Research in Developmental Disabilities 36 (2015) 366–375
Contents lists available at ScienceDirect
Research in Developmental Disabilities
Related factors and incidence risk of acute myocardial infarction among the people with disability: A national population-based study Ying-Ying Huang a,b, Pei-Tseng Kung c,1, Li-Ting Chiu b, Wen-Chen Tsai a,1,* a b c
Department of Health Services Administration, China Medical University, Taichung, Taiwan, ROC China Medical University Hospital, Taichung, Taiwan, ROC Department of Healthcare Administration, Asia University, Taichung, Taiwan, ROC
A R T I C L E I N F O
A B S T R A C T
Article history: Received 23 August 2014 Received in revised form 11 October 2014 Accepted 15 October 2014 Available online
Cardiovascular disease has always been a leading cause of death worldwide. Because the mobility of people with disability is relatively decreased, their risk of cardiovascular disease is increased. This study investigated the risks and relevant factors of acute myocardial infarction (AMI) among people with disability. This is a retrospective cohort study based on secondary data analysis. This study focused on 798,328 people with disability who were aged 35 and above during 2002–2008 and were registered in the National Disability Registration Database; the relevant medical data from 2000 to 2011 were acquired from the National Health Insurance Research Database. A Cox proportional hazards model was adopted for analyzing the relative AMI risks among different disability types and finding latent risk factors. The results indicated that the AMI incidence rate (per 1000 patient-years) among people with disability was 2.48. Men had an AMI incidence rate of 2.68 per 1000 patient-years, which was significantly higher than that of women (2.21; p < .05). The AMI risk for people with mental disabilities was 0.76 times the risk for people with physical disabilities (95% confidence interval [CI] = 0.71–0.82). The AMI risk for people with profound disabilities was 2.04 times (95% CI = 1.93–2.16) the risk for people with mild disabilities. AMI risk increased with age. People with disability aged 65 and above had an AMI risk that was 5.01–6.03 times the risk for people with disability aged below 45. Disabled indigenous people had a relatively higher AMI risk (HR = 1.35, 95% CI = 1.19–1.52). The AMI risk for people with disability with a Charlson comorbidity index (CCI) of 4 and above was 5.89 times (95% CI = 5.56–6.25) the risk for those with a CCI of 0. Compared with people with physical disabilities, people with visual impairment and people with dysfunctional primary organs had significantly higher AMI risks (HR = 1.15; HR = 1.66). This study found that people with disability who were male, aged 65 and above, married, indigenous, with physical disabilities, with high comorbidity, or with high disability levels had relatively higher AMI risks than other people with disability. The research outcomes can be used as references by public health authorities to improve the engagement of people with disability in AMI-prevention health services. ß 2014 Elsevier Ltd. All rights reserved.
Keywords: Disabilities Acute myocardial infarction Risk factor Physical disability Mental disability
* Corresponding author at: No. 91 Hsueh-Shih Road, Taichung 40402, Taiwan, ROC. Tel.: +886 422073070; fax: +886 422028895. E-mail address: [email protected] (W.-C. Tsai). 1 Authors had equal contribution to this work. http://dx.doi.org/10.1016/j.ridd.2014.10.019 0891-4222/ß 2014 Elsevier Ltd. All rights reserved.
