View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

H.C. Hsu and R.F. Pwu

TOO LATE TO QUIT? EFFECT OF AND ON MORBIDITY AND MORTALITY AMONG THE ELDERLY IN A LONGITUDINAL STUDY

Hui-Chuan Hsu and Raoh-Fang Pwu1 Department of Health Care Administration, Healthcare and Management University, Taichung, and 1Research Director, iStat Healthcare Consulting Co Ltd, , .

This prospective study of the elderly population estimated the risks of smoking for morbidity and mortality and identified whether cessation of smoking reduced the risk of disease. Data came from face-to-face interviews that used a population-based probability sample of those aged 60 years or over in Taiwan, provided by the Population and Health Research Center, Bureau of Health Promotion. In total, 4,049 subjects were included at the baseline year of 1989 and followed up in 1993 and 1996. Smoking-related variables included current smoking status, smoking history, daily consumption, and years since the cessation of smoking. Cox regression models were used to analyze the relative risks for morbidity and mortality, controlling for demographics, physical function, and comorbidities. The sample was made up of 50.2% nonsmokers, 15.2% ex-smokers, and 34.6% current smokers in the baseline year. Current smokers were more likely to have lower respiratory tract diseases throughout the study. Current smokers had a higher risk of stroke from 1989 to 1993. No dose-response relationship for smoking exposure or impact of years since smoking cessation was found. Whether cessation of smoking is protective should be investigated for middle-aged adults followed to old age. An effective strategy for smoking cessation in the elderly is suggested, and people should be encouraged to quit smoking at any time.

Key Words: smoking, smoking cessation, longitudinal studies, elderly, relative risk, successful aging ( J Med Sci 2004;20:484–91)

Smoking is an important health hazard in mortality. It is There is evidence that a reduction in risk occurs on estimated that about 3 million people die each year from quitting smoking. The reduction in risk for heart disease is smoking-related diseases, and this will increase to over 10 linked to the duration of cessation [14]. Smoking cessation million worldwide by 2025 [1]. Many prospective studies can reduce the risk of mortality, and smoking cessation af- have identified cigarette smoking as an important risk ter acute myocardial infarction can reduce the relative risk factor for cardiovascular disease, chronic bronchitis, several (RR) of death by between 15% and 61% [15]. Smoking types of cancer, and other diseases [2–13]. cessation may increase lung function and lower mortality for male smokers [16]; even an interrupted cessation of smoking is helpful and can improve lung function. Only a few studies of the impact of smoking on health in Received: May 17, 2004 Accepted: August 27, 2004 Taiwan have been reported. A prospective cohort study Address correspondence and reprint requests to: Dr. Hui-Chuan explored the attributable risk of cigarette smoking to death Hsu, Department of Health Care Administration, Taichung Health- from various diseases in township residents aged 40 or over care and Management University, 500 Lioufeng Road, Wufeng Shiang, Taichung 413, Taiwan. [17]. The RR of mortality for male smokers was 1.3, and 1.5 E-mail: [email protected] in all cancers. The RRs for stomach disease, ischemic and

