Skip to main content
Effective Health Care Program
Home » Products » Ventricular Arrhythmias and Cerebrovascular Events in the Elderly Using Conventional and Atypical Antipsychotic Medications » Ventricular Arrhythmias and Cerebrovascular Events in the Elderly Using Conventional and Atypical Antipsychotic Medications

Ventricular Arrhythmias and Cerebrovascular Events in the Elderly Using Conventional and Atypical Antipsychotic Medications

Research Report Archived

Note: This report is greater than 5 years old. Findings may be used for research purposes but should not be considered current.

Author affiliations

Philip S. Wang, M.D., Dr.P.H. *, **

Sebastian Schneeweiss, M.D., Sc.D. **

Soko Setoguchi, M.D., Sc.D. **

Amanda Patrick, M.S. **

Jerry Avorn, M.D. **

Helen Mogun, M.S. **

Niteesh Choudhry, M.D., Ph.D. **

M. Alan Brookhart, Ph.D **

*Division of Services and Intervention Research, National Institute of Mental Health.

**Department of Psychiatry and Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School.


The US Food and Drug Administration (FDA) issued an Advisory in 2005 that atypical antipsychotic medications (APMs) significantly increased the risk of death vs. placebo in 17 short-term randomized controlled trials among elderly with dementia.1 “Black box” warnings were added to labels of all atypical APMs (i.e., aripiprazole, clozapine, olanzapine, quetiapine, risperidone, and ziprasidone) describing these risks and advising that these agents are not approved for use in elderly patients with dementia. The Advisory did not cover conventional APMs (e.g., phenothiazines and butyrophenones) due to insufficient data on the mortality associated with them.1,2

However in the absence of warnings for conventional APMs, concern has mounted that clinicians may simply switch elderly patients to these older agents,3 particularly because their use had until recently been widespread.4 Extrapolating mainly from studies in younger populations, some have suggested that conventional APMs could pose risks greater than those of the newer drugs in older populations.5-8 In a recent observational study of elderly beginning use of APMs, patients prescribed conventional agents had a 37% greater, dose-dependent risk of short-term mortality than those prescribed atypicals.9 A recent meta-analysis of randomized trials among elderly with dementia found the conventional agent, haloperidol, increased short-term mortality vs. placebo by 107%, a risk greater than that for atypical agents.10 Another observational study found higher mortality in those given haloperidol vs. two atypical drugs (risperidone or olanzapine).11

An important next step in understanding whether conventional APMs truly pose greater hazards than atypical agents is to investigate potential mechanisms through which they may act. In the FDA’s analysis, heart-related events (heart failure and sudden death) and infections (mostly pneumonia) accounted for many deaths.1 Furthermore, anticholinergic properties affecting blood pressure and heart rate, Q-T prolongation causing conduction delays, as well as sedation and extrapyramidal symptoms causing potential swallowing problems, are all more common with conventional than atypical agents.5-8,12 For these reasons, cardiac (e.g., myocardial infarction and ventricular arrhythmias), cerebrovascular (e.g., stroke and transient ischemic events), and infection (e.g., aspiration pneumonia) outcomes may all be potential mediators of an increased risk of death from conventional vs. atypical agents.

The aims of the current study were to examine whether elderly newly started on conventional vs. atypical APMs, have greater risks of cardiac, cerebrovascular, and infection outcomes. Reasons for using both drug groups (e.g., dementia, delirium) may themselves be risk factors for these outcomes so we restricted analyses to only patients given an APM. We also restricted analyses to just new users to guard against selection bias among prevalent users from early symptom emergence, drug intolerance, or treatment failure.13 To control for potential differences in characteristics of those prescribed different APMs, we used traditional multivariable and propensity score adjusted Cox proportional hazards models14 as well as instrumental variable (IV) estimation.15-18 Because we observed previously9 that the relative risk of death from conventional vs. atypical APMs may not be uniform over time, we separately examined the hazards of developing conditions within 30, 60, and 120 days of initiating an APM. Identifying potential mechanisms is critical for answering whether conventional agents truly pose greater hazards than atypicals and should not simply replace the latter drugs stopped in response to recent FDA warnings.3


Data Sources

Pennsylvania Pharmaceutical Assistance Contract for the Elderly (PACE) program

Information from PACE, the largest US state prescription benefits program for the elderly, was available from 1/1/94-12/31/03.19 PACE has no deductibles or maximum annual benefit and charges a modest copayment of $6 for each prescription. The income ceiling for eligibility is $14,000 if single and $17,200 for a couple, resulting in a recipient population of both indigent and near-poor elderly. These generous benefits and requirements for financial need result in essentially no out-of-pocket (i.e., out-of-system) medication use.