Y.-Y. Huang et al. / Research in Developmental Disabilities 36 (2015) 366–375
367
1. Introduction According to the statistics publicized by the Ministry of the Interior in 2013, 1,117,521 people in Taiwan were registered as disabled by the end of 2012, which was 4.8% of the total population (Ministry of the Interior, R.O.C, 2013). A report by the World Health Organization (WHO, 2013) also indicated that people with disability comprised approximately 15% of the world population. The prevalence of disability increases with age (Klijs, Nusselder, Looman, & Mackenbach, 2011). A study conducted in Holland also found that an increasing life expectancy existed among people with disability aged 65 and above (Perenboom, Van Herten, Boshuizen, & Van Den Bos, 2004). Cardiovascular disease (CVD) is a major cause of death among people aged 65 and above, and the related medical expenses constantly increase (North & Sinclair, 2012). Acute myocardial infarction (AMI) is a type of ischemic heart disease that requires immediate medical treatment. The WHO has ranked CVD as the number one cause of death that is responsible for 30% of human deaths worldwide as well as disability-adjusted life years. The WHO predicted that CVD would cause 23 million deaths by 2030 (WHO, 2011; Yusuf et al., 2004). According to the National Heart, Lung, and Blood Institute (NHLBI, 2006) in the United States, approximately 1,100,000 cases of heart attack occur in the United States annually, within which 515,000 die of AMI and 51% of which are men. AMI often occurs without warning signs, or if symptoms existed, they tend to be neglected. Consequently, more than half of the patients with AMI died before arriving at hospitals (NHLBI, 2006). Statistics from 2008 indicated that the incidence rate of AMI in the United States was approximately 208 per 100,000 person-years (Yeh et al., 2010). The statistics announced by Denmark in 2012 showed that the AMI incidence rate was 213 per 1000 men and 131 per 1000 women 100,000 peoples in 2008 (Schmidt, Jacobsen, Lash, Bøtker, & Sørensen, 2012). In Japan, a 30-year follow-up study revealed that the AMI incidence rate (per 100,000 person-year) increased from 7.4 per 100,000 person-years in 1979 to 27 per 100,000 personyears in 2008 (Takii et al., 2010). According to the statistics publicized by the Health Promotion Administration in Taiwan in 2010, the AMI incidence rate in Taiwan increased from 39.2 per 100,000 person-years in 1996 to 79.8 per 100,000 personyears in 2009, indicating that it doubled within 13 years (Health Promotion Administration, 2010). A worldwide study on the risk factors for AMI in 2004 indicated that men had a higher AMI incidence rate and a lower age of onset than women did, and that the relevant risk factors involved smoking, dyslipidemia, diabetes, hypertension, abdominal obesity, diet, exercise, and sociopsychological factors (Yusuf et al., 2004). After adjusting age, gender, and city, the overall risk of coronary artery disease for individuals with moderate or severe disabilities was 2.2 times higher than that for people without disability (Plichart et al., 2010). In addition, variables such as region, ethnicity, income, and educational attainment increased AMI risk (Canto et al., 2011; Rosengren et al., 2009; Steyn et al., 2005). However, the frequency at which people with disability used preventive health services such as pap smear, mammography, and prostate-specific antigen tests was greatly lower than the ideal rate of use (Armour, Thierry, & Wolf, 2009; Huang, Tsai, & Kung, 2012; Kung, Tsai, & Chiou, 2012; Ramirez, Farmer, Grant, & Papachristou, 2005). The number of people with disability is increasing yearly, and as they age, get chronic diseases, or engage in decreased amounts of physical activity, AMI incidence rates increase. As AMI risk increases while the use of preventive health services remains insufficient, people with disability may experience higher morbidity than people without disability do. Therefore, this study investigated the AMI risk of people with disability and relevant factors for obtaining references for the formulation of preventive health policies. 