Examining the association of smoking with work productivity and associated costs in Japan

Figures & data

Table 1. Demographics and health characteristics by smoking status.

	Smoking status						One-way ANOVA p-value
	Never smoked (n = 13,238)		Former smoker (n = 5,055)		Current smoker (n = 5,445)
	Mean	SD	Mean	SD	Mean	SD
Age of the respondent	40.13a	12.22	45.34b	10.93	43.65c	10.86	<.001
Charlson Comorbidity Index Scoring (CCI)	0.07a	0.35	0.15b	0.45	0.17b	0.93	<.001
	n	%	n	%	n	%	χ^2 p-value
Sex
Female	8,115a	61.3%	2,030b	40.2%	1,879c	34.5%	<.001
Male	5,123a	38.7%	3,025b	59.8%	3,566c	65.5%
Education
No university degree	4,043a	30.5%	1,745b	34.5%	2,064c	37.9%	<.001
University degree	9,195a	69.5%	3,310b	65.5%	3,381c	62.1%
Annual household income
<¥3,000,000	2,270a	17.1%	757b	15.0%	844b	15.5%	<.001
¥3,000,000–<¥5,000,000	2,953a	22.3%	1,235b	24.4%	1,273a,b	23.4%
¥5,000,000–<¥8,000,000	3,377a	25.5%	1,429b	28.3%	1,545b	28.4%
>¥8,000,000	3,124a	23.6%	1,235a	24.4%	1,367a	25.1%
Decline to Answer	1,514a	11.4%	399b	7.9%	416b	7.6%
BMI category
Underweight	1,716a	13.0%	398b	7.9%	591c	10.9%	<.001
Normal weight	8,743a	66.0%	3,358a	66.4%	3,504a	64.4%
Overweight	1,557a	11.8%	927b	18.3%	1,000b	18.4%
Obese	427a	3.2%	207b	4.1%	202a,b	3.7%
Unknown	795a	6.0%	165b	3.3%	148b	2.7%
Alcohol consumption
Never	4,329a	32.7%	1,001b	19.8%	863c	15.8%	<.001
Any	8,909a	67.3%	4,054b	80.2%	4,582c	84.2%
Labor force participation
No	4,577a	34.6%	1,119b	22.1%	859c	15.8%	<.001
Yes	8,661a	65.4%	3,936b	77.9%	4,586c	84.2%

Abbreviations. SD, standard deviation; CCI, Charlson Comorbidity Index; BMI, body mass index.

For pairwise comparisons, refer to subscripts (e.g. a, b, c) attached to means or cell counts. Mean values or cell counts in the same row that do not share the same subscript (e.g. XX_a vs XX_b) are significantly different at p < .05 after adjusting for the Bonferroni correction. If they share the same subscript (e.g. XX_a vs XX_a or XX_a,_b vs XX_a) then they are not significantly different at p < .05.

Table 2. Adjusted means examining effect of smoking status on WPAI-GH and costs after controlling for demographic and health characteristics.

	Never Smoked (A) (N = 7,766)						Former Smoker (B) (N = 3,575)				Current Smoker (C) (N = 4,198)
			RR	p-value	RR	p-value			RR	p-value
WPAI-GH	Mean (SE)	95% LCI, UCI	B vs. A	B vs. A	C vs. A	C vs. A	Mean (SE)	95%LCI, UCI	C vs. B	C vs. B	Mean (SE)	95% LCI, UCI
Absenteeism (% work missed)	2.72 (.14)	2.47, 3.00	1.03	.729	1.15	.114	2.81 (.20)	2.44, 3.23	1.11	.285	3.12 (.21)	2.73, 3.56
Presenteeism (% impaired while working)	15.85 (.24)	15.38, 16.33	1.00	.951	1.07	.012	15.82 (.35)	15.15, 16.53	1.07	.026	16.92 (.35)	16.25, 17.61
Overall Work Impairment (% overall work impairment)	17.41 (.27)	16.89, 17.95	1.01	.863	1.07	.008	17.49 (.40)	16.73, 18.29	1.07	.030	18.69 (.39)	17.94, 19.48
Activity Impairment (% activity impairment)*	17.31 (.19)	16.93, 17.69	1.06	.011	1.10	<.001	18.28 (.33)	17.65, 18.94	1.04	.100	19.04 (.33)	18.40, 19.70
Annual indirect costs (Japanese Yen in thousands)
Absenteeism Costs	104.63 (5.76)	93.92, 116.56	1.05	.595	1.14	.161	110.27 (8.82)	94.27, 128.99	1.09	.450	119.72 (9.01)	103.31, 138.74
Presenteeism Costs	503.16 (8.78)	486.25, 520.66	1.02	.582	1.12	<.001	511.96 (13.05)	487.01, 538.19	1.10	.007	561.56 (13.20)	536.28, 588.04
Indirect Costs	608.16 (10.91)	587.14, 629.94	1.02	.462	1.12	<.001	622.79 (16.32)	591.62, 655.60	1.10	.012	680.65 (16.48)	649.10, 713.74