Pennsylvania Medicare

Medicare data included both Part A (hospitalizations and nursing home stays) and Part B (outpatient services and procedures) on all PACE enrollees during 1/1/94-12/31/03. Medicare data on mortality were drawn from the Death Master File, which undergoes extensive verification and weekly updates by the Social Security Administration.20

We assembled data on all filled prescriptions, procedures, physician encounters, hospitalizations, and long-term care into a relational database. All traceable person-specific identifiers were transformed into anonymous, coded study numbers to protect subjects’ privacy. This study was approved by the Brigham and Women’s Hospital IRB.

Study Population

All individuals were >65 and filled a first recorded (index) prescription for an oral APM from 1/1/94-12/31/03. To ensure a uniform 6-month eligibility period prior to filling the index APM prescription, all study subjects were required to have utilized >1 medical service and >1 prescription, both within the 6 months prior to the index date as well as in the >6 months before the index date.

Antipsychotic Medications

Atypical APM agents included aripiprazole, clozapine, olanzapine, quetiapine, risperidone, and ziprasidone. Other APMs were considered conventional APMs, including acetophenazine, chlorpromazine, fluphenazine, mesoridazine, perphenazine, promazine, thioridazine, trifluoperazine, triflupromazine, chlorprothixene, haloperidol, loxapine, molindone, pimozide, and thiothixene.21

Outcomes Potentially Mediating the Increased Hazards of Conventional vs. Atypical APMs

The following outcomes were assessed in the 180 days after the index day:

Acute myocardial infarction (AMI). Hospitalization with ICD-9 diagnoses22 (in the principal or secondary position) or DRG23 codes for acute myocardial infarction.24

Ventricular arrhythmia. Ventricular arrhythmia diagnosis25 plus use of a Group I-IV antiarrhythmia medication.

Cerebrovascular events. Diagnoses of cerebrovascular events (e.g., both cerebral hemorrhagic and ischemic events).26

Congestive heart failure. Hospitalization with a diagnostic code for congestive heart failure.27

Pneumonia. Diagnostic codes for pneumonia plus prescription for an antibiotic medication.28

Other serious bacterial infections. Hospitalization with a diagnostic code for bacteremia/septicemia, cellulitis, encephalitis/meningitis, endocarditis/myocarditis, pyelonephritis, septic arthritis, osteomyelitis, or opportunistic infection.28

Other Covariates

We defined the following patient characteristics in the 6 months prior to each subject’s index date:

Sociodemographic data. Age, gender, and race.

Comorbidities. We employed ICD-9 diagnostic codes, CPT procedure codes29, DRG hospitalization codes, and medication use to define the presence of clinical conditions prior to initiation of antipsychotic medication. Acute myocardial infarction, cardiac arrhythmia, cerebrovascular events, congestive heart failure, pneumonia, and other serious bacterial infections were defined as above. Diabetes was defined by the presence of diagnoses plus use of anti-diabetic medications. Additional conditions included other evidence of ischemic heart disease (e.g., angina, PTCA, CABG, or nitroglycerin use), other cardiovascular conditions (e.g., valvular disease, aneurysms, peripheral vascular disease), cancers, HIV, dementia, delirium, mood disorders, psychotic disorders, and other psychiatric disorders.