2. Materials and methods 2.1. Data sources and participants This study was a retrospective and longitudinal cohort study. The data were collected from two sources, the National Health Insurance Research Database published by the Ministry of Health and Welfare in Taiwan and the National Disability Registration Database of Taiwan compiled by the Ministry of the Interior. Since the National Health Insurance was implemented in Taiwan in 1995, the percentage of insured people has exceeded 99.5%. The National Health Insurance Research Database includes every patient’s medical information regarding outpatient records, emergency cases, hospitalizations, prescriptions, medical treatments, medical institutes, and doctors (National Health Insurance, 2013). Therefore, the data acquired from the Databases are considered complete and reliable (Cheng, 2003; Sun et al., 2012). This study focused on all the people with disability in the database who were aged 35 and above during 2002–2008 and were not in a vegetative state. The observation was conducted from 2000 to the end of 2011. The patients with AMI were primarily diagnosed by using the ICD-9-CM (The International Classification of Disease, Ninth Revision, Clinical Modification), and the codes 410, 410.x, 410.x0, or 410.x1 were assigned during hospitalization. The participants who had been diagnosed with AMI before becoming disabled were excluded. This study has been approved by the research ethics committee of China Medical University and Hospital (Institutional Review Board No. CMU-REC-101-012). 2.2. Variables description This study included the following variables: (1) demographics: gender, age, educational attainment, marital status, and indigenous status; (2) 17 disability types: people in vegetative states were excluded; physical disabilities included visual impairment, hearing impairment, balance disorder, speech and language disorder, dysfunction of the primary organs,
368
Y.-Y. Huang et al. / Research in Developmental Disabilities 36 (2015) 366–375
physical disability, facial impairment, intractable epilepsy, and rare diseases; mental disabilities included intellectual disabilities, dementia, autism, chronic mental disorders, chromosomal abnormalities, congenital metabolic abnormalities, and other congenital disorders; we further divided disabilities into the categories of physical, mental, or both (Verger et al., 2005); additionally, disability levels involved mild, moderate, severe, and extremely severe; (3) economic factors: 7 groups were formed based on monthly income; (4) environmental factors: levels of urbanization of residence area (7 levels, within which Level 1 represents the highest level of urbanization and Level 7 denotes the lowest level of urbanization); in this study, the levels were categorized into four levels, namely, Level 1, Levels 2–3, Levels 4–5, and Levels 6–7, to facilitate analysis and discussion; (5) health conditions: the Charlson comorbidity index (CCI; Deyo, Cherkin, & Ciol, 1992) was adopted in this study and 19 diseases were involved and assigned with a score of 0, 1, 2, 3, or 4. 2.3. Statistical analysis This study adopted the SAS 9.3 statistical software for data analysis. Regarding descriptive statistics, the number of people, percentages, and AMI incidence rates for people with disability were calculated based on the five dimensions of disability characteristics, demographics, economic factors, environmental factors, and health conditions. Regarding inferential statistics, we performed chi-square tests to analyze variables for verifying their correlation with AMI risk. A Cox proportional hazards model was employed for analyzing the AMI risk of people with disability and relevant factors. The study observation time was during 2002–2011, and month was used as the unit for time. Confirmed AMI diagnoses were considered as events, whereas nonoccurrence of AMI or death during the study was regarded as censored. Additionally, the Cox proportional hazards model was used again to analyze the relative risks and relevant factors for the incidence of AMI among people with different types of disabilities. 