G Model
EURPSY-3212; No. of Pages 7 European Psychiatry xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
European Psychiatry journal homepage: http://www.europsy-journal.com
Original article
Schizophrenia, antipsychotic drugs and cardiovascular risk: Descriptive study in primary care M. Castillo-Sa´nchez a,b,c,*, M. Fa`bregas-Escurriola a,d, D. Berge`-Baquero e,f, Q. Foguet-Boreu a,g,h, M.I. Ferna´ndez-San Martı´n a,i, A. Goday-Arno b,j,k a
Institut Universitari d’Investigacio` en Atencio` Primaria (IDIAP) Jordi Gol, Barcelona, Spain Departamento de Medicina, Universitat Auto`noma de Barcelona, Barcelona, Spain Me´dico de familia, EAP Beso`s, Barcelona, Spain d Me´dico de familia EAP La Marina, Barcelona, Spain e Institut de Neuropsiquiatria i Addiccions, Parc de Salut Mar, Barcelona, Spain f Unitat de Recerca en Neurcocie`ncia Cognitiva, Universitat Auto`noma de Barcelona, Barcelona, Spain g Hospital de Campdeva`nol, Girona, Spain h Departamento de Ciencias Me´dicas, Facultad de Medicina, Universitat de Girona, Girona, Spain i Te´cnica de Salud ICS, Unitat Docent AFiC, Barcelona, Spain j Servicio de Endocrinologı´a y Nutricio´n, Hospital del Mar, Barcelona, Spain k CIBER Obn, Centro de Investigaciones Biomedicas en obesidad y nutricio´n. Parc de Salut Mar, Barcelona, Spain b c
A R T I C L E I N F O
Article history: Received 4 August 2014 Received in revised form 12 December 2014 Accepted 12 December 2014 Available online xxx Keywords: Schizophrenia and psychosis Antipsychotics Addiction (consumption/abuse/ dependence) Quality of care Miscellaneous
1. Background Patients affected by schizophrenia (SZ) have an increased cardiovascular morbidity and mortality [15,33]. Cardiovascular risk (CVR) has been extensively studied in the general population [34], but the aetiology in patients with SZ is extraordinarily complex [7], and has not been extensively studied. There is an association between SZ and unhealthy lifestyles [5,13,37], but some studies suggest that it is the treatment with antipsychotic drugs (TAD) that leads to weight gain [28], consequently altering the metabolism of carbohydrates and lipids [10,29,32] and increasing the prevalence of metabolic syndrome
* Corresponding author. CS Orihuela I, C. Madre Elisea 2b, cp 03300, Orihuela (Alicante), Spain. Tel.: +34966904053. E-mail address: [email protected] (M. Castillo-Sa´nchez).
and the expected CVR [3,17,20]. Similarly, a recent review [18] concluded that the increased CVR could be caused solely by the antipsychotics drugs’ side effects. However, some intrinsic factors of psychotic illness could also be involved. At the genetic level, SZ is associated with ‘‘classic’’ cardiovascular risk factors (CVRF) such as diabetes [22]. This association has been clinically confirmed [16,26,35]. Also, at the biochemical level, other markers suggest that a relationship exists between psychosis and ‘‘non-classical’’ CVRF such as differing levels of cortisol [35], interleukin-6 [16] or glutathione [1]; regardless of the TAD received. TAD improves the control of psychotic illness and it could perhaps reduce the CVR that this illness could generate per se, decreasing overall mortality [6,38]. Furthermore, drugs such as clozapine, which leads to major alterations of lipid and carbohydrate metabolism as compared to other neuroleptics, does not possess a greater CVR than other atypical antipsychotic drugs [23]. This calls into question the relationship between true CVR and the alteration of CVRF generated by the TAD. Finally, under-diagnosis and under-treatment of physical illnesses in patients with severe mental illness also seem crucial [14]. Differences can be found in both primary [2,36] and secondary [27,31] prevention in the management of these patients as compared to the general population. To mitigate this problem, various clinical guidelines and consensus-based recommendations have been developed [11]. However, there are few studies conducted in Spain and in primary care, even though they would fall within the ideal scope of these recommendations. The objective of this paper was to describe the presence of CVRF in SZ patients and also in non-SZ patients receiving TAD (NS-TAD group), in the primary care population, without established cardiovascular disease (cardiovascular primary prevention).
http://dx.doi.org/10.1016/j.eurpsy.2014.12.010 0924-9338/ß 2015 Published by Elsevier Masson SAS.
Please cite this article in press as: Castillo-Sa´nchez M, et al. Schizophrenia, antipsychotic drugs and cardiovascular risk: Descriptive study in primary care. European Psychiatry (2015), http://dx.doi.org/10.1016/j.eurpsy.2014.12.010
G Model
EURPSY-3212; No. of Pages 7 M. Castillo-Sa´nchez et al. / European Psychiatry xxx (2015) xxx–xxx
2
2. Methods Cross-sectional study from records stored in the digitized clinical historial in primary care. The data used came from six consecutive years (2006–2011). 2.1. Source of data Individual morbidity data were obtained from the digitized clinical history (eCAP) of the Catalan Health Institute, which manages 274 of the 358 (76.5%) primary care centers in Catalonia (Spain), the remainder being managed by other entities. The Database from the Information System for the Development of Research in Primary Care (SIDIAP) [4] has eCAP’s anonymous clinical information from 4,859,725 adults. Billing data from pharmacies were used for the analysis of treatments. This study was achieved by using a sub-sample of clinical logs from the SIDIAP, the SIDIAPQ, consisting of 1,936,443 adult patients (40% of the population in the SIDIAP database). SIDIAPQ integrates the records of 1365 general practitioners, whose records obtained the highest score (relative to expected prevalence values) in a matching process validated for clinical research [19]. In this validation process, a prevalence of SZ, obesity and diabetes (among other pathologies) similar to that expected according to the studies of reference was described. This manuscript has not been prepared in collaboration with the institutions holding these records, so the quality and accuracy of the results are the responsibility of the authors of the manuscript alone.
although classified as antipsychotic drugs according to the ATC classification (Anatomical, Therapeutic Chemical classification system), had different recommendations for use in psychosis treatments. This group is miscellaneous (Table 1), although the prevalence of some disorders related to the consumption of psychotropic substances, affective disorders and disorders of neurotic nature are highlighted. At no time during the study period did the patients of this group present SZ (F20 and subtypes) diagnoses; control group: non-exposed group (without SZ or TAD), randomly chosen from the rest of the study population (simple randomization). Formed by patients without SZ or TAD, randomly chosen from the rest of the study population. A sample that was twice as large as the total exposed population (SZ and NS-TAD) was generated before selecting the SIDIAPQ sub-sample.
2.3. Exclusion criteria People with F28 (‘‘other psychotic disorders not organic’’) or F29 (‘‘non-organic psychosis without specifying’’) codes without SZ (n = 8593) were excluded from the three groups at the early stage of study. Patients with dementia (F0 and subtypes) were also excluded. To determine the CVR without established cardiovascular disease, those people with a prior diagnosis of ischemic heart disease (I20, I21 I22, I23, I24 and subtypes), cerebrovascular disease (I25, I61, I63, I64, I65 and subtypes) or intermittent claudication (I73.9) were excluded.
2.2. Study population 2.4. Study variables Patients over 18 years of age, assigned to the SIDIAPQ medical teams and attended to at some point during the study period were used. Three groups were established: SZ, NS-TAD subjects and control group. The inclusion criteria were: SZ group: prevalent SZ and diagnoses of schizophrenic instances during the period of study. Selecting the ICD-10 (International Statistical Classification of Diseases and Related Health Problems, tenth version) codes included in F20 (‘‘Schizophrenia’’) and subtypes. They could or not have a prescription for TAD; NS-TAD group (no SZ, but TAD): patients that had withdrawn at least three prescriptions for any antipsychotic drug from the pharmacy during the study period were considered to be receiving TAD. We discarded sulpiride and lithium, which,
For each patient, age, sex, and the health problems of interest that were active at some time during the period of study were collected. These problems were defined according to codes found in the ICD-10, such as those related to consumption of alcohol (F10 and subtypes), cocaine (F14 and subtypes), and tobacco (smoker or non-smoker, as well as code F17). The following CVRF were collected: type 1 and 2 diabetes mellitus (E10, E11, E12, E13, E14), obesity (E66 and subtypes or BMI greater than 30), dyslipidemia (E78.0, E78.5, E78.1, E78.3, E78.4) and arterial hypertension (I10, I11, I12, I13, I15); and the average number of individual, modifiable CVRF (from 0 to 5, including smoking). The Framingham-Regicor equation [30] was applied, and expressed as a percentage of CVR after 10 years. This equation is
Table 1 Diagnoses found in the ICD-10 classification that could justify the prescription of antipsychotic drugs in the NS-TAD group (patients without schizophrenia but in treatment with antipsychotic drugs). ICD-10 codes
Mental and behavioral disorders
Total number in the NS-TAD groupa
%
F10-19 F20-29 F30-39 F40-49 F50-59 F60-69 F70-79 F80-89 F90-98 F99 F10-99
Mental and behavioral disorders due to psychoactive substance use Schizophreniab, schizotypal and delusional disorders Mood (affective) disorders Neurotic, stress-related and somatoform disorders Behavioral syndromes associated with physiological disturbances and physical factors Disorders of adult personality and behavior Mental retardation Disorders of psychological development Behavioral and emotional disorders with onset usually occurring in childhood and adolescence Unspecified mental disorder Totalc
9180 1772 6739 6456 2275 1288 965 122 654 460 29,911
30.7 5.9 22.5 21.6 7.6 4.3 3.2 0.4 2.2 1.5 100
a Diagnoses included have been active at some time during the period; some of them may be developmental diagnoses (in which case the initial diagnosis as well as the evolved recounts diagnosis). b By definition, in NS-TAD group the diagnoses of schizophrenia (F20 and subtypes) are excluded. c Total of diagnoses from ‘‘Mental and behavioral disorders’’ in ICD-10, excluding ‘‘organic mental disorders, including symptomatic disorders’’ (whose presence is exclusion criterion).
