Combined Effects of Smoking and Systolic Blood Pressure on Risk of Coronary Heart Disease: A Cohort Study in Chinese Women

Abstract

Objectives:

To quantify the combined effects of systolic blood pressure (SBP) and cigarette smoking on incident coronary heart disease (CHD) in women.

Methods:

Overall, 86,338 women aged ≥40 years were enrolled in 1991. The follow-up evaluation was conducted in 1999–2000, with a response rate of 92.9%.

Results:

A total of 829 CHD events (fatal and nonfatal) were observed among the participants who were free of cardiovascular diseases (CVD) at baseline. Higher SBP was significantly associated with more risk of CHD in both nonsmokers and current smokers (all p < 0.0001 for linear trends). Comparing with never smoking, both low and high levels of cigarettes smoked per day (1–7, and ≥8 cigarettes per day) and pack-years (<10, and ≥10 pack-years) were associated with increased risk of CHD in those with normal and high SBP. The multivariate adjusted relative risks (RRs) of CHD were 2.54 (95% confidence interval [CI] 2.00-3.23), 1.28 (1.01-1.63), and 1.57 (1.33-1.86) for current smokers with high SBP, current smokers with normal SBP, and nonsmokers with high SBP, respectively, compared with nonsmokers with normal SBP. The present study identified a statistically significant additive interaction between these two factors on CHD.

Conclusions:

Our study indicated that the combined effects of cigarette smoking and high SBP could be expected to have extra adverse effects on CHD in women, which highlights the importance of multifactorial interventions to decrease the risk of CHD, for example, quitting smoking and treatment of high blood pressure in Chinese women.

Introduction

Both cigarette smoking and high blood pressure (BP) are important modifiable risk factors that play significant roles in the development of coronary heart disease (CHD), which accounts for a striking global burden of premature death in both developing and industrialized regions.^1,2 Data from the International Collaborative Study of Cardiovascular Disease in Asia (InterASIA) indicate that the age-specific prevalence of hypertension was 10.7%, 26.8%, 38.9%, and 50.2% in participants aged 35–44 years, 45–54 years, 55–64 years, and 65–74 years, respectively, and the prevalence of current cigarette smoking was 6.9% in Chinese women aged 35–74 years in the 2000–2001.^3,4 A previous study identified that smoking and increased BP interact to increase markers of cardiovascular risk, for example, intimal-medial thickness of the carotid artery.⁵ Hence, a combination of raised BP and cigarette smoking might have a synergistic impact on CHD incidence. To the extent that smoking and BP interact in their impact on CHD, multifactorial interventions aimed at both risk factors will potentially contribute more to reducing CHD than occurred in former studies where their interaction was not quantified.

Interest in multifactorial interventions for CHD is increasing.⁶ Some studies assessing the combined effect of the BP and cigarette smoking on incident CHD were limited to either western populations⁷ or secondary analyses.⁸ The interaction of systolic blood pressure (SBP) in combination with cigarette smoking on CHD incidence has not been well estimated in Chinese women. We focus on these two factors rather than other combinations because high BP and smoking are the first and second most common risk factors for ischemic cardiovascular disease (CVD) among Chinese adults. Given the paucity of data and public health implications of determining this interaction on the risk of CHD incidence, we investigated the combined relationship of the two risk factors and the magnitude of the interaction of cigarette smoking and SBP on CHD incidence in a large population-based prospective cohort of Chinese women.

Materials and Methods

Study population

Study subjects were all female participants of the China National Hypertension Survey,⁹ a multistage random cluster sampling design, except for those whose contact information was not available. From this population, 86,338 women aged ≥40 years at their baseline examination were eligible to participate in the study, and a total of 80,166 (92.9%) study participants (or their proxies) were identified and interviewed as part of the follow-up study. In this report, study participants missing information on cigarette smoking status (n = 7,254) and BP (n = 116) were excluded from all analyses, as were those with prevalent CHD and stroke (n = 1,942) at baseline. Participants included in the final analysis were not different from the overall female study population in 1991 in regard to baseline characteristics. We examined the baseline characteristics of age, body weight and height, BP, high school education, alcohol consumption, physical inactivity, CVD, and inhabitation (north vs. south, and urban vs. rural).

