This study uses the Danish nationwide health care registries to investigate the association of secondary preventive cardiovascular treatment with all-cause mortality after myocardial infarction in patients with schizophrenia compared with those without schizophrenia.
Key Points
Question
Is use of cardioprotective medication after myocardial infarction associated with reduced mortality rates among patients with schizophrenia?
Findings
In this Danish nationwide cohort study of 105 018 patients with myocardial infarction, patients with schizophrenia who were not exposed to cardioprotective medication had an excess mortality rate; however, exposure to medication was associated with a reduced mortality.
Meaning
The increased cardiovascular mortality among patients with schizophrenia was reduced in the current study if patients were efficiently administered cardioprotective treatment after cardiac events.
Abstract
Importance
Cardioprotective medication use is an important secondary preventive treatment after cardiovascular events. Patients with schizophrenia have excess cardiovascular morbidity and mortality, but no studies have investigated whether taking recommended cardioprotective medication can reduce this excess mortality.
Objective
To assess the association of exposure to secondary cardiovascular treatment with all-cause mortality after myocardial infarction among patients with schizophrenia compared with the general population.
Design, Setting, and Participants
This nationwide cohort study included individuals admitted with first-time myocardial infarction in Denmark from January 1, 1995, to December 31, 2015. The cohort was dichotomously divided by a diagnosis of schizophrenia. Data on the prescription of guideline-recommended cardioprotective medication, including antithrombotics, β-blockers, vitamin K antagonist, angiotensin-converting enzyme inhibitors, and statins, were obtained from the nationwide registries.
Exposures
Time exposed to cardioprotective medication.
Main Outcomes and Measure
Cox proportional hazards regression was used to calculate hazard ratios (HRs) with 95% CIs for the association between treatment exposure and all-cause mortality after myocardial infarction between patients with and without schizophrenia.
Results
The cohort included 105 018 patients with myocardial infarction, including 684 patients with schizophrenia (0.7%; 483 male [70.6%]; mean [SD] age at diagnosis, 57.3 [10.6] years) and 104 334 general population patients (99.3%; 73 454 male [70.4%]; mean [SD] age at diagnosis, 61.0 [10.6] years), with a total follow-up of 796 435 person-years and 28 059 deaths. Patients diagnosed with schizophrenia who did not receive cardioprotective treatment had the highest mortality rate (HR, 8.78; 95% CI, 4.37-17.64) compared with the general population treated, with treated patients diagnosed with schizophrenia having an increased HR of 1.97 (95% CI, 1.25-3.10). The analyses of the associations of different cardiac therapy strategies with mortality rates revealed that patients with schizophrenia who were treated with any combination of triple therapy had mortality rates similar to those observed in the general population (HR, 1.05; 95% CI, 0.43-2.52) in the multivariable analysis.
Conclusions and Relevance
Cardioprotective medication after myocardial infarction should be carefully managed to improve prognosis. The study results suggest that some of the increased cardiac mortality among patients with schizophrenia can be reduced if these patients are efficiently administered combinations of secondary preventive treatments after cardiac events.
Introduction
Cardiovascular disease, including ischemic heart disease and stroke, is the leading cause of death worldwide.1 Deaths from cardiovascular disease have decreased during the past decades, especially in Western Europe, North America, and Japan.2 This decrease in cardiovascular death is believed to be a result of multiple factors, including improvement in lifestyle behavior (eg, smoking cessation, physical activity, and balanced diet), implementation of interventional cardiologic procedures (eg, coronary artery bypass grafting and percutaneous coronary intervention [PCI]), and increases in prophylactic cardiovascular treatment (antithrombotics, β-blockers, angiotensin-converting enzyme inhibitors [ACEIs], and statins).3
Patients with schizophrenia have an excess cardiac mortality, which contributes to a shortened life expectancy of 15 to 20 years compared with the general population.4,5 Studies6,7 on survival after ischemic heart disease have reported worse outcomes in patients with schizophrenia compared with people without mental illness. The causes of the differences in outcome are unclear, although earlier findings suggest deficits in quality of medical care and fewer cardiac interventions performed in patients with schizophrenia compared with the general population.8,9,10 Recent studies6,11 indicate that patients with schizophrenia who receive cardiac revascularization procedures have poorer outcome after ischemic heart disease, and a recent Danish population-based study11 found that patients with severe mental illness (SMI) are less likely to receive recommended, long-term, secondary preventive medications after PCI; this finding is further supported by a study12 that found no differences in short-term mortality but an increased long-term mortality rate after ischemic heart disease in patients with schizophrenia compared with the general population.
Taking recommended cardioprotective medication after myocardial infarction (MI) reduces hospital admissions,13 health care costs, and all-cause mortality in patients without psychiatric conditions.14 Studies11,15 on the prescription rate of cardioprotective medication in patients with SMI have been limited, and no studies have examined the association of prophylactic cardiac treatment exposure with mortality rates after MI in patients with schizophrenia compared with the rates in the general population. In the current study, we investigated the association of secondary preventive cardiovascular treatment with all-cause mortality after MI in patients with schizophrenia compared with those without schizophrenia using the Danish nationwide health care registries.