Please cite this article in press as: Castillo-Sa´nchez M, et al. Schizophrenia, antipsychotic drugs and cardiovascular risk: Descriptive study in primary care. European Psychiatry (2015), http://dx.doi.org/10.1016/j.eurpsy.2014.12.010
G Model
EURPSY-3212; No. of Pages 7 M. Castillo-Sa´nchez et al. / European Psychiatry xxx (2015) xxx–xxx
based on sex, age, biochemical values (plasma cholesterol), anthropometric measurements (blood pressure, body mass index) and dichotomous variables (diabetes yes/no, smoking yes/no). Other CVR assessment models exist, such as the Systemic COronary Risk Evaluation (SCORE) [8], but in 2005, a consensus was reached in Catalonia for the use of the Framingham-Regicor equation, due to its safety and its costeffectiveness [12]. The Framingham-Regicor estimation was calculated using two variables: recording of the calculation in the clinical history (recorded or not in each patient); the categorization of the CVR calculated in three categories: high (above 10%), moderate (between 5 and 10%) and low (less than 5%).
Frequentation (number of annual visits) of the patient to his primary care team was collected, as it probably influenced the recording of health-related problems. Prescriptions of the different marketed antipsychotic drugs withdrawn from pharmacies were collected. This study was approved by the Primary Care Research Institute IDIAP Jordi Gol Clinical Research Ethics Committee. 2.5. Analysis plan The population was separated into three groups, where the mean and standard deviations or medians and the 25 and 75 percentiles were used as the quantitative variables, and percentages were used for the prevalence of CVRF or sex (categorical variables). Demographics and clinical characteristics were analyzed using Chi-square (x2) tests for categorical variables and t-tests for continuous variables. An ANOVA test was used for comparing the three groups. After significant differences were obtained, unadjusted and adjusted comparisons for each CVRF were assessed based on a logistic regression model with each group (SZ vs control, NS-TAD vs control) as the binary outcome. Multivariate logistic regression was used for the odds ratios, adjusted for age, sex and frequentation. Multivariate logistic regression of age data was adjusted for using the other two variables (gender and frequentation), and the same method was used for these two. Odds ratios and their confidence intervals were produced. The R software v 2.15.1 and RStudio v 0.97.248 were used.
3
3. Results A total of 129,706 subjects (Fig. 1) were used and were distributed into the three study groups as follows: 98,715 in the control group, 7051 in the SZ group (2140 incidents), and 23,940 NS-TAD group (19,783 incidents). The age distribution of each group is shown in Fig. 2. The general characteristics (Tables 2a and 2b) showed significant differences. The control and SZ groups had a different gender distribution but were similar in average age. All CVRF were more frequently associated with the SZ group, except for hypertension, with the CVR estimation and a moderate-high percentage of CVR found to be similar in both SZ and control groups. The NS-TAD group was generally older and had a higher number of visits to their primary care team. It was the group with the highest proportion of all CVRF studied, except for smoking and obesity. A higher proportion of patients with high CVR was evident, although less frequently recorded. Given the demographic differences, an adjusted analysis (Table 3) was performed. The NS-TAD group had a greater presence of alcohol and cocaine use than SZ, but not for smoking. Hypertension was lower in SZ and NS-TAD. The SZ group showed obesity, diabetes, and dyslipidemia more often. By contrast, the NS-TAD group had less association with obesity and dyslipidemia, without differences in diabetes. These differences resulted in an average of CVRF which remained elevated in the SZ group; but not in the NSTAD group. The Framingham-Regicor function was used less often on the NS-TAD group, although the percentage of the estimated CVR was similar in all three groups. According to the distribution of the types of antipsychotics prescribed (Table 4), 20% of the patients with SZ had no TAD. About one-fourth to one-fifth of the patients received mixed treatment (typical and atypical TAD). The atypical TAD was the most common in both groups. The biggest difference was in the use of typical TAD, which was more than double in the NS-TAD group. Within the SZ group, we compared those patients without evidence of TAD versus those that had received some kind of TAD (Table 5). Almost all modifiable CVRF appeared more associated with the SZ with TAD subgroup, which had the largest record of CVR estimation. 4. Discussion The SZ and NS-TAD groups were associated with a greater presence of the CVRF studied; but differences in age, sex and frequentation forced an adjustment of the analysis.
Fig. 1. Flowchart of participant inclusion. NS-TAD: patients without schizophrenia but in treatment with antipsychotic drugs; SIDIAP not Q: patients followed by doctors with Registry Quality Score [19] negative.
Please cite this article in press as: Castillo-Sa´nchez M, et al. Schizophrenia, antipsychotic drugs and cardiovascular risk: Descriptive study in primary care. European Psychiatry (2015), http://dx.doi.org/10.1016/j.eurpsy.2014.12.010
G Model
EURPSY-3212; No. of Pages 7 M. Castillo-Sa´nchez et al. / European Psychiatry xxx (2015) xxx–xxx
4
Fig. 2. Age distribution (18–99 years) by strata in each group, expressed as percentage of total patients included in the group. Age at the beginning of period: January 1st 2006. Controls: control group; NS-TAD: patients without schizophrenia but in treatment with antipsychotic drugs.
Table 2a Description of the population. Non-adjusted values.
Age (years)a, mean (DS) Gender (men) Annual visits, median [ICR] Alcoholb Cocaineb Tobaccob Arterial hypertensionb Obesityb Diabetes mellitusb Dyslipidemiab Mean of CVRF for person (DS) CVR calculation, recorded Result of the calculated CVR Lower than 5% Between 5% and 10% Higher than 10%
Control group (n = 98,715)
Schizophrenia (n = 7051)
NS-TAD (n = 23,940)
P
44.7 (17.9) 47,554 (48.2%) 2.7 [1.2–5.3] 2857 (2.9%) 297 (0.3%) 27,191 (27.5%) 23,181 (23.5%) 21,707 (22.0%) 8871 (9.0%) 22,879 (23.2%) 1.05 (1.11) 24,884 (25.2%)
42.5 (14.5) 4466 (63.3%) 4.3 [2.2–7.5] 504 (7.1%) 146 (2.1%) 3697 (52.4%) 1109 (15.7%) 2315 (32.8%) 813 (11.5%) 1951 (27.7%) 1.40 (1.13) 2038 (28.9%)
52.1 (19.7) 10,546 (44.1%) 6.2 [3.3–10.2] 2110 (8.8%) 857 (3.6%) 8545 (35.7%) 7451 (31.1%) 6918 (28.9%) 3599 (15.0%) 7407 (30.9%) 1.24 (1.15) 6654 (27.8%)
< 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001
17,407 (70.0%) 5756 (23.1%) 1721 (6.9%)
1.420 (69.7%) 488 (23.9%) 130 (6.4%)
4.615 (69.4%) 1562 (23.5%) 477 (7.2%)
< 0.001 < 0.001 < 0.001
CVR: cardiovascular risk estimated with Framingham-Regicor function; CVRF: cardiovascular risk factors; DS: standard deviation; ICR: interquartile range (p25–p75); NSTAD: patients without schizophrenia but in treatment with antipsychotic drugs. a Age at the beginning of period: January 1st 2006. b Health problems activated at any moment of the study period (2006–2011).
Table 2b Crude prevalence of control group and age-standardized prevalencea of cardiovascular risk factors, expressed in percentage.
Alcohol, % Cocaine, % Tobacco, % Arterial hypertension, % Obesity, % Diabetes mellitus, % Dyslipidemia, % CVR calculation, %
Control group (n = 98,715)
Schizophreniaa (n = 7051)
NS-TADa (n = 23,940)
2.89 0.3 27.54 23.48 21.99 8.99 23.18 25.21
6.75 1.83 48.71 18.79 33.27 13.06 28.43 29.00
10.15 4.51 41.72 24.75 28.71 12.42 29.48 28.78
CVR: cardiovascular risk estimated with Framingham-Regicor function; NS-TAD: patients without schizophrenia but in treatment with antipsychotic drugs. a Direct standardization by age (3 age groups: < 50 years, 50–75 years, > 75 years), with regard to control group.
After adjustment, only hypertension occurred less frequently in SZ than in the control group. The rest of the CVRF were associated with SZ, and the average of CVRF remained high; these results were increased by TAD. The NS-TAD group continued to be associated with drug use (tobacco, alcohol and cocaine); but hypertension, obesity and dyslipidemia had a lower prevalence than in the control group, without differences in diabetes. Consequently, the CVRF average decreased. The estimation of CVR was less common in the NS-TAD group. 4.1. Limitations and strengths In addition to those risks inherent to any population-based study of an electronic database, our study had other limitations.
Please cite this article in press as: Castillo-Sa´nchez M, et al. Schizophrenia, antipsychotic drugs and cardiovascular risk: Descriptive study in primary care. European Psychiatry (2015), http://dx.doi.org/10.1016/j.eurpsy.2014.12.010
G Model
EURPSY-3212; No. of Pages 7 M. Castillo-Sa´nchez et al. / European Psychiatry xxx (2015) xxx–xxx
5
Table 3 Multivariate logistic regression adjusted by age, gender and frequentation, with respect to control group. Schizophrenia (n = 7051) OR (95% CI) Age Gender (men) Alcohol Cocaine Tobacco Arterial hypertension Obesity Diabetes mellitus Dyslipidemia Mean of CVRF for person CVR registered Result of the calculated CVR
0.98 2.20 1.82 4.15 2.32 0.49 1.61 1.18 1.15 1.25 0.99 1.00
(0.98–0.98)a (2.09–2.31)b (1.64–2.02)b (3.36–5.11)b (2.20–2.44)b (0.45–0.53)a (1.52–1.70)b (1.09–1.28)b (1.08–1.22)b (1.22–1.28)b (0.94–1.06) (1.00–1.01)
P
NS-TAD (n = 23,940) OR (95% CI)
< 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.02 < 0.01
1.01 1.09 3.00 14.70 1.68 0.64 0.88 0.97 0.84 0.97 0.62 1.00
P
(1.01–1.01)b (1.05–1.12)b (2.81–3.19)b (12.79–16.94)b (1.62–1.74)b (0.62–0.67)a (0.85–0.91)a (0.93–1.02) (0.81–0.87)a (0.96–0.99)a (0.60–0.64)a (1.00–1.01)
< 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 0.06 < 0.001 < 0.001 < 0.001 < 0.02
95% CI: 95% confidence interval; CVR: cardiovascular risk estimated with Framingham-Regicor function; CVRF: cardiovascular risk factors; NS-TAD: patients without schizophrenia but in treatment with antipsychotic drugs; OR: Odds ratio. a Variables less associated with study group than with control group (P < 0.05). b Variables with greater association with the study group with respect to the control group (P < 0.05).