3. Results The analysis included 798,328 people with disability from 2002 to 2008. The average observation period for each individual was 6.46 3.34 years, during which 12,783 people suffered from AMI, generating an annual incidence rate of 2.48 per one thousand people. The incidence rates mentioned hereafter were calculated based on the unit of per 1000 person-years (Table 1). According to the demographic analysis, men had an AMI incidence rate of 2.68, which was higher than that of women, which was 2.21. People aged 35–44 had an AMI incidence rate of 0.49. The AMI incidence rate increased with age, and people aged 75 and above had the highest rate (4.54). Regarding educational attainment, the AMI incidence rate (2.79) among the people who were illiterate or had elementary school educations was the highest, followed by that among the people with college educations and above (2.12). Compared with the unmarried people, who had an AMI incidence rate of 1.38, married and divorced/widowed people had higher incidence rates (2.80 and 2.42, respectively). Regarding indigenous status, no significant difference (p > .05) existed between the AMI incidence rate of indigenous people (2.46) and that of nonindigenous people (2.48). Concerning disability types, people with intellectual disabilities or mental disorders had relatively lower incidence rates (0.30 and 0.58, respectively). Regarding disability characteristics, people with physical disabilities had an incidence rate of 2.73, whereas people with mental disabilities had a significantly lower rate of 0.99. In addition, AMI incidence rates increased with disability levels. Regarding economic factors, people with monthly salary lower than NT$17,280 had the lowest AMI incidence rate (1.29), whereas people with monthly salary of NT$22,801–28,800 had the highest AMI incidence rate (3.26), followed by those with monthly salary of NT$45,801 and above (3.04). The AMI incidence rate among the people with disability in low-income families was 1.74. Regarding urbanization of residence area, the AMI incidence rate was the highest (3.02) in Level-1 highly developed and urbanized areas, and the rates in the areas of the other three levels were 2.34, 2.28, and 2.53. People with higher CCIs exhibited significantly higher incidence rates. The AMI incidence rate among the people with a CCI of more than 3 was eight times higher than that among the people without comorbidity (7.26 vs. 0.87). The Cox proportional hazards model was adopted in this study (Table 2). The disability characteristics were categorized into physical disabilities, mental disabilities, and both. According to Model A in Table 2, when the other variables were controlled, the AMI risk for people with mental disabilities was 0.76 times the risk for those with physical disabilities (95% CI = 0.71–0.82, p < .05); conversely people with physical disabilities had 1.32 times the risk of AMI that those with mental disabilities did. The AMI risk for people with both types of disabilities was 0.60 times the risk for those with physical disabilities (95% CI = 0.51–0.72, p < .05). Regarding the variables (Table 2; Model A), relatively higher AMI incidence rates occurred among people with profound disabilities, men, older people, married people, indigenous people, people with moderate incomes, and people with high CCIs. The AMI risk for the people with profound disabilities was 2.04 times the risk for those with mild disabilities (95% CI = 1.93–2.16, p < .05). Women had a lower AMI risk than men did (HR = .70, 95% CI = 0.67–0.73, p < .05). AMI risk increased with age; the AMI risk among people aged 65 and above was 5.01–6.03 times higher than the risk for those aged below 45. The AMI risk among people with high school educations was 12% greater than that among people who were illiterate or had elementary school educations (HR = 0.12, 95% CI = 1.05–1.20, p < .05). Married people had a higher AMI risk than unmarried people did (HR = 1.17, 95% CI = 1.10–1.25, p < .05), and indigenous people had comparatively higher AMI risk as well (HR = 1.35, 95% CI = 1.19–1.52). The AMI risk for people with a monthly salary of NT$22,801–28,800 was 21% higher than the risk for those in low-income families (HR = 1.21, 95% CI = 1.09–1.34, p < .05).
Y.-Y. Huang et al. / Research in Developmental Disabilities 36 (2015) 366–375
369
Table 1 The incidence rates of acute myocardial infarction among people with disability (per 1000 person-years). Variable
N
AMI Case
Average observed person-years
SD
Incidence rate
