Mortality Rate and Causes of Death in a Canadian Tertiary Schizophrenia Program

People with schizophrenia have an increased mortality rate ¹ and decreased life expectancy ² relative to the general population. Mortality is elevated across a range of causes, including medical illness, suicide, and accidents.^1,2 Despite an urgent need to understand mortality trends in order to design protective interventions, the mortality rate and causes of death of people with schizophrenia spectrum disorders (SSDs) have not been studied within the context of a Canadian tertiary mental healthcare centre.

We conducted a retrospective chart review of patients (≥18 years) diagnosed with an SSD (schizophrenia, schizoaffective disorder, or psychosis not otherwise specified [NOS]) who died during their enrollment with The Royal Ottawa Mental Health Centre's (ROMHC) Inpatient or Outpatient Schizophrenia Program from June 1, 2014, to July 20, 2022. Outpatients died as outpatients, regardless of admission history. Inpatients died during admission. Death was determined if it occurred during the study period and was documented in the ROMHC's incident reporting system. Patients were not followed after discharge to ascertain mortality. Those with medical or psychiatric comorbidities were eligible for inclusion. Patients were excluded from chart review if they had been admitted to a ROMHC Schizophrenia Program without an SSD diagnosis. Data were extracted from admission, progress, consult, nursing, and social work notes, stored forms or applications, and discharge summaries. Causes of death were obtained from medical records or incident reports.

Mortality rate was calculated for all patients enrolled in the ROMHC schizophrenia program at any point during the study period, regardless of chart access, completeness, or diagnosis. This study was approved by the ROMHC Research Ethics Board; informed consent was not required.

A total of 109 of the 3035 ROMHC schizophrenia program patients died during the 8-year, 49-day study period, an overall mortality rate of 3.59% (5 inpatient deaths [total inpatients: n = 563], inpatient mortality rate = 0.89%; 104 outpatient deaths [total outpatients: n = 2472], outpatient mortality rate = 4.21%).

Seventy-four deceased patients with an SSD diagnosis (schizophrenia: n = 59, 79.7%; schizoaffective disorder: n = 14, 18.9%; psychosis NOS, n = 1, 1.4%; mean [SD] age of first episode psychosis = 24.4 [8.2] years) had medical records containing sufficient information for chart review. The most common cause of death was medical illness (n = 46, 62.2%), followed by undetermined causes (n = 20, 27.0%), suicide (n = 5, 6.8%), and accidents (n = 3, 4.1%); 19% of deaths of undetermined cause (n = 4, three overdose) were deemed to be unnatural, while the remainder could not be classified due to insufficient data. Because chart access was sealed for six patients (three inpatients) for undisclosed reasons, the cause of most inpatient deaths was unknown.

Deceased patients were typically single (79.7%) males (75.7%), with a mean age of 54.6 (12.7) years, managed on an average of 1.6 (0.9) anti-psychotic medications (Table 1). They were older and more often male than typical program participants (mean age = 49.0 [14.8], 64.3% male). Among outpatients, mean duration in the program was 10.9 (8.7) years (n = 39).

Table 1.

Demographic Characteristics of our Chart Review Sample (N = 74).

	n	%
Age
18–29	5	6.8
30–39	6	8.1
40–49	9	12.2
50–59	27	36.5
60–69	20	27.0
≥70	7	9.5
Program
Inpatient	1	1.4
Outpatient	73	98.6
Marital status
Single	59	79.7
Married	6	8.1
Common law	1	1.4
Divorced	6	8.1
Unknown	1	1.4
Income source
ODSP	40	54.1
Unknown	20	27.0
Senior's benefit	9	12.2
Employment	7	9.5
Family support	7	9.5
Smoking status
Smoker	45	60.8
Non-smoker	29	39.2
Clozapine
Yes	20	27.0
No	54	73.0

For income source, patients may belong to multiple categories. ODSP = Ontario Disability Support Program.

This report is the first to quantify the mortality rate and causes of death of patients admitted to a Canadian tertiary schizophrenia program, providing a benchmark for other centres. Deaths typically occurred among men in their 50s and 60s. Patients were most likely to die of a medical illness, a finding consistent with previous work. ¹ However, our mortality rate (3.59%, 13.40 deaths/year) is lower than both the 20% mortality rate (180.80 deaths/year) reported in a large, population-based sample of Canadians with schizophrenia from 2014 ³ and a 1991 study of 3623 schizophrenia patients (301 deaths, 8.3% mortality rate, 33.44 deaths/year) in Alberta. ⁴ One explanation may be the highly personalized treatment and risk factor management provided by specialist clinicians in our tertiary program. However, decreased mortality may have also occurred secondary to changes in psychiatric practice over time or better access to care, regardless of quality or setting, relative to population-based samples. This lack of clarity highlights a need for comparison with other centres.

Similar to previous work, ² deaths occurred at a younger age in our sample of SSD patients relative to the general population. ⁵ This strengthens our argument for urgency in designing protective interventions specific to tertiary settings. Integration of primary care within our outpatient psychiatric clinic could reduce mortality by improving recognition and management of medical comorbidities among patients with SSDs.

A definitive cause of death was not available for 27% of our sample: 51 incident reports were inaccessible and others remained incomplete. A report capturing more data, including known risk factors for medical death or suicide, with an established protocol for follow-up with external services could improve reporting compliance and prevention strategies.

Additional limitations of this work include the inherent methodological restraints of retrospective studies with respect to fully capturing clinically relevant information and the inability to track care delivered outside our program.