|Home | About | Journals | Submit | Contact Us | Français|
We have previously documented inequalities in the quality of medical care provided to those with mental ill health but the implications for mortality are unclear. We aimed to test whether disparities in medical treatment of cardiovascular conditions, specifically receipt of medical procedures and receipt of prescribed medication, are linked with elevated rates of mortality in people with schizophrenia and severe mental illness. We undertook a systematic review of studies that examined medical procedures and a pooled analysis of prescribed medication in those with and without comorbid mental illness, focusing on those which recruited individuals with schizophrenia and measured mortality as an outcome. From 17 studies of treatment adequacy in cardiovascular conditions, eight examined cardiac procedures and nine examined adequacy of prescribed cardiac medication. Six of eight studies examining the adequacy of cardiac procedures found lower than average provision of medical care and two studies found no difference. Meta-analytic pooling of nine medication studies showed lower than average rates of prescribing evident for the following individual classes of medication; angiotensin converting enzyme inhibitors (n=6, aOR=0.779, 95% CI=0.638–0.950, p=0.0137), beta-blockers (n=9, aOR=0.844, 95% CI=0.690–1.03, p=0.1036) and statins (n=5, aOR=0.604, 95% CI=0.408–0.89, p=0.0117). No inequality was evident for aspirin (n=7, aOR=0.986, 95% CI=0.7955–1.02, p=0.382). Interestingly higher than expected prescribing was found for older non-statin cholesterol-lowering agents (n=4, aOR=1.55, 95% CI=1.04–2.32, p=0.0312). A search for outcomes in this sample revealed ten studies linking poor quality of care and possible effects on mortality in specialist settings. In half of the studies there was significantly higher mortality in those with mental ill health compared with controls but there was inadequate data to confirm a causative link. Nevertheless, indirect evidence supports the observation that deficits in quality of care are contributing to higher than expected mortality in those with severe mental illness (SMI) and schizophrenia. The quality of medical treatment provided to those with cardiac conditions and comorbid schizophrenia is often suboptimal and may be linked with avoidable excess mortality. Every effort should be made to deliver high-quality medical care to people with severe mental illness.
There is longstanding and conclusive evidence that the physical health of people with severe mental illness (SMI) is poor compared with the general population (Mitchell and Malone, 2006; Prince et al., 2007). Amongst those with mental illness there is a higher than expected mortality rate and a lower life-expectancy with a persistent mortality gap over several decades (Saha et al., 2007). Individuals with schizophrenia appear to have higher rates of hypothyroidism, dermatitis, eczema, obesity, epilepsy, viral hepatitis, diabetes (type II), essential hypertension, chronic obstructive pulmonary disease and fluid/electrolyte disorders (Carney et al., 2006; Weber et al., 2009). The presence of this medical comorbidity adversely effects quality of life and recovery from the underlying psychiatric disorder (Kisely and Simon, 2005). Although many of these chronic conditions may be unavoidable given our current state of knowledge, many deaths in SMI are considered avoidable (Amaddeo et al., 2007; Crompton et al., 2010). In addition up to half of all chronic conditions go unrecognized in schizophrenia (Bernardo, 2006; Farmer, 1987; Fallow et al., 1995; Korany, 1979; Klilbourne et al., 2006). For example, in a study of the homeless mentally ill 43.6% had an unmet need for medical care (Desai and Rosenheck, 2005). Many studies suggest that mental health professionals often miss physical conditions in their patients (Felker et al., 1996; Koran et al., 1989) and rarely undertake physical examinations of their patients (McIntyre and Romano, 1977; Patterson, 1978).