484 Kaohsiung J Med Sci October 2004 ¥ Vol 20 ¥ No 10 © 2004 Elsevier. All rights reserved. Effect of smoking on morbidity & mortality among the elderly other heart diseases, and chronic obstructive pulmonary for 1989, 1993 and 1996 were 91.8%, 91.0%, and 88.9%. There disease (COPD) ranged from 1.4 to 1.9, with higher RRs for were no differences in age, sex, and smoking status among females. The attributable risks for mortality associated with the completed cases and those missing through loss or smoking were 21.3% and 2.9% for males and females, re- death from the initial sample when examined for goodness- spectively. In another case-control study, the risk of lung of-fit. cancer was significantly higher in smokers than in non- smokers [18]. Even for those who had stopped smoking, the Variables RRs were higher at 6.39 and 4.00. The risk for every pack of Dependent variables were self-reported chronic diseases cigarettes smoked was 1.04. The risk of a smoking history and mortality. Diseases included heart disease, stroke, lower over 25 years for nasopharyngeal cancer in Taiwan was 1.7 respiratory tract diseases (asthma, bronchitis, tuberculosis), compared to nonsmokers [19]. In Taiwan, smoking cessation diabetes, digestive system diseases (ulcer, gastrointestinal projects and intervention programs have been developed problems, liver/gallbladder diseases), kidney diseases, and [20–23]. However, evidence that cessation of smoking re- cancer (only surveyed in 1993 and 1996). duces the risk of disease and mortality is limited [24]. The major independent variables related to smoking The question of whether smoking cessation reduces were current smoking status (nonsmokers, who had never health hazard risks for the elderly is important for the smoked; ex-smokers, who quit smoking; and current attractive strategy of health promotion as a way of achieving smokers), smoking history (starting age of smoking, years successful aging [25]. However, few studies have shown of quitting), and smoking exposure (cumulative smoking that smoking cessation has an impact on morbidity and years, amount smoked every day). The cumulative smoking mortality in old age. Also, there have been no prospective years were defined from the age of starting smoking to the follow-up studies targeting the elderly population to esti- survey year. If the subject had quit smoking, their smoking mate the risks of smoking and the effect of smoking cessation history was also counted. The cumulative smoking years on morbidity and mortality. Therefore, to explore the effects for those who never smoked was defined as zero. of smoking cessation on the risk of disease will be useful as Controlled risk factors included age (60–74, ≥ 75 years), a basis for the development of a health promotion strategy gender, education (low = illiterate, no formal education, for the elderly and to attain the goal of successful aging. elementary school; medium = junior high school; high = The purpose of this study was to use prospective elderly senior high school or over), ethnic group (Ming-Nan, others samples with 7-year follow-up to assess the impact of including Hakka, mainland provinces, etc), disability in smoking status on morbidity and mortality, and examine physical function in 1989 (any difficulties caused by health whether cessation of smoking reduced the risks of diseases problems in shopping for personal items, management of in old age. money, making phone calls, and taking a bath alone).

Analysis METHODS The incidence of each disease was estimated. The RRs and population-attributable risk (PAR) of smoking status on Data and samples morbidity and mortality were estimated using Cox Data came from a face-to-face interview, the “Study of proportional hazards regression models. Three kinds of Health and Living Status of the Elderly” a national repre- smoking effects were analyzed. First, the risk of smoking sentative elderly sample from a survey conducted by the status (nonsmokers, ex-smokers, and current smokers) in Population and Health Research Center, Bureau of Health all samples. Second, for current smokers, the risks of smok- Promotion. Samples were of people aged 60 years or over ing exposure (daily cigarette consumption, age at starting and drawn randomly from the entire elderly population in smoking, cumulative smoking years) were analyzed. Final- Taiwan with 1989 as the baseline, and then followed up in ly, for ex-smokers, the risk of smoking history (cumulative 1993 and 1996. Sampling was made up of a three-stage smoking years, years after quitting) on morbidity and probability sample. All interviews were conducted by senior mortality were analyzed. We did not estimate the 7-year interviewers under high quality control. In this study, we morbidity from cancer because of the absence of cancer used the 1989 sample as the baseline (n = 4,049), together data for the baseline year. Multiple Cox regression analysis with follow-up data for 1993 (n = 3,155) and 1996 (n = 2,669, of smoking status relative to morbidity and mortality, aged 67 years or over). Completion rate (excluding deaths) controlling for related factors, was also done.