Baseline examination

Baseline data were collected at a single clinic visit by specially trained physicians and nurses using standardized methods, with stringent levels of quality control.⁹ Data on demographic characteristics, medical history, and lifestyle risk factors were obtained using a standard questionnaire administered by trained staff. The smoking status in all analyses was assessed according to the smoking history at baseline examination. A current smoker was defined as one who smoked at least 1 cigarette per day for 1 year or more at baseline examination, and a former smoker was defined as one who did not smoke for 1 year at the baseline examination, with the others being nonsmokers. For participants who reported past or current cigarette smoking, information on the number of cigarettes smoked per day along with the duration of cigarette smoking was also collected. Alcohol consumption was defined as drinking alcohol at least 12 times during the past year. High school education was defined as high school education or higher (≥12 years of education received). Work-related physical activity was assessed because leisure time physical activity was uncommon in 1991 among the Chinese population. Body mass index (BMI) was calculated as weight in kilograms divided by height in square meters. Three BP measurements were taken after the study participant had been seated quietly for 5 minutes using a standard mercury sphygmomanometer according to a standard protocol.¹⁰ The mean of three SBP measures was used in all analyses.

Follow-up data collection

The follow-up investigation was carried out between 1999 and 2000. The method is described in detail in other publications.^11,12 If a study participant reported a hospitalization or emergency room overnight stay attributable to CHD during the in-person interview, this participant's hospital records, including medical history, physical examination findings, laboratory test results, electrocardiogram and coronary angiography results, and discharge diagnosis or autopsy reports, were abstracted by trained staff using a standard form. All deaths reported by their proxies of the participants during the in-person interview were verified by obtaining death certificates from the local public health department or police department. If death occurred during a hospitalization, the participant's hospital records and autopsy results were abstracted by trained staff using a standard form. If death occurred outside of the hospital, detailed information on medical history was obtained from a family member or healthcare provider.

Methods for the end point assessment have been described in full.^11,12 Causes of death were coded according to the International Classification of Diseases, Ninth Revision (ICD-9). For this analysis, CHD incidence was defined as a confirmed diagnosis of CHD during the follow-up period or CHD listed as an underlying cause of death (ICD-9: 410.0–414.9) among those without a history of CHD or stroke.

This study was approved by the Cardiovascular Institute and FuWai Hospital Ethics Committee and the Tulane University Health Sciences Center Institutional Review Board. Written informed consent was obtained from all study participants at their follow-up visit.

Statistical methods

Baseline characteristics were compared between former and current smokers and nonsmokers using χ² tests for categorical variables and ANOVA for continuous variables. Person-years of follow-up were calculated from the date of baseline examination until the date of CHD, death, or follow-up interview for each participant. The age-standardized rate of CHD was calculated using the 5-year age-specific incidence and the age distribution of the Chinese population from year 2000 census data for providing a comparable result in Chinese women with that be found in other population.

Cox proportional hazards models were used to estimate the association between the risk factor (SBP or smoking) and the risk of CHD, adjusted for baseline age, education, alcohol consumption, physical inactivity, BMI, geographic region (north vs. south), urbanization (rural vs. urban), and time-dependent history of diabetes. The associations between SBP and CHD incidence were explored by stratified analyses in which participants were classified into four groups according to levels of baseline SBP (<120, 120–139, 140–159, and ≥160 mm Hg) in both nonsmokers and current smokers. These cutoff points were chosen on the basis of BP classification according to the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC7).¹³ Relative risks (RRs) and 95% confidence intervals (CIs) for a 10 mm Hg increment in the level of SBP were also estimated across smoking status. Dose-response relationships for current smokers were investigated using nonsmokers as the reference category compared with low and high levels of daily cigarettes smoked (1–7 and ≥8 cigarettes per day), as well as of the pack-years (<10 and ≥10 pack-years) in both high (≥140 mm Hg) and normal SBP study participants.