Methods
Study Population
This nationwide, retrospective cohort study included all individuals admitted with first-time MI (International Statistical Classification of Diseases and Related Health Problems, Tenth Revision [ICD-10] code I21) in Denmark from January 1, 1995, to December 31, 2015, identified from the Danish National Patient Registry (NPR). The NPR covers all somatic hospitalizations from January 1, 1977, to December 31, 1994, and from January 1, 1995, onward, covering all somatic and psychiatric inpatient and outpatient contacts with all public hospitals in Denmark.16 Persons in the NPR with a diagnosis of MI (International Classification of Diseases, Eighth Revision [ICD-8] code 410 or ICD-10 code I21) before 1995 were excluded from the study population to include only patients with incident MI in this study. The Danish Data Protection Agency and the National Board of Health approved the use of the data. Retrospective studies do not require ethical approval in Denmark. All data were deidentified.
From the initial sample, we established a cohort of patients with a previous diagnosis of schizophrenia, defined as an ICD-8 or ICD-10 code of 295 or a Danish Psychiatric Central Research Register (DPCRR) or NPR code of F20. The DPCRR was established in 1969 and contains information on every psychiatric inpatient hospitalization from January 1, 1969, to December 31, 1994, and from January 1, 1995, onward, covering all inpatient and outpatient contacts with all psychiatric hospitals in Denmark.17 The remaining patients with MI from the cohort served as controls representing patients with MI in the general population. Incident cases of schizophrenia after MI diagnosis in the control cohort were excluded from the analyses. We excluded patients younger than 30 years who had an MI to minimize inclusion of patients with potential congenital heart defects.
Drug Exposure
From the NPR, we identified all dispensed prescriptions for guideline-recommended secondary preventive treatments after MI using the Anatomical Therapeutic Chemical (ATC) code in agreement with the World Health Organization.18 Data on every dispensed prescription, including dispensing date, defined daily dose, and the total amount of pills dispensed, are available from January 1, 1995, in the NPR for the entire Danish population.19
We defined 5 therapeutic drug groups: antiplatelets (ATC code B01AC), vitamin K antagonists (ATC codes B01AA, B01AE, and B01AF), β-blockers (ATC code C07), ACEIs (ATC code C09), and statins (ATC codes C10). Because of the natural progress of change in treatment regimens during follow-up, we allowed patients to switch among therapeutic groups at any time.
The treatment exposure groups were created as time-varying covariates. We assumed the use of 1 defined daily dose per day with a grace period of 14 days. The grace period was introduced to allow for some degree of nonadherence and for irregular dispensing attributable to stockpiling. Periods of concomitant treatment were collapsed to coherent periods during the follow-up. For each individual, we defined exposed and unexposed periods during the entire follow-up period, and these periods where used in the calculation of hazard ratios (HRs) for mortality estimates.
Monotherapy was defined as treatment periods with use of one of the previously described treatment groups. Dual therapy was defined as 2 overlapping periods of concurrent treatments, whereas triple therapy was defined as 3 or more overlapping treatment periods. Patients were allowed to switch groups at any time during the follow-up, according to the use of time-varying covariates in the Cox proportional hazards regression model.
Outcome Measures
The primary study outcome was time to all-cause mortality using time of medication exposure as a time-dependent covariate. Mortality data were retrieved from the Danish Causes of Death Register, which contains data on age of death and underlying cause of death.20 The discharge date of the MI hospitalization was used as the index date. All individuals were followed up from the time of index until the time of death or until December 31, 2015, whichever came first. Patients were allowed to initiate and discontinue use of medications throughout follow-up and thus change exposure status during the study period. The secondary outcome was time to all-cause mortality, comparing exposure to monotherapy, dual therapy, and triple therapy in patients diagnosed or not diagnosed with schizophrenia.
Covariates
This study included the demographic covariates sex, age, and calendar year of the index date. The number of MI readmissions after index were divided into 0, 1, or 2 or more. Furthermore, we defined patients who had received PCI according to the Danish health care classification system codes KFNG02 and KFNG05.
Cardiovascular Risk Factors
Patients were defined as having diabetes if they had a registered ICD-8 diagnosis code of 249 or 250 or an ICD-10 diagnosis code of E10 to E14 (diabetes). In addition, patients who claimed prescriptions for glucose-lowering drugs (ATC code A10B or A10XA) or insulins (ATC code A10A) without a registered diagnosis in the NPR were also considered as having diabetes because patients treated by the general practitioner are not registered in the NPR. Patients who had a registered ICD-8 diagnosis code of 491 or 492, an ICD-10 diagnosis code of J41 to J44, or any prescription of drugs for chronic obstructive pulmonary disease (COPD) (ATC code R03) were considered to have COPD. We defined hypertension as an ICD-8 diagnosis code of 400 to 404, ICD-10 diagnosis code of I10 to I15, or prescriptions for antihypertensives (ATC code C02), diuretics (ATC code C03) (except loop diuretics [code C03C]), or calcium antagonists (ATC code C08) (except verapamil hydrochloride [code C08DA01] and diltiazem hydrochloride [code C08DB01]). We also included substance abuse (ie, alcohol-related [ICD-10 code F10.x] or drug-related substance abuse [ICD-10 codes F11.x-F19.x]), excluding tobacco abuse (ICD-10 code F17.x), as explanatory variables.
