Tobacco Documents Online

,J. L4\~ v Av, a~- ~2 + CHATTER 6: PASSIVE SMOKING 6.1 Introduction. Ak*~4~ -17~ c-1 89 /I keKA~(dyw Passive smoking is increasingly recog-nised as an important public health hazard (We11s, 1988). In addition to the long -established adverse effects.on respiratory diseases especially in children, the association between lung cancer and passive smoking now appears to satisfy epidemiological criteria: for causality wi~th 10' out of 13 published' studies from 6 countries showing a positive association (Repace Lowrey, 1985)~. an& More recently the evildence has begun to accumulate implicating passive smoking as a causal factor in CHD. There are now at least 5 cohort studies and 2 case-control studies which Weils (1988) has summarized. Using pooled relative risks Wells has estimated that the risk of CHD mortality due to home exposure in never smokiing men is 1.3 (95% CI: 1.1 - 1.6) and in women 1.2' (1.1 - 1.4)1. Apparently only one published study to date (Svendsen et al, 19'87), has separately examined the effect of work exposure to passive smokiing on1CHD risk and only in men; this study shows a greater effect on CHD risk of work than home exposure. In this chapter CHD risk due to passive smoking both at home and work ils examined in men and women. Data on passive smoking was only collected in the second year of the Auckland Heart Study when reports on C2-ID and passive smoking,in the medical literature stimulated the investigators to examine this issue, therefore the number of peoplie included in these

158 analyses are considerably less than are examined in ocher chapters. The analyses have also been limited to people with no history of MI or angina because any effect of passive smoking is likel'y to be swamped~ by the risks associated with symptomatic CHD. All risk calculations use never smokers as the reference group; both current and ex-smokers excluded-from these calculations., Whereas all examined home exposure to spousal smoking, only, are previousT studies have thi~s study includes exposure to smoking from any cohabitant. 6.2 Results. 6.2'.1 Passive smoking status in cases and~controls. Table 6.1 shows the numbers and proportions of people exposed to passive smoking at home only, at both home and work and at work only in the case-control categories. Also included in this Table are the number of ex and current smokers identified during the period that passive smoking data was colliecte&. As noted above„ people with a past history of CHD and angina are excluded from this Table and subsequent analyses. These data show that among never smokers in the Auckland population aged 35-64 years approximately 7% of men are exposed to passive smoking at home only, 5% at work and home, and 26% at work only. The corresponding figures for women are 13%, 4% and 19% respectively, indicating that women are more likely to be exposed at home than men whereas for work exposure the situation is reversed (these percentages are not shown in

Table 6. 1. Passive sm,_.ny exposure by sex ari<.l case-contro: aCec3ory. MIC I CASE-CONTROL CATEGORY MI CDC I CD Number' t INumberl t INumber+ k tNumber, h SEX IPASSIVE SMOKING MEN home only 9 3 . 5 3 1 1 --- home+work 6 2 1 4 1 3 4 1 3 -H- 1 2 1 2 work only ~ 32 1 10~- 8 ~ 6 ~ 18 ~ 12~ 71 9 no exposure 1 76 1 24 1 16 1 13 1 34 1 22 11 1 13 current smoker ~ 63 1 20 t 58 1 47 1 30 1 19 t 40 1 49 ex-smoker 1 132 1 42 1 38 1 31 ~ 64~ 41 ~ 211 26 Total fi 3181 1001 124 1 100 1 155 1 100 f 82 1 100 WOMEN PASSIVE SMOKING home only 15 7 3 B 2 2 3 14 home+wor-k 1 4 1 2 t 1 1 3+ -_ 4 1 4 1 . I work only 1 21 1 10 1 2 1 51 9 1 9 1 3 1 14 no exposure ~ 72 1 35 ~ 5 ~ 13~ 471 45 ~ 3 ~ 14 current smoker 1 50 1 24 [ 23 1 591 251 241 8f 36 ex-smoker ~ 451 22 [ 51 13[ 171 161 51 23 Total 1 2071 100 1 39 1 100 F1041 1001 22 1 100 Total 1 5251 1001 1631 1001 2591 1001 1'04fi 100 ~- ~~ w . / A ~ ~- w-•Lc..~c_ 4-ucC4~~ . /! r~ (n n,l)) F.~,. (;r r. ..N 1 p ' ~~) ^ .~..M1 M 1 , r r . . \ I r r r sozEvvszo%

i60 the Table but are calculated using never smokers as the denominator). No majjor case-control differences are obvious in thils Table however, it can be seen ttiat the number of subjects in each category are relatively smal!l and! therefore the precision of relative risk estimates in subsequent analyses is poor. © Table 6:.2 shows the average number of cigarettes smoke& each day in the presence of never smoking study participants by cohabitants in each of the case-control categories. This is likely to be an~underestimate of the passive smoking exposure at home because, for logistic reasons information was only collected for the cohabiitant who smoked the most cigarettes. However,, only 1'68 of participants reported that more than, one cohabitant smoked regularly in their presence. Cases were exposed'on average to 1.5 times as many cigarettes as controls. In addition women were exposed to more cigarettes than men in three of the four case-control categories. No data was collected on, the number of cigarettes participants were exposed to at work. 6.2.2 Coronary heart disease risk associiated with passive smoking. Table 6.3 shows the crude relative risks of CHD associated with passive smoking exposure at home and at work; the categories used are exclusive (i.e.„ home exposure only, work only,, or both~ home and work). The reference categ©ry was the unexposed group. Women exposed at home only were at greater risk of CHD than those exposed at work whereas for men the opposite appeared to occur. For men and women those exposed both at