Total Disability characteristics Physical (reference) Mental Both Disability type Physical disability (reference) Visual impairment Hearing impairment Speech disorder Intellectual disability Multiple disabilities Dysfunction of primary organs Dementia Mental disorder Othera Disability level Mild (reference) Moderate Severe Profound Gender Male (reference) Female Ageb 35–44 (reference) 45–54 55–64 65–74 75 Educational attainment Illiterate or elementary school (reference) Junior high school High school College or above Unknown Marital status Unmarried (reference) Married Divorced or widowed Unknown Indigenous status No (reference) Yes Monthly salary (NT$) Low-income family (reference) 17,280 17,281–22,800 22,801–28,800 28,801–36,300 36,301–45,800 45,801 Urbanization level 1 (reference) 2+3 4+5 6+7 CCI 0 (reference) 1 2 3 4
798,328
12,783
6.46
3.34
2.48
674,564 113,330 10,434
11,922 730 131
6.47 6.50 5.23
3.38 3.06 3.11
2.73 0.99 2.40
–
Contents lists available at ScienceDirect
Research in Developmental Disabilities
Related factors and incidence risk of acute myocardial infarction among the people with disability: A national population-based study Ying-Ying Huang a,b, Pei-Tseng Kung c,1, Li-Ting Chiu b, Wen-Chen Tsai a,1,* a b c
Department of Health Services Administration, China Medical University, Taichung, Taiwan, ROC China Medical University Hospital, Taichung, Taiwan, ROC Department of Healthcare Administration, Asia University, Taichung, Taiwan, ROC
A R T I C L E I N F O
A B S T R A C T
Article history: Received 23 August 2014 Received in revised form 11 October 2014 Accepted 15 October 2014 Available online
Cardiovascular disease has always been a leading cause of death worldwide. Because the mobility of people with disability is relatively decreased, their risk of cardiovascular disease is increased. This study investigated the risks and relevant factors of acute myocardial infarction (AMI) among people with disability. This is a retrospective cohort study based on secondary data analysis. This study focused on 798,328 people with disability who were aged 35 and above during 2002–2008 and were registered in the National Disability Registration Database; the relevant medical data from 2000 to 2011 were acquired from the National Health Insurance Research Database. A Cox proportional hazards model was adopted for analyzing the relative AMI risks among different disability types and finding latent risk factors. The results indicated that the AMI incidence rate (per 1000 patient-years) among people with disability was 2.48. Men had an AMI incidence rate of 2.68 per 1000 patient-years, which was significantly higher than that of women (2.21; p < .05). The AMI risk for people with mental disabilities was 0.76 times the risk for people with physical disabilities (95% confidence interval [CI] = 0.71–0.82). The AMI risk for people with profound disabilities was 2.04 times (95% CI = 1.93–2.16) the risk for people with mild disabilities. AMI risk increased with age. People with disability aged 65 and above had an AMI risk that was 5.01–6.03 times the risk for people with disability aged below 45. Disabled indigenous people had a relatively higher AMI risk (HR = 1.35, 95% CI = 1.19–1.52). The AMI risk for people with disability with a Charlson comorbidity index (CCI) of 4 and above was 5.89 times (95% CI = 5.56–6.25) the risk for those with a CCI of 0. Compared with people with physical disabilities, people with visual impairment and people with dysfunctional primary organs had significantly higher AMI risks (HR = 1.15; HR = 1.66). This study found that people with disability who were male, aged 65 and above, married, indigenous, with physical disabilities, with high comorbidity, or with high disability levels had relatively higher AMI risks than other people with disability. The research outcomes can be used as references by public health authorities to improve the engagement of people with disability in AMI-prevention health services. ß 2014 Elsevier Ltd. All rights reserved.
Keywords: Disabilities Acute myocardial infarction Risk factor Physical disability Mental disability
* Corresponding author at: No. 91 Hsueh-Shih Road, Taichung 40402, Taiwan, ROC. Tel.: +886 422073070; fax: +886 422028895. E-mail address: [email protected] (W.-C. Tsai). 1 Authors had equal contribution to this work. http://dx.doi.org/10.1016/j.ridd.2014.10.019 0891-4222/ß 2014 Elsevier Ltd. All rights reserved.