Statistical Analysis
Descriptive analysis of the study population’s demographics and illness history was initially conducted. The frequency and distribution of patient characteristics were described by calculating means (SDs) for continuous variables and numbers (percentages) for categorical variables.
The primary analysis was a Cox proportional hazards regression model that evaluated all-cause mortality using drug exposure as defined previously as a time-varying covariate. We adjusted for sex, year of birth, age at MI, calendar year, and PCI in the first adjusted model and then further adjusted for substance abuse, diabetes, COPD, and hypertension in the second adjusted model. We calculated unadjusted and adjusted HRs and their 95% CIs by using treated patients from the general population as reference and compared these data with those of patients diagnosed with schizophrenia.
The secondary outcome was the association of each therapeutic class from the guideline-recommended secondary preventive treatments using a Cox proportional hazards regression model with specific therapy exposure as a time-varying covariate. We used triple therapy in patients from the general population as a reference because this protocol is currently recommended in the Danish guidelines of treatment after MI (depending on risk profile).
The assumption for linearity in the Cox proportional hazards regression was met in the primary and secondary analyses. The significance level was defined as 2-sided P < .05. Statistical analyses were performed with Stata statistical software, version 14 (StataCorp) at the Statistics Denmark server with remote access.
Sensitivity Analysis
We evaluated the role of prior exposure to any of those medications investigated because many patients might already have initiated the treatment before incident MI. Thus, we applied the primary analysis considering only patients who were naive to these medications at the time of MI and started the follow-up from 1996 (because we did not have data on prior medication use for patients included in 1995). In addition, we compared patients with schizophrenia with all others experiencing an MI excluding all with any previous psychiatric diagnosis (comparing patients diagnosed with schizophrenia with a group of psychiatrically healthy persons).
Results
Population Characteristics
The cohort included 105 018 patients with MI, including 684 patients with schizophrenia (0.7%; 483 male [70.6%]; mean [SD] age at diagnosis, 57.3 [10.6] years) and 104 334 general population patients (99.3%; 73 454 male [70.4%]; mean [SD] age at diagnosis, 61.0 [10.6] years), yielding 3827.09 and 792 608.14 person-years of follow-up, respectively. The proportion of individuals readmitted with MI was higher in the general population compared with patients with schizophrenia (1 readmission: 173 patients with schizophrenia [25.3%] vs 28 340 general population patients [27.2%]; ≥2 readmissions: 215 patients with schizophrenia [31.4%] vs 40 598 general population patients [38.9%]; P < .001). Fewer patients with schizophrenia had PCI compared with patients in the general population (134 [19.6%] vs 24 334 [23.3%]; P = .02). Furthermore, patients with schizophrenia had a higher prevalence of diabetes (222 [32.5%] vs 24 871 [23.8%]; P < .001), COPD (314 [45.9%] vs 39 094 [37.5%]; P < .001), and substance abuse (171 [25.0%] vs 4809 [4.6%]; P < .001) compared with the general population. The prevalence of hypertension (489 [71.5%] vs 77 991 [74.8%]; P = .051) was similar in both populations. In patients diagnosed with schizophrenia, a lower proportion of prescriptions of antiplatelets (581 [84.9%]), vitamin K antagonists (109 [15.9%]), β-blockers (507 [74.1%]), ACEIs (485 [70.9%]), and statins (494 [72.2%]) and the highest proportion of patients with absence of any prescriptions after MI (53 [7.8%]) were observed compared with the general population (P < .001 for all) (Table 1).
Table 1. Patient Characteristicsa.