What has only recently become clear is that the quality of medical care provided to those with known mental health diagnoses is also less than ideal. Medical care can be divided into the care for those with established medical comorbidity and screening/preventive practices for those at risk of medical illness. Medical care includes medical treatment and processes of care such as investigations and monitoring. Patients with SMI appear to suffer in all areas. Also in question is the delivery of appropriate preventive screening services. Lack of screening and related services is important not just for the reduction in future morbidity but low receipt of preventive care is associated with lower quality of life (Mackell et al., 2005). An analysis of National Ambulatory Medical Care Survey data showed that psychiatrists provided preventive services to people with SMI during only 11% of visits (Daumit et al., 2002). Our group recently evaluated 27 studies that examined receipt of medical care in those with and without mental illness (Mitchell et al., 2009). The majority of studies demonstrated significant inequalities in the quality of care received by patients with SMI against those without SMI. We also reviewed 25 studies that examined preventive care in individuals with versus without psychiatric illness. For those individuals with schizophrenia eight of nine analyses showed inferior preventive care in several areas including osteoporosis screening, blood pressure monitoring, vaccinations, mammography and cholesterol monitoring (Lord et al., 2010). The magnitude deficits in quality of care vary considerably according to the setting and method of data collection. In some of these studies the magnitude of the inequality approaches 50% of the comparative standard but in other domains differences were quite subtle.
Given that mortality rates are high and quality of care is less than optimal, an important question is are these two factors linked? We already know that successful medical mass screening programmes have the potential to improve survival (Kerlikowske et al., 1995; Heresbach et al., 2006; Lindholt and Norman 2008). If deficits in quality of medical care or preventive services directly influence survival in those with schizophrenia then some responsibility for poor outcomes must rest with healthcare professionals. In this review, we aim to examine the hypothesis that suboptimal cardiac treatment (one facet of medical care) influences mortality in those with schizophrenia.
We searched Medline and Embase databases from inception to May 2010. In these databases the keywords/MeSH terms (‘comorbidity or comorbidity’ or ‘organic or physical or medic* or cardia’) and (‘psychi* or mental or depress* or schizophr* or severe mental illness or SMI’) were used. We included those with SMI provided there was a subgroup with schizophrenia. Four full-text collections were searched: Science Direct, Ingenta Select, Springer-Verlag’s LINK and Blackwell-Wiley. In these online databases the same search terms were used but as a full text search and as a citation search. The abstract databases Web of Knowledge and Scopus were searched, using the above terms as a text word search, and using key papers in a reverse citation search. Finally, a number of journals were hand-searched and several experts contacted. Using this strategy we located 440 references but only 80 were primary data studies. From these we identified 18 studies regarding treatment adequacy and eight studies that examined both mortality and quality of care by psychiatric diagnosis. We excluded studies purely focusing on depression (Goodwin et al., 2004). We pooled data from treatment data using adjusted odds ratios (aORs) provided by the original authors and converting hazard ratios to odds ratios (ORs) where needed. We used random effects meta-analysis as I2 heterogeneity was above 75%.
Looking at cardiac procedures first, Druss et al. (2000) examined cardiovascular care following an acute myocardial infarction. After adjusting for demographic, clinical, hospital and regional factors, those with mental disorders were only 41% (for schizophrenia) to 78% (for substance use) as likely to undergo cardiac catheterization as those without mental disorders. Druss et al. (2001) also found patients hospitalized for myocardial infarction with mental health diagnoses were less likely to have reperfusion conducted. Young and Foster (2000) found those with a mental illness post-myocardial infarction had significantly lower levels of all three revascularization procedures (cardiac catheterization, percutaneous transluminal coronary angioplasty [PTCA] and coronary artery bypass graft [CABG]) compared with those without mental illness, with the lowest rates seen in those older than 64 years. Petersen et al. (2003) examined the records of 4340 male veterans discharged after a clinically confirmed myocardial infarction. In this study those with mental illness were less likely to undergo inpatient diagnostic angiography (age adjusted relative risk [RR]=0.9; 95% confidence interval [CI] 0.83–0.98) but with no difference in CABG. Lawrence et al. (2003) conducted a population-based record-linkage study of 210,129 users of mental health services in Western Australia during 1980–1998. They found revascularization rates were low for people with dementia followed by those with schizophrenia, substance disorder, other psychoses and affective psychoses (rate ratios 0.14, 0.31, 0.60, 0.66, 0.77, respectively) but significant only for men (Figure 1). Kisely et al. (2007) carried out a population-based record-linkage analysis of related data from 1995 to 2001 compared with the general public for each outcome (n=215,889). In psychiatric inpatients the adjusted rate ratios for cardiac catheterization, PTCA and CABG were 0.41, 0.22 and 0.34, respectively. However, Plomondon et al. (2007) found no differences in cardiac procedure after acute coronary syndromes presenting to Veterans Health Administration (VHA) hospitals and similarly Jones and Carney (2005) found no difference in the rates of revascularization. Thus, six from eight studies found lower than average rates of cardiac procedures in those with a psychiatric history treated for cardiac complaints.
