Mortality Attributable to Tobacco Use in Canada and its Regions, 1994 and 1996

Eva M Makomaski Illing and Murray J Kaiserman

Abstract

Using data from the National Population Health Survey and the Canadian Mortality Database, we applied the Smoking-Attributable Mortality, Morbidity and Economic Cost method to estimate national and regional smoking-attributable mortality for 1994 and 1996. The results indicate that 29,229 men and 15,986 women died in 1996 as a result of smoking, including 105 children under the age of 1. This total of approximately 45,200 deaths represents an increase of 3,807 deaths since 1991, of which 2,445 occurred in women. The increase in female mortality is almost entirely due to adult diseases, divided between cancers (1,026), cardiovascular diseases (743) and respiratory diseases (870).

Key words: Canada; mortality; smoking; tobacco

Introduction

In the past, Health Canada has calculated smoking-attributable mortality (SAM) for the survey years of 1985,¹ 1989^2,3 and 1991⁴ using a modification of the Smoking-Attributable Mortality, Morbidity and Economic Cost  (SAMMEC) method.⁵ The SAMMEC method, with its potential for estimating SAM for 26 categories known to be attributable to cigarette smoking, is considered to be one of the most reliable methods available. Since 1991, new smoking behaviour data for 1994⁶ and 1996⁷ have become available. The purpose of the present report is to update this earlier work and to determine whether SAM estimates have changed nationally and regionally.

Data Sources

For both 1994⁸ and 1996,⁹ mortality data on 22 adult smoking-related diseases and four pediatric diseases linked with maternal smoking were drawn from the Canadian Mortality Database, maintained at the Health Statistics Division, Statistics Canada. Deaths were categorized by diagnosis, region, sex and five-year age group for persons aged 35-64 and 65 or over, and infants aged less than 1 year. Diagnoses were designated by code from the ninth revision of the International Classification of Diseases.

Data on fire deaths caused by smokers' material, which includes cigarettes, cigars, pipes, matches and lighters, were obtained from the 1994 Annual Report of Fire Losses in Canada,¹⁰ classified by region and sex (unpublished provincial tabulations were provided through the courtesy of Human Resources Development Canada). For 1996, information on fire deaths due to smoking was not available (at the time of writing this report) and was not included in our 1996 SAM calculations.

Calculation of SAM required smoking prevalence rates for adults 35 years of age or older and for women of childbearing age (15-44 years). Current, former and never smoker rates for 1994 and 1996 by region (province), sex and age group were obtained from the National Population Health Survey (NPHS) 1994/95⁶ and the NPHS 1996/97.⁷

The NPHS is a longitudinal household-based survey conducted every two years by Statistics Canada. The survey is designed to collect information about health status and health determinants, including health behaviour, use of health services and sociodemographic information. The target population of the NPHS is household residents in all provinces and territories, except persons living on Indian reserves, on Canadian Forces bases or in some remote areas of Ontario and Quebec.

Prevalence rates of current smokers aged 35 and older married to non-smokers were not available in either NPHS. Rates were obtained from the Survey on Smoking in Canada 1994/95, cycle 3,¹¹ and were used to calculate passive smoking deaths for both 1994 and 1996.

Diagnosis-specific relative risks for smoking-related diseases¹² were determined from the Cancer Prevention Study II (CPS-II) of the American Cancer Society and were previously presented in Collishaw and Leahy's 1989 report.² These relative risk estimates were based on a four-year follow-up study, from 1982 to 1986, of 1.2 million entrants in the CPS-II and were used to calculate regional SAM. The relative risks obtained were those for current and former smokers aged 35-64 and 65 or over by diagnosis and sex.

Methods

The Canadian mortality data^8,9 and smoking prevalence rates^6,7 for 1994 and 1996, combined with the CPS-II relative risks,¹² provide the basis for the smoking-related mortality estimates presented in this report. These estimates were calculated according to methods presented in Makomaski Illing and Kaiserman's 1991 report.⁴

Results

The 1996 results will be discussed in detail throughout this paper, but the 1994 estimates will be presented only in tables and figures. Total SAM estimates, by disease and region, are provided in Tables 1A (males) and 1B (females) for 1994 and in Tables 2A (males) and 2B (females) for 1996.

Overall, 29,229 males and 15,986 females died of smoking-attributable causes in 1996, including 63 boys and 42 girls under the age of 1. Cancers accounted for 17,703 of the total number of such deaths, cardiovascular diseases accounted for another 17,762 and respiratory diseases for the remaining 9,498 deaths.