Y.-Y. Huang et al. / Research in Developmental Disabilities 36 (2015) 366–375
367
1. Introduction According to the statistics publicized by the Ministry of the Interior in 2013, 1,117,521 people in Taiwan were registered as disabled by the end of 2012, which was 4.8% of the total population (Ministry of the Interior, R.O.C, 2013). A report by the World Health Organization (WHO, 2013) also indicated that people with disability comprised approximately 15% of the world population. The prevalence of disability increases with age (Klijs, Nusselder, Looman, & Mackenbach, 2011). A study conducted in Holland also found that an increasing life expectancy existed among people with disability aged 65 and above (Perenboom, Van Herten, Boshuizen, & Van Den Bos, 2004). Cardiovascular disease (CVD) is a major cause of death among people aged 65 and above, and the related medical expenses constantly increase (North & Sinclair, 2012). Acute myocardial infarction (AMI) is a type of ischemic heart disease that requires immediate medical treatment. The WHO has ranked CVD as the number one cause of death that is responsible for 30% of human deaths worldwide as well as disability-adjusted life years. The WHO predicted that CVD would cause 23 million deaths by 2030 (WHO, 2011; Yusuf et al., 2004). According to the National Heart, Lung, and Blood Institute (NHLBI, 2006) in the United States, approximately 1,100,000 cases of heart attack occur in the United States annually, within which 515,000 die of AMI and 51% of which are men. AMI often occurs without warning signs, or if symptoms existed, they tend to be neglected. Consequently, more than half of the patients with AMI died before arriving at hospitals (NHLBI, 2006). Statistics from 2008 indicated that the incidence rate of AMI in the United States was approximately 208 per 100,000 person-years (Yeh et al., 2010). The statistics announced by Denmark in 2012 showed that the AMI incidence rate was 213 per 1000 men and 131 per 1000 women 100,000 peoples in 2008 (Schmidt, Jacobsen, Lash, Bøtker, & Sørensen, 2012). In Japan, a 30-year follow-up study revealed that the AMI incidence rate (per 100,000 person-year) increased from 7.4 per 100,000 person-years in 1979 to 27 per 100,000 personyears in 2008 (Takii et al., 2010). According to the statistics publicized by the Health Promotion Administration in Taiwan in 2010, the AMI incidence rate in Taiwan increased from 39.2 per 100,000 person-years in 1996 to 79.8 per 100,000 personyears in 2009, indicating that it doubled within 13 years (Health Promotion Administration, 2010). A worldwide study on the risk factors for AMI in 2004 indicated that men had a higher AMI incidence rate and a lower age of onset than women did, and that the relevant risk factors involved smoking, dyslipidemia, diabetes, hypertension, abdominal obesity, diet, exercise, and sociopsychological factors (Yusuf et al., 2004). After adjusting age, gender, and city, the overall risk of coronary artery disease for individuals with moderate or severe disabilities was 2.2 times higher than that for people without disability (Plichart et al., 2010). In addition, variables such as region, ethnicity, income, and educational attainment increased AMI risk (Canto et al., 2011; Rosengren et al., 2009; Steyn et al., 2005). However, the frequency at which people with disability used preventive health services such as pap smear, mammography, and prostate-specific antigen tests was greatly lower than the ideal rate of use (Armour, Thierry, & Wolf, 2009; Huang, Tsai, & Kung, 2012; Kung, Tsai, & Chiou, 2012; Ramirez, Farmer, Grant, & Papachristou, 2005). The number of people with disability is increasing yearly, and as they age, get chronic diseases, or engage in decreased amounts of physical activity, AMI incidence rates increase. As AMI risk increases while the use of preventive health services remains insufficient, people with disability may experience higher morbidity than people without disability do. Therefore, this study investigated the AMI risk of people with disability and relevant factors for obtaining references for the formulation of preventive health policies. 