Although a long-term study of cross-sectional design could create doubts, we believed the use of 6 years to be correct for two reasons. Firstly, the implementation of the eCAP software was carried out starting from 2005, and data from the first year could be unreliable. Working with chronic health problems (hypertension, diabetes, dyslipidemia) and/or of difficult final resolution (substance abuse) would reduce the bias. Secondly, if the cross-section had been made at the end of the period, the health problems resolved during the period (perhaps present since long before) would not have been taken into account, ignoring their effect on the patient’s CVR. Including current patients (both SZ and NS-TAD) implies a survival bias, since all the subjects in the control group remained in the cohort throughout the period. This bias mostly affected the NSTAD group, which, besides being representatively older, over 80% were new cases. The NS-TAD group was older and had no previous cardiovascular events, which was another bias: it may be a group of patients with increased predisposition to good cardiovascular health, which would be consistent with the lower CVRF frequency obtained. Another limitation was not having requested biochemical data or measurements of blood pressure, which forced a simplified diagnosis, such as yes/no categories. The negative effect decreased by selecting ‘‘good recorder’’ doctors (SIDIAPQ), which also allowed us to indirectly include measurements not taken at the health center (mutual insurance companies, external medical clinics not belonging to ICS), but which had generated a diagnostic label. The definition of ‘‘patient in TAD’’ does not take into account the length of time the patients were in TAD or if recruited patients were in early phase of treatment, middle or long standing treatment. The data on TAD previous to the study period was not available either. It is an important limitation because the side Table 4 Percentage of patients treated with typical and atypical antipsychotics during the study period, according to the group.
Without antipsychotics, % Typical antipsychotics, % Atypical antipsychotics, % Typical and atypical antipsychotics, %
Schizophreniaa (n = 7051)
NS-TADa (n = 23,940)
1437 564 3141 1909
– 4314 (18.0%) 14,960 (62.5%) 4666 (19.5%)
(20.4%) (8.0%) (44.5%) (27.1%)
NS-TAD: patients without schizophrenia but in treatment with antipsychotic drugs. a Chi-squared test (schizophrenia and NS-TAD versus type of antipsychotic treatment) (P < 0.05).
effects are often dose-dependant and dependant on the time the patient is on the medications. Data on drugs used in TAD were obtained from the ICS pharmacy billing. Therefore, the drugs prescribed by external practitioners were not taken into account. The accuracy of these data was questionable, as we could not be sure of how many patients were really drug-free. Lastly, the adherence to the treatment by the patients was not assessed. Finally, the population cared for in primary care does not necessarily represent the general population, especially in the study groups. Patients who sporadically visited their primary care team lacked the necessary follow-up for the proper management of CVRF. In the SZ group, the selection bias could be accentuated with the loss of young patients having low comorbidity, who did not visit their primary care team. However, it is interesting to examine the characteristics of this sub-population, as it will allow us to focus on preventive strategies with greater efficiency in the primary care field. On the other hand, this study possesses a large sample and a wide population base, as well as a reliable source of data [4,19]. A NS-TAD group study on a large population base is a way of
Table 5 Multivariate logistic regression adjusted by age, gender and frequentation of the schizophrenia subgroup with antipsychotic treatment, with respect to the schizophrenia subgroup without evidence of treatment.
Age Gender (men) Alcohol Cocaine Tobacco Arterial hypertension Obesity Diabetes mellitus Dyslipidemia Mean of CVRF for person CVR registered Result of the calculated CVR
Schizophrenia with antipsychotic treatment (n = 5614) OR (95% CI)
P
1.00 0.99 0.94 1.65 1.17 1.06 1.68 1.28 1.19 1.19 1.26 0.99
0.34 0.85 0.62 < 0.05 < 0.02 0.56 < 0.001 < 0.03 < 0.03 < 0.001 < 0.003 0.16
(1.00–1.01) (0.86–1.12) (0.75–1.19) (1.03–2.77)a (1.03–1.33)a (0.88–1.27) (1.46–1.94)a (1.03–1.59)a (1.03–1.38)a (1.12–1.27)a (1.08–1.46)a (0.97–1.01)
95% CI: 95% confidence interval; CVR: cardiovascular risk estimated with Framingham-Regicor function; CVRF: cardiovascular risk factors; NS-TAD: patients without schizophrenia but in treatment with antipsychotic drugs; OR: Odds ratio. a Variables more associated with the study group with respect to the control group (P < 0.05).
Please cite this article in press as: Castillo-Sa´nchez M, et al. Schizophrenia, antipsychotic drugs and cardiovascular risk: Descriptive study in primary care. European Psychiatry (2015), http://dx.doi.org/10.1016/j.eurpsy.2014.12.010
G Model
EURPSY-3212; No. of Pages 7 M. Castillo-Sa´nchez et al. / European Psychiatry xxx (2015) xxx–xxx
6
dichotomizing the effect of NS and the TAD drugs. The perspective of primary care could be useful and complementary for a multidisciplinary approach.
possible under-diagnosis. However, SZ could impair metabolism of carbohydrates [22], even before starting the TAD [26,35], which would aggravate this alteration.
4.2. Drug abuse: alcohol and cocaine
4.7. Dyslipidemia
The consumption of cocaine and alcohol in the SZ group were lower than those reported in other studies [21], leading to the belief that there was an under-recording of the use of these substances. This could be due to the professionals not asking questions on a topic that may discomfort the patient, or it could be that the patient hid their habits unless they generated any clinical complications later on. However, the high proportions within the study groups were credible.
The prevalence of dyslipidemia is not comparable to the other Spanish studies (based on biochemical values), but we found similar results (29.4%) in international literature [25]. The SZ group had an increased frequency of dyslipidemia than the control group, unlike the NS-TAD group. It has been postulated that dyslipidemia would increase with the evolution of SZ, with an unclear role of the TAD [32,39]. 4.8. Estimation of CVR (mean of CVRF and Framingham-Regicor)
4.3. Smoking The SZ group results were assessed according to two individual Spanish studies. The first is the CLAMORS study [3], with 1452 patients with psychosis and TAD, monitored by psychiatrists at the outpatient level. The average age was slightly lower (40.7 years old) and it did not have a similar gender distribution (60.9% male). Their results on smoking (53.7%) were consistent with ours. The newer RICAVA study [2], with 733 patients with SZ, hospitalized and treated with atypical antipsychotics, of lower mean age (37.8 years old) and higher percentage of males (71.8%) had a higher proportion of smokers (71%). Smoking remained associated with both the SZ and NS-TAD groups even after the adjusted analysis, resulting in about twice as many smokers as in the control group, in line with the literature [9]. 4.4. Hypertension The SZ population in the present study had a lower proportion of hypertension than the control group, and was similar to the RICAVA study [2]. Both the SZ and the NS-TAD groups were strongly associated with a lesser presence of hypertension. It could be that blood pressure was not measured at the physician’s office with the same frequency. Other similarly designed studies obtained similar figures, and this result was attributed to under-diagnosis [36]. 4.5. Obesity The SZ group had similar values as previous studies on the Spanish population (between 24% and 30.8%) [2,3]. The NS-TAD group also had a higher prevalence than the control group. After adjusted analysis, SZ was associated with an increased prevalence of obesity, contrary to the NS-TAD group. This could be due to the increased susceptibility of patients with SZ. Another hypothesis could be that the profile of the patients with SZ who visited the general practitioners is that of a psychiatrist-monitored, mentally-stable patient, allowing the primary care team to focus on aspects of physical health. In other TAD-treatable pathologies, the prioritization of other problems that motivate visiting primary care physicians would justify a lower diagnostic record. 4.6. Diabetes mellitus We found a similar prevalence in the CLAMORS study (14.2%) [3]. After adjusting, only the SZ group had more diabetes, with no differences between the control and the NS-TAD groups. Some studies [18,24] concluded that the TAD was an independent factor that caused diabetes, so we contemplated a
After the adjusted analysis, the NS-TAD group showed half the frequency as the control group. We concluded that there are three reasons for this: firstly, the exclusion of more than the 16% of the patients older than 75 from the CVR estimation affected the results. Secondly, the bias described in ‘‘Limitations’’ related to their good cardiovascular health. Finally, a relationship between the scarce CVR calculation and the low prevalence of obesity, dyslipidemia, and hypertension may exist, because quantitative data that defines the diagnosis of these disorders (body mass index, cholesterol, blood pressure) are necessary for the CVR calculation. The scarce CVR calculation would imply that the CVRF have been evaluated with lower frequency in the NS-TAD group than in the control group, corroborating the lesser attention that the patients with severe mental illnesses received in primary prevention of CVR [14]. As for the risk itself, the SZ and NS-TAD groups had a mean value similar to the control group, unlike other studies [3,17,20]. Adjusted analysis counteracts the great weight that age has on the calculation of CVR values, especially in the NS-TAD group. 4.9. Type of TAD prescribed The typical antipsychotics drugs prevailed in the NS-TAD group, which could be associated with the lower registering of CVRF, given that atypical antipsychotics have been linked with increased incidences of metabolic syndrome [10]. However, this relationship is not clear [18,28]. A fifth of the SZ patients did not have TAD on record, but we cannot be sure that they did not receive TAD through other external means. We noticed a greater presence of modifiable CVRF in SZ patients with TAD than in SZ without TAD (except for hypertension, which was similar). We postulated that the TAD would act synergistically with the predisposition associated with SZ [39], although we must assess the under-recording of the CVR calculation in the SZ without TAD subgroup as a sign of worse management in cardiovascular prevention (either because they received TAD and/or management by external services).
5. Conclusions The differences between the SZ and control groups in obesity, diabetes and dyslipidemia support the hypothesis of SZ as a systemic disease associated with CVR. Within the SZ group, the TAD subgroup had an increased CVR, unlike other diseases treated with TAD (NS-TAD group). We believe that, despite having a multifactorial origin, the under-recording of the CVR calculation estimated in the NS-TAD group could reflect under-diagnosis of CVRF that has been related
Please cite this article in press as: Castillo-Sa´nchez M, et al. Schizophrenia, antipsychotic drugs and cardiovascular risk: Descriptive study in primary care. European Psychiatry (2015), http://dx.doi.org/10.1016/j.eurpsy.2014.12.010
G Model
EURPSY-3212; No. of Pages 7 M. Castillo-Sa´nchez et al. / European Psychiatry xxx (2015) xxx–xxx
to patients with severe mental illness in the literature, particularly hypertension, which also showed a low association with SZ. Our results could probably be extrapolated to other populations attended in primary care, but cannot be extended to all patients with SZ or TAD. Studies that avoid the effect of under-diagnosis of CVRF are required to clarify the role of the TAD and SZ in the development of CVRF and the real CVR. Multidisciplinary assessment of medical and behavioral conditions is required to improve the CVR profile of patients with SZ and with TAD, because they have special needs of care.