Health Care Utilization. Health care utilization potentially predictive of developing the outcomes of interest was also assessed, including hospitalizations, nursing home stays, use of other psychiatric medications, and total number of medications used (excluding APMs and drugs used to define covariates).30


Distributions of sociodemographic, clinical, and utilization characteristics among conventional and atypical APM users were calculated. Unadjusted and multivariable (controlling for calendar year and all covariates listed above) Cox proportional hazards models were constructed of developing individual outcomes of interest. Models of developing outcomes within 30, 60, and 120 days of APM initiation were constructed based upon our earlier finding of roughly proportional hazards of death between conventional and atypical APM users within these time intervals.9

In confirmatory analyses, we repeated Cox models using propensity score adjustments to balance independent risk factors for outcomes between drug user groups.14 Propensity scores were derived from predicted probabilities in logistic regression models of conventional vs. atypical APM use. Final non-parsimonious models contained all covariates shown in Table 1 and strongly predicted the type of APM used (c-statistic = 0.845). We then stratified Cox models of individual outcomes across deciles of the propensity score. We also used instrumental variable (IV) analysis to provide estimates that would remain unbiased even if important confounding variables were unmeasured.15-17 An IV is an observable factor related to treatment choice but unrelated to patient characteristics and outcomes. As in other recent work18, we employed the prescribing physician’s preference for conventional vs. atypical APMs (as indicated by their most recent new APM prescription) as the instrument. Using two-stage linear regression for the IV estimation and additional adjustment for measured patient characteristics, we calculated the risk difference of developing 180-day outcomes between conventional vs. atypical APM users. All confirmatory analyses were limited to outcomes found to be significant in the primary analysis.


Patients who began use of conventional APM agents (N=9,142) were slightly younger and more likely to be male and non-white than those who began use of atypical APMs (N=13,748) (Table 1). New users of the conventional agents were less likely than new users of the atypical agents to have cerebrovascular disease, dementia, delirium, psychoses, and other psychiatric disorders, but more likely to have CHF, non-MI ischemic heart disease, and cancer. Conventional APM users had lower rates of using antidepressants, other psychotropic medications, total number of drugs, hospitalizations, and nursing home stays. In the first 180 days of use, 17.9% of patients who began conventional APMs died, compared to 14.6% of those who began atypical APMs. Counts of events by cohort and follow-up period are given in Table 2.

Unadjusted hazard ratios comparing the risk of developing conditions for new users of conventional vs. atypical APMs are shown in Table 3. Adjusted hazard ratios comparing the risk of developing conditions for new users of conventional vs. atypical APMs are shown in Table 4. By 30 days, adjusted hazards were significantly higher for conventional than atypical APM use only in models of developing ventricular arrhythmias (adjusted HR 1.20; 95% CI 1.03-1.39). By 60 days, conventional vs. atypical APM use was only associated with a significantly increased hazard of developing cerebrovascular events (adjusted HR 1.10; 95% CI 1.02-1.19). Similarly, only the hazards of developing cerebrovascular events were significantly greater for conventional than atypical APM use at 120 days (adjusted HR 1.09; 95% CI 1.02-1.16).

Confirmatory analyses using propensity score adjustments14 yielded nearly identical results to the traditional multivariable Cox analyses. Results from the IV analysis agreed with the direction and statistical significance of the traditional multivariable Cox analyses for all outcomes examined.


In this study of over 20,000 elderly patients initiating APM treatments, use of conventional agents was associated with significantly increased hazards of developing ventricular arrhythmias and cerebrovascular events relative to use of atypical agents. The increased hazards for these two conditions may explain at least in part the greater risk of mortality that has been observed from conventional vs. atypical APMs in elderly patients.9

On one hand, our finding of a potential hazard of ventricular arrhythmias from conventional APM use is not new and provides some reassurance concerning our analyses’ ability to identify known effects. Soon after their introduction, conventional APMs were associated with the development of arrhythmias, cardiac arrest, and sudden death.31-35 Prolongation of cardiac repolarization and QTc intervals is thought to be responsible and generally more common with conventional than atypical agents, an exception being ziprasidone.36 Most36,37 but not all32 earlier epidemiological data comparing APM agents have found higher risks of ventricular arrhythmia and cardiac arrest with conventional vs. atypical use. Possible explanations for any discrepancy include adequacy of study power and control for “channeling” of patients at highest risk for arrhythmias away from conventional agents due to earlier warnings.