2. Materials and methods 2.1. Data sources and participants This study was a retrospective and longitudinal cohort study. The data were collected from two sources, the National Health Insurance Research Database published by the Ministry of Health and Welfare in Taiwan and the National Disability Registration Database of Taiwan compiled by the Ministry of the Interior. Since the National Health Insurance was implemented in Taiwan in 1995, the percentage of insured people has exceeded 99.5%. The National Health Insurance Research Database includes every patient’s medical information regarding outpatient records, emergency cases, hospitalizations, prescriptions, medical treatments, medical institutes, and doctors (National Health Insurance, 2013). Therefore, the data acquired from the Databases are considered complete and reliable (Cheng, 2003; Sun et al., 2012). This study focused on all the people with disability in the database who were aged 35 and above during 2002–2008 and were not in a vegetative state. The observation was conducted from 2000 to the end of 2011. The patients with AMI were primarily diagnosed by using the ICD-9-CM (The International Classification of Disease, Ninth Revision, Clinical Modification), and the codes 410, 410.x, 410.x0, or 410.x1 were assigned during hospitalization. The participants who had been diagnosed with AMI before becoming disabled were excluded. This study has been approved by the research ethics committee of China Medical University and Hospital (Institutional Review Board No. CMU-REC-101-012). 2.2. Variables description This study included the following variables: (1) demographics: gender, age, educational attainment, marital status, and indigenous status; (2) 17 disability types: people in vegetative states were excluded; physical disabilities included visual impairment, hearing impairment, balance disorder, speech and language disorder, dysfunction of the primary organs,
368
Y.-Y. Huang et al. / Research in Developmental Disabilities 36 (2015) 366–375
physical disability, facial impairment, intractable epilepsy, and rare diseases; mental disabilities included intellectual disabilities, dementia, autism, chronic mental disorders, chromosomal abnormalities, congenital metabolic abnormalities, and other congenital disorders; we further divided disabilities into the categories of physical, mental, or both (Verger et al., 2005); additionally, disability levels involved mild, moderate, severe, and extremely severe; (3) economic factors: 7 groups were formed based on monthly income; (4) environmental factors: levels of urbanization of residence area (7 levels, within which Level 1 represents the highest level of urbanization and Level 7 denotes the lowest level of urbanization); in this study, the levels were categorized into four levels, namely, Level 1, Levels 2–3, Levels 4–5, and Levels 6–7, to facilitate analysis and discussion; (5) health conditions: the Charlson comorbidity index (CCI; Deyo, Cherkin, & Ciol, 1992) was adopted in this study and 19 diseases were involved and assigned with a score of 0, 1, 2, 3, or 4. 2.3. Statistical analysis This study adopted the SAS 9.3 statistical software for data analysis. Regarding descriptive statistics, the number of people, percentages, and AMI incidence rates for people with disability were calculated based on the five dimensions of disability characteristics, demographics, economic factors, environmental factors, and health conditions. Regarding inferential statistics, we performed chi-square tests to analyze variables for verifying their correlation with AMI risk. A Cox proportional hazards model was employed for analyzing the AMI risk of people with disability and relevant factors. The study observation time was during 2002–2011, and month was used as the unit for time. Confirmed AMI diagnoses were considered as events, whereas nonoccurrence of AMI or death during the study was regarded as censored. Additionally, the Cox proportional hazards model was used again to analyze the relative risks and relevant factors for the incidence of AMI among people with different types of disabilities. 3. Results The analysis included 798,328 people with disability from 2002 to 2008. The average observation period for each individual was 6.46 3.34 years, during which 12,783 people suffered from AMI, generating an annual incidence rate of 2.48 per one thousand people. The incidence rates mentioned hereafter were calculated based on the unit of per 1000 person-years (Table 1). According to the demographic analysis, men had an AMI incidence rate of 2.68, which was higher than that of women, which was 2.21. People aged 35–44 had an AMI incidence rate of 0.49. The AMI incidence rate increased with age, and people aged 75 and above had the highest rate (4.54). Regarding educational attainment, the AMI incidence rate (2.79) among the people who were illiterate or had elementary school educations was the highest, followed by that among the people with college educations and above (2.12). Compared with the unmarried people, who had an AMI incidence rate of 1.38, married and divorced/widowed people had higher incidence rates (2.80 and 2.42, respectively). Regarding indigenous status, no significant difference (p > .05) existed between the AMI incidence rate of indigenous people (2.46) and that of nonindigenous people (2.48). Concerning disability types, people with intellectual disabilities or mental disorders had relatively lower incidence rates (0.30 and 0.58, respectively). Regarding disability characteristics, people with physical disabilities had an incidence rate of 2.73, whereas people with mental disabilities had a significantly lower rate of 0.99. In addition, AMI incidence rates increased with disability levels. Regarding economic factors, people with monthly salary lower than NT$17,280 had the lowest AMI incidence rate (1.29), whereas people with monthly salary of NT$22,801–28,800 had the highest AMI incidence rate (3.26), followed by those with monthly salary of NT$45,801 and above (3.04). The AMI incidence rate among the people with disability in low-income families was 1.74. Regarding urbanization of residence area, the AMI incidence rate was the highest (3.02) in Level-1 highly developed and urbanized areas, and the rates in the areas of the other three levels were 2.34, 2.28, and 2.53. People with higher CCIs exhibited significantly higher incidence rates. The AMI incidence rate among the people with a CCI of more than 3 was eight times higher than that among the people without comorbidity (7.26 vs. 0.87). The Cox proportional hazards model was adopted in this study (Table 2). The disability characteristics were categorized into physical disabilities, mental disabilities, and both. According to Model A in Table 2, when the other variables were controlled, the AMI risk for people with mental disabilities was 0.76 times the risk for those with physical disabilities (95% CI = 0.71–0.82, p < .05); conversely people with physical disabilities had 1.32 times the risk of AMI that those with mental disabilities did. The AMI risk for people with both types of disabilities was 0.60 times the risk for those with physical disabilities (95% CI = 0.51–0.72, p < .05). Regarding the variables (Table 2; Model A), relatively higher AMI incidence rates occurred among people with profound disabilities, men, older people, married people, indigenous people, people with moderate incomes, and people with high CCIs. The AMI risk for the people with profound disabilities was 2.04 times the risk for those with mild disabilities (95% CI = 1.93–2.16, p < .05). Women had a lower AMI risk than men did (HR = .70, 95% CI = 0.67–0.73, p < .05). AMI risk increased with age; the AMI risk among people aged 65 and above was 5.01–6.03 times higher than the risk for those aged below 45. The AMI risk among people with high school educations was 12% greater than that among people who were illiterate or had elementary school educations (HR = 0.12, 95% CI = 1.05–1.20, p < .05). Married people had a higher AMI risk than unmarried people did (HR = 1.17, 95% CI = 1.10–1.25, p < .05), and indigenous people had comparatively higher AMI risk as well (HR = 1.35, 95% CI = 1.19–1.52). The AMI risk for people with a monthly salary of NT$22,801–28,800 was 21% higher than the risk for those in low-income families (HR = 1.21, 95% CI = 1.09–1.34, p < .05).
Y.-Y. Huang et al. / Research in Developmental Disabilities 36 (2015) 366–375
369
Table 1 The incidence rates of acute myocardial infarction among people with disability (per 1000 person-years). Variable
N
AMI Case
Average observed person-years
SD
Incidence rate
Total Disability characteristics Physical (reference) Mental Both Disability type Physical disability (reference) Visual impairment Hearing impairment Speech disorder Intellectual disability Multiple disabilities Dysfunction of primary organs Dementia Mental disorder Othera Disability level Mild (reference) Moderate Severe Profound Gender Male (reference) Female Ageb 35–44 (reference) 45–54 55–64 65–74 75 Educational attainment Illiterate or elementary school (reference) Junior high school High school College or above Unknown Marital status Unmarried (reference) Married Divorced or widowed Unknown Indigenous status No (reference) Yes Monthly salary (NT$) Low-income family (reference) 17,280 17,281–22,800 22,801–28,800 28,801–36,300 36,301–45,800 45,801 Urbanization level 1 (reference) 2+3 4+5 6+7 CCI 0 (reference) 1 2 3 4
798,328
12,783
6.46
3.34
2.48
674,564 113,330 10,434
11,922 730 131
6.47 6.50 5.23
3.38 3.06 3.11
2.73 0.99 2.40
–
