Effect of Antipsychotics on Mortality Risk in Patients With Dementia With and Without Comorbidities

Ane Nørgaard MD, PhD; Christina Jensen-Dahm MD, PhD; Theresa Wimberley MSc, PhD; Jesper Hastrup Svendsen MD, DMSc; Kazi Ishtiak-Ahmed PhD; Thomas Munk Laursen MSc, PhD; Gunhild Waldemar MD, DMSc; Christiane Gasse RPharm, PhD

Disclosures

J Am Geriatr Soc. 2022;70(4):1169-1179.

Abstract and Introduction

Abstract

Background: We investigated the mortality risk associated with the initiation of antipsychotic treatment among patients with dementia and whether comorbidities related to the cardiovascular system and diabetes interact with antipsychotic treatment to increase the mortality risk beyond the risk of death independently associated with antipsychotics and comorbidity alone.

Methods: We designed a matched cohort study using nationwide registry data. All Danish residents aged 65–95 years diagnosed with dementia between 2009 and 2014 were included. Dementia was assessed as a first-time registered dementia diagnosis in the Danish National Patient Register or the Danish Psychiatric Central Research Register and/or a first-time prescription for antidementia medication. Patients exposed to antipsychotics were matched with up to three unexposed patients. Cox proportional hazards models were used to compare rates of death within 180 days after the initiation of antipsychotic treatment. The models were adjusted for potential confounders. Analyses were stratified for diabetes, heart disease, and cerebrovascular disease, and we calculated the relative excess risk due to interaction (RERI).

Results: The study cohort included 8244 exposed patients and 24,730 unexposed patients. A total of 5938 patients died during the first 180 days of follow-up. Patients exposed to antipsychotics had a significantly higher adjusted risk of death (hazard ratio: 1.35, 95% confidence interval: 1.27–1.43) than unexposed patients. Crude mortality rates were higher among patients with heart disease and diabetes when antipsychotic treatment was initiated compared with patients without comorbidities. Relative risk estimates did not differ between patients with and without heart disease, cerebrovascular disease, and diabetes, and RERI suggested no positive additive interaction. Risk analysis suggested higher mortality in patients without cerebrovascular disease who initiated antipsychotics.

Conclusion: This nationwide study adds to the evidence that antipsychotic treatment is associated with increased mortality and suggests that attention should be paid to all initiators of antipsychotics irrespective of cardiovascular disease and diabetes.

Introduction

Antipsychotic drugs are commonly used to treat behavioral and psychological symptoms in patients with dementia. However, the treatment may cause side effects and can lead to serious adverse events and death.^[1,2] Clinical trials (typical trial duration of 10 weeks^[3]) found an increased mortality during antipsychotic treatment which led to warnings against antipsychotic treatment for patients with dementia.^[4] Population-based studies found an increased short-term (6–12 months) mortality among patients treated with antipsychotic drugs.^[5–7]

The side effects of antipsychotic treatment are manifold and include QT prolongation, orthostatic hypotension, sedation, Parkinsonism, and metabolic disturbances. The possible mechanisms leading to increased mortality may be linked to these side effects. QT prolongation may lead to ventricular tachyarrhythmias and sudden cardiac death,^[8,9] orthostatic hypotension may lead to falls and ischemic stroke,^[10] and sedation may lead to aspiration pneumonia.^[11,12] Thus, antipsychotic treatment may negatively influence the cardiovascular and metabolic systems, and patients with preexisting cardiovascular comorbidity could be at higher risk of adverse events and death during antipsychotic treatment.

A Danish study of older adults found that people with preexisting cardiovascular disease had a greater risk of major cardiovascular events during antipsychotic treatment compared with people without cardiovascular disease.^[13] A small Finnish study investigated the mortality risk during antipsychotic treatment among community-dwelling older people in different comorbidity groups. People with respiratory disease at baseline had the highest risk of death associated with antipsychotic treatment.^[14] However, none of the studies focused on older adults with dementia.

To date, no large-scale studies have explicitly investigated whether the effect of antipsychotic treatment on the risk of death among patients with dementia varies across groups of patients with or without preexisting comorbidities. In fact, many observational studies have adjusted for comorbidity in the statistical analyses^[1,5–7] and, thus, did not evaluate the potential interaction between comorbidity and antipsychotic treatment. Therefore, the aims of this study were to investigate (i) the 180-day mortality risk associated with antipsychotic drug treatment in patients with dementia and (ii) whether specific preexisting comorbid conditions (heart disease, cerebrovascular disease, and diabetes) interact with antipsychotic treatment leading to an increased mortality beyond the risk of death independently associated with antipsychotic treatment and comorbidity alone.

