Mortality, cause of death and life expectancy in patients with schizophrenia study cohort.
The patient population consisted of all patients with a diagnosis of schizophrenia (ICD-9 code: 295) in NCKUH from Jan 1, 1988, to Dec 31, 2010. This cohort population of schizophrenia was established from the medical record in the NCKUH. The medical record has been regularly maintained in electric format since
1988 and contains information on patients' demographics, including name, date of birth, gender, personal identification (ID) number, diagnosis, prescription and exam result, both outpatients and inpatient services.
The causes of death for each deceased patients was obtained through the National Mortality Registry (NMR) via the personal ID number. These two dataset will be connected via personal ID.
Data Analysis:
Standardized mortality ratio (SMR)
There were 4, 377 patients with schizophrenia identified from NCKUH medical record, and 408 of them were deceased until the end of 2010. The cohort was followed since their enrolled from 1988.1.1 until 2010.12.31, and the total person years were calculated based on whether each patients was deceased or censored. Mortality rate for the Taiwanese population from 1988-2011 were used to estimate the expected number of deaths. We will calculate SMRs for each ICD class and for natural (ICD-9 codes:1-799) and unnatural (ICD-9 codes:E800-E999: suicide, accidents, homicide, external and unspecified violence) causes of death, in male and female separately.
Loss of life expectancy and lifetime medical cost estimation
We'll apply the iSQoL, software for quality adjusted survival estimation and extrapolation of survival for follow-up studies (Hwang and Wang, 1999, 2004) to estimate the expected years of life lost of the study cohort and the expected lifetime medical cost of schizophrenia. The expected year of life loss will be the difference in life expectancy between the schizophrenia and the ager-gender matched reference group. The semiparametric extrapolation method used in iSQol, was based on age-gender specific reference group stimulation by Monte Carlo Method and the Kaplan-Meier Method to estimate lifetime survival function based on the follow-up data from 1988 to 2011.
(五) 成果與討論（含結論與建議）
The longitudinal comparison on the symptoms, executive function was shown in Table 1. These findings indicated that the severity decreased (Figure 1, general PANSS), the general cognitive function remains (Figure 2, full IQ), while the memory improved (Figure 3, delay recall).
Table 1: Severity and cognitive function. (N=37)
Baseline (age = 26.4) Folllow up (age
=34.9)
Mean SD Mean SD Z P
PANSS_P 22.78 6.64 13.94 6.46 3.64 0.00 PANSS_N 21.32 7.36 17.58 5.09 2.94 0.00 PANSS_G 38.68 11.75 27.84 8.23 3.37 0.00
PANSS_S 5.79 2.55 4.05 2.04 2.33 0.00
PANSS_Total 88.61 24.71 64.00 18.35 3.66 0.00
BPRS 43.11 8.25 28.44 10.80 3.48 0.00
GAS 39.32 15.27 56.26 13.90 3.57 0.00
FTT_R 43.99 12.25 41.21 11.12 1.37 0.17 FTT_L 41.29 10.99 39.77 8.17 1.13 0.26 WCST_error 12.94 7.15 15.41 13.75 0.38 0.74 WCST_complete 2.41 1.70 2.76 1.56 1.30 0.19
PIQ 94.00 14.26 97.24 13.54 1.06 0.29
VIQ 106.12 20.57 98.06 11.34 1.58 0.11
FIQ 101.24 17.51 96.94 9.23 0.78 0.44
Verbal_memory 82.35 15.00 91.18 14.50 2.42 0.02 Visual_memory 102.88 19.09 104.88 15.73 0.36 0.72 General_memory 86.81 16.29 94.69 15.02 2.15 0.03 Attention 102.47 13.43 104.35 12.55 0.00 1.00 Delay recall 88.71 17.69 97.94 16.05 2.65 0.01
Figure 1: The trajectory of PANSS general score.
Figure 2: The trajectory of full IQ.
Figure 3: The trajectory of delayed recall performance.
One of our recent published paper probe the improvement in the BPRS psychopathological rating is related with dopamine transporter availability (measured by the SPECT) at baseline. Significant improvement in the BPRS psychopathological rating was observed. (Z=−0.48, p=0.66, Table 1). The striatal DAT availability was significantly lower at follow-up (Z=2.67, p=0.008), but the significance diminished after controlling for years of follow-up (p>0.20). A significant negative correlation was observed between the baseline striatal DAT availability and psychiatric symptoms at follow-up, controlling for age (r=−0.73, p=0.04) (see Tseng et al., 2017)
Our preliminary study on the CPT in the fMRI indicated patients with schizophrenia have a tendency of hyper-activity on some parietal and occipital sites (as illustrated in Figure 4a and b). This finding is novelty.
Figure 4a: the BOLD response (schizophrenia > controls)
Figure 4b: the BOLD response (schizophrenia > controls)
To confirmed the validity of this CPT paradigm in fMRI, we probed the association between the baseline
CPT performance and BOLD signal in the follow-up, we found that the CPT performance under challenge (masked d`), is related with the subcortical area (Table 2), and frontal area activity (Table 3).