Figure 1 shows the proportions of smoking-related deaths in 1996, by disease category, among 29,166 men, 15,944 women and 105 children (of both sexes). Of these deaths, lung cancer caused 8,973 (31%) male deaths and 4,519 (28%) female deaths, while ischemic heart disease caused 6,441 (22%) male deaths and 3,137 (22%) female deaths.

 In 1996, 111,405 males and 101,476 females died of all causes in Canada; the top three leading causes of death in both men and women were cancer, heart disease and cerebrovascular disease.¹³ Overall, 21% of these deaths were attributable to smoking in 1996.

Cigarette smoking, the primary risk factor for the top three causes of death,⁴ was estimated to be responsible for 26% of all male deaths and 16% of all female deaths in 1996. It is interesting to note that, among men, smoking accounted for 10% of all deaths from cardiovascular disease (CVD), 11% of all deaths from smoking-related cancers and 5% of all deaths from respiratory diseases; among women, these proportions were 6%, 6% and 4% respectively.

TABLE 1A Smoking-attributable mortality (SAM) estimates for current and former smokers, by disease category and region, MALES, Canada, 1994
		MALES: SAM BY REGION
Disease category	ICD-9 code	Canada	Atlantic	Quebec	Ontario	Prairies	BC
ADULT DISEASES (35+ years)		29,657	3,024	8,467	10,251	4,528	3,386
Cancers		11,895	1,168	3,722	4,105	1,575	1,326
Lip, oral cavity, pharynx	140-149	646	54	204	231	76	81
Esophagus	150	673	63	138	271	105	96
Pancreas	157	405	33	130	139	57	45
Larynx	161	352	28	133	110	40	41
Trachea, lung, bronchus	162	9,004	912	2,891	3,064	1,177	960
Cervix uteri	180	N/A	N/A	N/A	N/A	N/A	N/A
Urinary bladder	188	450	38	123	173	62	53
Kidney, other urinary	189	365	40	102	116	57	50
Cardiovascular diseases		12,005	1,235	3,174	4,166	2,022	1,408
Rheumatic heart disease	390-398	44	3	14	13	8	7
Hypertension	401-405	148	13	31	58	28	16
Ischemic heart disease     Ages 35-64     Ages 65+	410-414	2,609 4,300	246 434	823 1,093	932 1,565	375 691	231 517
Pulmonary heart disease	415-417	103	10	29	27	23	14
Other heart disease	420-429	1,423	180	378	381	291	193
Cerebrovascular disease     Ages 35-64     Ages 65+	430-438	417 1,389	48 138	115 324	134 479	66 248	54 201
Atherosclerosis	440	432	34	89	179	89	39
Aortic aneurysm	441	810	86	194	283	148	99
Other arterial disease	442-448	331	44	84	114	54	35
Respiratory diseases		5,578	622	1,571	1,981	931	653
Respiratory tuberculosis	010-012	20	2	5	8	4	1
Pneumonia/influenza	480-487	1,193	123	220	443	221	185
Bronchitis/emphysema	491-492	799	70	292	220	129	88
Asthma	493	61	8	10	16	16	11
Chronic airways obstruction	496	3,685	418	1,043	1,294	562	367
PEDIATRIC DISEASES (<1 year)		90	6	21	34	21	8
Low birth weight	765	21	1	4	10	4	1
Respiratory distress syndrome	769	16	1	6	5	3	1
Respiratory conditions-newborn	770	16	2	3	8	3	1
Sudden infant death syndrome	798.0	35	2	8	10	11	4
FIRE DEATHS (all ages)		29	3	0	12	7	7
PASSIVE SMOKING DEATHSa,b		215	25	92	79	10	9
TOTAL		29,991	3,059	8,580	10,376	4,566	3,410
a Deaths due to ETS (environmental tobacco smoke) were calculated using rr_ets  = 1.3 and prevalence of current smokers, aged 35+, married to non-smokers, taken from SOSIC 1994/95, cycle 3 (Reference 11). b The estimates presented here include only lung cancer mortality from exposure to second-hand smoke. It is recognized that exposure to second-hand smoke is associated with other diseases, including heart disease and breast cancer. Until authoritative, consensus-based relative risks are made available, estimates of mortality remain unavailable.