| Characteristic | Patients With Schizophrenia (n = 684) | General Population (n = 104 334) | P Value |
|---|---|---|---|
| Age at diagnosis, mean (SD), y | 57.34 (10.64) | 61.00 (10.65) | <.001 |
| Male sex | 483 (70.6) | 73 454 (70.4) | .90 |
| No. of person-years | 3827.09 | 792 608.14 | NA |
| No. of deaths | 307 (44.9) | 27 752 (26.6) | <.001 |
| Causes of death | |||
| Cardiovascular | 209 (68.1) | 18 595 (67.0) | .69 |
| Noncardiovascular | 98 (31.9) | 9157 (33.0) | |
| MI diagnosis in calendar period | |||
| 1995-1999 | 95 (14.8) | 17 518 (17.3) | .29 |
| 2000-2004 | 150 (22.0) | 23 805 (22.9) | |
| 2005-2009 | 170 (24.8) | 26 498 (25.3) | |
| 2010-2015 | 269 (38.4) | 36 513 (34.5) | |
| No. of MI readmissions after index | |||
| 0 | 296 (43.3) | 35 396 (33.9) | <.001 |
| 1 | 173 (25.3) | 28 340 (27.2) | |
| ≥2 | 215 (31.4) | 40 598 (38.9) | |
| PCI | 134 (19.6) | 24 334 (23.3) | .02 |
| Physical comorbidity | |||
| Diabetes | 222 (32.5) | 24 871 (23.8) | <.001 |
| COPD | 314 (45.9) | 39 094 (37.5) | <.001 |
| Hypertension | 489 (71.5) | 77 991 (74.8) | .05 |
| Substance abuse | 171 (25.0) | 4809 (4.6) | <.001 |
| Prescriptions | |||
| Antiplatelets | 581 (84.9) | 95 814 (91.8) | <.001 |
| Vitamin K antagonists | 109 (15.9) | 25 216 (24.2) | <.001 |
| β-Blockers | 507 (74.1) | 88 617 (84.9) | <.001 |
| ACEIs | 485 (70.9) | 90 351 (86.6) | <.001 |
| Statins | 494 (72.2) | 91 045 (87.3) | <.001 |
| None | 53 (7.8) | 3455 (3.3) | <.001 |
Abbreviations: ACEI, angiotensin-converting enzyme inhibitors; COPD, chronic obstructive pulmonary disease; MI, myocardial infarction; NA, not applicable; PCI, percutaneous coronary intervention.
Data are presented as number (percentage) of patients unless otherwise indicated.
Drug Exposure and Mortality Rates After MI
A total of 307 patients with schizophrenia (44.9%) and 27 752 patients from the general population (26.6%) died after index (P < .001). Causes of death were equally distributed between groups, with 66.2% of all deaths being defined as cardiovascular deaths. Patients diagnosed with schizophrenia who did not receive cardioprotective treatment had the highest mortality rate (HR, 8.78; 95% CI, 4.37-17.64) compared with the general population treated, with treated patients diagnosed with schizophrenia having an increased HR of 1.97 (95% CI, 1.25-3.10). No difference was found in mortality rates between nonexposed patients from the general population (HR, 2.95; 95% CI, 2.62-3.32) and patients with schizophrenia exposed to treatment (HR, 1.97; 95% CI, 1.25-3.10) when adjusting for baseline characteristics (Table 2).
Table 2. Cox Proportional Hazards Regression Model Investigating Mortality Rates Using Treatment Exposure as a Time-Varying Covariate in Patients From the General Population and Patients With Schizophrenia.
| Treatment | Crude HR (95% CI) | Adjusted HR (95% CI)a | Adjusted HR (95% CI)b | |||
|---|---|---|---|---|---|---|
| General Population | Patients With Schizophrenia | General Population | Patients With Schizophrenia | General Population | Patients With Schizophrenia | |
| Any Treatment | ||||||
| Treated | 1 [Reference] | 2.27 (2.00-2.58) | 1 [Reference] | 2.34 (1.49-3.67) | 1 [Reference] | 1.97 (1.25-3.10) |
| Untreated | 1.22 (1.19-1.26) | 2.45 (1.93-3.11) | 2.78 (2.47-3.13) | 9.80 (4.89-19.66) | 2.95 (2.62-3.32) | 8.78 (4.37-17.64) |
| Antiplatelets | ||||||
| Treated | 1 [Reference] | 2.44 (2.12-2.82) | 1 [Reference] | 2.00 (1.16-3.46) | 1 [Reference] | 1.71 (0.99-2.95) |
| Untreated | 1.55 (1.51-1.59) | 2.86 (2.38-3.43) | 2.57 (2.38-2.77) | 9.80 (5.78-16.60) | 2.54 (2.35-2.74) | 7.97 (4.69-13.55) |
| Vitamin K Antagonists | ||||||
| Treated | 1 [Reference] | 1.63 (0.92-2.87) | 1 [Reference] | 2.53 (1.43-4.46) | 1 [Reference] | 2.21 (1.25-3.90) |
| Untreated | 0.71 (0.68-0.74) | 1.62 (1.44-1.83) | 0.94 (0.90-0.98) | 2.85 (2.52-3.22) | 0.97 (0.93-1.02) | 2.53 (2.24-2.86) |
| β-Blockers | ||||||
| Treated | 1 [Reference] | 2.20 (1.80-2.68) | 1 [Reference] | 2.23 (1.16-4.30) | 1 [Reference] | 1.96 (1.02-3.79) |
| Untreated | 1.45 (1.41-1.49) | 3.23 (2.82-3.71) | 1.67 (1.56-1.80) | 5.53 (3.47-8.82) | 1.71 (1.59-1.84) | 4.40 (2.75-7.04) |
| ACEIs | ||||||
| Treated | 1 [Reference] | 2.35 (1.95-2.82) | 1 [Reference] | 2.45 (1.27-4.73) | 1 [Reference] | 2.06 (1.07-3.99) |
| Untreated | 1.13 (1.10-1.15) | 2.45 (2.11-2.80) | 1.62 (1.51-1.74) | 4.56 (2.86-7.27) | 1.78 (1.66-1.91) | 4.08 (2.56-6.52) |
| Statins | ||||||
| Treated | 1 [Reference] | 2.16 (1.80-2.58) | 1 [Reference] | 1.87 (1.03-3.38) | 1 [Reference] | 1.56 (0.86-2.83) |
| Untreated | 1.81 (1.77-1.85) | 3.69 (3.20-4.27) | 2.46 (2.29-2.65) | 8.34 (5.10-13.65) | 2.45 (2.28-2.64) | 7.04 (4.30-11.55) |
Abbreviations: ACEI, angiotensin-converting enzyme inhibitors; HR, hazard ratio.