1 2 3 4

Next Section

Abstract and Introduction
Methods
Results
Discussion
References
Sidebar

Table 1. Characteristics of exposed and unexposed patients with dementia
Characteristic	Treated with antipsychotics (exposed), n = 8244	Not treated with antipsychotics at the beginning of follow-up (unexposed), n = 24,730
Age at the time of dementia diagnosis, years	81.7 (76.2–86.4)	81.7 (76.2–86.4)
Sex (female)	4671 (56.7%)	14,013 (56.7%)
Nursing home resident	1995 (24.2%)	5454 (22.1%)
Other somatic comorbidities, Charlson Comorbidity Index score^a
0	4563 (55.3%)	13,604 (55.0%)
1	1307 (15.9%)	4027 (16.3%)
≥2	2374 (28.8%)	7099 (28.7%)
Psychiatric disease prior to dementia diagnosis	1689 (20.5%)	3955 (16.0%)
Heart disease	2512 (30.5%)	7743 (31.3%)
Cerebrovascular disease	1732 (21.0%)	4868 (19.7%)
Diabetes	1294 (15.7%)	4072 (16.5%)
Number of hospital admission 5 years prior to the index date	4 (2–7)	3 (1–6)
Number of somatic drugs^b	6 (3–9)	5 (3–8)
Prior benzodiazepine treatment^c	2841 (34.5%)	3158 (12.8%)
Prior antidepressant drug treatment^c	4442 (53.9%)	5862 (23.7%)
Time from dementia diagnosis to first antipsychotic prescription, days	261 (54–683)	Not relevant

Note: Numbers are given as n (%) or median (25% quartile to 75% quartile) as appropriate.
^aExcluding dementia and comorbidities under investigation.
^bCounted at the Anatomical Therapeutic Chemical level 5 within 3 months prior to the index date excluding N05A, N05B, N05C, N06A, N06D, and A10.
^cRedeemed a prescription within 180 days prior to the index date.

Table 2. Mortality within 180 days for patients with dementia treated with antipsychotics (exposed) versus patients with dementia not treated with antipsychotics (unexposed)
	Exposed (n = 8244)	Unexposed (n = 24,730)
Number of deaths	1993	3945
Person-years	3418.8	10,866.8
Crude mortality rate per 100 person-years	58.3 (55.8–60.9)	36.3 (35.2–37.5)
Age- and sex-adjusted HR (95% CI)	1.49 (1.41–1.57)	1.0 (Ref.)
Adjusted HR (95% CI)^a	1.35 (1.27–1.43)	1.0 (Ref.)
Interaction term (heart disease × antipsychotic treatment)	p = 0.2433
Interaction term (cerebrovascular disease × antipsychotic treatment)	p = 0.0010
Interaction term (diabetes × antipsychotic treatment)	p = 0.4913

Note: Follow-up time, 180-days mortality rates, and HRs with 95% CIs are presented.
Abbreviations: 95% CI, 95% confidence interval; HR, hazard ratio.
^aAdjusted for: age at dementia diagnosis, sex, calendar year, nursing home residency, heart disease, cerebrovascular disease, diabetes mellitus, Charlson Comorbidity Index score (excluding dementia, acute myocardial infarction, congestive heart disease, cerebrovascular disease, diabetes, and diabetes with end-organ damage), psychiatric comorbidity, treatment with benzodiazepines and antidepressants within 180 days before the index date, and number of hospitalizations within 5 years before the index date.

Table 3. Mortality within 180 days for patients with dementia treated with antipsychotics (exposed) versus patients with dementia not treated with antipsychotics (unexposed) stratified for specific preexisting comorbidities
		Number of deaths	Person-years	Crude mortality rate per 100 person-years (95% CI)	Age- and sex-adjusted HR (95% CI)	Adjusted HR (95% CI)^a
Heart disease
No	Unexposed, n = 20,649 (83.5%)	2286	7609.0	30.0 (28.8–31.3)	1.0 (Ref.)	1.0 (Ref.)
No	Exposed, n = 6892 (83.6%)	1222	3418.8	50.2 (47.5–53.1)	1.55 (1.45–1.66)	1.26 (1.17–1.36)
Yes	Unexposed, n = 4081 (16.5%)	1659	3257.9	50.9 (48.5–53.4)	1.0 (Ref.)	1.0 (Ref.)
Yes	Exposed, n = 1352 (16.4%)	771	985.6	78.2 (72.9–83.9)	1.41 (1.29–1.53)	1.19 (1.08–1.30)
Cerebrovascular disease
No	Unexposed, n = 19,862 (80.3%)	2898	8815.2	32.9 (31.7–34.1)	1.0 (Ref.)	1.0 (Ref.)
No	Exposed, n = 6512 (79.0%)	1516	2719.6	55.7 (53.0–58.6)	1.57 (1.47–1.67)	1.41 (1.31–1.50)
Yes	Unexposed, n = 4868 (19.7%)	1047	2051.6	51.0 (48.0–54.2)	1.0 (Ref.)	1.0 (Ref.)
Yes	Exposed, n = 1732 (21.0%)	477	699.2	68.2 (62.4–74.6)	1.24 (1.16–1.38)	1.17 (1.05–1.32)
Diabetes
No	Unexposed, n = 20,658 (83.5%)	3169	9121.4	34.7 (33.6–36.0)	1.0 (Ref.)	1.0 (Ref.)
No	Exposed, n = 6950 (84.3%)	1630	2896.5	56.3 (53.6–59.1)	1.50 (1.41–1.59)	1.25 (1.18–1.34)
Yes	Unexposed, n = 4072 (16.5%)	776	1745.4	44.5 (41.4–47.7)	1.0 (Ref.)	1.0 (Ref.)
Yes	Exposed, n = 1294 (15.7%)	363	522.3	69.5 (62.7–77.0)	1.43 (1.26–1.62)	1.15 (1.01–1.32)