Kaohsiung J Med Sci October 2004 ¥ Vol 20 ¥ No 10 485 H.C. Hsu and R.F. Pwu

RESULTS at starting smoking, and cumulative smoking years on the 7-year incidence of disease and mortality (Table 2). How- In 1989, 36.6% of the initial sample were aged 60–64 ever, no dose-response relationship was found. We also years, 28.5% were aged 65–69 years, 17.9% were aged 70– analyzed the risk of cumulative smoking years and the 74 years, and 17.0% were aged 75 years or above. There years since quitting on the 7-year incidence of disease and were more men (57.1%, n = 2,311) than women (42.9%, mortality for ex-smokers (Table 3). Smoking history did n = 1,738). Half (50.3%) of the elderly were illiterate or not have a significant impact on the relative risks of mor- without formal education, 30.7% had attended elementary bidity or mortality. school, and 18.9% were educated to junior high school or Table 4 shows the multiple Cox regression models for above. Of the sample, 60.9% were Ming-Nan, 15.0% were smoking status relative to morbidity and mortality, Hakka, 22.4% came from mainland provinces, and 1.7% controlling for demographic and related health factors. were of other ethnic groups. Overall, 18.2% were physically Smoking status influenced only lower respiratory tract disabled. disease: current (RR, 1.7) and ex-smokers (RR, 1.4) were Of the initial sample of 4,049, 21.7% had heart disease, more likely to have respiratory disease. Comorbidity with 4.3% had had a stroke, 8.4% had diabetes, 18.5% had lower other diseases was also more likely to increase the incidence respiratory tract disease, 27.2% had digestive system disease, of morbidity. The older elderly with comorbidity and and 6.3% had kidney disease. In 1993, the prevalence of physical disability had a higher risk of mortality. cancer was 1.7%. In 1989, there were 2,033 (50.2%) nonsmokers, 616 (15.2%) ex-smokers, and 1,399 (34.6%) current smokers (1 DISCUSSION case was missing). In 1996, 670 of 2,669 (25.1%) complete respondents still smoked. Most smokers were male. In This study used a 7-year national longitudinal dataset from 1989, 55.3% of men were current smokers, 24.4% were ex- Taiwan to explore the RRs of morbidity and mortality with smokers, and 20.4% were nonsmokers. Of the women, only respect to smoking status, smoking history, and the cessation 7.0% were current smokers, 3.0% were ex-smokers, and of smoking among the elderly. The smoking elderly had 90.0% were nonsmokers. The younger elderly had a higher a higher RR of lower respiratory tract disease (1.6; PAR, rate of smoking. In 1989, the current smoking rate was 17.2%), as did ex-smokers (RR, 1.4). Ex-smokers had a 36.8% for those aged 60–74 and 24.0% for those aged 75 or higher risk of cancer morbidity. However, for both current above. smokers and quitters at the 7-year follow-up, the dose- Of the current smokers, 1,326 of 1,385 (95.7%) complete response relationship of smoking consumption, smoking cases were long-term smokers (smoking history > 20 years); history, and years of cessation was not associated signifi- 483 of 594 (81.3%) ex-smokers had smoked for more than cantly with morbidity and mortality risks. 20 years. Among the 1,390 current smokers (9 cases were The prevalence of elderly smoking in Taiwan was stable missing), the daily cigarette consumption was less than from 1989 to 1996 in this study. According to the 1989 10 cigarettes for 44.1% and 11–20 cigarettes for 50.9%, while survey, smoking prevalence among the elderly was 34.6%, only 5.0% smoked more than 20 cigarettes per day. For with 15.2% who had quit smoking and 50.2% who had ex-smokers, about 40% had quit smoking within the past never smoked. The 1996 prevalence was similar to that in 5 years, while 25.5% had quit for 20 years or more. 1989. Prospective panel data suggest that there may have The univariate RR of morbidity and mortality for dif- been changes in smoking behavior during the 7 years of ferent smoking status is shown in Table 1. Smokers were follow-up. However, the three surveys did not all include more likely to have lower respiratory tract disease (RR, 1.6 the complete set of variables for smoking exposure, so we for current smokers, 1.4 for ex-smokers; PAR, 17.2% from could only estimate the exposure according to the baseline 1989 to 1996). Current smokers also had a higher risk over year. the two intervals 1989–1993 and 1993–1996. In addition, Current smoking status was associated with the current smokers had a higher risk of stroke from 1989 to incidence of stroke, lower respiratory tract disease, and 1993 (RR, 1.6; PAR, 17.2%). The 1993–1996 RR of cancer was cancer using univariate Cox models, and the effect on higher for ex-smokers. respiratory disease persisted in the multivariate model. For current smokers, univariate Cox regression was Respiratory disease can be very sensitive to smoking used to analyze the risk of daily cigarette consumption, age behavior, as other case-control studies and prospective