To evaluate additive interaction between high SBP and smoking status, dichotomizing SBP at 140 mm Hg and smoking status as nonsmokers and current smokers, we used relative excess risk due to interaction (RERI), attributable proportion due to interaction (AP), and synergy index (SI).¹⁴ These measures are defined as follows: \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland,xspace}\usepackage{amsmath,amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6} \begin{document} \begin{align*}{\rm RERI} = {\rm RR_{{AB}}} - {\rm RR_{A}} - {\rm RR_{B}} + 1 \end{align*} \end{document} \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland,xspace}\usepackage{amsmath,amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6} \begin{document} \begin{align*}{\rm AP} = {\rm RERI}/{\rm RR_{AB}} \end{align*} \end{document} \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland,xspace}\usepackage{amsmath,amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6} \begin{document} \begin{align*}{\rm SI} = [{\rm RR_{AB}} - 1]/[({\rm RR_{A}} - 1) + ({\rm RR_{B}} - 1)] \end{align*} \end{document}

Where RR_A and RR_B are the adjusted RRs associated with risk factors A and B alone, and RR_AB is the RR for those exposed to both risk factors. No interaction means that RERI and AP are equal to 0 and SI is equal to 1. Both the point estimation and the 95% CI of the RERI, AP, and SI were assessed using a method accounting for the asymmetric distribution of confidence limits for risk ratio.¹⁵ We ignored the former smokers in the stratified analyses because of the limited number of CHD events (13 CHD events, including 6 fatal CHD) in the present study. Methods to estimate variances that take into account sample clustering were used in Cox proportional hazards models.¹⁶ Statistical analyses were conducted using SAS statistical software (version 9.0, SAS Institute, Cary, NC).

Results

Baseline characteristics of study population

Baseline characteristics of study participants according to their smoking status are shown in Table 1. There were 68,198 (86.3%) nonsmokers, 568 (0.7%) former smokers, and 10,231 (13.0%) current smokers in the present study. Current and former smokers tended to be older, to have less education, to be less physically active, and to have higher alcohol consumption and higher SBP compared with nonsmokers.

Table 1.

Baseline Characteristics by Smoking Status

Characteristic	Nonsmokers	Current smokers	Former smokers
No. of participants	68,198	10,231	568
Age, mean (SD)	55.7 (11.0)	59.3 (9.5)^*	64.2 (9.9)^*
≥High school education, %	18.5	9.7^*	5.3^*
Alcohol consumption, %	1.8	9.2^*	7.6^*
Physical inactivity, %	34.2	38.9^*	43.7^*
BMI, mean (SD)	22.8 (3.9)	22.4 (4.2)^*	22.9 (4.7)
SBP, mean (SD)	126.7 (23.9)	129.6 (24.4)^*	136.1 (28.0)^*
DBP, mean (SD)	76.5 (12.2)	77.1 (12.7)^*	76.7 (14.2)
Cardiovascular disease, %	2.44	4.41^*	6.51^*
North, %	62.3	82.7^*	60.0
Urban, %	58.2	63.2^*	66.6^*

p < 0.05 for chi-square test or ANOVA compared with nonsmokers.

SBP, systolic blood pressure; DBP, diastolic blood pressure.

Rates of CHD incidence across smoking status and SBP

During an average of 8.3 years of follow-up, we recorded a total of 829 CHD events, of which 487 were fatal CHD. The age-standardized rates for total CHD were 123.0 and 186.5 per 100,000 person-years in nonsmokers and current smokers with normal SBP, respectively, with 222.8 and 411.8 per 100,000 person-years in nonsmokers and current smokers with high SBP, respectively. After adjustment for baseline age, education, alcohol consumption, BMI, physical inactivity, history of diabetes, urban or rural residence, and northern or southern China, current cigarette smoking remained a significant predictor of CHD incidence, with the RRs being 1.34 (95% CI 1.05-1.72) and 1.52 (95% CI 1.19-1.94) compared with nonsmokers in both normal and high SBP participants, respectively (Table 2).

Table 2.

Age-Standardized Rates and Relative Risks of Coronary Heart Disease Incidence and Fatal Coronary Heart Disease by Smoking Status and Dichotomizing Systolic Blood Pressure at 140 mm Hg