Disclosure of interest The authors declare that they have no conflicts of interest concerning this article. Acknowledgement Thanks to the XB Aid for Primary Care Research in ICSBarcelona. Thank you to Mario Fon ([email protected]) for revising the manuscript. Funding: There are no funding sources. References [1] Ballesteros A, Jiang P, Summerfelt A, Du X, Chiappelli J, O’Donnell P, et al. No evidence of exogenous origin for the abnormal glutathione redox state in schizophrenia. Schizophr Res 2013;146(1–3):184–9. ˜ as F, Banegas JR, Casademont J, Riesgo Y, Varela C. Prevalence [2] Bernardo M, Can and awareness of cardiovascular risk factors in patients with schizophrenia: a cross-sectional study in a low cardiovascular disease risk geographical area. Eur Psychiatry 2009;24(7):431–41. [3] Bobes J, Arango C, Aranda P, Carmena R, Garcia-Garcia M, Rejas J. Cardiovascular and metabolic risk in outpatients with schizophrenia treated with antipsychotics: results of the CLAMORS Study. Schizophr Res 2007;90(1– 3):162–73. [4] Bolı´bar B, Fina Avile´s F, Morros R, Garcia-Gil Mdel M, Hermosilla E, Ramos R, et al. [SIDIAP database: electronic clinical records in primary care as a source of information for epidemiologic research]. Med Clin (Barc) 2012;138(14):617– 21 [Spanish]. [5] Brown S, Birtwistle J, Roe L, Thompson C. The unhealthy lifestyle of people with schizophrenia. Psychol Med 1999;29:697–701. [6] Bushe CJ, Taylor M, Haukka J. Mortality in schizophrenia: a measurable clinical endpoint. J Psychopharmacol 2010;24(4 Suppl.):17–25. [7] Castillo Sa´nchez M, Fa`bregas Escurriola M, Berge` Baquero D, Goday Arno A, Valle`s Callol JA. [Psychosis, cardiovascular risk and associated mortality: are we on the right track?] Clin Investig Arterioscler 2014;26(1):23–32 [Review. Spanish]. [8] Conroy RM, Pyo¨ra¨la K, Fitzgerald AP, Sans S, Menotti A, De Backer G, et al. Estimation of ten-year risk of fatal cardiovascular disease in Europe: the SCORE project. Eur Heart J 2003;24:987–1003. [9] Correll CU. Balancing efficacy and safety in treatment with antipsychotics. CNS Spectr 2007;12(Suppl. 17):12–20. [10] De Hert M, Schreurs V, Sweers K, Van Eyck D, Hanssens L, Sinko S, et al. Typical and atypical antipsychotics differentially affect long-term incidence rates of the metabolic syndrome in first-episode patients with schizophrenia: a retrospective chart review. Schizophr Res 2008;101(1–3):295–303. [11] De Hert M, Dekker JM, Wood D, Kahl KG, Holt RI, Mo¨ller HJ. Cardiovascular disease and diabetes in people with severe mental illness position statement from the European Psychiatric Association (EPA), supported by the European Association for the Study of Diabetes (EASD) and the European Society of Cardiology (ESC). Eur Psychiatry 2009;24(6):412–24. [12] Departament de Salut - Generalitat de Catalunya. Llibre blanc: consens sobre les activitats preventives a l’edat adulta dins l’atencio´ prima`ria, 2nd ed, Barcelona: Direccio´ General de Salut Pu´blica; 2005. [13] Dipasquale S, Pariante CM, Dazzan P, Aguglia E, McGuire P, Mondelli V. The dietary pattern of patients with schizophrenia: a systematic review. J Psychiatr Res 2013;47(2):197–207. [14] Fagiolini A, Goracci A. The effects of undertreated chronic medical illnesses in patients with severe mental disorders. J Clin Psychiatry 2009;70(3 Suppl.): 22–9.
7
[15] Fan Z, Wu Y, Shen J, Ji T, Zhan R. Schizophrenia and the risk of cardiovascular diseases: a meta-analysis of thirteen cohort studies. J Psychiatr Res 2013; 47(11):1549–56. [16] Fernandez-Egea E, Bernardo M, Donner T, Conget I, Parellada E, Justicia A, et al. Metabolic profile of antipsychotic-naive individuals with non-affective psychosis. Br J Psychiatry 2009;194(5):434–8. [17] Fleischhacker WW, Siu CO, Bode´n R, Pappadopulos E, Karayal ON, Kahn RS, et al. Metabolic risk factors in first-episode schizophrenia: baseline prevalence and course analysed from the European First-Episode Schizophrenia Trial. Int J Neuropsychopharmacol 2013;16(5):987–95. [18] Foley DL, Morley KI. Systematic review of early cardiometabolic outcomes of the first treated episode of psychosis. Arch Gen Psychiatry 2011;68:609–16. [19] Garcı´a-Gil Mdel M, Hermosilla E, Prieto-Alhambra D, Fina F, Rosell M, Ramos R, et al. Construction and validation of a scoring system for the selection of highquality data in a Spanish population primary care database (SIDIAP). Inform Prim Care 2011;19(3):135–45. [20] Goff DC, Sullivan LM, McEvoy JP, Meyer JM, Nasrallah HA, Daumit GL, et al. A comparison of ten-year cardiac risk estimates in schizophrenia patients from the CATIE study and matched controls. Schizophr Res 2005;80(1):45–53. [21] Green AI, Brown ES. Comorbid schizophrenia and substance abuse. J Clin Psychiatry 2006;67(9):e08. [22] Hansen T, Ingason A, Djurovic S, Melle I, Fenger M, Gustafsson O, et al. At-risk variant in TCF7L2 for type II diabetes increases risk of schizophrenia. Biol Psychiatry 2011;70(1):59–63. [23] Kelly DL, McMahon RP, Liu F, Love RC, Wehring HJ, Shim JC, et al. Cardiovascular disease mortality in patients with chronic schizophrenia treated with clozapine: a retrospective cohort study. J Clin Psychiatry 2010;71(3):304–11. [24] Kessing LV, Thomsen AF, Mogensen UB, Andersen PK. Treatment with antipsychotics and the risk of diabetes in clinical practice. Br J Psychiatry 2010;197(4):266–71. [25] Kilbourne AM, Morden NE, Austin K, Ilgen M, McCarthy JF, Dalack G, et al. Excess heart-disease-related mortality in a national study of patients with mental disorders: identifying modifiable risk factors. Gen Hosp Psychiatry 2009;31(6):555–63. [26] Kirkpatrick B, Miller BJ, Garcia-Rizo C, Fernandez-Egea E, Bernardo M. Is abnormal glucose tolerance in antipsychotic-naive patients with nonaffective psychosisconfounded by poor health habits? Schizophr Bull 2012;38(2): 280–4. [27] 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): 713–20. [28] Leucht S, Corves C, Arbter D, Engel RR, Li C, Davis JM. Second-generation versus first-generation antipsychotic drugs for schizophrenia: a meta-analysis. Lancet 2009;373(9657):31–41. [29] Manu P, Correll CU, Wampers M, van Winkel R, Yu W, Shiffeldrim D, et al. Dysmetabolic features of the overweight patients receiving antipsychotic drugs: a comparison with normal weight and obese subjects. Eur Psychiatry 2014;29(3):179–82. [30] Marrugat J, Solanas P, D’Agostino R, Sullivan L, Ordovas J, Cordo´n F, et al. [Coronary risk estimation in Spain using a calibrated Framingham function]. Rev Esp Cardiol 2003;56(3):253–61 [Spanish]. [31] Mitchell AJ, Lord O. Do deficits in cardiac care influence high mortality rates in schizophrenia? A systematic review and pooled analysis. J Psychopharmacol 2010;24(4 Suppl.):69–80. [32] Mitchell AJ, Vancampfort D, Sweers K, van Winkel R, Yu W, De Hert M. Prevalence of metabolic syndrome and metabolic abnormalities in schizophrenia and related disorders–a systematic review and meta-analysis. Schizophr Bull 2013;39(2):306–18. [33] Nordentoft M, Wahlbeck K, Ha¨llgren J, Westman J, Osby U, Alinaghizadeh H, et al. Excess mortality, causes of death and life expectancy in 270,770 patients with recent onset of mental disorders in Denmark, Finland and Sweden. PLoS One 2013;8(1):e55176. [34] O’Donnell CJ, Elosua R. [Cardiovascular risk factors. Insights from Framingham Heart Study]. Rev Esp Cardiol 2008;61(3):299–310 [Review. Spanish]. [35] Ryan MC, Collins P, Thakore JH. Impaired fasting glucose tolerance in firstepisode, drug-naive patients with schizophrenia. Am J Psychiatry 2003; 160(2):284–9. [36] Smith DJ, Langan J, McLean G, Guthrie B, Mercer SW. Schizophrenia is associated with excess multiple physical-health comorbidities but low levels of recorded cardiovascular disease in primary care: cross-sectional study. BMJ Open 2013;3(4). [37] Spelman LM, Walsh PI, Sharifi N, Collins P, Thakore JH. Impaired glucose tolerance in first-episode drug-naı¨ve patients with schizophrenia. Diabet Med 2007;24:481–5. [38] Tiihonen J, Lo¨nnqvist J, Wahlbeck K, Klaukka T, Niskanen L, Tanskanen A, et al. 11-year follow-up of mortality in patients with schizophrenia: a populationbased cohort study (FIN11 study). Lancet 2009;374(9690):620–7. [39] Vancampfort D, Wampers M, Mitchell AJ, Correll CU, De Herdt A, Probst M, et al. A meta-analysis of cardio-metabolic abnormalities in drug naı¨ve, firstepisode and multi-episode patients with schizophrenia versus general population controls. World Psychiatry 2013;12(3):240–50.