486 Kaohsiung J Med Sci October 2004 ¥ Vol 20 ¥ No 10 Effect of smoking on morbidity & mortality among the elderly

Table 1. Incidence of disease and mortality by smoking status among the elderly between 1989 and 1996: relative risk (95% confidence interval)

1989–1993 1993–1996 1989–1996

Heart disease Current smokers 0.95 (0.8–1.2) 0.8 (0.6–1.1) 0.9 (0.7–1.1) Ex-smokers 0.96 (0.69–1.3) 0.8 (0.6–1.2) 0.9 (0.7–1.2) PAR of smoking –1.8% –7.4% –3.6% Stroke Current smokers 1.6 (1.1–2.3) 0.7 (0.4–1.2) 1.1 (0.8–1.6) Ex-smokers 1.3 (0.8–2.2) 1.8 (1.0–3.1) 1.5 (1.0–2.3) PAR of smoking 17.2% –11.6% 3.3% Diabetes Current smokers 0.9 (0.6–1.2) 0.8 (0.6–1.3) 0.9 (0.7–1.2) Ex-smokers 1.0 (0.6–1.6) 0.8 (0.5–1.5) 1.1 (0.7–1.6) PAR of smoking –3.6% –7.4% –3.6% Cancer Current smokers – 1.4 (0.6–3.2) – Ex-smokers –2.8 (1.2–6.7) – PAR of smoking – 12.2% – Lower respiratory tract disease Current smokers 1.5 (1.2–2.0) 1.7 (1.2–2.3) 1.6 (1.3–2.0) Ex-smokers 1.2 (0.9–1.8) 1.9 (1.3–2.8) 1.4 (1.0–1.9) PAR of smoking 14.7% 19.5% 17.2% Digestive disease Current smokers 1.2 (0.9–1.6) 0.8 (0.7–1.1) 1.0 (0.8–1.2) Ex-smokers 1.0 (0.7–1.5) 1.0 (0.7–1.4) 0.8 (0.6–1.2) PAR of smoking6.5%–7.4% 0% Kidney disease Current smokers 0.7 (0.5–1.0) 0.9 (0.6–1.3) 0.8 (0.6–1.1) Ex-smokers 0.7 (0.4–1.1) 0.7 (0.4–1.2) 0.7 (0.4–1.0) PAR of smoking –11.6% –3.6% –7.4% Mortality Current smokers 1.0 (0.9–1.2) 1.2 (1.0–1.5) 1.2 (1.0–1.5) Ex-smokers 1.3 (1.0–1.6) 1.3 (1.0–1.7) 1.3 (1.0–1.7) PAR of smoking 0% 6.5% 6.5%

Reference group is nonsmokers. PAR = population-attributable risk.

Table 2. Incidence of disease and mortality by cigarette consumption and smoking history among elderly current smokers during 7-year follow-up, 1989–1996: relative risk (95% confidence interval)

Lower Heart Digestive Kidney Stroke Diabetes respiratory Mortality disease disease disease tract disease