Table 2: The association between CPT performance at baseline, and estimated parameters in fMRI for a CPT at the 8 years follow-up: subcortical areas (N = 37)
CPT unmask d' CPT mask d'
Table 3: The association between CPT performance at baseline, and estimated parameters in fMRI for a CPT at the 8 years follow up: frontal cortex (N = 37)
MNI_Frontal_Inf_Oper_L_roi.mat RHO -0.05 0.31
P 0.85 0.26
MNI_Frontal_Inf_Oper_R_roi.mat RHO 0.17 0.56
P 0.50 0.03
MNI_Frontal_Inf_Orb_L_roi.mat RHO 0.13 0.61
P 0.60 0.02
MNI_Frontal_Inf_Orb_R_roi.mat RHO 0.12 0.63
P 0.64 0.01
MNI_Frontal_Inf_Tri_L_roi.mat RHO -0.16 0.16
P 0.54 0.56
MNI_Frontal_Inf_Tri_R_roi.mat RHO 0.05 0.57
P 0.85 0.03
MNI_Frontal_Med_Orb_L_roi.mat RHO 0.21 0.48
P 0.41 0.07
MNI_Frontal_Med_Orb_R_roi.mat RHO 0.01 0.29
P 0.98 0.30
MNI_Frontal_Mid_L_roi.mat RHO -0.15 0.03
P 0.55 0.91
MNI_Frontal_Mid_Orb_L_roi.mat RHO -0.08 0.10
P 0.75 0.72
MNI_Frontal_Mid_Orb_R_roi.mat RHO 0.16 0.25
P 0.52 0.36
MNI_Frontal_Mid_R_roi.mat RHO 0.13 0.33
P 0.61 0.23
MNI_Frontal_Sup_L_roi.mat RHO 0.26 0.47
P 0.30 0.08
MNI_Frontal_Sup_Medial_L_roi.mat RHO 0.11 0.64
P 0.67 0.01
MNI_Frontal_Sup_Medial_R_roi.mat RHO 0.15 0.56
P 0.55 0.03
MNI_Frontal_Sup_Orb_L_roi.mat RHO 0.12 0.28
P 0.65 0.32
MNI_Frontal_Sup_Orb_R_roi.mat RHO 0.07 0.16
P 0.77 0.58
MNI_Frontal_Sup_R_roi.mat RHO 0.23 0.43
P 0.36 0.11
To understanding the mechanism of the above finding, we probe the association with DAT availability.
We found that the frontal activity is related with negatively related with baseline DAT availability, instead of the follow up period, as shown in Table 4. This finding is novelty which could reconfirm the importance of baseline DAT availability and patients’ 8-year outcome.
Table 4: The association between baseline DAT availability, and estimated parameters in fMRI for a CPT at the 8 years follow up: frontal cortex (N = 37)
DAT_baseline DAT_follow_up
In this project, we also conducted the second-hand data analysis. In one subproject, we estimate the lifetime duration of disability in personal and social functioning in patients with schizophrenia. A nationwide database and cross-sectional schizophrenia survey data were used. Lifetime duration of disability was
obtained by multiplying the proportion of patients with functional disability by the survival probability over time and summing throughout the patient’s lifetime. We found that the average lifetime during manifested disability of global functioning was estimated to be 20.9 years, which represents approximately 73% of the whole lifetime of patients. The duration of disability in socially-useful activity was estimated to be 15.6 years, while that of personal and social relationships was 17.5 years. The female patients had a longer duration of manifested disability (22.9 years) than the male patients (19.5 years), as illustrated in Figure 5
Figure 5: Dynamic changes in proportions of functional disabilities in patients with schizophrenia (manuscript submitted)
We also compared the risk of mortality and the years of potential life lost (YPLL) associated with various causes of death in patients with schizophrenia and the general population. A total of 4,298 patients with schizophrenia were included. The cohort was linked to the Taiwan Death Register between 1998 and 2010 using personal identification numbers, which showed that 367 patients with schizophrenia had died by the end of 2010. The standard mortality ratios (SMRs) and YPLL were analyzed by age, sex and cause of death, as shown in Table 5. The overall SMR was significantly higher in patients with schizophrenia (SMR, 8.8; 95%
CI, 7.8–9.6). Suicide had the most significantly greater SMR (SMR, 31.3; 95% CI, 23.3–38.0), and the SMRs for malignant neoplasms, cardiovascular diseases, cerebrovascular diseases, diabetes mellitus , pneumonia, hepatitis and liver cirrhosis , and accidents and injuries were all significantly greater in schizophrenic subjects.
The suicide-specific SMR was significantly higher in subjects with schizophrenia in all age groups. Suicide had the largest YPLL/deaths among all causes of mortality in schizophrenic subjects (mean±SD, 39.6±12.2).
Table 2
Cause-, sex- and age-specific SMRs.