On the other hand, our finding of a greater hazard of cerebrovascular accidents from conventional vs. atypical APMs has not been established. Because of the paucity of trials involving conventional agents, the FDA has warned only of increased risks for strokes and transient ischemic events from the atypical agents risperidone, olanzapine, and aripiprazole.38-40 One potential mechanism through which conventional agents might increase risks of cerebrovascular events beyond those seen with atypical agents, is suggested by their known anticholinergic effects on heart rate and blood pressure. Hypoperfusion has been shown to lead to microinfarcts in dementia patients and investigators have proposed that conventional APMs, in particular, may accentuate this process.41 Two epidemiologic studies comparing conventional vs. atypical agents have not found statistically significant differences.42,43 One possible explanation for the discrepancy with our results is raised by our finding that the greater hazards from conventionals did not emerge immediately and may require separately examining intermediate or longer periods of follow-up.

These results should be interpreted with the following potential limitations kept in mind. Conventional agents may have been more likely than atypical APMs to be given to patients at risk of developing arrhythmias and cerebrovascular accidents. For this reason, we controlled for sociodemographic, clinical, and health care utilization factors which may be independent predictors of developing these conditions through traditional multivariable, propensity score, and instrumental variable techniques.14-18 We also restricted analyses to only APM users as well as to just new users, to control for underlying reasons for APM use and any selection bias among prevalent users from early symptom emergence, drug intolerance, or treatment failure.13 Because there have been long-standing warnings concerning conventional APMs and ventricular arrhythmias,36 any residual confounding may have led to underestimation of hazards from conventional agents on this outcome. For the same reason, we limited our examination to data from essentially before the first warnings regarding atypical agents and stroke, to avoid overestimating cerebrovascular risks from conventional APMs.38-40 Non-differential exposure misclassification (e.g., not consuming filled prescriptions or switching APM classes) would bias results towards the null; differential misclassification (e.g., worse adherence with conventional APMs, as has been found)44 again may have led to underestimation of hazards from conventional agents. Although we attempted to use or adapt established outcome definitions,24-28 misclassification is still possible and would presumably bias our findings towards the null. Finally, we controlled for calendar time to adjust for any improvements in the prevention of cardiac or cerebrovascular events, which could otherwise lead to reduced risks in later years when atypical use was more common. However, in spite of these safeguards and convergence of results from confirmatory and sensitivity analyses, it is important to keep in mind that our study is based upon non-experimental administrative claims data, in which the completeness and accuracy of clinical information is uncertain. There still may be other aspects of patients newly prescribed conventional APMs that we were unable to control for, leaving open the possibility of residual confounding. For this reason, circumspect interpretation of these findings is required.

Other potential limitations include the possibility that patients who did not fill antipsychotic medications in the prior six months may not have been new initiators, but instead former users beginning a new episode of care. Our study may also have had both inadequate statistical power to observe true associations as well as the potential to detect spurious associations due to having made multiple comparisons. Finally, the generalizability of these findings to other elderly populations may be uncertain.

If confirmed, these results add to growing evidence that conventional APMs may not be safer than atypical APMs for the elderly and should not simply replace the latter drugs stopped in response to recent FDA warnings.9,10,45 They suggest that conventional APMs may raise the risks of mortality through the development of ventricular arrhythmias and cerebrovascular accidents. However beyond suggesting caution regarding conventional APM use in older populations, our results leave many important questions unanswered. First, any greater risk for ventricular arrhythmias or cerebrovascular accidents from conventional agents may only partially mediate the greater mortality observed among conventional vs. atypical APM users. More research is needed to identify other possible intervening conditions and hence potentially vulnerable subpopulations in which conventional agents should be especially avoided. Furthermore, our study sheds no light on other pharmacologic or non-pharmacologic interventions that could preferentially be used to manage the many conditions and symptoms in older populations for which APMs are currently used.12 Well-designed studies shedding light on optimal care for the elderly are sorely needed.


1. U.S. Food and Drug Administration. FDA Public Health Advisory: Deaths with antipsychotics in elderly patients with behavioral disturbances. Available at: Accessed April 15, 2005.

2. Kuehn BM. FDA warns antipsychotic drugs may be risky for elderly. JAMA 2005;293:2462.

3. Strong C. Antipsychotic use in elderly patients with dementia prompts new FDA warning. Neuropsychiatry Rev 2005;6(5):1-17.

4. Dewa CS, Remington G, Herrmann N, et al. How much are atypical antipsychotic agents being used, and do they reach the populations who need them?: A Canadian experience. Clin Therapeutics 2002;24(9):1466-76.