Daily cigarette consumption ≤ 10 1.0 1.0 1.0 1.0 1.0 1.0 1.0 11–20 0.7 (0.5–1.0) 0.8 (0.5–1.3) 1.0 (0.6–1.6) 1.2 (0.8–1.6) 0.9 (0.6–1.3) 0.6 (0.4–1.0) 1.0 (0.7–1.4) ≥ 21 0.9 (0.5–1.7) 1.0 (0.3–2.8) 1.0 (0.3–2.8) 1.2 (0.6–2.4) 1.4 (0.7–2.7) 0.7 (0.3–2.0) 0.6 (0.3–1.4) Age of starting smoking (yr) < 20 1.0 1.0 1.0 1.0 1.0 1.0 1.0 ≥ 20 0.8 (0.6–1.2) 0.9 (0.5–1.5) 0.8 (0.5–1.3) 0.9 (0.7–1.3) 1.2 (0.8–1.8) 0.9 (0.5–1.4) 0.8 (0.6–1.1) Cumulative smoking years 0–20 1.0 1.0 1.0 1.0 1.0 1.0 1.0 ≥ 21 0.9 (0.4–2.0) 1.0 (0.2–3.9) 0.7 (0.2–2.2) 0.9 (0.4–2.1) 1.3 (0.4–4.0) 0.6 (0.2–1.6) 1.7 (0.5–5.4)

Only those who were smokers according to their 1989 smoking status are included in this table.

Kaohsiung J Med Sci October 2004 ¥ Vol 20 ¥ No 10 487 H.C. Hsu and R.F. Pwu studies have shown [8,18,26]. Ex-smokers had a lower risk smoking cessation is not significant [14–16]. This can of morbidity than current smokers. Smoking cessation possibly be explained as follows. may thus be a protective factor. A previous study did not First, all subjects in our study were aged 60 years and find a dose-response relationship between smoking ex- over. Some elderly might have had diseases attributed to posure and mortality in the elderly [17], and the effect of smoking before they entered their old age, or smoking

Table 3. Incidence of disease and mortality by smoking history among ex-smokers during 7-year follow-up, 1989–1996: relative risk (95% confidence interval)

Lower Heart Digestive Kidney Stroke Diabetes respiratory Mortality disease disease disease tract disease

Cumulative smoking years 1–5 1.0 1.0 1.0 1.0 1.0 1.0 1.0 6–20 1.5 (0.3–6.7) ∞ 0.4 (0.1–1.6) 1.0 (0.3–3.7) 0.6 (0.2–1.7) 1.0 (0.2–4.6) 1.6 (0.2–13.4) ≥ 21 1.1 (0.3–4.7) ∞ 0.6 (0.2–2.0) 1.2 (0.4–3.8) 0.4 (0.2–1.1) 0.6 (0.1–2.6) 4.3 (0.6–31.2) Years since quitting 1–5 1.0 1.0 1.0 1.0 1.0 1.0 1.0 6–20 0.8 (0.4–1.5) 1.5 (0.6–3.5) 1.0 (0.5–2.2) 0.6 (0.3–1.1) 1.2 (0.5–2.6) 0.9 (0.4–2.3) 0.9 (0.4–2.3) ≥ 21 1.1 (0.6–2.0) 2.1 (0.9–4.8) 0.8 (0.3–1.9) 0.9 (0.5–1.6) 1.8 (0.9–3.9) 0.9 (0.4–2.4) 0.9 (0.4–2.4)

Only those who had quit smoking according to their smoking status in 1989 are included in this table. Some cells show “∞” because of non- convergence of parameters in iterations.

Table 4. Incidence of disease and mortality by smoking status and other risk factors among the elderly by Cox multiple regression models, 1989–1996: relative risk (95% confidence interval)

Lower Heart Digestive Kidney Stroke Diabetes respiratory Mortality disease disease disease tract disease