TABLE 1B Smoking-attributable mortality (SAM) estimates for current and former smokers, by disease category and region, FEMALES, Canada, 1994
		FEMALES: SAM BY REGION						TOTAL SAM
Disease category	ICD-9 code	Canada	Atlantic	Quebec	Ontario	Prairies	BC	M+F
ADULT DISEASES (35+ years)		15,287	1,455	3,966	5,484	2,311	2,071	44,945
Cancers		5,523	476	1,472	2,002	808	766	17,418
Lip, oral cavity, pharynx	140-149	175	15	43	61	26	31	820
Esophagus	150	230	15	55	91	31	38	902
Pancreas	157	522	54	134	185	81	68	927
Larynx	161	80	6	31	24	9	10	432
Trachea, lung, bronchus	162	4,148	349	1,107	1,508	603	580	13,151
Cervix uteri	180	144	11	33	52	32	17	144
Urinary bladder	188	167	20	51	62	17	16	617
Kidney, other urinary	189	58	6	17	18	10	6	423
Cardiovascular diseases		6,228	644	1,634	2,183	963	803	18,233
Rheumatic heart disease	390-398	69	5	24	21	10	9	113
Hypertension	401-405	145	17	41	50	25	13	293
Ischemic heart disease     Ages 35-64     Ages 65+	410-414	615 2,588	65 242	183 719	245 882	78 387	45 358	3,223 6,889
Pulmonary heart disease	415-417	87	7	23	27	17	12	189
Other heart disease	420-429	1,012	137	265	278	176	156	2,435
Cerebrovascular disease     Ages 35-64     Ages 65+	430-438	338 465	28 58	97 106	120 138	61 77	34 86	755 1,854
Atherosclerosis	440	443	31	73	253	55	31	875
Aortic aneurysm	441	270	30	55	97	45	42	1,081
Other arterial disease	442-448	196	25	49	72	32	17	527
Respiratory diseases		3,536	335	861	1,299	540	502	9,294
Respiratory tuberculosis	010-012	8	2	2	3	1	1	28
Pneumonia/influenza	480-487	1,093	116	203	437	179	158	2,286
Bronchitis/emphysema	491-492	410	30	130	119	68	63	1,209
Asthma	493	72	7	16	27	9	13	133
Chronic airways obstruction	496	1,954	180	511	714	282	267	5,639
PEDIATRIC DISEASES (<1 year)		60	5	13	23	14	6	150
Low birth weight	765	18	1	3	8	5	1	39
Respiratory distress syndrome	769	12	2	3	3	2	2	28
Respiratory conditions-newborn	770	12	1	2	5	3	2	30
Sudden infant death syndrome	798.0	18	1	4	6	5	2	53
FIRE DEATHS (all ages)		12	0	0	9	3	0	41
PASSIVE SMOKING DEATHSa,b		121	11	32	40	16	22	336
TOTAL		15,481	1,471	4,011	5,555	2,344	2,100	45,472
a Deaths due to ETS (environmental tobacco smoke) were calculated using rr_ets  = 1.3 and prevalence of current smokers, aged 35+, married to non-smokers, taken from SOSIC 1994/95, cycle 3 (Reference 11). b The estimates presented here include only lung cancer mortality from exposure to second-hand smoke. It is recognized that exposure to second-hand smoke is associated with other diseases, including heart disease and breast cancer. Until authoritative, consensus-based relative risks are made available, estimates of mortality remain unavailable.

Of the total 45,215 smoking-attributable deaths in Canada in 1996, 15,642 occurred in Ontario;12,328 in Quebec; 7,080 in the Prairies; 5,860 in British Columbia; and 4,305 in the Atlantic region (Tables 2A and 2B).

