this cohort (82%) had a diagnosis of substance abuse (6%), possessed prescription medications (49%), or both (72%). No 2 variables in the regression model were highly correlated, with Pearson correlation coefficients ranging from 0.04 to 0.38. The high P value for the regression model’s Hosmer and Lemeshow goodness-of-fit (P = 0.92) indicates a reasonable fit of the data. Although not statistically significant in the multivariate model, younger patients appeared more likely to choose OTC medications for overdose. The mean age was significantly lower for those using OTC medications: 35 (SD 9) years vs 42 (SD 14) years (P = 0.01). Age did not correlate highly with possession of prescription medications (r = 0.11, P = 0.29) or with a diagnosis of substance abuse (r = 0.07, P = 0.52), suggesting that age itself may be important to consider when assessing the risk of OTC overdose.
DiscussionRecent studies have identified OTC medications as a common component in self-poisonings (1,4,12). Further, the incidence of acetaminophen overdose has increased dramatically since the early 1980s (1,12) and accounts for the largest number of deaths from drug overdose in some studies (6). Despite these statistics, OTC medications remain freely accessible to patients who have demonstrated suicidal behaviour. Indeed, most patients in this study had previous overdose attempts, including more than 73% of those who used OTC medications. Knowledge about which patients are more likely to choose OTC medications in overdose may alert clinicians to the risk of OTC overdose and could be used to develop strategies aimed at lowering the frequency of OTC medication overdose. The results of this study indicate that patients with a diagnosis of substance abuse or those who possessed prescription medications at the time of overdose may be less likely to choose OTC medications in their suicide-overdose attempts. The risk markers identified in this study appear logical and provide insight into patient situations that prompt the use of OTC medications in overdose. Patients with a history of substance abuse may pursue and stockpile, thus increasing availability of prescription medication. It should be noted, however, that the availability of prescription medications was controlled in the regression model and therefore would not completely explain why substance abuse was independently associated with a decreased risk of OTC use. There appear to be other reasons why individuals who suffer from substance abuse were less likely to choose OTC medications. Perhaps this reflects a misconception among this population that prescription medications are more lethal in overdose. Patients who do not possess prescription medications at the time of overdose would logically be more likely to use agents available without a prescription. Because most of the study population had a psychiatric diagnosis and had previously attempted suicide, it is conceivable that medications were not prescribed, owing to the perceived risk of prescription drug overdose. Although this cannot be proven from the data presented here, it reminds us that the availability of OTC medications is an important consideration in suicidal patients. Given the potential lethality of OTC medication overdose, assertive prescribing of indicated psychiatric medications in suicidal patients may be an appropriate way to limit suicide attempts in some patients. Still, there are several limitations to this study. First, the study population was limited to hospitalized patients and to those who were brought to the hospital emergency room. Consequently, completed suicide and suicide attempts that did not result in presentation to a hospital were not included in the analysis. Possibly, the risk markers for OTC medication use in overdose for these patients differ significantly from those in the study population. The study population, however, likely represents most hospitals located in the city core, with a significant proportion of patients having a diagnosis of substance abuse. In addition, data were collected by retrospective chart review and therefore may be less accurate than if collected prospectively. The subjectivity of psychiatric diagnosis is especially prone to intraindividual variability and is thus difficult to accurately assess retrospectively. To help limit this problem, a psychiatrist confirmed the diagnosis in one-third of the charts reviewed. Psychiatric diagnoses