Smoking status Current smoker 1.1 (0.8–1.5) 0.8 (0.5–1.3) 1.2 (0.8–1.8) 1.7 (1.2–2.3) 0.7 (0.5–1.0) 0.8 (0.6–1.2) 1.2 (0.9–1.6) Ex-smoker 1.1 (0.8–1.5) 1.0 (0.6–1.7) 1.4 (0.9–2.3) 1.4 (1.0–2.1) 0.8 (0.6–1.1) 0.7 (0.4–1.1) 1.2 (0.9–1.6) Nonsmoker 1.0 1.0 1.0 1.0 1.0 1.0 1.0 Age (yr) ≥ 75 1.0 (0.7–1.3) 1.0 (0.6–1.7) 0.5 (0.3–0.96) 1.3 (1.0–1.8) 0.9 (0.6–1.3) 0.6 (0.4–1.0) 2.8 (2.3–3.4) 60–74 1.0 1.0 1.0 1.0 1.0 1.0 1.0 Gender Male 0.7 (0.6–1.0) 1.7 (1.1–2.6) 0.6 (0.4–0.96) 1.1 (0.8–1.5) 1.3 (1.0–1.8) 1.0 (0.7–1.4) 1.3 (1.0–1.7) Female 1.0 1.0 1.0 1.0 1.0 1.0 1.0 Education level High 1.0 (0.7–1.2) 0.8 (0.6–1.3) 0.8 (0.6–1.2) 0.5 (0.4–0.7) 1.1 (0.8–1.4) 0.8 (0.5–1.1) 0.8 (0.6–1.0) Low 1.0 1.0 1.0 1.0 1.0 1.0 1.0 Ethnic group Ming-Nan 1.0 (0.8–1.2) 0.7 (0.5–0.98) 0.9 (0.7–1.2) 0.8 (0.7–1.0) 1.2 (1.0–1.5) 1.2 (0.9–1.5) 1.2 (1.0–1.4) Others 1.0 1.0 1.0 1.0 1.0 1.0 1.0 Comorbidity 1.3 (1.2–1.5) 1.4 (1.2–1.6) 1.1 (0.95–1.3) 1.2 (1.1–1.4) 1.2 (1.0–1.3) 1.4 (1.2–1.6) 1.3 (1.2–1.4) Physical function Disabled 1.0 (0.7–1.3) 1.2 (0.7–1.9) 0.8 (0.5–1.3) 1.2 (0.8–1.6) 0.9 (0.6–1.3) 0.9 (0.6–1.4) 1.8 (1.4–2.3) Normal 1.0 1.0 1.0 1.0 1.0 1.0 1.0

Education level: low = illiterate, no formal education, elementary school, junior high school; high = senior high school and over. Ethnic group: others include Hakka, mainland provinces, and others. Incidence of disease models, comorbidity = cumulative prevalent numbers of other 5 diseases in 1989; mortality model, comorbidity = cumulative numbers of 6 diseases in 1989. Smoking status, age, and physical function according to 1989 status.