	Normal SBP (<140 mm Hg)		High SBP (≥140 mm Hg)
	Nonsmokers	Current smokers	Nonsmokers	Current smokers
CHD incidence
Person-years of follow-up	374,332.5	51,327.8	107,287.9	18,163.4
No. of events	333	96	293	94
Age-standardized rate^a	123.0	186.5	222.8	411.8
Age-adjusted RR	1.00 (ref)	1.35 (1.06-1.71)	1.00 (ref)	1.50 (1.18-1.90)
Multivariate adjusted RR1^b,c	1.00 (ref)	1.34 (1.05-1.72)	1.00 (ref)	1.52 (1.19-1.94)
Multivariate adjusted RR2^b,d	1.00 (ref)	1.28 (1.01-1.63)	1.57 (1.33-1.86)	2.54 (2.00-3.23)
Fatal CHD
Person-years of follow-up	374,914.2	51,490.9	107,679.2	18,277.1
No. of events	171	51	194	65
Age-standardized rate^a	78.6	115.0	129.4	257.2
Age-adjusted RR	1.00 (ref)	1.28 (0.93-1.77)	1.00 (ref)	1.61 (1.20-2.14)
Multivariate adjusted RR1^b,c	1.00 (ref)	1.26 (0.91-1.76)	1.00 (ref)	1.59 (1.18-2.14)
Multivariate adjusted RR2^b,d	1.00 (ref)	1.19 (0.86-1.64)	1.63 (1.31-2.03)	2.77 (2.06-3.72)

Age-standardized rate per 100,000 person-years.

The RRs were adjusted for baseline age, education, alcohol consumption, BMI, physical inactivity, history of diabetes, urban or rural residence, and northern or southern China.

RR1, The reference categories were nonsmokers with normal SBP and nonsmokers with high SBP in subset analyses.

RR2, The reference category was nonsmokers with normal SBP, compared with the other 3 categories, and the RERI, AP and SI can be evaluated according to the RR2.

RR, relative risks; CHD, coronary heart disease.

Risk for SBP and smoking in stratified analysis

The RRs of incident CHD for SBP and cigarette smoking in stratified analyses are shown in Tables 3 and 4. Higher level of SBP was associated with greater risk of CHD incidence, irrespective of smoking status (Table 3). For instance, the multivariate adjusted RRs were 1.37 (95% CI 1.09-1.72), 1.60 (95% CI 1.24-2.06), and 2.31 (95% CI 1.79-2.98) for those with SBP of 120–139, 140–159, and ≥160 mm Hg compared with those <120 mm Hg, respectively, and a 10 mm Hg increment of SBP was associated with 1.11 (95% CI 1.07-1.14) in nonsmokers. The corresponding figures were 1.17 (95% CI 0.77-1.77), 1.91 (95% CI 1.23-2.96), 2.37 (95% CI 1.50-3.75), and 1.14 (95% CI 1.08-1.20) in current smokers. There were significant linear associations between the level of SBP and both total and fatal CHD incidence in nonsmokers and current smokers (all p < 0.0001 for linear trends).

Table 3.

Relative Risks of Coronary Heart Disease Incidence and Fatal Coronary Heart Disease by Different Categories of Systolic Blood Pressure in Nonsmokers and Current Smokers

	<120 mm Hg	120–139 mm Hg	140–159 mm Hg	≥160 mm Hg	p values for trends	Increment of 10 mm Hg
CHD
Nonsmokers	1.00 (ref)	1.37 (1.09-1.72)	1.60 (1.24-2.06)	2.31 (1.79-2.98)	<0.0001	1.11 (1.07-1.14)
Current smokers	1.00 (ref)	1.17 (0.77-1.77)	1.91 (1.23-2.96)	2.37 (1.50-3.75)	<0.0001	1.14 (1.08-1.20)
Total	1.00 (ref)	1.33 (1.09-1.62)	1.69 (1.36-2.11)	2.35 (1.88-2.94)	<0.0001	1.12 (1.09-1.15)
Fatal CHD
Nonsmokers	1.00 (ref)	1.43 (1.03-1.97)	1.76 (1.25-2.48)	2.47 (1.76-3.48)	<0.0001	1.12 (1.08-1.17)
Current smokers	1.00 (ref)	1.34 (0.75-2.40)	2.44 (1.35-4.41)	3.03 (1.64-5.57)	<0.0001	1.18 (1.11-1.26)
Total	1.00 (ref)	1.41 (1.07-1.87)	1.92 (1.43-2.59)	2.61 (1.94-3.52)	<0.0001	1.14 (1.10-1.17)

The relative risks were adjusted for baseline age, education, alcohol consumption, BMI, physical inactivity, history of diabetes, urban or rural residence, and northern or southern China.