Please cite this article in press as: Castillo-Sa´nchez M, et al. Schizophrenia, antipsychotic drugs and cardiovascular risk: Descriptive study in primary care. European Psychiatry (2015), http://dx.doi.org/10.1016/j.eurpsy.2014.12.010
EURPSY-3212; No. of Pages 7 European Psychiatry xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
European Psychiatry journal homepage: http://www.europsy-journal.com
Original article
Schizophrenia, antipsychotic drugs and cardiovascular risk: Descriptive study in primary care M. Castillo-Sa´nchez a,b,c,*, M. Fa`bregas-Escurriola a,d, D. Berge`-Baquero e,f, Q. Foguet-Boreu a,g,h, M.I. Ferna´ndez-San Martı´n a,i, A. Goday-Arno b,j,k a
Institut Universitari d’Investigacio` en Atencio` Primaria (IDIAP) Jordi Gol, Barcelona, Spain Departamento de Medicina, Universitat Auto`noma de Barcelona, Barcelona, Spain Me´dico de familia, EAP Beso`s, Barcelona, Spain d Me´dico de familia EAP La Marina, Barcelona, Spain e Institut de Neuropsiquiatria i Addiccions, Parc de Salut Mar, Barcelona, Spain f Unitat de Recerca en Neurcocie`ncia Cognitiva, Universitat Auto`noma de Barcelona, Barcelona, Spain g Hospital de Campdeva`nol, Girona, Spain h Departamento de Ciencias Me´dicas, Facultad de Medicina, Universitat de Girona, Girona, Spain i Te´cnica de Salud ICS, Unitat Docent AFiC, Barcelona, Spain j Servicio de Endocrinologı´a y Nutricio´n, Hospital del Mar, Barcelona, Spain k CIBER Obn, Centro de Investigaciones Biomedicas en obesidad y nutricio´n. Parc de Salut Mar, Barcelona, Spain b c
A R T I C L E I N F O
Article history: Received 4 August 2014 Received in revised form 12 December 2014 Accepted 12 December 2014 Available online xxx Keywords: Schizophrenia and psychosis Antipsychotics Addiction (consumption/abuse/ dependence) Quality of care Miscellaneous
1. Background Patients affected by schizophrenia (SZ) have an increased cardiovascular morbidity and mortality [15,33]. Cardiovascular risk (CVR) has been extensively studied in the general population [34], but the aetiology in patients with SZ is extraordinarily complex [7], and has not been extensively studied. There is an association between SZ and unhealthy lifestyles [5,13,37], but some studies suggest that it is the treatment with antipsychotic drugs (TAD) that leads to weight gain [28], consequently altering the metabolism of carbohydrates and lipids [10,29,32] and increasing the prevalence of metabolic syndrome
* Corresponding author. CS Orihuela I, C. Madre Elisea 2b, cp 03300, Orihuela (Alicante), Spain. Tel.: +34966904053. E-mail address: [email protected] (M. Castillo-Sa´nchez).
and the expected CVR [3,17,20]. Similarly, a recent review [18] concluded that the increased CVR could be caused solely by the antipsychotics drugs’ side effects. However, some intrinsic factors of psychotic illness could also be involved. At the genetic level, SZ is associated with ‘‘classic’’ cardiovascular risk factors (CVRF) such as diabetes [22]. This association has been clinically confirmed [16,26,35]. Also, at the biochemical level, other markers suggest that a relationship exists between psychosis and ‘‘non-classical’’ CVRF such as differing levels of cortisol [35], interleukin-6 [16] or glutathione [1]; regardless of the TAD received. TAD improves the control of psychotic illness and it could perhaps reduce the CVR that this illness could generate per se, decreasing overall mortality [6,38]. Furthermore, drugs such as clozapine, which leads to major alterations of lipid and carbohydrate metabolism as compared to other neuroleptics, does not possess a greater CVR than other atypical antipsychotic drugs [23]. This calls into question the relationship between true CVR and the alteration of CVRF generated by the TAD. Finally, under-diagnosis and under-treatment of physical illnesses in patients with severe mental illness also seem crucial [14]. Differences can be found in both primary [2,36] and secondary [27,31] prevention in the management of these patients as compared to the general population. To mitigate this problem, various clinical guidelines and consensus-based recommendations have been developed [11]. However, there are few studies conducted in Spain and in primary care, even though they would fall within the ideal scope of these recommendations. The objective of this paper was to describe the presence of CVRF in SZ patients and also in non-SZ patients receiving TAD (NS-TAD group), in the primary care population, without established cardiovascular disease (cardiovascular primary prevention).
http://dx.doi.org/10.1016/j.eurpsy.2014.12.010 0924-9338/ß 2015 Published by Elsevier Masson SAS.
Please cite this article in press as: Castillo-Sa´nchez M, et al. Schizophrenia, antipsychotic drugs and cardiovascular risk: Descriptive study in primary care. European Psychiatry (2015), http://dx.doi.org/10.1016/j.eurpsy.2014.12.010
G Model
EURPSY-3212; No. of Pages 7 M. Castillo-Sa´nchez et al. / European Psychiatry xxx (2015) xxx–xxx
2
2. Methods Cross-sectional study from records stored in the digitized clinical historial in primary care. The data used came from six consecutive years (2006–2011). 2.1. Source of data Individual morbidity data were obtained from the digitized clinical history (eCAP) of the Catalan Health Institute, which manages 274 of the 358 (76.5%) primary care centers in Catalonia (Spain), the remainder being managed by other entities. The Database from the Information System for the Development of Research in Primary Care (SIDIAP) [4] has eCAP’s anonymous clinical information from 4,859,725 adults. Billing data from pharmacies were used for the analysis of treatments. This study was achieved by using a sub-sample of clinical logs from the SIDIAP, the SIDIAPQ, consisting of 1,936,443 adult patients (40% of the population in the SIDIAP database). SIDIAPQ integrates the records of 1365 general practitioners, whose records obtained the highest score (relative to expected prevalence values) in a matching process validated for clinical research [19]. In this validation process, a prevalence of SZ, obesity and diabetes (among other pathologies) similar to that expected according to the studies of reference was described. This manuscript has not been prepared in collaboration with the institutions holding these records, so the quality and accuracy of the results are the responsibility of the authors of the manuscript alone.
although classified as antipsychotic drugs according to the ATC classification (Anatomical, Therapeutic Chemical classification system), had different recommendations for use in psychosis treatments. This group is miscellaneous (Table 1), although the prevalence of some disorders related to the consumption of psychotropic substances, affective disorders and disorders of neurotic nature are highlighted. At no time during the study period did the patients of this group present SZ (F20 and subtypes) diagnoses; control group: non-exposed group (without SZ or TAD), randomly chosen from the rest of the study population (simple randomization). Formed by patients without SZ or TAD, randomly chosen from the rest of the study population. A sample that was twice as large as the total exposed population (SZ and NS-TAD) was generated before selecting the SIDIAPQ sub-sample.
2.3. Exclusion criteria People with F28 (‘‘other psychotic disorders not organic’’) or F29 (‘‘non-organic psychosis without specifying’’) codes without SZ (n = 8593) were excluded from the three groups at the early stage of study. Patients with dementia (F0 and subtypes) were also excluded. To determine the CVR without established cardiovascular disease, those people with a prior diagnosis of ischemic heart disease (I20, I21 I22, I23, I24 and subtypes), cerebrovascular disease (I25, I61, I63, I64, I65 and subtypes) or intermittent claudication (I73.9) were excluded.
2.2. Study population 2.4. Study variables Patients over 18 years of age, assigned to the SIDIAPQ medical teams and attended to at some point during the study period were used. Three groups were established: SZ, NS-TAD subjects and control group. The inclusion criteria were: SZ group: prevalent SZ and diagnoses of schizophrenic instances during the period of study. Selecting the ICD-10 (International Statistical Classification of Diseases and Related Health Problems, tenth version) codes included in F20 (‘‘Schizophrenia’’) and subtypes. They could or not have a prescription for TAD; NS-TAD group (no SZ, but TAD): patients that had withdrawn at least three prescriptions for any antipsychotic drug from the pharmacy during the study period were considered to be receiving TAD. We discarded sulpiride and lithium, which,
For each patient, age, sex, and the health problems of interest that were active at some time during the period of study were collected. These problems were defined according to codes found in the ICD-10, such as those related to consumption of alcohol (F10 and subtypes), cocaine (F14 and subtypes), and tobacco (smoker or non-smoker, as well as code F17). The following CVRF were collected: type 1 and 2 diabetes mellitus (E10, E11, E12, E13, E14), obesity (E66 and subtypes or BMI greater than 30), dyslipidemia (E78.0, E78.5, E78.1, E78.3, E78.4) and arterial hypertension (I10, I11, I12, I13, I15); and the average number of individual, modifiable CVRF (from 0 to 5, including smoking). The Framingham-Regicor equation [30] was applied, and expressed as a percentage of CVR after 10 years. This equation is
Table 1 Diagnoses found in the ICD-10 classification that could justify the prescription of antipsychotic drugs in the NS-TAD group (patients without schizophrenia but in treatment with antipsychotic drugs). ICD-10 codes
Mental and behavioral disorders
Total number in the NS-TAD groupa
%
F10-19 F20-29 F30-39 F40-49 F50-59 F60-69 F70-79 F80-89 F90-98 F99 F10-99
Mental and behavioral disorders due to psychoactive substance use Schizophreniab, schizotypal and delusional disorders Mood (affective) disorders Neurotic, stress-related and somatoform disorders Behavioral syndromes associated with physiological disturbances and physical factors Disorders of adult personality and behavior Mental retardation Disorders of psychological development Behavioral and emotional disorders with onset usually occurring in childhood and adolescence Unspecified mental disorder Totalc
9180 1772 6739 6456 2275 1288 965 122 654 460 29,911
30.7 5.9 22.5 21.6 7.6 4.3 3.2 0.4 2.2 1.5 100
a Diagnoses included have been active at some time during the period; some of them may be developmental diagnoses (in which case the initial diagnosis as well as the evolved recounts diagnosis). b By definition, in NS-TAD group the diagnoses of schizophrenia (F20 and subtypes) are excluded. c Total of diagnoses from ‘‘Mental and behavioral disorders’’ in ICD-10, excluding ‘‘organic mental disorders, including symptomatic disorders’’ (whose presence is exclusion criterion).