Adjusted for sex, birth year, age at myocardial infarction, calendar year, and percutaneous coronary intervention.
Adjusted for sex, birth year, age at myocardial infarction, calendar year, percutaneous coronary intervention, diabetes, chronic obstructive pulmonary disease, hypertension, and substance abuse.
The subanalysis of mortality rates for each cardioprotective treatment showed a generally increased mortality rate among patients diagnosed with schizophrenia who were treated compared with treated patients in the general population with the exception of antiplatelets and statins (Table 2). Similarly, patients diagnosed with schizophrenia who were not exposed to cardioprotective treatment had increased rates of mortality compared with patients from the background population who were not exposed to cardioprotective treatment (Table 2) for both the unadjusted and adjusted models. Patients diagnosed with schizophrenia who were exposed to cardioprotective treatment did not have a lower mortality rate than nonexposed patients from the background population (Table 2). In the sensitivity analysis that included only patients who initiated treatment without any prior use (n = 87 355), we observed patterns similar to the findings in the main analysis.
Medication Therapy and Mortality Rates
In the adjusted model, we found that patients diagnosed with schizophrenia did not have an increased HR for all-cause mortality compared with patients from the general population who received triple therapy as cardioprotective treatment (adjusted HRs: 1.86 [95% CI, 1.71-2.02] vs 6.65 [95% CI, 3.56-12.40] for dual therapy and 4.38 [95% CI, 3.87-4.95] vs 13.10 [95% CI, 6.51-26.35] for no treatment). With lowered treatment intensity differences being observed for dual therapy and no treatment, the difference in monotherapy did not reach statistical significance (adjusted HRs: 3.90 [95% CI, 3.53-4.30] vs 6.89 [95% CI, 2.57-18.42] for monotherapy) (Table 3). We furthermore conducted a post hoc sensitivity analysis to investigate a potential association of SMI in the comparison group. When comparing only patients diagnosed with schizophrenia with a comparison group in which all with a previous psychiatric diagnosis were excluded, we found similar results as in our main analysis.
Table 3. Cox Proportional Hazards Regression Model Investigating the Association of Different Cardiac Therapy Strategies and Mortality Rates.
| Therapy Strategy | Crude HR (95% CI) | Adjusted HR (95% CI)a | Adjusted HR (95% CI)b | |||
|---|---|---|---|---|---|---|
| General Population | Patients With Schizophrenia | General Population | Patients With Schizophrenia | General Population | Patients With Schizophrenia | |
| Triple therapy | 1 [Reference] | 2.07 (1.68-2.56) | 1 [Reference] | 1.19 (0.50-2.87) | 1 [Reference] | 1.05 (0.43-2.52) |
| Dual therapy | 1.48 (1.43-1.52) | 3.82 (3.09-4.72) | 1.76 (1.62-1.92) | 7.66 (4.11-14.28) | 1.86 (1.71-2.02) | 6.65 (3.56-12.40) |
| Monotherapy | 2.25 (2.18-2.32) | 4.37 (3.43-5.58) | 3.72 (3.38-4.10) | 8.39 (3.14-22.39) | 3.90 (3.53-4.30) | 6.89 (2.57-18.42) |
| No treatment | 1.63 (1.57-1.69) | 3.25 (2.56-4.13) | 3.99 (3.53-4.50) | 13.95 (6.95-28.01) | 4.38 (3.87-4.95) | 13.10 (6.51-26.35) |
Abbreviation: HR, hazard ratio.
Adjusted for sex, birth year, age at myocardial infarction, calendar year, and percutaneous coronary intervention.
Adjusted for sex, birth year, age at myocardial infarction, calendar period, percutaneous coronary intervention, diabetes, chronic obstructive pulmonary disease, hypertension, and substance abuse.
Discussion
In this large, nationwide, retrospective cohort study of 105 018 patients with MI with a total follow-up of 796 435 person-years, we found that when patients with schizophrenia were not exposed to cardioprotective treatment, the mortality rate was high compared with patients exposed, using treatment-exposed patients from the general population as reference. By comparing the associations between different cardiac therapy strategies and mortality rates, we found that patients with schizophrenia who received any combination of triple therapy had the lowest mortality rates.