5. Chan YC, Pariser SF, Neufeld G. Atypical antipsychotics in older adults. Pharmacotherapy 1999;19:811-22.

6. Tariot PN. The older patient: the ongoing challenge of efficacy and tolerability. J Clin Psychiatry 1999;60(Suppl 23):29-33.

7. Maixner SM, Mellow AM, Tandon R. The efficacy, safety, and tolerability of antipsychotics in the elderly. J Clin Psychiatry 1999;60 Suppl 8:29-41.

8. Lawlor B. Behavioral and psychological symptoms in dementia: the role of atypical antipsychotics. J Clin Psychiatry 2004;65 Suppl 11:5-10.

9. Wang PS, Schneeweiss S, Avorn J, et al. Risk of death in elderly users of conventional vs. atypical antipsychotic medications. New England Journal of Medicine 2005;353:2335-2341.

10. Schneider LS, Dagerman KS, Insel P. Risk of death with atypical antipsychotic drug treatment for dementia: meta-analysis of randomized placebo-controlled trials. JAMA 2005;294:1934-1943.

11. Nasrallah HA, White T, Nasrallah AT. Lower mortality in geriatric patients receiving risperidone and olanzapine versus haloperidol: preliminary analysis of retrospective data. Am J Geriatr Psychiatry 2004;12:437-9.

12. Alexopoulos GS, Streim J, Carpenter D, et al. Using antipsychotic agents in older patients: expert consensus panel for using antipsychotic drugs in older patients. J Clin Psychiatry 2004;65 Suppl 2:5-99.

13. Ray W. Evaluating medication effects outside of clinical trials: new-user designs. Am J Epidemiol 2003;158:915-20.

14. Rosenbaum PR, Rubin DB. The central role of the propensity score in observational studies for causal effects. J Am Stat Assoc 1983;79:516-524.

15. Bowden RJ, Turkington DA. Instrumental Variables. Cambridge, UK: Cambridge University Press; 1984.

16. Angrist JD, Imbens GW, Rubin DB. Identification of causal effects using instrumental variables. J Am Stat Soc 1996;91:444-455.

17. McClellan M, McNeil BJ, Newhouse JP. Does more intensive treatment of acute myocardial infarction in the elderly reduce mortality? Analysis using instrumental variables. JAMA 1994;272:859-866.

18. Brookhart MA, Wang PS, Solomon DH, Schneeweiss S. Evaluating short-term drug effects in claims databases using physician-specific prescribing preferences as an instrumental variable. Epidemiol 2006;17:268-275.

19. Commonwealth of Pennsylvania. Pharmacy Assistance Contract for the Elderly Act. 62 P.S. 2901-2908. Available at: Accessed August 25, 2005.

20. The Social Security Death Master File Database. Global Internet Management Corp; 2004. Available at: Accessed August 25, 2005.

21. Salzman, C, ed. Clinical Geriatric Psychopharmacology, 4th edition. New York: McGraw-Hill; 2004.

22. International Classification of Diseases, 9th Revision, Clinical Modification. Washington, DC: Public Health Service, U.S. Dept. of Health and Human Services. 1988.

23. St. Anthony’s Diagnosis Related Group Working Guidebook. Alexandria, VA: St. Anthony Publishing; 1993.

24. Kiyota Y, Schneeweiss S, Glynn RJ, et al. Accuracy of Medicare claims-based diagnosis of acute myocardial infarction: estimating positive predictive value on the basis of review of hospital records. Am Heart Journal 2004;148:99-104.

25. De Bruin ML, van Hemel NM, Leufkens HGM, Hoes AW. Hospital discharge diagnoses of ventricular arrhythmias and cardiac arrest were useful for epidemiological research. J Clin Epi 2005;58:1325-1329.

26. Birman-Deych E, Waterman AD, Yan Y, Nilasena DS, Radford MJ, Gage BF.Accuracy of ICD-9-CM codes for identifying cardiovascular and stroke risk factors. Medical Care 2005;43:480-485.

27. Lee DS, Donovan L, Austin PC, Gong Y, Liu PP, Rouleau JL, Tu JV. Comparison of coding of heart failure and comorbidities in administrative and clinical data for use in outcomes research. Med Care 2005;43:182-8.