TABLE 2A Smoking-attributable mortality (SAM) estimates for current and former smokers, by disease category and region, MALES, Canada, 1996
		MALES: SAM BY REGION
Disease category	ICD-9 code	Canada	Atlantic	Quebec	Ontario	Prairies	BC
ADULT DISEASES (35+ years)		28,952	2,838	8,144	9,841	4,535	3,594
Cancers		11,844	1,087	3,772	3,981	1,615	1,389
Lip, oral cavity, pharynx	140-149	614	37	201	226	82	69
Esophagus	150	702	63	141	278	118	103
Pancreas	157	410	35	121	140	63	51
Larynx	161	329	22	136	110	32	29
Trachea, lung, bronchus	162	8,973	844	2,935	2,974	1,184	1,036
Cervix uteri	180	N/A	N/A	N/A	N/A	N/A	N/A
Urinary bladder	188	435	39	131	144	67	53
Kidney, other urinary	189	381	47	107	110	68	49
Cardiovascular diseases		11,429	1,182	2,877	3,964	1,959	1,448
Rheumatic heart disease	390-398	38	1	13	13	5	6
Hypertension	401-405	148	16	30	52	27	22
Ischemic heart disease     Ages 35-64     Ages 65+	410-414	2,392 4,049	237 407	682 1,020	867 1,443	359 677	248 503
Pulmonary heart disease	415-417	111	10	29	33	23	17
Other heart disease	420-429	1,473	184	346	423	302	218
Cerebrovascular disease     Ages 35-64     Ages 65+	430-438	378 1,389	34 140	110 307	137 460	55 288	41 194
Atherosclerosis	440	326	17	76	138	57	39
Aortic aneurysm	441	778	96	176	275	110	121
Other arterial disease	442-448	346	42	87	122	57	39
Respiratory diseases		5,679	569	1,495	1,897	961	757
Respiratory tuberculosis	010-012	15	1	5	5	1	2
Pneumonia/influenza	480-487	1,183	114	213	429	238	187
Bronchitis/emphysema	491-492	710	55	242	195	122	95
Asthma	493	63	5	9	22	14	14
Chronic airways obstruction	496	3,708	393	1,025	1,245	586	63
PEDIATRIC DISEASES (<1 year)		63	4	16	24	15	4
Low birth weight	765	18	1	5	8	3	1
Respiratory distress syndrome	769	15	1	5	4	3	1
Respiratory conditions-newborn	770	13	1	2	6	2	1
Sudden infant death syndrome	798.0	17	2	3	5	7	1
FIRE DEATHS (all ages)		N/A	N/A	N/A	N/A	N/A	N/A
PASSIVE SMOKING DEATHSa,b		214	23	94	77	10	10
TOTAL		29,229	2,865	8,254	9,942	4,560	3,608
SAM as percentage of all deaths		26	3	7	9	4	3
Male-to-female SAM ratio		1.8	2.0	2.0	1.7	1.8	1.6
NOTE: Estimates in bold should be interpreted with caution. a Deaths due to ETS (environmental tobacco smoke) were calculated using rr_ets  = 1.3 and prevalence of current smokers, aged 35+, married to non-smokers, taken from SOSIC 1994/95, cycle 3 (Reference 11). b The estimates presented here include only lung cancer mortality from exposure to second-hand smoke. It is recognized that exposure to second-hand smoke is associated with other diseases, including heart disease and breast cancer. Until authoritative, consensus-based relative risks are made available, estimates of mortality remain unavailable.

TABLE 2B Smoking-attributable mortality (SAM) estimates for current and former smokers, by disease category and region, FEMALES, Canada, 1996
		FEMALES: SAM BY REGION						TOTAL SAM
Disease category	ICD-9 code	Canada	Atlantic	Quebec	Ontario	Prairies	BC	M+F
ADULT DISEASES (35+ years)		15,811	1,423	4,027	5,645	2,492	2,224	44,763
Cancers		5,859	508	1,577	2,047	905	822	17,703
Lip, oral cavity, pharynx	140-149	160	10	31	65	31	24	775
Esophagus	150	235	20	44	101	35	36	937
Pancreas	157	527	53	145	168	88	73	936
Larynx	161	66	1	23	27	11	5	395
Trachea, lung, bronchus	162	4,519	391	1,250	1,555	687	636	13,492
Cervix uteri	180	146	17	30	62	20	16	146
Urinary bladder	188	142	10	37	51	21	23	577
Kidney, other urinary	189	63	8	17	19	12	7	444
Cardiovascular diseases		6,133	557	1,538	2,257	982	798	17,562
Rheumatic heart disease	390-398	65	5	25	18	9	8	104
Hypertension	401-405	150	15	36	54	26	19	298
Ischemic heart disease     Ages 35-64     Ages 65+	410-414	568 2,569	68 192	163 644	208 1,003	78 372	50 359	2,960 6,618
Pulmonary heart disease	415-417	85	6	20	28	20	12	197
Other heart disease	420-429	1,007	138	271	319	204	145	2,550
Cerebrovascular disease     Ages 35-64     Ages 65+	430-438	307 511	25 44	85 113	111 192	51 79	34 83	685 1,900
Atherosclerosis	440	315	17	61	144	63	29	641
Aortic aneurysm	441	270	29	66	93	49	32	1,048
Other arterial disease	442-448	216	19	53	87	31	27	562
Respiratory diseases		3,819	358	912	1,341	605	604	9,498
Respiratory tuberculosis	010-012	11	2	2	3	2	3	26
Pneumonia/influenza	480-487	1,074	105	189	399	200	180	2,257
Bronchitis/emphysema	491-492	414	31	132	120	68	63	1,124
Asthma	493	66	5	12	29	10	11	130
Chronic airways obstruction	496	2,254	216	577	790	325	347	5,962
PEDIATRIC DISEASES (<1 year)		42	4	11	13	10	4	105
Low birth weight	765	14	0	4	5	4	2	32
Respiratory distress syndrome	769	8	2	3	1	1	1	23
Respiratory conditions-newborn	770	7	1	2	3	2	0	20
Sudden infant death syndrome	798.0	13	1	3	4	4	1	30
FIRE DEATHS (all ages)		N/A	N/A	N/A	N/A	N/A	N/A	N/A
PASSIVE SMOKING DEATHSa,b		133	13	36	42	18	24	347
TOTAL		15,986	1,440	4,075	5,700	2,520	2,251	45,215
SAM as percentage of all deaths		16	1	4	6	3	2	21
NOTE: Estimates in bold should be interpreted with caution. a Deaths due to ETS (environmental tobacco smoke) were calculated using rr_ets  = 1.3 and prevalence of current smokers, aged 35+, married to non-smokers, taken from SOSIC 1994/95, cycle 3 (Reference 11). b The estimates presented here include only lung cancer mortality from exposure to second-hand smoke. It is recognized that exposure to second-hand smoke is associated with other diseases, including heart disease and breast cancer. Until authoritative, consensus-based relative risks are made available, estimates of mortality remain unavailable.