488 Kaohsiung J Med Sci October 2004 ¥ Vol 20 ¥ No 10 Effect of smoking on morbidity & mortality among the elderly might have shortened their life. These persons with early the three surveys: in 1993 and 1996, physical function was morbidity or immature mortality were lost or not eligible measured using six activities of daily living (ADL) (eating, for our analysis at baseline. Thus, the different results ob- dressing, walking indoors, transferring from bed or chair, tained here may be attributed to a different study design using the toilet, taking a bath), but in 1989, the physical and sample. Our study was based on a national represen- function included only one ADL and three instrumental tative and prospective elderly population, rather than a ADL items. The physical function variable was a more case-controlled one or targeted at specific patients. Using advanced activity. Third, the variables in the surveys for such a longitudinal study design limited the incidence each year were not the same, and some information was and case numbers and may have made the observation of not available. For example, there was no quitting smoking a dose-response relationship difficult. Thus, the risk of and cigarette consumption data in the 1993 survey, so the smoking might have been underestimated. dynamic changes in years since quitting, cumulative smok- Second, smoking behavior is changeable. Patients with ing years, and daily consumption of cigarettes during severe diseases often have stronger motivation to quit these 7 years were unavailable. Fourth, some cases were smoking [27]. The elderly might have sensed their frailty or lost to follow-up or died during the longitudinal study. Al- the presence of disease and quit smoking before the sur- though we compared the demographic and smoking sta- veys. Cessation of smoking might thus be combined with tus of the lost cases and the initial cases and found no dif- other changes in health behavior, which might create a ference, the morbidity and mortality of the lost cases is survival advantage. Further, quitting smoking might re- unknown. Additionally, if smoking shortened the time duce risks immediately (such as after surgery or myocar- from morbidity to death or increased the mortality from dial infarction) for some patients [14,15]. The effect of ces- diseases, or smoking induced cancer during the subject’s sation over the long-term thus cannot be observed. middle age, or individuals were dead before their old Third, the variation in smoking exposure among these age, then the RR of smoking for such a longitudinal study elderly was limited (elderly smokers were mostly long- might be underestimated. term smokers and also low-consumption smokers), so the In this study, we analyzed the RRs of smoking on dose-response relationship might have been masked. morbidity and mortality among the elderly population in Subjects were divided into two ethnic groups, Ming- Taiwan. Smoking has its greatest effect on morbidity from Nan and others, in the multivariate model. Initially, it had respiratory disease in the elderly. The effect of years since been separated into Ming-Nan, Hakka, mainland provinces, cessation was not clear. We suggest that the protective and others, but because of the limited cases in this longi- effect of smoking cessation should be investigated from tudinal study and the many parameters in the multivari- middle age to old age. Further studies should emphasize ate model, we categorized independent variables into fewer the observation of dynamic smoking behavior and be based groups to avoid convergence problems in Cox regression on a life-course perspective because health status and health estimation. We considered the sample numbers in each behavior is changeable. In terms of promotion of successful ethnic group, the lower incidence of morbidity in the Ming- aging, we suggest that health promotion for the elderly Nan group, and because the purpose of using ethnic groups should start at an earlier age and continue throughout old was only for control variables, we conducted the analysis age. with these two groups. Readers should be careful to note that this might mingle the effects of Hakka, mainland province or aboriginal ethnic origin. ACKNOWLEDGMENTS This study is subject to a number of limitations. First, the data came from a face-to-face self-report survey, and the This study was supported by the project “Attributable risks accuracy of disease prevalence and recall bias about smoking of smoking to multiple diseases of the elderly in Taiwan” history should be considered. In addition, the perception of (BHP-92-Anti-Tobacco-E204) and grants from the Bureau hypertension, heart disease, stroke, and so on, might be in- of Health Promotion, Department of Health, Taiwan, ROC. accurate. Second, cancer prevalence was not available in This study is based on data from the “Study of Health and the 1989 survey, and the different kinds of cancers could not Living Status of the Elderly”, provided by the Population be differentiated in the 1993 and 1996 data; therefore, the and Health Research Center, Bureau of Health Promotion. relative risks of cancer over the 7 years could not be estimated. The interpretation and conclusions contained herein do not Measures of physical function were not consistent across represent those of the Bureau of Health Promotion.