28. Schneeweiss S, Robicsek A, Scranton R, Zuckerman D, Solomon DH. Health care utilization databases code serious bacterial infections reasonably accurately. In submission.

29. Physicians' Current Procedural Terminology, 4th Edition. Chicago, Ill.: American Medical Association; 1989.

30. Schneeweiss S, Seeger J, Maclure M, Wang P, Avorn, J, Glynn RJ. Performance of comorbidity scores to control for confounding in epidemiologic studies using claims data. Am J Epidemiol 2001;154:854-64.

31. Kelly HG, Fay JE, Lavery SG. Thioridazine hydrochloride (Mellaril): its effects on the electrocardiogram and a report of two fatalities with electrocardiographic abnormalities. Can Med Assoc J 1963;89:546-554.

32. Hennessy S, Bilker WB, Knauss JS, et al. Cardiac arrest and ventricular arrhythmia in patients taking antipsychiotic drugs: cohort study using administrative data. British Medical Journal 2002;325:1070-74.

33. Straus SMJM, Bleumink GS, Dieleman JP, et al. Antipsychotics and the risk of sudden cardiac death. Archives of Internal Medicine 2004;164:1293­97.

34. Reilly JG, Ayis SA, Ferrier IN, Jones SJ, Thomas SHL. Thioridazine and sudden unexplained death in psychiatric inpatients. The British Journal of Psychiatry 2002;180:515-522.

35. Ray WA, Meredith S, Thapa PB, Meador KG, Hall K, Murray KT. Antipsychotics and the risk of sudden cardiac death. Arch Gen Psychiatry 2001;58:1161-1167.

36. Glassman AH, Bigger JT. Antipsychotic drugs: prolonged QTc Interval, Torsade de Pointes, and sudden death. American Journal of Psychiatry 2001;158:11:1774-82.

37. Liperoti R, Gambassi G, Lapane KL, Chiang C, Pedone C, Mor V, Bernabei R. Conventional and atypical antipsychotics and the risk of hospitalization for ventricular arrhythmias or cardiac arrest. Archives of Internal Medicine 2005;165:696-701.

38. Sponsor’s website for Risperdal. Available at: Accessed March 24, 2006.

39. Sponsor’s website for Zyprexa. Available at: Accessed March 24, 2006.

40. Sponsor’s website for Abilify. Available at: Accessed March 24, 2006.

41. Miklossy J. Cerebral hypoperfusion induces cortical watershed microinfarcts which may further aggravate cognitive decline in Alzheimer’s disease. Neurol Res 2003;25:605-610

42. Herrmann N, Mamdani M, D P, Lanctot KL. Atypical antipsychotics and the risk of cerebrovascular accidents. Am J Psychiatry 2004;161:6:1113-15.

43. Liperoti R, Gambassi G, Lapane KL, Chiang C, Pedone C, Mor V, Bernabei R. Cerebrovascular events among elderly nursing home patients treated with conventional or atypical antipsychotics. J Clin Psychiatry 2005;66:1090-1096.

44. Dolder CR, Lacro JP, Dunn LB, Jeste DV. Antipsychotic medication adherence: is there a difference between typical and atypical agents? Am J Psychiatry 2002;159:103-108.

45. Rabins PV, Lyketsos CG. Antipsychotic drugs in dementia: what should be made of the risks? JAMA 2005;294:1963-1965.

Table 1. Characteristics of elderly new users of conventional and atypical antipsychotic medications (n=22,890).

% of Conventional APM Users % of Atypical APM Users
Age (mean): 83.2 83.5
Gender: Female 77.6 83.0
Male 22.4 17.0
Race: White 92.8 94.7
Non-white 7.2 5.3
Diagnoses: Cardiac arrhythmia 1.4 1.4
Cerebrovascular disease 29.1 30.9
CHF 32.6 31.1
Diabetes 25.8 26.8
Hypertension 58.1 65.1
Myocardial infarction 3.5 3.5
Other ischemic heart disease 29.3 24.4
Other cardiovascular disorders 12.7 12.3
Cancer 15.6 14.0
HIV <0.1 <0.1
Dementia 40.8 52.5
Delirium 12.2 16.1
Mood disorders 22.2 36.3
Psychotic disorders 21.3 24.7
Other psychiatric disorders 5.9 8.3
Other drug use: Antidepressants 28.0 43.5
Other psychotropic medications 11.5 13.5
Number of different drugs (mean) 6.8 7.9
Hospitalization in prior 180 days 51.2 53.5
Nursing home residence in prior 180 days 15.9 21.4

Table 2. Counts of events by cohort and follow-up time.