Cause of death
Age (years) Male Female Total
Observed SMR 95% CI Observed SMR 95% CI Observed SMR 95% CI
Malignant neoplasms 15–39 8 8.1 2.8–14.7 2 3.2 0.1–10.5 10 6.2 2.5–10.6
40–64 19 4.2 2.3–6.2 15 4.7 2.3–7.3 34 4.4 2.8–5.9
65+ 6 4.2 1.2–8.5 10 5.5 2.2–9.4 16 4.9 2.5–7.6
Total 33 4.8 3.0–64 27 4.8 2.9–6.6
Cardiovascular diseases 15–39 4 12.5 2.2–29.3 0 0.0 – 4 9.7 1.7–22.7
40–64 12 10.3 4.5–16.8 8 16.9 5.8–30.9 20 12.2 6.7–17.8
65+ 3 5.1 0.5–13.5 3 3.0 0.3–8.0 6 3.8 1.0–7.6
Total 19 9.1 4.9–13.5 11 7.0 3.0–11.8
Cerebrovascular diseases 15–39 0 0.0 – 0 0.0 – 0 0.0 –
40–64 5 5.5 1.3–11.7 3 6.1 0.7–16.2 8 5.7 2.0–10.4
65+ 3 6.5 0.7–17.3 5 6.6 1.5–14.2 8 6.6 2.3–12
Total 8 5.1 1.8–9.3 8 6.1 2.1–11.1
Diabetes mellitus 15–39 2 24.0 0.7–78.4 1 28.1 0.8–141.0 3 25.2 2.7–67.1
40–64 5 8.4 1.9–18.0 8 15.8 5.5–28.9 13 11.8 5.4–18.9
65+ 2 6.7 0.2–21.9 2 2.5 0.1–8.0 4 3.6 0.6–8.4
Total 9 9.2 3.4–16.3 11 8.1 3.4–13.6
Pneumonia 15–39 2 28.1 0.9–91.8 0 0.0 – 2 20.9 0.6–68.2
40–64 2 9.2 0.3–30.2 2 21.6 0.7–70.8 4 12.9 2.2–30.3
65+ 1 3.4 0.1–17.0 4 12.1 2.1–28.5 5 8.0 1.8–17.2
Total 5 8.6 2.0–18.4 6 13.4 3.7–27.0
Hepatitis and liver cirrhosis 15–39 2 3.8 0.1–12.3 0 0.0 – 2 3.4 0.1–11.2
40–64 7 5.2 1.6–9.8 4 13.7 2.4–32.1 11 6.7 2.8–11.1
65+ 1 8.6 0.2–42.9 2 8.8 0.3–28.7 3 8.7 0.9–23.2
Total 10 5.0 2.0–8.6 6 10.5 2.9–21.0
Nephritis and renal failure 15–39 0 0.0 – 1 40.3 1.1–202.1 1 14.3 0.4–71.6
40–64 1 4.7 0.1–23.6 1 4.9 0.1–24.6 2 4.8 0.1–15.7
65+ 0 0.0 – 1 3.0 0.1–15.0 1 2.1 0.1–10.3
Total 1 2.4 0.1–12.3 3 5.3 0.6–14.2
Accidents and injury 15–39 17 8.2 4.2–12.4 5 10.5 2.4–22.6 22 8.6 4.9–12.4
40–64 8 5.8 2.0–10.7 4 8.2 1.4–19.3 12 6.5 2.8–10.6
65+ 3 16.9 1.8–45.0 0 0.0 – 3 8.0 0.9–21.4
Total 28 7.7 4.7–10.6 9 7.8 2.9–13.7
Suicide 15–39 23 28.1 16.1–40.1 24 65.3 37.9–92.3 47 39.7 27.3–50.7
40–64 13 18.3 8.4–29.3 12 35.1 15.4–57.4 25 23.7 13.9–33.3
65+ 2 34.4 1.1–112.6 0 0.0 – 2 15.5 0.5–50.7
Total 38 23.9 15.7–31.5 36 46.1 29.9–61.1
Homicide 15–39 0 0 – 1 38.0 1.1–190.7 1 9.2 0.3–46.1
40–64 0 0 – 1 54.3 1.5–272.2 1 14.2 0.4–71.2
65+ 0 0 – 0 0 – 0 0.0 –
Total 0 0.0 – 2 42.9 1.3–140.1
Our finding demonstrated the trajectory of the symptoms, and outcomes of schizophrenia. Their severity decreased on multiple domains. In particular, the memory function is improved after 8 years treatment, although their IQ remains similar. Baseline dopaminergic activity, measured by SPECT, is predictive to the treatment outcome. Evidence supports the validity of our novel CPT paradigm, which is related with the DAT availability at the baseline. The mechanism of this finding is unclear. We are probing other effects, according to the aims of our study. Our second-hand analysis indicated that the functionality could be an important issue, for those who concern the economic cost of this disorder. The high relapse rate for patients with schizophrenia after first hospitalization was found. A need for lifelong care for patients with schizophrenia should be noticed.
Suicide had the most significantly greater risk of mortality among patients with schizophrenia as compared with the general population. Greater SMRs were also observed for other causes of mortality related to physical illnesses. Patients with schizophrenia are highly vulnerable in terms of increased mortality and require special attention. These information will be published, and integrate for our development of the new model for estimating medical cost/outcome in treating patients with schizophrenia in Taiwan.