Regarding adequacy of prescribed medication we found nine analyses relating to five classes of medication (Table 2). These comprised statins, non-statin cholesterol agents, beta-blockers, aspirin and angiotensin converting enzyme (ACE) inhibitors/angiotensin receptor blockers (ARBs). After rates were adjusted for potential confounders and pooled using meta-analysis the following results were apparent. Of these studies low prescribing was evident for the following individual classes of medication. ACE/ARBs (n=6, aOR=0.779, 95% CI=0.638–0.950, p=0.0137), beta-blockers (n=9, aOR=0.844, 95% CI=0.690–1.03, p=0.1036) statins (n=5, aOR=0.604, 95% CI=0.408–0.89, p=0.0117). No inequality was evident for aspirin (n=7, aOR=0.986, 95% CI=0.955–1.02, p=0.3819). Provisionally, higher than expected prescribing was found for older non-statin cholesterol-lowering agents (n=4, aOR=1.55, 95% CI=1.04–2.32, p=0.0312).
In the study from Druss et al. (2000), patients with mental disorders had a small but statistically significantly lower risk of mortality at baseline and in unadjusted analysis 12.8% of those with schizophrenia died within 30 days compared with 10.8% in the comparator population but this was not significant after adjustments. Yet in their replication study Druss et al. (2001) found that mental disorder of all types was associated with a 19% increase in mortality at one year. Importantly, when the five quality indicators were added to the model this association was no longer significant, suggesting that elevated mortality is related to poor quality of care. Young and Foster (2000) found that in the older cohort (older than 65 years) with mental illness there was a 21% lower risk adjusted likelihood of death (p<0.001) compared with those without mental illness. In the younger cohort those with schizophrenia and substance abuse had higher inpatient mortality rates (both p<0.001). Thus, age may play an important role in modifying risk. Petersen et al. (2003) examined the records of 4340 male veterans discharged after a clinically confirmed myocardial infarction. The authors found a trend towards higher rate of death at one year in those with mental illness; risk for death within one year was 1.25 (95% CI 1.00–1.53). In the study from Lawrence et al. (2003) ischemic heart disease (not suicide) was the major cause of excess mortality in psychiatric patients. Standardized mortality rates (SMRs) due to ischemic heart disease in mental health users were almost twice that in the overall population (SMR 1.91 total ischemic heart disease, 1.74 acute myocardial infarction). Males with schizophrenia were only 60% as likely to be admitted for ischemic heart disease compared with males in the general population, despite being 1.8 times as likely to die from ischemic heart disease. Kisely et al. (2007) carried out a population-based record-linkage analysis of related data from 1995 through 2001 compared with the general public for each outcome (n=215,889). The age-standardized mortality-rate ratio for psychiatric patients was 1.31 (95% CI 1.25–1.36). Psychiatric patients were often cases less likely to undergo any of the cardiac procedures than were the general population. Plomondon et al. (2007) studied 14,194 patients (including 18% with SMI) with acute coronary syndromes presenting to VHA hospitals between October 2003 and September 2005 and although one year mortality was lower for patients with SMI (15.8% vs. 19.1%, p<0.001) there was no difference in quality of care. Interestingly Li et al. (2007) analysed New York State hospital discharge data between 2001 and 2003 and New York's publicly-released Cardiac Surgery Report of surgeons' risk-adjusted mortality rate. After adjustments patients with both substance-use and psychiatric disorders (n=113), but not substance-use alone (n=447), were more likely to receive care from surgeons in the high-mortality quintile group (OR=1.76, p=0.024).
