Correlates of risk factors for reduced life expectancy in schizophrenia: Is it possible to develop a predictor profile?

Abstract

Patients with schizophrenia have significantly greater mortality rates than the general population, with an estimated reduced lifespan of 10–20 years. We previously reported on a link between impairment in cognition and premature death in a prospective 20-year study. Patients who had died prematurely showed neurocognitive impairment in nine different cognitive tests compared to those who did not. Based on those findings, in this study the surviving patients in the cohort were divided into three different groups based on neurocognitive impairment and compared on symptom severity including remission status, RAND-36, weight and BMI at onset of illness and baseline of the study, and medical/physical symptomatology (i.e., blood pressure, symptom awareness, vertigo and orthostatic symptoms). Differences were most prominent between the cognitively unimpaired and severely cognitively impaired (SCI) groups, with remission, negative symptoms, general symptoms and PANSS total scores differing. For SF-36 (RAND) Physical functioning and Role limitations due to physical health subscales the SCI were worst. The findings indicate that greater impairments in cognitive ability during the illness are associated with several potential indicators of risk for early mortality. Together these factors may be of guidance for establishing an algorithm to detect patients at risk of premature death early in their illness.

Introduction

Compared to the general population, patients with schizophrenia spectrum disorder have a shortened life expectancy of up 10–20 years (Tiihonen et al., 2009; Laursen et al., 2014). While Palmer et al. (2005) found a lifetime suicide rate of 4.9% for patients with schizophrenia, Osborn et al. (2007) found an increased odds ratio of 3.22 for people with schizophrenia under 50 for cardiovascular death and an increased odds ratio of 2.53 for stroke. These findings suggest that medical co-morbidities confer more risk than suicide in the group as a whole. Despite the large knowledge base on cognitive impairment and functional disability in this group, its relation to the excess morality has not been studied until recently. Helldin et al. (2015) found a prospective link between lower neurocognitive abilities and premature death in patients with schizophrenia, with differences in neurocognitive performance in patients who later died prematurely compared to the survivors in the same study. There was no association between symptom severity and premature death in that study sample.

Brown et al. (2000) showed that patients with schizophrenia die due to the same natural causes (heart diseases, respiratory diseases, cancer, etc.) as seen in the general population. Osby et al. (2000) found that cardiovascular (CV) disease contributed to as much as 50% of the excess mortality in patients with schizophrenia. Arango et al. (2008) found a high prevalence of metabolic syndrome including central obesity, which is a risk factor for cardiovascular disease. An unhealthy lifestyle, including poor diet and sedentary behaviour (Brown et al., 1999), but also the negative impacts of some antipsychotics (Maayan and Correll, 2010), are known to be contributing factors for the obesity in patients with schizophrenia.

In a recent study of schizophrenia and bipolar disorder, patients found to be overweight 6 months after their first episode had a risk of obesity of over 60% at the 20-year follow-up of the sample. Interestingly, the risk for obesity did not differ across the diagnoses in patients who entered the study overweight, but for patients with normal weight at the baseline assessment, the risk of development of obesity was greater for the schizophrenia patients over 20 years (Strassnig et al., 2017a). In another set of analyses on this sample, for schizophrenia patients as a group, the mean waist circumference was 46 in. for men and 45 in. for females. When the impact of obesity and associated physical limitations was examined, it was found that reduced mobility (indexed by a test of chair stands) was an important predictor of unemployment at the 20 year follow-up, with greater predictive power than cognitive performance. Although the patients were only 48 years of age on average, 38% were unable to meet minimal standards for geriatric patients in rising from a chair (Strassnig et al., 2017b).

In the Helldin et al., 2015 study, baseline characteristics of those patients still living were compared with those who died prematurely, with the finding that only cognitive performance significantly differed between the two groups. In this study we evaluate the association of impaired cognitive performance and other factors such as poor symptomatic control, somatic factors such as high weight, BMI and blood pressure, in order to determine if there are reliably detectable correlates of cognition as a risk factor for premature death. Finally, physical functioning was evaluated with patient self-reports. The present data are collected in the “Clinical Long-term Investigation of Psychosis in Sweden (CLIPS) –study”, an ongoing study following patients over 20 years and the data used were collected at baseline of the study. As the study did not begin at the onset of illness, we can also examine changes in certain variables between illness onset and the beginning of the study.