Previous studies have found that patients with schizophrenia have increased cardiovascular mortality5,21 and a lower prescription rate for cardioprotective treatment11,15 compared with the general population. The current study was the first, to our knowledge, to investigate the association between exposure to cardioprotective medication after MI and mortality rates among patients with schizophrenia compared with the general population. Mitchell et al15 found lower odds of prescribing ACEIs (odds ratio [OR], 0.89; 95% CI, 0.81-0.98), β-blockers (OR, 0.90; 95% CI, 0.84–0.96), and statins (OR, 0.61; 95% CI 0.39–0.94) among patients with SMI compared with patients without psychiatric conditions. A recent study by Jakobsen et al11 found similar results of lower prescription rates for β-blockers, statins, and ACEIs 1 year after MI among patients with SMI compared with the general population. These findings are similar to our finding of a lower use of antiplatelets, vitamin K antagonists, β-blockers, ACEIs, and statins in patients with schizophrenia compared with the general population. Nevertheless, studies from the United States have found that patients with schizophrenia were more adherent to statins than patients without psychiatric conditions,22,23 although these studies were limited by the inclusion of patients who had at least 2 refills during a 12-month period in payer-provided care, which might have biased results toward an overrepresentation of well-functioning patients diagnosed with schizophrenia. A number of factors contribute to reduced medication adherence, including poor insight into physical illness, negative attitudes toward medication, or lack of understanding the purpose of the medications.24 These contributing factors may be even worse in patients with schizophrenia, who have cognitive impairment, which has been linked to reduced medication adherence in general.25 One study26 found that adherence to hypoglycemic and antihypertensive medications was worse than adherence to antipsychotics in patients with schizophrenia. This explanation is further supported by earlier findings that suggest that patients with schizophrenia are less likely to consult their cardiologist after cardiac events.27,28 Together, we believe that these results could point toward a deficit in establishing an appropriate cardiac treatment plan in patients with schizophrenia that causes failure in initiation and maintenance of cardioprotective treatment. This hypothesis could explain the results of a generally increased mortality rate among patients with schizophrenia, especially when the results suggest that these patients die of cardiovascular causes that might be treatable. The current data do not show causality as a result of the study design, and as such, more studies are needed to confirm the results.
Prior observational studies3,29 have suggested that concomitant exposure to multiple cardioprotective medications results in lowered mortality rates compared with monotherapy or no treatment in people experiencing cardiovascular events. The results of the current study indicate a similar cardioprotective association with mortality rates in patients diagnosed with schizophrenia and the general population when exposed to multiple secondary preventive treatments. These findings are similar to those of the study by Korhonen et al,3 who compared the association of multiple preventive therapies with mortality rates after MI in populations without psychiatric conditions and found a generally favorable association between triple therapy and mortality rates. According to European and US guidelines for the treatment of patients with MI, cardioprotective triple therapy should be considered in patients with high ischemic and bleeding risk.30,31 We do not have information on the specific risk variables in each patient or on any changes in treatment strategy provided by the cardiologists during follow-up. Therefore, the reason for persons not taking their medication may be their own initiative or their practitioner’s recommendation. In our study, treatment with cardiac triple therapy was associated with reduced excess mortality rates among patients with schizophrenia, which could suggest a necessity of implementing and monitoring multiple secondary preventive cardiac treatments in patients with schizophrenia.
The current findings might be explained by confounding by indication because patients with increased severity of MI might be more intensively treated with triple therapy, but the finding that all-cause mortality diminishes as treatment intensity increases may suggest that increased treatment intensity, despite increased cardiac risks, affects long-term outcome. The bias of increased cardiac risk associated with increased treatment intensity would be conservative and bias the results toward an increased mortality rate associated with an increased treatment intensity. On the other hand, more severely ill patients may have difficulties complying with planned treatment because of the complexity of taking multiple drugs, which could bias the results in the opposite direction. Given the increased cardiovascular risk among patients with schizophrenia, we believe that the current findings support the use of intensive cardioprotective treatments in patients with schizophrenia. In addition, these patients should be provided with accurate treatment that is closely monitored during follow-up, and treatment intensity should be increased after cardiac events in patients with schizophrenia in general because a diagnosis of schizophrenia may be associated with an increased cardiac risk, which potentially can be countered by secondary preventive cardiac treatment.
Limitations
There are several limitations that need to be addressed. First, treatment exposure was based on whether patients redeemed prescriptions; thus, we cannot be certain that the patients were actually taking the prescribed medication for the full amount of days that we calculated. However, we assume that when patients redeem prescriptions, they have the intention to take their medications. Second, we did not have information on the severity of MI (eg, left ventricular function and symptom status) or information on lifestyle factors, such as diet habits, physical activity, and smoking status; therefore, we were unable to adjust for these confounders in the regression models. Future research should attempt to assess the degree to which these factors contribute to the increased mortality in patients with schizophrenia. Despite adjusting for multiple confounders in the regression models, it is possible that some residual confounding may have occurred. We also did not have information on in-hospital medication, but because we included patients with MI from the general population in the main analyses, we expect that any bias would affect both populations equally. Third, we included patients across a broad period, and we are aware that treatment patterns and guideline recommendations for preventive cardiovascular assessment have changed significantly. However, we indirectly accounted for this difference by including patients without schizophrenia and adjusting to calendar year in the Cox proportional hazards regression model.