Table 4 shows RRs of current smoking compared with never smoking according to both the number of cigarettes smoked per day and the pack-years among those participants with high and normal SBP. Both low and high levels of cigarettes smoked per day were associated with an increased risk of CHD in current smokers with high SBP (≥140 mm Hg). The association between both the cigarettes smoked per day and the pack-years with the total CHD incidence is shown in a dose-response fashion (all p < 0.05 for linear trends).

Table 4.

Relative Risks of Coronary Heart Disease Incidence in Current Smokers According to Number of Cigarettes Smoked per Day and Pack-Years

	Number of cigarettes smoked per day				Pack-years
	0	1–7	≥8	p values for trends	0	<10	≥10	p values for trends
CHD
Normal SBP	1.00 (ref)	1.30 (0.94-1.80)	1.39 (1.01-1.90)	0.0204	1.00 (ref)	1.27 (0.89-1.80)	1.40 (1.04-1.89)	0.0169
High SBP	1.00 (ref)	1.41 (1.01-1.97)	1.63 (1.19-2.22)	0.0007	1.00 (ref)	1.27 (0.87-1.85)	1.69 (1.27-2.26)	0.0003
Total	1.00 (ref)	1.35 (1.07-1.70)	1.50 (1.20-1.87)	<0.0001	1.00 (ref)	1.27 (0.98-1.64)	1.53 (1.25-1.89)	<0.0001
Fatal CHD
Normal SBP	1.00 (ref)	1.27 (0.83-1.94)	1.25 (0.80-1.96)	0.2055	1.00 (ref)	1.13 (0.69-1.86)	1.36 (0.91-2.03)	0.1326
High SBP	1.00 (ref)	1.61 (1.09-2.37)	1.58 (1.07-2.33)	0.0043	1.00 (ref)	1.60 (1.04-2.47)	1.59 (1.11-2.28)	0.0036
Total	1.00 (ref)	1.44 (1.08-1.92)	1.41 (1.05-1.89)	0.0038	1.00 (ref)	1.37 (0.99-1.90)	1.46 (1.12-1.91)	0.0021

The relative risks were adjusted for baseline age, education, alcohol consumption, BMI, physical inactivity, history of diabetes, urban or rural residence, and northern or southern China.

Combined effects and interaction of SBP and smoking on CHD

The combined effects of current smoking and high SBP on total and fatal CHD incidence are shown in Table 2. The multivariate adjusted RRs of total CHD were 2.54 (95% CI 2.00-3.23), 1.28 (95% CI 1.01-1.63), and 1.57 (95% CI 1.33-1.86) for current smokers with high SBP, current smokers with normal SBP, and nonsmokers with high SBP, respectively, with the reference category of nonsmokers with normal SBP. There was evidence of a significant interaction effect between high SBP and current smoking for the risk of CHD after adjusting for other risk factors in the present study. The RERI, AP, and SI were 0.69 (95% CI 0.10-1.34), 27% (95% CI 3%-43%), and 1.80 (95% CI 1.06-3.06), respectively, which meant there would be 0.69 relative excess risk due to the additive interaction, 27% of total CHD exposed to both risk factors was attributable to the additive interaction, and the risk of total CHD incidence in those who had high SBP (≥140 mm Hg) and currently smoked was 1.80 times as high as the sum of risks in the participants exposed to a single risk factor alone. The RERI, AP, and SI were 0.95 (95% CI 0.19-1.83), 34% (95% CI 6%-52%), and 2.15 (95% CI 1.09-4.26) for the additive interaction of current smoking and high SBP on fatal CHD, respectively.

We conducted a sensitivity analysis for the interaction by dichotomizing SBP at 130 and 150 mm Hg on total CHD incidence. A statistically significant interaction was shown for 150 mm Hg but not for 130 mm Hg. The RERI, AP, and SI were 0.53 (95% CI −0.01-1.08), 23% (95% CI −2%-41%), and 1.71 (95% CI 0.94-3.12) when the cutoff point was 130 mm Hg for high SBP, with 0.88 (95% CI 0.21-1.68), 32% (95% CI 7%-48%) and 2.03 (95% CI 1.17-3.52) for 150 mm Hg. We also evaluated the interaction of high diastolic BP (DBP) and current smoking on CHD incidence but found no significant interaction irrespective of dichotomizing DBP at 80, 85, 90 mm Hg or the higher cutoff points (data not shown).