161 Table 6.2 Average daily number of cigarettes ehat never smoking people are exposed to at tiome, by case-control category. CASE-CONTROL CATEGORY MEN ~;OMENI d MIC ~ 5. 9 8 "3 :11 9.9 11.3 CDC 8.4 6.5 CD~ 6.6 11.5

162' Table 6.3 Relative risk* of CHD associated with passive smoking, by exclusive exposure categories ME1*I WOMEN EXPOSURE MI/MIC CD/CDC MI/MIC CD/CDC S home only - 0.6 --2.4 ? ~ 23.5 ~_ (&.96 -` 5.9) (0.4 - 14.1) (,2.8 - 199.0) . ~ work only r 5 1 Z 1.2 " 0.9 5.2 (0.6 - 3.8) (0.4 - 3.6) (0.09 - 8.0) (0.9 - 30.1) home and work 2~? 3~ 1!.6 (&.6 - 12.1) (0.3:- 9.6) (0.4 - 4,9.5) * Crude relative risks and test-based 95% confidence intervals.

163 work and at home had the highest risk. Civen the small numbers, the confidence intervals were wide. These same data are recategorised in Table 6.4 into two groups; home exposure and' work exposure with no account taken of whether those exposed at home are also exposed at work and' vice versa..,, Most other studies have presented home exposure in this form because work exposure has not been. examine&. This Table shows exposure associated with a higher CHD risk than work exposure in the same pattern with home women and vice versa for men. It also: shows a possible small adverse effect of home exposure in men whichl was not seen when home exposure only was ex.ami~ned. Table 6.4 also shows relative risk estimates adjusted for age in 5 year groups and social class in three categories by logistic regression. Although there are some small di~fferences between the adjusted and unadjusted estimates the pattern described'above remains. As controlling for other risk factors did not affect the CHD risk of cigarette smoking described in the previous chapter were adjusted for i~n these analyses. no other factors Because of the small numbers included in this study and given that the relative risk estimates were similar for MI and CD within each exposure and sex category, the above data is shown i~n Table 6.5 for the combined MI and CD categories (it is debatable whether the MIC and CDC groups can be combined because they include the same people but two separate data sources). Again the same pattern emerges and overall these data are

164 Table 6.4 Fzlative risk of CHD associated with passive smoking by combined exposure categories. MEN WOM EN' EXPOSURE MI/t'fIC CD/CDC MI/MIC CD/CDC - ° Home + ~- ork 1.1* 1.0 2.8 - 7.8 (0.3' - 4.3) : (0.2 - 4.5) (0.6 - 13. 6) (1.3 - 48.0) 1.03+ 1.1 2.7 5.8 Work + irocne 1.7 1.3 1.4 3.6 (0~.7 - 4.0) (0.5 - 3.6) (0.3- 8.2 ) (0.7 - 20.1) 1.1 2'.2' Crude relative risks and test-based 95% confidence intervals. + Relative risks estimated using logistic regression adjusted for age and social class.

165 Table 6.5 Relative risk* of C1iD associated with~passive smoking by combined exposure categories and combined disease categories.. EXP05URE MEN WOMEN Home + work 1.1 4.0 .. ~- (0.4 - 3.0) (1.35 - 13.1) Work + home 1.5 2,,2 (0.8'- 3.0) (0.7 - 7.4) * Crude relative risks and'test-based 95% confidence intervalis.

166 consi~stent with the hypothesis that passive smokirng increases tYie risk of fatal and'nonfatal CHD in, never smokers except for home exposure men. 6.2'.3 Coronary, heart disease risk associated'with degree of passive -cmoking exposure. 4~1 in To examine the possibility of a dose response relationship, relative risks were calculated for two home exposure groups. High exposure was defined as more than one cohabitant smoking or exposure to more than 7 cigarettes per day at home. The remaining people were classified as exposure. This classification was low, arbitrarily chosen to ensure similar numbers in each category. Table 6.6 shows the crtide relative risks of fatal and nonfatal CHD combined related to high and low exposure by sex. In women there appeared to be a substantial increase in risk in the high exposure group although because of the small numbers the confidence intervals were wide. In men there was little difference in risk between the two exposure categories; given the lack of any overall effect on CHD risk of home exposure to passive smoking in men this was perhaps not surprising. 6.3 Discussion. The findings of this study are consistent with the accumulating evidence that exposure to passive smoking increases the risk of CHD. This is the .first study to directly examine passive smoking in New Zealand and one of the first world-wide to examine the relationship between work