Conclusions
Our study suggests that patients with schizophrenia who are treated with cardioprotective treatment after MI have a lower mortality risk compared with patients who are not treated, similar to those treated in the general population. This finding suggests that some of the increased cardiac mortality among patients with schizophrenia can be reduced if these patients are efficiently administered secondary preventive treatment after cardiac events.
References
- 1.Murray CJL, Barber RM, Foreman KJ, et al. ; GBD 2013 DALYs and HALE Collaborators . Global, regional, and national disability-adjusted life years (DALYs) for 306 diseases and injuries and healthy life expectancy (HALE) for 188 countries, 1990-2013: quantifying the epidemiological transition. Lancet. 2015;386(10009):2145-2191. doi: 10.1016/S0140-6736(15)61340-X [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Roth GA, Forouzanfar MH, Moran AE, et al. Demographic and epidemiologic drivers of global cardiovascular mortality. N Engl J Med. 2015;372(14):1333-1341. doi: 10.1056/NEJMoa1406656 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Korhonen MJ, Robinson JG, Annis IE, et al. Adherence tradeoff to multiple preventive therapies and all-cause mortality after acute myocardial infarction. J Am Coll Cardiol. 2017;70(13):1543-1554. doi: 10.1016/j.jacc.2017.07.783 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Jayatilleke N, Hayes RD, Dutta R, et al. Contributions of specific causes of death to lost life expectancy in severe mental illness. Eur Psychiatry. 2017;43:109-115. doi: 10.1016/j.eurpsy.2017.02.487 [DOI] [PubMed] [Google Scholar]
- 5.Correll CU, Solmi M, Veronese N, et al. Prevalence, incidence and mortality from cardiovascular disease in patients with pooled and specific severe mental illness: a large-scale meta-analysis of 3,211,768 patients and 113,383,368 controls. World Psychiatry. 2017;16(2):163-180. doi: 10.1002/wps.20420 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Bodén R, Molin E, Jernberg T, Kieler H, Lindahl B, Sundström J. Higher mortality after myocardial infarction in patients with severe mental illness: a nationwide cohort study. J Intern Med. 2015;277(6):727-736. doi: 10.1111/joim.12329 [DOI] [PubMed] [Google Scholar]
- 7.Westman J, Eriksson SV, Gissler M, et al. Increased cardiovascular mortality in people with schizophrenia: a 24-year national register study. Epidemiol Psychiatr Sci. 2018;27(5):519-527. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Laursen TM, Munk-Olsen T, Agerbo E, Gasse C, Mortensen PB. Somatic hospital contacts, invasive cardiac procedures, and mortality from heart disease in patients with severe mental disorder. Arch Gen Psychiatry. 2009;66(7):713-720. doi: 10.1001/archgenpsychiatry.2009.61 [DOI] [PubMed] [Google Scholar]
- 9.Schulman-Marcus J, Goyal P, Swaminathan RV, et al. Comparison of trends in incidence, revascularization, and in-hospital mortality in ST-elevation myocardial infarction in patients with versus without severe mental illness. Am J Cardiol. 2016;117(9):1405-1410. doi: 10.1016/j.amjcard.2016.02.006 [DOI] [PubMed] [Google Scholar]
- 10.Druss BG, Bradford WD, Rosenheck RA, Radford MJ, Krumholz HM. Quality of medical care and excess mortality in older patients with mental disorders. Arch Gen Psychiatry. 2001;58(6):565-572. doi: 10.1001/archpsyc.58.6.565 [DOI] [PubMed] [Google Scholar]
- 11.Jakobsen L, Terkelsen CJ, Christiansen EH, et al. Severe mental illness and clinical outcome after primary percutaneous coronary intervention. Am J Cardiol. 2017;120(4):550-555. doi: 10.1016/j.amjcard.2017.05.021 [DOI] [PubMed] [Google Scholar]
- 12.Kugathasan P, Laursen TM, Grøntved S, Jensen SE, Aagaard J, Nielsen RE. Increased long-term mortality after myocardial infarction in patients with schizophrenia [published online March 16, 2018]. Schizophr Res. doi: 10.1016/j.schres.2018.03.015 [DOI] [PubMed] [Google Scholar]
- 13.Kuepper-Nybelen J, Hellmich M, Abbas S, Ihle P, Griebenow R, Schubert I. Association of long-term adherence to evidence-based combination drug therapy after acute myocardial infarction with all-cause mortality: a prospective cohort study based on claims data. Eur J Clin Pharmacol. 2012;68(10):1451-1460. doi: 10.1007/s00228-012-1274-x [DOI] [PubMed] [Google Scholar]
- 14.Ruppar TM, Cooper PS, Mehr DR, Delgado JM, Dunbar-Jacob JM. Medication adherence interventions improve heart failure mortality and readmission rates: systematic review and meta-analysis of controlled trials. J Am Heart Assoc. 2016;5(6):e002606. doi: 10.1161/JAHA.115.002606 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Mitchell AJ, Lord O, Malone D. Differences in the prescribing of medication for physical disorders in individuals with v. without mental illness: meta-analysis. Br J Psychiatry. 2012;201(6):435-443. doi: 10.1192/bjp.bp.111.094532 [DOI] [PubMed] [Google Scholar]