Discussion

Both cigarette smoking and elevated BP are important modifiable risk factors for CHD that account for a dramatic global and regional burden of premature death, suggesting the need for epidemiological research on how smoking interacts with other risks.^1,2 Only a few studies focus on CHD risk factors in women.^17
–19 Data on CHD were substantially scarce in Chinese women, in whom the burden of CHD has been increasing, although its incidence was lower than the levels in western female populations.²⁰ One report suggests that high BP in combination with cigarette smoking may promote atherosclerosis, a prevailing cause of occlusion of the coronary arteries and CHD.⁵ The present study documented evidence that a higher level of SBP is associated with the risk of CHD incidence, irrespective of smoking status, with a significant and linear association between the level of SBP and CHD incidence among both nonsmokers and current smokers. Compared with never smoking, both low and high levels of cigarettes smoked per day (1–7 and ≥8 cigarettes per day) and pack-years (<10 and ≥10 pack-years) were associated with increased risk of fatal CHD in the participants with high SBP. We evaluated the combined effects of high SBP (≥140 mm Hg) and current smoking as well as the magnitude of the potential interaction on CHD incidence in this large, nationally representative sample of the Chinese female population. There was a significant additive interaction between current smoking and high SBP, and the risk of total CHD incidence for current smokers with high SBP was 1.80 times as high as what could be expected from the sum of risks in participants with high SBP or current smoking alone. The interaction on fatal CHD also tended to be higher than expected from risk addition. The present results illustrate the need for a global risk factor assessment to identify individuals who need intensified risk factor management, including a targeted program to stop smoking and to treat hypertension in Chinese women.

A positive interaction between cigarette smoking and hypertension on CHD incidence has been established in western female populations,⁷ but a contradictory finding has been reported in an Asian-Pacific region population study.⁸ Our study found a significantly increased risk of CHD incidence associated with the interaction of current smoking with high SBP in Chinese women, consistent with what was evaluated for the additive interaction.⁷ The study from the Asia-Pacific region did not identify an interaction of SBP and smoking on the risk for CHD using a model of multiplicative interaction,⁸ which is different from the present analysis. It has been suggested that the interaction term in the Cox regression model, which was exponential and inherently multiplicative, had no direct relevance for the issue of whether or not biological interaction exists, and the degree of biological interaction between risk factors should be measured as deviation from additivity using the corresponding disease rates and not, for example, as a deviation from multiplicativity.^14,21

A statistically significant interaction between hypertension and smoking for the risk of cardiac events has been identified in the western female population, where hypertension was defined as an SBP≥160 mm Hg or a DBP ≥95 mm Hg or pharmacological treatment of hypertension.⁷ To distinguish the combined effect of DBP and current smoking on CHD, we also investigated the interaction. The fact that no such interaction between high DBP and current smoking was found to be statistically significant in the present analysis suggested that SBP is stronger than DBP as a predictor of CHD risk in the relatively old female population, consistent with the previous study in a western population.²² The present study found that the effect of the interaction was stronger on fatal CHD than on CHD incidence. One reason might be the fact that the mean SBP level was higher among those with fatal CHD compared with the SBP level among those who suffered from CHD incidence (145.0 mm Hg vs. 141.0 mm Hg). Another reason was that women with fatal CHD smoked for a longer time than did those with CHD incidence (42.0 years vs. 39.0 years).

Our study has certain limitations. Data on serum lipids, diet, and leisure time physical activity were not obtained. Lack of adjustment for these variables may have caused a slight overestimate of the RR of CHD. However, our study also has several important strengths. It is a large prospective cohort study examining the interaction between cigarette smoking and high SBP on the risk of CHD incidence in a nationally representative sample of Chinese women. Information on baseline cigarette smoking, SBP, and CHD events was collected using stringent quality control procedures, and a very high follow-up rate was attained.

Conclusions

These results suggest that both cigarette smoking and high SBP are important risk factors for incident CHD in Chinese women. Furthermore, this study provided epidemiological evidence that there was an additive interaction between current smoking and high SBP on CHD incidence, which implied lowering BP and quitting cigarette smoking would contribute more to reducing CHD incidence than the effect of each change alone. Our study highlighted the importance of multifactorial interventions to reduce the risk of incident CHD.