Kaohsiung J Med Sci October 2004 ¥ Vol 20 ¥ No 10 489 H.C. Hsu and R.F. Pwu

REFERENCES 14. Rea T, Heckbert SR, Kaplan RC, et al. Smoking status and risk for recurrent coronary events after myocardial infarction. Ann 1. Emmons K. Smoking cessation and : an over- Intern Med 2002;137:494–500. view. Chest 1999;116:490S–2S. 15. Wilson P, D’Agostino RB, Levy D, et al. Prediction of coronary 2. Roget E, Murray JL. Smoking and cause of death among U.S. heart disease using risk factor categories. Circulation 1998;97: veterans: 16 years of observation. Public Health Rep 1980;95: 1837–47. 213–22. 16. Pelkonen M, Notkola IL, Tukiainen H, et al. Smoking cessation, 3. Carstensen JM, Pershagen G, Eklund G. Mortality in rela- decline in pulmonary function and total mortality: a 30 year tion to cigarette and pipe smoking: 16 years observation of follow up study among the Finnish cohorts of the Seven 25,000 Swedish men. J Epidemiol Community Health 1987;41: Countries Study. Thorax 2001;56:703–7. 166–72. 17. Liaw KM, Cheng CJ. Mortality attributable to cigarette smoking 4. Freund KM, Belanger AJ, D’Agostino RB, et al. The health risk in Taiwan: a 12-year follow-up study. Tobacco Control 1998;7: of smoking. The Framingham study: 34 years of follow-up. 141–8. Ann Epidemiol 1993;3:417–24. 18. Lee LT, Chen CJ, Suo J, et al. Pulmonary tuberculosis and lung 5. McLaughlin JK, Hrubez Z, Blot WJ, et al. Smoking and cancer cancer: a case-control study. Formosan J Med 1997;2:176–84. mortality among U.S. veterans: a 26-year follow-up. Intl J 19. Cheng YJ, Hildesheim A, Hsu MM, et al. Cigarette smoking, Cancer 1995;60:190–3. alcohol consumption and risk of nasopharyngeal carcinoma 6. Suminori A, Hirayama T. Cigarette smoking and cancer in Taiwan. Cancer Causes Control 1999;10:201–7. mortality risk in Japanese men and women: results from 20. Lee L, Tsai Y. A Community-Based Intervention Study on Smoking reanalysis of the six-prefecture cohort study data. Environ Cessation. Project of Department of Health, DOH89-TD-1180, Health Perspect 1990;87:19–26. 1999. 7. Yuan JM, Ross RK, Wang XL, et al. Morbidity and mortality in 21. Lee WC, Lin RS, Chen WJ. An Epidemiology Study on Health relation to cigarette smoking in Shanghai, China: a prospective Hazard of Smoking. Project of Department of Health, DOH84- male cohort study. JAMA 1996;275:1646–50. HP-015-3M24, 1995. 8. Wald NJ, Watt HC. Prospective study of effect of switching 22. Tsen CH. The impact of cigarette smoking and smoking ces- from cigarettes to pipes or cigars on mortality from three sation on cardiovascular disease in diabetic patients. Chinese smoking related diseases. BMJ 1997;314:1860–3. J Public Health (Taiwan) 1999;18:241–6. 9. Christen WG, Glynn RJ, Ajani UA, et al. Smoking cessation 23. Yang SC, Yang SP. Bronchial responsiveness and lung function and risk of age-related cataract in men. JAMA 2000;284:713–6. related to cigarette smoking and smoking cessation. Chang 10. He J, Ogden LG, Bazzano LA, et al. Risk factors for congestive Gung Med J 2002;25:645–55. heart failure in US men and women: NHANES I Epidemiologic 24. Pan LY, Lee L. Adult smoking prevalence and its relationships Follow-up Study. Arch Intern Med 2001;161:996–1002. with education and occupation in Taiwan: conditions before 11. Hu SC, Lanese RR. The applicability of the theory of planned the implementation of the Tobacco Hazards Control Act (1993– behavior to the intention to quit smoking across workplaces 1996). Chinese J Public Health (Taiwan) 1999;18:189–98. in southern Taiwan. Addict Behav 1998;23:225–37. 25. Rowe JW, Kahn RL. Successful aging. Gerontologist 1997;37: 12. Ruigrok Y, Buskens E, Rinkel GJE. Attributable risk of common 433–40. and rare determinants of subarachnoid hemorrhage. Stroke 26. Almirall J, Gonzalez C, Balanzo X, et al. Proportion of 2001;32:1173–5. community-acquired pneumonia cases attributable to tobacco 13. Sauer W, Berlin JA, Strom BL, et al. Cigarette yield and the risk smoking. Chest 1999;116:375–9. of myocardial infarction in smokers. Arch Intern Med 2002;162: 27. Bak S, Sindrup SH, Alslev T, et al. Cessation of smoking after 300–6. first-ever stroke: a follow-up study. Stroke 2002;33:2263–9.

490 Kaohsiung J Med Sci October 2004 ¥ Vol 20 ¥ No 10 Effect of smoking on morbidity & mortality among the elderly

– –

Kaohsiung J Med Sci October 2004 ¥ Vol 20 ¥ No 10 491