- 16.Lynge E, Sandegaard JL, Rebolj M. The Danish National Patient Register. Scand J Public Health. 2011;39(7)(suppl):30-33. doi: 10.1177/1403494811401482 [DOI] [PubMed] [Google Scholar]
- 17.Mors O, Perto GP, Mortensen PB. The Danish Psychiatric Central Research Register. Scand J Public Health. 2011;39(7)(suppl):54-57. doi: 10.1177/1403494810395825 [DOI] [PubMed] [Google Scholar]
- 18.WHO Collaborating Centre for Drug Statistics Methodology Guidelines for ATC Classification and DDD Assignment 2018. Oslo, Norway: WHO Collaborating Centre for Drug Statistics Methodology; 2017. [Google Scholar]
- 19.Kildemoes HW, Sørensen HT, Hallas J. The Danish National Prescription Registry. Scand J Public Health. 2011;39(7)(suppl):38-41. doi: 10.1177/1403494810394717 [DOI] [PubMed] [Google Scholar]
- 20.Helweg-Larsen K. The Danish Register of Causes of Death. Scand J Public Health. 2011;39(7)(suppl):26-29. doi: 10.1177/1403494811399958 [DOI] [PubMed] [Google Scholar]
- 21.Ringen PA, Engh J, Birkenaes AB, Dieset I, Andreassen O. Increased mortality in schizophrenia due to cardiovascular disease: a non-systematic review of epidemiology, possible causes, and interventions. Front Psychiatry. 2014;5:1-11. doi: 10.3389/fpsyt.2014.00137 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Owen-Smith A, Stewart C, Green C, et al. Adherence to common cardiovascular medications in patients with schizophrenia vs patients without psychiatric illness. Gen Hosp Psychiatry. 2016;38:9-14. doi: 10.1016/j.genhosppsych.2015.07.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Nelson LA, Graham MR, Lindsey CC, Rasu RS. Medication adherence and glycemic control in patients with psychotic disorders in the Veterans Affairs healthcare system. Pharm Pract (Granada). 2011;9(2):57-65. doi: 10.4321/S1886-36552011000200001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Ho PM, Lambert-Kerzner A, Carey EP, et al. Multifaceted intervention to improve medication adherence and secondary prevention measures after acute coronary syndrome hospital discharge: a randomized clinical trial. JAMA Intern Med. 2014;174(2):186-193. doi: 10.1001/jamainternmed.2013.12944 [DOI] [PubMed] [Google Scholar]
- 25.Dolansky MA, Hawkins MAW, Schaefer JT, et al. Association between poorer cognitive function and reduced objectively monitored medication adherence in patients with heart failure. Circ Heart Fail. 2016;9(12):e002475. doi: 10.1161/CIRCHEARTFAILURE.116.002475 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Piette JD, Heisler M, Ganoczy D, McCarthy JF, Valenstein M. Differential medication adherence among patients with schizophrenia and comorbid diabetes and hypertension. Psychiatr Serv. 2007;58(2):207-212. doi: 10.1176/ps.2007.58.2.207 [DOI] [PubMed] [Google Scholar]
- 27.Kurdyak P, Vigod S, Calzavara A, Wodchis WP. High mortality and low access to care following incident acute myocardial infarction in individuals with schizophrenia. Schizophr Res. 2012;142(1-3):52-57. doi: 10.1016/j.schres.2012.09.003 [DOI] [PubMed] [Google Scholar]
- 28.Kisely S, Smith M, Lawrence D, Cox M, Campbell LA, Maaten S. Inequitable access for mentally ill patients to some medically necessary procedures. CMAJ. 2007;176(6):779-784. doi: 10.1503/cmaj.060482 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Lafeber M, Spiering W, van der Graaf Y, et al. The combined use of aspirin, a statin, and blood pressure-lowering agents (polypill components) and the risk of vascular morbidity and mortality in patients with coronary artery disease. Am Heart J. 2013;166(2):282-289.e1. doi: 10.1016/j.ahj.2013.04.011 [DOI] [PubMed] [Google Scholar]
- 30.Amsterdam EA, Wenger NK, Brindis RG, et al. 2014 AHA/ACC guideline for the management of patients with non-ST-elevation acute coronary syndromes: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol. 2014;64(24):e139-e228. doi: 10.1016/j.jacc.2014.09.017 [DOI] [PubMed] [Google Scholar]
- 31.Ibanez B, James S, Agewall S, et al. ; ESC Scientific Document Group . 2017 ESC Guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation: the Task Force for the management of acute myocardial infarction in patients presenting with ST-segment elevation of the European Society of Cardiology (ESC). Eur Heart J. 2018;39(2):119-177. doi: 10.1093/eurheartj/ehx393 [DOI] [PubMed] [Google Scholar]