Footnotes

Acknowledgments

This study was supported by a grant (2006BAI01A01) from the Ministry of Science and Technology, Beijing, China, and by a national Grant-in-Aid (9750612N) from the American Heart Association, Dallas, Texas, and partially by a grant (1999-272) from the Chinese Ministry of Health, Beijing, China, and by the Chinese Academy of Medical Sciences, Beijing, China.

Disclosure Statement

The authors have no conflicts of interest to report.

References

Ezzati

, Henley

, Thun

, Lopez

. Role of smoking in global and regional cardiovascular mortality. Circulation, 2005; 112:489–497.

Lawes

, Vander Hoorn

, Rodgers

. Global burden of blood pressure-related disease, 2001. Lancet, 2008; 371:1513–1518.

, Reynolds

, Wu

et al. Prevalence, awareness, treatment, and control of hypertension in China. Hypertension, 2002; 40:920–927.

, Wu

, Reynolds

et al. Cigarette smoking and exposure to environmental tobacco smoke in China: The International Collaborative Study of Cardiovascular Disease in Asia. Am J Public Health, 2004; 94:1972–1976.

Howard

, Wagenknecht

, Burke

et al. Cigarette smoking and progression of atherosclerosis: The Atherosclerosis Risk in Communities (ARIC) Study. JAMA, 1998; 279:119–124.

Ezzati

, Hoorn

, Rodgers

, Lopez

, Mathers

, Murray

. Estimates of global and regional potential health gains from reducing multiple major risk factors. Lancet, 2003; 362:271–280.

Janzon

, Hedblad

, Berglund

, Engstrom

. Tobacco and myocardial infarction in middle-aged women: A study of factors modifying the risk. J Intern Med, 2004; 256:111–118.

Nakamura

, Barzi

, Lam

et al. Cigarette smoking, systolic blood pressure, and cardiovascular diseases in the Asia-Pacific region. Stroke, 2008; 39:1694–1702.

, Duan

, Gu

, Hao

, Tao

, Fan

. Prevalence of hypertension and its trends in Chinese populations. Int J Cardiol, 1995; 52:39–44.

10.

Frohlich

. Recommendations for blood pressure determination by sphygmomanometry. Ann Intern Med, 1988; 109:612.

11.

, Gu

, Wu

et al. Major causes of death among men and women in China. N Engl J Med, 2005; 353:1124–1134.

12.

, He

, Duan

et al. Body weight and mortality among men and women in China. JAMA, 2006; 295:776–783.

13.

Chobanian

, Bakris

, Black

et al. The seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: The JNC 7 report. JAMA, 2003; 289:2560–2572.

14.

Andersson

, Alfredsson

, Kallberg

, Zdravkovic

, Ahlbom

. Calculating measures of biological interaction. Eur J Epidemiol, 2005; 20:575–579.

15.

Zou

. On the estimation of additive interaction by use of the four-by-two table and beyond. Am J Epidemiol, 2008; 168:212–224.

16.

Ingram

, Makuc

. Statistical issues in analyzing the NHANES I Epidemiologic Follow-up Study. Series 2: Data evaluation and methods research. Vital Health Stat, 2. 1994; 1–30.

17.

Mosca

, Manson

, Sutherland

, Langer

, Manolio

, Barrett-Connor

. Cardiovascular disease in women: A statement for healthcare professionals from the American Heart Association Writing Group. Circulation, 1997; 96:2468–2482.

18.

Al-Delaimy

, Manson

, Solomon

et al. Smoking and risk of coronary heart disease among women with type 2 diabetes mellitus. Arch Intern Med, 2002; 162:273–279.

19.

, Stampfer

, Manson

et al. Trends in the incidence of coronary heart disease and changes in diet and lifestyle in women. N Engl J Med, 2000; 343:530–537.

20.

Zhang

, Lu

, Liu

. Coronary heart disease in China. Heart, 2008; 94:1126–1131.

21.

Ahlbom

, Alfredsson

. Interaction: A word with two meanings creates confusion. Eur J Epidemiol, 2005; 20:563–564.

22.

Franklin

, Larson

, Khan

et al. Does the relation of blood pressure to coronary heart disease risk change with aging? The Framingham Heart Study. Circulation, 2001; 103:1245–1249.