National guidelines are agreed that the medical care of patients with mental disorders and schizophrenia in particular is paramount (Department of Health, 1999; De Hert et al., 2009; National Institute for Clinical Excellence, 2009; Unützer et al., 2006). Unfortunately there is little evidence that this advice is being heeded. Indeed serious concerns have been raised about the quality of medical (and screening) services offered to patients with SMI (Mitchell et al., 2009; Lord et al., 2010). In spite of higher than average risks of physical ill health and premature mortality, individuals with schizophrenia receive as little as half of the monitoring offered to people without schizophrenia in some studies (Roberts et al., 2007). Our previous work found that they also receive less adequate quality of care for established medical conditions (Desai et al., 2002a; Redelmeier et al., 1998; Vahia et al., 2008). These disparities exist in some of the most critical areas of patient care such as general medicine, cardiovascular and cancer care (Mateen et al., 2008). In this review, we extend these findings to cardiac treatment as well as associated poor outcomes in terms of elevated mortality. In particular, we found that six of eight studies examining the adequacy of cardiac procedures in patients with schizophrenia and related conditions found lower than average provision of medical care although two studies found no difference. From nine medication studies lower than average rates of prescribing were evident for the following individual classes of medication: ACE/ARBs, beta-blockers and statins but not for aspirin and higher than expected prescribing was found for older non-statin cholesterol-lowering agents.
These deficits in medical treatment appear to exist alongside worrying elevations in mortality. Indeed patients with schizophrenia may also have higher rates of post-operative complications (Li et al., 2008) and post-operative mortality (Copeland et al., 2008). However the direction of this relationship is not clearly established from the design of these studies, which are largely observational. Three possible hypotheses link poor medical care and high mortality. Either the poor medical care directly contributes to excess mortality, or a confounding factor indirectly links poor medical care and excess mortality, or the two observations are independent. Even in the latter case, less than average medical care in the face of excess mortality would be concerning. That said Druss et al. (2001) found that mental disorder of all types was associated with a 19% increase in mortality at one year in an unadjusted analysis, but when the five quality indicators were added to the model the association was no longer significant, suggesting that elevated mortality is in fact related to poor quality of care. Is there any wider evidence that links medical care and mortality in mental illness? Several factors may be suggestive. Firstly the types of deaths are often considered avoidable according to the European Community Classification (Amaddeo et al., 2007). Standardized mortality ratios appear to be highest for deaths preventable with adequate health promotion policies. They are also highest in young males, with comorbid alcohol/drug disorder. Elevated rates of poverty, unemployment and lack of insurance (where applicable) are linked with excess mortality and these factors may also hinder these individuals’ access to basic medical services (Saitz et al., 2004; Wells et al., 2002). In those with severe mental illness at risk of hepatitis C, one group found that ethnicity (being black), gender (male) or living in a community with high exposure to community violence lowered those odds of receiving appropriate care (Swartz et al., 2003). This elevated mortality does not appear to be linked with changes in psychiatric care (Heila et al., 2005). Indeed medical care is linked with medical complications in SMI. For example, Bishop et al. found a low rate of receipt of osteoporosis services along with a greater number of total fractures in women with schizophrenia (Bishop et al., 2004).