Condition Counts of Events within 30 days

Count (%)
Counts of Events within 60 days

Count (%)
Counts of Events within 120 days

Count (%)
Conventional Atypical Conventional Atypical Conventional Atypical
Acute myocardial infarction 60 (0.66) 96 (0.70) 99 (1.08) 151 (1.10) 164 (1.79) 249 (1.81)
Cerebrovascular event 1319 (14.43) 1746 (12.70) 1759 (19.24) 2375 (17.28) 2314 (25.31) 3169 (23.05)
Congestive heart failure 1517 (16.59) 2170 (15.78) 2065 (22.59) 2990 (21.75) 2677 (29.28) 3891 (28.30)
Pneumonia 75 (0.82) 96 (0.70) 121 (1.32) 162 (1.18) 183 (2.00) 279 (2.03)
Other serious infection 258 (2.82) 313 (2.28) 395 (4.32) 537 (3.91) 616 (6.74) 882 (6.42)
Ventricular arrhythmia 472 (5.16) 615 (4.47) 721 (7.89) 999 (7.27) 1065 (11.65) 1518 (11.04)

Table 3. Unadjusted relative hazards of developing conditions among elderly new users of conventional vs. atypical antipsychotic medications.

Condition HR (95%CI) in 30 days HR (95%CI) in 60 days HR (95%CI) in 120 days
Acute myocardial Infarction 0.93 (0.66, 1.32) 1.00 (0.76, 1.30) 1.03 (0.85, 1.27)
Cerebrovascular event 1.19 (1.10, 1.27) 1.18 (1.11, 1.26) 1.17 (1.11, 1.24)
Congestive heart failure 1.09 (1.02, 1.17) 1.09 (1.02, 1.15) 1.09 (1.03, 1.14)
Pneumonia 1.30 (0.96, 1.78) 1.24 (0.98, 1.58) 0.84 (0.66, 1.05)
Other serious infection 1.33 (1.12, 1.57) 1.03 (0.87, 1.21) 1.12 (1.01, 1.25)
Ventricular arrhythmia 1.21 (1.07, 1.37) 1.15 (1.04, 1.27) 1.12 (1.04, 1.21)

Table 4. Adjusted Relative Hazards of Developing Conditions Among Elderly New Users of Conventional vs. Atypical Antipsychotic Medications.

Condition Adjusted HR* (95%CI)

in 30 days
Adjusted HR* (95%CI)

in 60 days
Adjusted HR* (95%CI)

in 120 days
Acute myocardial Infarction 0.89 (0.59, 1.33) 1.02 (0.75, 1.40) 1.16 (0.91, 1.48)
Cerebrovascular event 1.08 (0.99, 1.18) 1.10 (1.02, 1.19) 1.09 (1.02, 1.16)
Congestive heart failure 1.04 (0.95, 1.11) 1.00 (0.93, 1.07) 1.01 (0.95, 1.07)
Pneumonia 1.11 (0.76, 1.63) 1.03 (0.76, 1.38) 0.84 (0.66, 1.05)
Other serious infection 1.20 (0.98, 1.48) 1.03 (0.87, 1.21) 1.00 (0.88, 1.14)
Ventricular arrhythmia 1.20 (1.03, 1.39) 1.10 (0.98, 1.24) 1.06 (0.96, 1.17)

*Controlled for calendar year, age, gender, race, prior arrhythmias, cerebrovascular disease, CHF, diabetes, MI, hypertension, other evidence of ischemic heart disease, other cardiovascular conditions, cancers, HIV, pneumonia, other serious infection, dementia, delirium, mood disorders, psychotic disorders, other psychiatric disorders, use of other psychiatric medications, total number of medications, hospitalizations, and nursing home stays.