Note: Follow-up time, 180-days mortality rates, and HRs with 95% CIs are presented.
Abbreviations: 95% CI, 95% confidence interval; HR, hazard ratio.
^aAdjusted for: age at dementia diagnosis, sex, calendar year, nursing home residency, heart disease, cerebrovascular disease, and diabetes mellitus [not adjusted for the comorbidity in strata of interest]), Charlson Comorbidity Index score (excluding dementia, acute myocardial infarction, congestive heart disease, cerebrovascular disease, diabetes, and diabetes with end-organ damage), psychiatric comorbidity, treatment with benzodiazepines and antidepressants within 180 days before the index date, and number of hospitalizations within 5 years before the index date.

Authors and Disclosures

Ane Nørgaard MD, PhD^1,2, Christina Jensen-Dahm MD, PhD^1,2, Theresa Wimberley MSc, PhD³, Jesper Hastrup Svendsen MD, DMSc^2,4, Kazi Ishtiak-Ahmed PhD^5,6, Thomas Munk Laursen MSc, PhD^3,7, Gunhild Waldemar MD, DMSc^1,2 and Christiane Gasse RPharm, PhD⁶

¹Danish Dementia Research Centre, Department of Neurology, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark
²Department of Clinical Medicine, Faculty and Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark
³The National Centre for Register-Based Research, Aarhus University, Aarhus, Denmark
⁴Department of Cardiology, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark
⁵Department of Public Health, University of Copenhagen, Copenhagen, Denmark
⁶Department of Affective Disorders, Aarhus University Hospital Psychiatry, Aarhus, Denmark
⁷Centre for Integrated Register-based Research (CIRRAU), Aarhus University, Aarhus, Denmark

Correspondence
Ane Nørgaard, Danish Dementia Research Centre, Department of Neurology, Neuroscience Centre Rigshospitalet, University of Copenhagen, Blegdamsvej 9, #8007, 2100 Copenhagen, Denmark. Email: ane.noergaard.christensen.02@regionh.dk

Conflict of Interest
Gunhild Waldemar reports to have received consultancy fees for her department from Hoffmann-La Roche outside this work. Jesper Hastrup Svendsen reports to be a member of Medtronic advisory boards and to have received speaker honoraria and research grants from Medtronic outside this work. The rest of the authors report no conflicts of interest.

Author Contributions
Dr. Ane Nørgaard had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Ane Nørgaard, Christina Jensen-Dahm, Jesper Hastrup Svendsen, Kazi Ishtiak-Ahmed, Thomas Munk Laursen, Gunhild Waldemar, Christiane Gasse. Acquisition, analysis, or interpretation of data: Ane Nørgaard, Christina Jensen-Dahm, Theresa Wimberley, Thomas Munk Laursen, Gunhild Waldemar, Christiane Gasse. Drafting of the manuscript: Ane Nørgaard, Gunhild Waldemar, Christiane Gasse. Critical revision of the manuscript for important intellectual content: Ane Nørgaard, Christina Jensen-Dahm, Theresa Wimberley, Jesper Hastrup Svendsen, Kazi Ishtiak-Ahmed, Thomas Munk Laursen, Gunhild Waldemar, Christiane Gasse. Study supervision: Christiane Gasse.

processing....

Effect of Antipsychotics on Mortality Risk in Patients With Dementia With and Without Comorbidities

Abstract and Introduction

Abstract

Introduction

Tables

Tables

Tables

References

Authors and Disclosures

Authors and Disclosures

Sidebar

Sidebar

Key Points

Why Does This Paper Matter?

Most Popular Articles

Email This

Feedback