Figure 2 shows the changes in SAM for both sexes from 1985 to 1996. The number of smoking-attributable deaths among women rose from 9,009 in 1985 to 15,986 in 1996, an increase of 77%. The number of deaths among men remained relatively constant throughout this period (from 28,321 in 1985 to 29,229 in 1996), representing a drop in the male-to-female SAM ratio, from 3.1 to 1.8.

The SAM trends by disease category for both sexes are presented in Figure 3. From 1989 to 1996, female SAM increased by 48% (5,166 deaths), whereas male SAM increased only slightly, by 6% (1,692 deaths). The large increase in female SAM is mainly due to large jumps in cancers (19%; 2,016 deaths), cardiovascular diseases (15%; 1,670 deaths) and respiratory diseases (16%; 1,691 deaths). Among men aged 35 and over, smoking-related deaths from both cancers and respiratory diseases rose by 3% and, from cardiovascular diseases, rose minimally by 0.2%. Figure 4 compares increases in smoking-related deaths from lung cancer, ischemic heart disease and chronic airways obstruction between 1989 and 1996 among men and women.

Discussion

In 1996, smoking remained the number one preventable cause of death and disease in Canada. Accounting for over 45,200 deaths that year, smoking far exceeded the second most important preventable cause of death-accidents (8,603 deaths).¹³ Since 1991, the number of smoking-attributable deaths in Canada has increased by an estimated 8% (3,807 deaths); of these, almost two thirds (64%; 2,445 deaths) occurred in females. On a regional basis, there were 1,188 more smoking-related deaths (+20%) in the Prairies over the same five years, followed by an increase of 1,149 deaths (+8%) in Ontario.

The number of smoking-related deaths among females is rising faster than among males. In 1985, the ratio of male-to-female deaths attributable to smoking was approximately 3.1; in 1989, this ratio had fallen to 2.5; in 1991, to 2.1; and in 1996, it had declined further to 1.8.

The World Health Organization predicted that 16,000 Canadian women and 30,000 Canadian men would die in 1995 from smoking-related causes.^14,15 As can be seen from our results (Figure 2), this forecast was reasonable.

The smoking behaviour of the population two decades earlier is reflected in the present trends in smoking-attributable mortality. Among women, smoking rates peaked in the late 1970s, and lung cancer death rates are now more than four times as high as rates in 1969.¹⁶ Among men, however, smoking rates peaked in the mid 1960s; after decades of continuous increases, lung cancer death rates reached a peak in the late 1980s and have fallen slightly since then.²

The number of smoking-related deaths is also influenced by the growth and aging of the population.¹⁷ As "baby boomers" age, it can be expected that large numbers of Canadians will continue to die from smoking-related causes, in particular from lung cancer, heart disease and cerebrovascular disease.¹⁴

Given the levelling off of male SAM over the past few years coupled with the decrease in smoking prevalence among males, there is little reason to change earlier predictions that male SAM may start to decline by the new millennium and eventually level off, while female SAM will continue to rise and may eventually reach, or even exceed, male levels.