Please cite this article in press as: Castillo-Sa´nchez M, et al. Schizophrenia, antipsychotic drugs and cardiovascular risk: Descriptive study in primary care. European Psychiatry (2015), http://dx.doi.org/10.1016/j.eurpsy.2014.12.010
G Model
EURPSY-3212; No. of Pages 7 M. Castillo-Sa´nchez et al. / European Psychiatry xxx (2015) xxx–xxx
based on sex, age, biochemical values (plasma cholesterol), anthropometric measurements (blood pressure, body mass index) and dichotomous variables (diabetes yes/no, smoking yes/no). Other CVR assessment models exist, such as the Systemic COronary Risk Evaluation (SCORE) [8], but in 2005, a consensus was reached in Catalonia for the use of the Framingham-Regicor equation, due to its safety and its costeffectiveness [12]. The Framingham-Regicor estimation was calculated using two variables: recording of the calculation in the clinical history (recorded or not in each patient); the categorization of the CVR calculated in three categories: high (above 10%), moderate (between 5 and 10%) and low (less than 5%).
Frequentation (number of annual visits) of the patient to his primary care team was collected, as it probably influenced the recording of health-related problems. Prescriptions of the different marketed antipsychotic drugs withdrawn from pharmacies were collected. This study was approved by the Primary Care Research Institute IDIAP Jordi Gol Clinical Research Ethics Committee. 2.5. Analysis plan The population was separated into three groups, where the mean and standard deviations or medians and the 25 and 75 percentiles were used as the quantitative variables, and percentages were used for the prevalence of CVRF or sex (categorical variables). Demographics and clinical characteristics were analyzed using Chi-square (x2) tests for categorical variables and t-tests for continuous variables. An ANOVA test was used for comparing the three groups. After significant differences were obtained, unadjusted and adjusted comparisons for each CVRF were assessed based on a logistic regression model with each group (SZ vs control, NS-TAD vs control) as the binary outcome. Multivariate logistic regression was used for the odds ratios, adjusted for age, sex and frequentation. Multivariate logistic regression of age data was adjusted for using the other two variables (gender and frequentation), and the same method was used for these two. Odds ratios and their confidence intervals were produced. The R software v 2.15.1 and RStudio v 0.97.248 were used.
3
3. Results A total of 129,706 subjects (Fig. 1) were used and were distributed into the three study groups as follows: 98,715 in the control group, 7051 in the SZ group (2140 incidents), and 23,940 NS-TAD group (19,783 incidents). The age distribution of each group is shown in Fig. 2. The general characteristics (Tables 2a and 2b) showed significant differences. The control and SZ groups had a different gender distribution but were similar in average age. All CVRF were more frequently associated with the SZ group, except for hypertension, with the CVR estimation and a moderate-high percentage of CVR found to be similar in both SZ and control groups. The NS-TAD group was generally older and had a higher number of visits to their primary care team. It was the group with the highest proportion of all CVRF studied, except for smoking and obesity. A higher proportion of patients with high CVR was evident, although less frequently recorded. Given the demographic differences, an adjusted analysis (Table 3) was performed. The NS-TAD group had a greater presence of alcohol and cocaine use than SZ, but not for smoking. Hypertension was lower in SZ and NS-TAD. The SZ group showed obesity, diabetes, and dyslipidemia more often. By contrast, the NS-TAD group had less association with obesity and dyslipidemia, without differences in diabetes. These differences resulted in an average of CVRF which remained elevated in the SZ group; but not in the NSTAD group. The Framingham-Regicor function was used less often on the NS-TAD group, although the percentage of the estimated CVR was similar in all three groups. According to the distribution of the types of antipsychotics prescribed (Table 4), 20% of the patients with SZ had no TAD. About one-fourth to one-fifth of the patients received mixed treatment (typical and atypical TAD). The atypical TAD was the most common in both groups. The biggest difference was in the use of typical TAD, which was more than double in the NS-TAD group. Within the SZ group, we compared those patients without evidence of TAD versus those that had received some kind of TAD (Table 5). Almost all modifiable CVRF appeared more associated with the SZ with TAD subgroup, which had the largest record of CVR estimation. 4. Discussion The SZ and NS-TAD groups were associated with a greater presence of the CVRF studied; but differences in age, sex and frequentation forced an adjustment of the analysis.
Fig. 1. Flowchart of participant inclusion. NS-TAD: patients without schizophrenia but in treatment with antipsychotic drugs; SIDIAP not Q: patients followed by doctors with Registry Quality Score [19] negative.
Please cite this article in press as: Castillo-Sa´nchez M, et al. Schizophrenia, antipsychotic drugs and cardiovascular risk: Descriptive study in primary care. European Psychiatry (2015), http://dx.doi.org/10.1016/j.eurpsy.2014.12.010
G Model
EURPSY-3212; No. of Pages 7 M. Castillo-Sa´nchez et al. / European Psychiatry xxx (2015) xxx–xxx
4
Fig. 2. Age distribution (18–99 years) by strata in each group, expressed as percentage of total patients included in the group. Age at the beginning of period: January 1st 2006. Controls: control group; NS-TAD: patients without schizophrenia but in treatment with antipsychotic drugs.
Table 2a Description of the population. Non-adjusted values.
Age (years)a, mean (DS) Gender (men) Annual visits, median [ICR] Alcoholb Cocaineb Tobaccob Arterial hypertensionb Obesityb Diabetes mellitusb Dyslipidemiab Mean of CVRF for person (DS) CVR calculation, recorded Result of the calculated CVR Lower than 5% Between 5% and 10% Higher than 10%
Control group (n = 98,715)
Schizophrenia (n = 7051)
NS-TAD (n = 23,940)
P
44.7 (17.9) 47,554 (48.2%) 2.7 [1.2–5.3] 2857 (2.9%) 297 (0.3%) 27,191 (27.5%) 23,181 (23.5%) 21,707 (22.0%) 8871 (9.0%) 22,879 (23.2%) 1.05 (1.11) 24,884 (25.2%)
42.5 (14.5) 4466 (63.3%) 4.3 [2.2–7.5] 504 (7.1%) 146 (2.1%) 3697 (52.4%) 1109 (15.7%) 2315 (32.8%) 813 (11.5%) 1951 (27.7%) 1.40 (1.13) 2038 (28.9%)
52.1 (19.7) 10,546 (44.1%) 6.2 [3.3–10.2] 2110 (8.8%) 857 (3.6%) 8545 (35.7%) 7451 (31.1%) 6918 (28.9%) 3599 (15.0%) 7407 (30.9%) 1.24 (1.15) 6654 (27.8%)
< 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001
17,407 (70.0%) 5756 (23.1%) 1721 (6.9%)
1.420 (69.7%) 488 (23.9%) 130 (6.4%)
4.615 (69.4%) 1562 (23.5%) 477 (7.2%)
< 0.001 < 0.001 < 0.001
CVR: cardiovascular risk estimated with Framingham-Regicor function; CVRF: cardiovascular risk factors; DS: standard deviation; ICR: interquartile range (p25–p75); NSTAD: patients without schizophrenia but in treatment with antipsychotic drugs. a Age at the beginning of period: January 1st 2006. b Health problems activated at any moment of the study period (2006–2011).
Table 2b Crude prevalence of control group and age-standardized prevalencea of cardiovascular risk factors, expressed in percentage.
Alcohol, % Cocaine, % Tobacco, % Arterial hypertension, % Obesity, % Diabetes mellitus, % Dyslipidemia, % CVR calculation, %
Control group (n = 98,715)
Schizophreniaa (n = 7051)
NS-TADa (n = 23,940)
2.89 0.3 27.54 23.48 21.99 8.99 23.18 25.21
6.75 1.83 48.71 18.79 33.27 13.06 28.43 29.00
10.15 4.51 41.72 24.75 28.71 12.42 29.48 28.78
CVR: cardiovascular risk estimated with Framingham-Regicor function; NS-TAD: patients without schizophrenia but in treatment with antipsychotic drugs. a Direct standardization by age (3 age groups: < 50 years, 50–75 years, > 75 years), with regard to control group.
After adjustment, only hypertension occurred less frequently in SZ than in the control group. The rest of the CVRF were associated with SZ, and the average of CVRF remained high; these results were increased by TAD. The NS-TAD group continued to be associated with drug use (tobacco, alcohol and cocaine); but hypertension, obesity and dyslipidemia had a lower prevalence than in the control group, without differences in diabetes. Consequently, the CVRF average decreased. The estimation of CVR was less common in the NS-TAD group. 4.1. Limitations and strengths In addition to those risks inherent to any population-based study of an electronic database, our study had other limitations.
Please cite this article in press as: Castillo-Sa´nchez M, et al. Schizophrenia, antipsychotic drugs and cardiovascular risk: Descriptive study in primary care. European Psychiatry (2015), http://dx.doi.org/10.1016/j.eurpsy.2014.12.010
G Model
EURPSY-3212; No. of Pages 7 M. Castillo-Sa´nchez et al. / European Psychiatry xxx (2015) xxx–xxx
5
Table 3 Multivariate logistic regression adjusted by age, gender and frequentation, with respect to control group. Schizophrenia (n = 7051) OR (95% CI) Age Gender (men) Alcohol Cocaine Tobacco Arterial hypertension Obesity Diabetes mellitus Dyslipidemia Mean of CVRF for person CVR registered Result of the calculated CVR
0.98 2.20 1.82 4.15 2.32 0.49 1.61 1.18 1.15 1.25 0.99 1.00
(0.98–0.98)a (2.09–2.31)b (1.64–2.02)b (3.36–5.11)b (2.20–2.44)b (0.45–0.53)a (1.52–1.70)b (1.09–1.28)b (1.08–1.22)b (1.22–1.28)b (0.94–1.06) (1.00–1.01)
P
NS-TAD (n = 23,940) OR (95% CI)
< 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.02 < 0.01
1.01 1.09 3.00 14.70 1.68 0.64 0.88 0.97 0.84 0.97 0.62 1.00
P
(1.01–1.01)b (1.05–1.12)b (2.81–3.19)b (12.79–16.94)b (1.62–1.74)b (0.62–0.67)a (0.85–0.91)a (0.93–1.02) (0.81–0.87)a (0.96–0.99)a (0.60–0.64)a (1.00–1.01)
< 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 0.06 < 0.001 < 0.001 < 0.001 < 0.02
95% CI: 95% confidence interval; CVR: cardiovascular risk estimated with Framingham-Regicor function; CVRF: cardiovascular risk factors; NS-TAD: patients without schizophrenia but in treatment with antipsychotic drugs; OR: Odds ratio. a Variables less associated with study group than with control group (P < 0.05). b Variables with greater association with the study group with respect to the control group (P < 0.05).