were grouped, owing to the small number of patients within each diagnostic subgroup. This may limit extrapolation of results to populations wherein the diagnostic mix differs substantially or to those wherein individual components of our diagnostic groupings (for example, mania) are particularly common. Finally, the study population was relatively small and was restricted to a single institution. Consequently, these results may not reflect the patterns of intentional drug overdose in the community at large. Results from 2 previous studies, however, suggest that similarities exist between our study population and other overdose cohorts. Stein and others (1) describe a sample of 103 hospitalized drug overdose patients in the US: 48% were men (57% in our sample), and 59% had a diagnosis of depression (52% in our sample). In an English study carried out by Crome, 48% of drug suicide cases reviewed were men (6). ConclusionSuicide-prone patients with a diagnosis of substance abuse who possess prescription medications are unlikely to use OTC medications in overdose. For this cohort, this represents a relatively small proportion of patients whom clinicians should consider to be at greater risk for OTC medication overdose. In particular, younger patients meeting these criteria may use OTC medications. These results could be used to develop strategies that aim to lower the frequency of suicides involving OTC medications. Clinicians may consider alerting family and health care professionals to the potential use of OTC medications in suicide-prone patients who meet these criteria. For these patients, limiting access to OTC medications may be appropriate. Funding and SupportThis research was carried out by the Pharmacy Department at St Paul’s Hospital, Vancouver, British Columbia, with no external funding. References1. Stein MD, Bonanno J, O’Sullivan PS. Changes in the pattern of drug overdoses. J Gen Intern Med 1993;8:179–84. 2. Bland RC, Newman SC, Dyck RJ. The epidemiology of parasuicide in Edmonton. Can J Psychiatry 1994;39:391–6. 3. Vital Statistics Agency of British Columbia. Detailed cause of death by gender and age. Appendix 2. Victoria (BC): British Columbia Ministry of Health; 1997. 4. Neeleman J, Wessely S. Drugs taken in fatal and non-fatal self-poisoning: a study in South London. Acta Psychiatr Scand 1997;95:283–7. 5. Teicher MH, Glod CA, Cole JO. Antidepressant drugs and the emergence of suicidal tendencies. Drug Saf 1993;8:186–212. 6. Crome P. The toxicity of drugs used for suicide. Acta Psychiatr Scand 1993;87:(Suppl 371):33–7. 7. Simpson SG, Jamison KR. The risk of suicide in patients with bipolar disorders. J Clin Psychiatry 1999;60(Suppl 2):53–6. 8. Angst J, Angst F, Stassen HH. Suicide risk in patients with major depressive disorder. J Clin Psychiatry 1999;60(Suppl 2):57–62. 9. Tondo L, Baldessarini RJ, Hennen J. Suicide attempts in major affective disorder patients with comorbid substance use disorders. J Clin Psychiatry 1999;60(Suppl 2):63–9. 10. SPSS 9.0. Chicago (IL): SPSS Inc. 11. Hosmer DW, Lemeshow S. Applied logistic regression. 1st ed. New York: John Wiley and Sons; 1989. 12. McLoone P, Crombie IK. Hospitalisation for deliberate self-poisoning in Scotland from 1981–1993: trends in rates and types of drugs used. Br J Psychiatry 1996;169:81–5. Author(s)Manuscript received April 2002 and accepted July 2002. Results of this research were previously presented at the American College of Clinical Pharmacy 2001 Annual Meeting; October 21–24, 2001; Tampa (FL). 1. Clinical Pharmacist, Department of Pharmacy, St Paul’s Hospital, Providence Health Care, Vancouver, British Columbia. 2. Research Coordinator, Department of Pharmacy, St Paul’s Hospital, Providence Health Care, Vancouver, British Columbia. 3. Clinical Pharmacist, Department of Pharmacy, St Paul’s Hospital, Providence Health Care, Vancouver, British Columbia. 4. Staff Psychiatrist, Department of Psychiatry, St Paul’s Hospital, Providence Health Care, Vancouver, British Columbia. 5. Associate Professor, Department of Public Health Sciences, University of Toronto and Division of Infectious Diseases, University Health Network, University of Toronto, Toronto, Ontario. Address for correspondence: Dr A Lo, Department of Pharmacy, St Paul’s Hospital, Providence Health Care, 1081 Burrard Street, Vancouver, BC V6Z 1Y6 e-mail: alo2@providencehealth.bc.ca
1 | 2
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||