There are many possible reasons for these apparent deficits in cardiac care for people with mental ill health (Kim et al., 2007). Possibilities include under-recognition, inadequate treatment, poor monitoring and systems issues (Campbell et al., 2002; Palmer, 1997). A much cited factor is lack of healthcare utilization. However, evidence supporting this as the primary explanation for inequality is not entirely clear (Cradock-O’Leary et al., 2002; Dickerson et al., 2003). Folsom et al. (2002) showed that patients with schizophrenia had fewer medical visits and fewer documented medical problems than those with depression. Salsberry et al. (2005) found that, compared with the general population, those with severe mental illness had more emergency department visits, visited a doctor more frequently but, despite this high healthcare utilization, had very low rates of cervical smears and mammograms. Of note is the fact that they also showed that patients with severe mental illness were less likely to have an appointment for general internal medicine in contrast to generally higher rates of visits, suggesting that visits targeted at delivering medical care may be deficient, even in the face of frequent attendance in primary care or emergency departments. In emergency departments they appear less likely to be offered hospital care than other people (Sullivan et al., 2006). Surveys suggest that people with mental ill health (including schizophrenia, bipolar disorder and major depressive disorder) perceive barriers to accessing primary physical healthcare (Bradford et al., 2008; Crews et al., 1998; Dickerson et al., 2003; Drapalski et al., 2008; Levinson Miller et al., 2003; Zeber et al., 2009). There is accumulating observational evidence to support this belief (Bradford et al., 2008; Chwastiak et al., 2008; Farmer, 1987; Rice and Duncan, 1995). Most commonly this involves availability of medical advice and quality of medical advice (Druss and Rosenheck 1998; Levinson Miller 2003; O’Day et al., 2005). For example, people with psychotic disorders are less likely to have a primary care physician (Bradford et al., 2008). Mental health status has also been linked with poorer GP (general practitioner) attitude and less time spent with the GP (Al-Mandhari et al., 2004).
However, until recently there has been no direct link between primary healthcare and mortality in psychiatry. In an important study Copeland et al. (2009) analysed whether patients’ reduced primary care use over time was a significant predictor of mortality over a four-year period among VHA groups: diabetes only (n=188,332), schizophrenia only (n=40,109), and schizophrenia with diabetes (n=13,025). Patients with schizophrenia only were likely to have low primary care use decreasing with time but most important increasing use was associated with improved survival.
Assuming these disparities are robust, what can be done to improve preventive care and ultimately reduce avoidable deaths in people with mental ill health? Anderson et al. (1998) reported a meta-analysis of 43 studies involving strategies to improve the delivery of preventive care. In general interventions were moderately effective in improving immunization, screening and counselling. Better communication between primary care providers and specialist mental health services could improve screening rates (Miller et al., 2007; Oud et al., 2009; Phelan et al., 2001) and might influence mortality (Crompton et al., 2010). However against this is a lack of consensus as to which healthcare professionals should be responsible for the prevention and management of co-morbid somatic illnesses in SMI patients (Cohn and Sernyak, 2006; Golomb et al., 2000). There is also no culture of ‘beyond routine’ medical care for people with current or past mental ill health. In addition the abilities of psychiatrists to look after physical health of patients is at best underdeveloped (Garden, 2005; Krummel and Kathol, 1987; Mitchell et al., 1998; Phelan and Blair, 2008). For example preventive services are provided on approximately 11% of psychiatric consultations (Daumit et al., 2002). In the UK the Royal College of Psychiatrists have recommended that primary care physicians set up specific clinics for people with mental disorders (The Royal College of Psychiatrists, 2009). It has also been suggested in the US that a reorganization of mental health services would help redefine responsibility for physical health (Horvitz-Lennon et al., 2006). There is some support for a collaborative model of care, co-locating psychiatric and primary care (Dombrovski and Rosenstock, 2004). In one trial of an integrated model of care for older people, the intervention helped with access but didn’t produce any significant treatment effects for depression or anxiety (Arean et al., 2008). In a second trial a nurse-led intervention increased screening for cardiovascular risk factors by about 30% people with SMI (Osborn et al., 2010). In spite of promising approaches to shared care there is a substantial gap in the routine medical care for many individuals with mental illness or substance use disorders (Knott et al., 2006; Weisner et al., 2001).
In conclusion, there is strong evidence that inequalities in cardiac treatment are disadvantaging those who have schizophrenia and this may be associated with subsequent elevated mortality, although a causative role is not yet established. Nevertheless, it is worrying that deficits in preventive care may prejudice the long-term health of those with mental ill health. Ultimately this may be reflected in higher mortality rates for those given lower frequency or low quality of care. This detrimental effect could be improved by a concerted effort to offer directed enhanced medical care. Currently there is little evidence to support the notion of widespread enhanced medical care for patients with mental illness that is recommended in national guidelines. Future work must focus on organizational changes (healthcare interventions) to increase the quality of medical care and medical treatment for those with schizophrenia and related disorders.
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.