Although a long-term study of cross-sectional design could create doubts, we believed the use of 6 years to be correct for two reasons. Firstly, the implementation of the eCAP software was carried out starting from 2005, and data from the first year could be unreliable. Working with chronic health problems (hypertension, diabetes, dyslipidemia) and/or of difficult final resolution (substance abuse) would reduce the bias. Secondly, if the cross-section had been made at the end of the period, the health problems resolved during the period (perhaps present since long before) would not have been taken into account, ignoring their effect on the patient’s CVR. Including current patients (both SZ and NS-TAD) implies a survival bias, since all the subjects in the control group remained in the cohort throughout the period. This bias mostly affected the NSTAD group, which, besides being representatively older, over 80% were new cases. The NS-TAD group was older and had no previous cardiovascular events, which was another bias: it may be a group of patients with increased predisposition to good cardiovascular health, which would be consistent with the lower CVRF frequency obtained. Another limitation was not having requested biochemical data or measurements of blood pressure, which forced a simplified diagnosis, such as yes/no categories. The negative effect decreased by selecting ‘‘good recorder’’ doctors (SIDIAPQ), which also allowed us to indirectly include measurements not taken at the health center (mutual insurance companies, external medical clinics not belonging to ICS), but which had generated a diagnostic label. The definition of ‘‘patient in TAD’’ does not take into account the length of time the patients were in TAD or if recruited patients were in early phase of treatment, middle or long standing treatment. The data on TAD previous to the study period was not available either. It is an important limitation because the side Table 4 Percentage of patients treated with typical and atypical antipsychotics during the study period, according to the group.
Without antipsychotics, % Typical antipsychotics, % Atypical antipsychotics, % Typical and atypical antipsychotics, %
Schizophreniaa (n = 7051)
NS-TADa (n = 23,940)
1437 564 3141 1909
– 4314 (18.0%) 14,960 (62.5%) 4666 (19.5%)
(20.4%) (8.0%) (44.5%) (27.1%)
NS-TAD: patients without schizophrenia but in treatment with antipsychotic drugs. a Chi-squared test (schizophrenia and NS-TAD versus type of antipsychotic treatment) (P < 0.05).
effects are often dose-dependant and dependant on the time the patient is on the medications. Data on drugs used in TAD were obtained from the ICS pharmacy billing. Therefore, the drugs prescribed by external practitioners were not taken into account. The accuracy of these data was questionable, as we could not be sure of how many patients were really drug-free. Lastly, the adherence to the treatment by the patients was not assessed. Finally, the population cared for in primary care does not necessarily represent the general population, especially in the study groups. Patients who sporadically visited their primary care team lacked the necessary follow-up for the proper management of CVRF. In the SZ group, the selection bias could be accentuated with the loss of young patients having low comorbidity, who did not visit their primary care team. However, it is interesting to examine the characteristics of this sub-population, as it will allow us to focus on preventive strategies with greater efficiency in the primary care field. On the other hand, this study possesses a large sample and a wide population base, as well as a reliable source of data [4,19]. A NS-TAD group study on a large population base is a way of
Table 5 Multivariate logistic regression adjusted by age, gender and frequentation of the schizophrenia subgroup with antipsychotic treatment, with respect to the schizophrenia subgroup without evidence of treatment.
Age Gender (men) Alcohol Cocaine Tobacco Arterial hypertension Obesity Diabetes mellitus Dyslipidemia Mean of CVRF for person CVR registered Result of the calculated CVR
Schizophrenia with antipsychotic treatment (n = 5614) OR (95% CI)
P
1.00 0.99 0.94 1.65 1.17 1.06 1.68 1.28 1.19 1.19 1.26 0.99
0.34 0.85 0.62 < 0.05 < 0.02 0.56 < 0.001 < 0.03 < 0.03 < 0.001 < 0.003 0.16
(1.00–1.01) (0.86–1.12) (0.75–1.19) (1.03–2.77)a (1.03–1.33)a (0.88–1.27) (1.46–1.94)a (1.03–1.59)a (1.03–1.38)a (1.12–1.27)a (1.08–1.46)a (0.97–1.01)
95% CI: 95% confidence interval; CVR: cardiovascular risk estimated with Framingham-Regicor function; CVRF: cardiovascular risk factors; NS-TAD: patients without schizophrenia but in treatment with antipsychotic drugs; OR: Odds ratio. a Variables more associated with the study group with respect to the control group (P < 0.05).
Please cite this article in press as: Castillo-Sa´nchez M, et al. Schizophrenia, antipsychotic drugs and cardiovascular risk: Descriptive study in primary care. European Psychiatry (2015), http://dx.doi.org/10.1016/j.eurpsy.2014.12.010
G Model
EURPSY-3212; No. of Pages 7 M. Castillo-Sa´nchez et al. / European Psychiatry xxx (2015) xxx–xxx
6
dichotomizing the effect of NS and the TAD drugs. The perspective of primary care could be useful and complementary for a multidisciplinary approach.
possible under-diagnosis. However, SZ could impair metabolism of carbohydrates [22], even before starting the TAD [26,35], which would aggravate this alteration.
4.2. Drug abuse: alcohol and cocaine
4.7. Dyslipidemia
The consumption of cocaine and alcohol in the SZ group were lower than those reported in other studies [21], leading to the belief that there was an under-recording of the use of these substances. This could be due to the professionals not asking questions on a topic that may discomfort the patient, or it could be that the patient hid their habits unless they generated any clinical complications later on. However, the high proportions within the study groups were credible.
The prevalence of dyslipidemia is not comparable to the other Spanish studies (based on biochemical values), but we found similar results (29.4%) in international literature [25]. The SZ group had an increased frequency of dyslipidemia than the control group, unlike the NS-TAD group. It has been postulated that dyslipidemia would increase with the evolution of SZ, with an unclear role of the TAD [32,39]. 4.8. Estimation of CVR (mean of CVRF and Framingham-Regicor)
4.3. Smoking The SZ group results were assessed according to two individual Spanish studies. The first is the CLAMORS study [3], with 1452 patients with psychosis and TAD, monitored by psychiatrists at the outpatient level. The average age was slightly lower (40.7 years old) and it did not have a similar gender distribution (60.9% male). Their results on smoking (53.7%) were consistent with ours. The newer RICAVA study [2], with 733 patients with SZ, hospitalized and treated with atypical antipsychotics, of lower mean age (37.8 years old) and higher percentage of males (71.8%) had a higher proportion of smokers (71%). Smoking remained associated with both the SZ and NS-TAD groups even after the adjusted analysis, resulting in about twice as many smokers as in the control group, in line with the literature [9]. 4.4. Hypertension The SZ population in the present study had a lower proportion of hypertension than the control group, and was similar to the RICAVA study [2]. Both the SZ and the NS-TAD groups were strongly associated with a lesser presence of hypertension. It could be that blood pressure was not measured at the physician’s office with the same frequency. Other similarly designed studies obtained similar figures, and this result was attributed to under-diagnosis [36]. 4.5. Obesity The SZ group had similar values as previous studies on the Spanish population (between 24% and 30.8%) [2,3]. The NS-TAD group also had a higher prevalence than the control group. After adjusted analysis, SZ was associated with an increased prevalence of obesity, contrary to the NS-TAD group. This could be due to the increased susceptibility of patients with SZ. Another hypothesis could be that the profile of the patients with SZ who visited the general practitioners is that of a psychiatrist-monitored, mentally-stable patient, allowing the primary care team to focus on aspects of physical health. In other TAD-treatable pathologies, the prioritization of other problems that motivate visiting primary care physicians would justify a lower diagnostic record. 4.6. Diabetes mellitus We found a similar prevalence in the CLAMORS study (14.2%) [3]. After adjusting, only the SZ group had more diabetes, with no differences between the control and the NS-TAD groups. Some studies [18,24] concluded that the TAD was an independent factor that caused diabetes, so we contemplated a
After the adjusted analysis, the NS-TAD group showed half the frequency as the control group. We concluded that there are three reasons for this: firstly, the exclusion of more than the 16% of the patients older than 75 from the CVR estimation affected the results. Secondly, the bias described in ‘‘Limitations’’ related to their good cardiovascular health. Finally, a relationship between the scarce CVR calculation and the low prevalence of obesity, dyslipidemia, and hypertension may exist, because quantitative data that defines the diagnosis of these disorders (body mass index, cholesterol, blood pressure) are necessary for the CVR calculation. The scarce CVR calculation would imply that the CVRF have been evaluated with lower frequency in the NS-TAD group than in the control group, corroborating the lesser attention that the patients with severe mental illnesses received in primary prevention of CVR [14]. As for the risk itself, the SZ and NS-TAD groups had a mean value similar to the control group, unlike other studies [3,17,20]. Adjusted analysis counteracts the great weight that age has on the calculation of CVR values, especially in the NS-TAD group. 4.9. Type of TAD prescribed The typical antipsychotics drugs prevailed in the NS-TAD group, which could be associated with the lower registering of CVRF, given that atypical antipsychotics have been linked with increased incidences of metabolic syndrome [10]. However, this relationship is not clear [18,28]. A fifth of the SZ patients did not have TAD on record, but we cannot be sure that they did not receive TAD through other external means. We noticed a greater presence of modifiable CVRF in SZ patients with TAD than in SZ without TAD (except for hypertension, which was similar). We postulated that the TAD would act synergistically with the predisposition associated with SZ [39], although we must assess the under-recording of the CVR calculation in the SZ without TAD subgroup as a sign of worse management in cardiovascular prevention (either because they received TAD and/or management by external services).
5. Conclusions The differences between the SZ and control groups in obesity, diabetes and dyslipidemia support the hypothesis of SZ as a systemic disease associated with CVR. Within the SZ group, the TAD subgroup had an increased CVR, unlike other diseases treated with TAD (NS-TAD group). We believe that, despite having a multifactorial origin, the under-recording of the CVR calculation estimated in the NS-TAD group could reflect under-diagnosis of CVRF that has been related
Please cite this article in press as: Castillo-Sa´nchez M, et al. Schizophrenia, antipsychotic drugs and cardiovascular risk: Descriptive study in primary care. European Psychiatry (2015), http://dx.doi.org/10.1016/j.eurpsy.2014.12.010
G Model
EURPSY-3212; No. of Pages 7 M. Castillo-Sa´nchez et al. / European Psychiatry xxx (2015) xxx–xxx
to patients with severe mental illness in the literature, particularly hypertension, which also showed a low association with SZ. Our results could probably be extrapolated to other populations attended in primary care, but cannot be extended to all patients with SZ or TAD. Studies that avoid the effect of under-diagnosis of CVRF are required to clarify the role of the TAD and SZ in the development of CVRF and the real CVR. Multidisciplinary assessment of medical and behavioral conditions is required to improve the CVR profile of patients with SZ and with TAD, because they have special needs of care.
Disclosure of interest The authors declare that they have no conflicts of interest concerning this article. Acknowledgement Thanks to the XB Aid for Primary Care Research in ICSBarcelona. Thank you to Mario Fon ([email protected]) for revising the manuscript. Funding: There are no funding sources. References [1] Ballesteros A, Jiang P, Summerfelt A, Du X, Chiappelli J, O’Donnell P, et al. No evidence of exogenous origin for the abnormal glutathione redox state in schizophrenia. Schizophr Res 2013;146(1–3):184–9. ˜ as F, Banegas JR, Casademont J, Riesgo Y, Varela C. Prevalence [2] Bernardo M, Can and awareness of cardiovascular risk factors in patients with schizophrenia: a cross-sectional study in a low cardiovascular disease risk geographical area. Eur Psychiatry 2009;24(7):431–41. [3] Bobes J, Arango C, Aranda P, Carmena R, Garcia-Garcia M, Rejas J. Cardiovascular and metabolic risk in outpatients with schizophrenia treated with antipsychotics: results of the CLAMORS Study. Schizophr Res 2007;90(1– 3):162–73. [4] Bolı´bar B, Fina Avile´s F, Morros R, Garcia-Gil Mdel M, Hermosilla E, Ramos R, et al. [SIDIAP database: electronic clinical records in primary care as a source of information for epidemiologic research]. Med Clin (Barc) 2012;138(14):617– 21 [Spanish]. [5] Brown S, Birtwistle J, Roe L, Thompson C. The unhealthy lifestyle of people with schizophrenia. Psychol Med 1999;29:697–701. [6] Bushe CJ, Taylor M, Haukka J. Mortality in schizophrenia: a measurable clinical endpoint. J Psychopharmacol 2010;24(4 Suppl.):17–25. [7] Castillo Sa´nchez M, Fa`bregas Escurriola M, Berge` Baquero D, Goday Arno A, Valle`s Callol JA. [Psychosis, cardiovascular risk and associated mortality: are we on the right track?] Clin Investig Arterioscler 2014;26(1):23–32 [Review. Spanish]. [8] Conroy RM, Pyo¨ra¨la K, Fitzgerald AP, Sans S, Menotti A, De Backer G, et al. Estimation of ten-year risk of fatal cardiovascular disease in Europe: the SCORE project. Eur Heart J 2003;24:987–1003. [9] Correll CU. Balancing efficacy and safety in treatment with antipsychotics. CNS Spectr 2007;12(Suppl. 17):12–20. [10] De Hert M, Schreurs V, Sweers K, Van Eyck D, Hanssens L, Sinko S, et al. Typical and atypical antipsychotics differentially affect long-term incidence rates of the metabolic syndrome in first-episode patients with schizophrenia: a retrospective chart review. Schizophr Res 2008;101(1–3):295–303. [11] De Hert M, Dekker JM, Wood D, Kahl KG, Holt RI, Mo¨ller HJ. Cardiovascular disease and diabetes in people with severe mental illness position statement from the European Psychiatric Association (EPA), supported by the European Association for the Study of Diabetes (EASD) and the European Society of Cardiology (ESC). Eur Psychiatry 2009;24(6):412–24. [12] Departament de Salut - Generalitat de Catalunya. Llibre blanc: consens sobre les activitats preventives a l’edat adulta dins l’atencio´ prima`ria, 2nd ed, Barcelona: Direccio´ General de Salut Pu´blica; 2005. [13] Dipasquale S, Pariante CM, Dazzan P, Aguglia E, McGuire P, Mondelli V. The dietary pattern of patients with schizophrenia: a systematic review. J Psychiatr Res 2013;47(2):197–207. [14] Fagiolini A, Goracci A. The effects of undertreated chronic medical illnesses in patients with severe mental disorders. J Clin Psychiatry 2009;70(3 Suppl.): 22–9.
7
[15] Fan Z, Wu Y, Shen J, Ji T, Zhan R. Schizophrenia and the risk of cardiovascular diseases: a meta-analysis of thirteen cohort studies. J Psychiatr Res 2013; 47(11):1549–56. [16] Fernandez-Egea E, Bernardo M, Donner T, Conget I, Parellada E, Justicia A, et al. Metabolic profile of antipsychotic-naive individuals with non-affective psychosis. Br J Psychiatry 2009;194(5):434–8. [17] Fleischhacker WW, Siu CO, Bode´n R, Pappadopulos E, Karayal ON, Kahn RS, et al. Metabolic risk factors in first-episode schizophrenia: baseline prevalence and course analysed from the European First-Episode Schizophrenia Trial. Int J Neuropsychopharmacol 2013;16(5):987–95. [18] Foley DL, Morley KI. Systematic review of early cardiometabolic outcomes of the first treated episode of psychosis. Arch Gen Psychiatry 2011;68:609–16. [19] Garcı´a-Gil Mdel M, Hermosilla E, Prieto-Alhambra D, Fina F, Rosell M, Ramos R, et al. Construction and validation of a scoring system for the selection of highquality data in a Spanish population primary care database (SIDIAP). Inform Prim Care 2011;19(3):135–45. [20] Goff DC, Sullivan LM, McEvoy JP, Meyer JM, Nasrallah HA, Daumit GL, et al. A comparison of ten-year cardiac risk estimates in schizophrenia patients from the CATIE study and matched controls. Schizophr Res 2005;80(1):45–53. [21] Green AI, Brown ES. Comorbid schizophrenia and substance abuse. J Clin Psychiatry 2006;67(9):e08. [22] Hansen T, Ingason A, Djurovic S, Melle I, Fenger M, Gustafsson O, et al. At-risk variant in TCF7L2 for type II diabetes increases risk of schizophrenia. Biol Psychiatry 2011;70(1):59–63. [23] Kelly DL, McMahon RP, Liu F, Love RC, Wehring HJ, Shim JC, et al. Cardiovascular disease mortality in patients with chronic schizophrenia treated with clozapine: a retrospective cohort study. J Clin Psychiatry 2010;71(3):304–11. [24] Kessing LV, Thomsen AF, Mogensen UB, Andersen PK. Treatment with antipsychotics and the risk of diabetes in clinical practice. Br J Psychiatry 2010;197(4):266–71. [25] Kilbourne AM, Morden NE, Austin K, Ilgen M, McCarthy JF, Dalack G, et al. Excess heart-disease-related mortality in a national study of patients with mental disorders: identifying modifiable risk factors. Gen Hosp Psychiatry 2009;31(6):555–63. [26] Kirkpatrick B, Miller BJ, Garcia-Rizo C, Fernandez-Egea E, Bernardo M. Is abnormal glucose tolerance in antipsychotic-naive patients with nonaffective psychosisconfounded by poor health habits? Schizophr Bull 2012;38(2): 280–4. [27] 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): 713–20. [28] Leucht S, Corves C, Arbter D, Engel RR, Li C, Davis JM. Second-generation versus first-generation antipsychotic drugs for schizophrenia: a meta-analysis. Lancet 2009;373(9657):31–41. [29] Manu P, Correll CU, Wampers M, van Winkel R, Yu W, Shiffeldrim D, et al. Dysmetabolic features of the overweight patients receiving antipsychotic drugs: a comparison with normal weight and obese subjects. Eur Psychiatry 2014;29(3):179–82. [30] Marrugat J, Solanas P, D’Agostino R, Sullivan L, Ordovas J, Cordo´n F, et al. [Coronary risk estimation in Spain using a calibrated Framingham function]. Rev Esp Cardiol 2003;56(3):253–61 [Spanish]. [31] Mitchell AJ, Lord O. Do deficits in cardiac care influence high mortality rates in schizophrenia? A systematic review and pooled analysis. J Psychopharmacol 2010;24(4 Suppl.):69–80. [32] Mitchell AJ, Vancampfort D, Sweers K, van Winkel R, Yu W, De Hert M. Prevalence of metabolic syndrome and metabolic abnormalities in schizophrenia and related disorders–a systematic review and meta-analysis. Schizophr Bull 2013;39(2):306–18. [33] Nordentoft M, Wahlbeck K, Ha¨llgren J, Westman J, Osby U, Alinaghizadeh H, et al. Excess mortality, causes of death and life expectancy in 270,770 patients with recent onset of mental disorders in Denmark, Finland and Sweden. PLoS One 2013;8(1):e55176. [34] O’Donnell CJ, Elosua R. [Cardiovascular risk factors. Insights from Framingham Heart Study]. Rev Esp Cardiol 2008;61(3):299–310 [Review. Spanish]. [35] Ryan MC, Collins P, Thakore JH. Impaired fasting glucose tolerance in firstepisode, drug-naive patients with schizophrenia. Am J Psychiatry 2003; 160(2):284–9. [36] Smith DJ, Langan J, McLean G, Guthrie B, Mercer SW. Schizophrenia is associated with excess multiple physical-health comorbidities but low levels of recorded cardiovascular disease in primary care: cross-sectional study. BMJ Open 2013;3(4). [37] Spelman LM, Walsh PI, Sharifi N, Collins P, Thakore JH. Impaired glucose tolerance in first-episode drug-naı¨ve patients with schizophrenia. Diabet Med 2007;24:481–5. [38] Tiihonen J, Lo¨nnqvist J, Wahlbeck K, Klaukka T, Niskanen L, Tanskanen A, et al. 11-year follow-up of mortality in patients with schizophrenia: a populationbased cohort study (FIN11 study). Lancet 2009;374(9690):620–7. [39] Vancampfort D, Wampers M, Mitchell AJ, Correll CU, De Herdt A, Probst M, et al. A meta-analysis of cardio-metabolic abnormalities in drug naı¨ve, firstepisode and multi-episode patients with schizophrenia versus general population controls. World Psychiatry 2013;12(3):240–50.
Please cite this article in press as: Castillo-Sa´nchez M, et al. Schizophrenia, antipsychotic drugs and cardiovascular risk: Descriptive study in primary care. European Psychiatry (2015), http://dx.doi.org/10.1016/j.eurpsy.2014.12.010