The calculation of costs takes respondent age and gender into consideration, and the calculation varies based on these respondent characteristics. Rate ratio indicates the increase in the outcome associated with a change in the first group to the second group, as indicated in smoking status.

Abbreviations. LCI, lower confidence interval; UCI, upper confidence interval; SE, standard error, RR, rate ratio.

* Calculated for the entire sample.

Table A1. Percent work productivity and activity impairment by smoking status.

Mean	Smoking status						One-way ANOVA p-value
	Never smoked (n = 13,238)		Former smoker (n = 5,055)		Current smoker (n = 5,445)
		SD	Mean	SD	Mean	SD
Absenteeism* (% work missed)	2.96a	12.20	2.76a	12.40	3.31a	12.57	.126
Presenteeism* (% impaired while working)	16.34a,b	21.53	15.48a	21.18	17.09b	22.09	.004
Overall work impairment* (% overall work impairment)	18.00a,b	23.79	17.05a	23.63	18.91b	24.41	.003
Activity impairment (% activity impairment)	18.05a	22.56	18.03a,b	23.19	19.00b	23.41	.026

Abbreviations. SD, standard deviation.

For pairwise comparisons, refer to subscripts (e.g. a, b, c) attached to means. Mean values in the same row that do not share the same subscript (e.g. XX_a vs XX_b) are significantly different at p < .05 after adjusting for the Bonferroni correction. If they share the same subscript (e.g. XX_a vs XX_a or XX_a,_b vs XX_a) then they are not significantly different at p < .05.

* Only asked of those in the labor force: Never Smoked (n = 7,766), Former Smoker (n = 3,575), Current Smoker (n = 4,198).

Table A2. Annual indirect costs (Japanese Yen in thousands) overall for those in the labor force by smoking status.

	Smoking status						One-way ANOVA p-value
	Never smoked (n = 7,766)		Former smoker (n = 3,575)		Current smoker (n = 4,198)
	Mean	SD	Mean	SD	Mean	SD
Absenteeism costs	109.69a	515.07	109.36a	525.76	125.75a	541.17	.229
Presenteeism costs	516.17a	799.52	512.29a	799.90	581.61b	850.12	<.001
Indirect costs	624.35a	990.15	619.87a	988.67	705.12b	1,047.55	<.001

Abbreviations. SD, standard deviation.

For pairwise comparisons, refer to subscripts (e.g. a, b, c) attached to means. Mean values in the same row that do not share the same subscript (e.g. XX_a vs XX_b) are significantly different at p < .05 after adjusting for the Bonferroni correction. If they share the same subscript (e.g. XX_a vs XX_a or XX_a,_b vs XX_a) then they are not significantly different at p < .05.

Table A3. Annual indirect costs (Japanese Yen in thousands), taking age and gender into account for those in the labor force by smoking status.

	Smoking Status						One-way ANOVA p-value
	Never smoked (n = 7,766)		Former smoker (n = 3,575)		Current smoker (n = 4,198)
	Mean	SD	Mean	SD	Mean	SD
Absenteeism costs	103.63a	503.38	118.60a,b	592.22	136.59b	619.61	.008
Presenteeism costs	498.56a	819.11	562.16b	923.95	624.25c	949.03	<.001
Indirect costs	600.75a	1,006.15	678.83b	1,130.90	758.41c	1,186.22	<.001

The calculation of costs takes respondent age and gender into consideration, and the calculation varies based on these respondent characteristics.

Abbreviations. SD, standard deviation.

For pairwise comparisons, refer to subscripts (e.g. a, b, c) attached to means. Mean values in the same row that do not share the same subscript (e.g. XX_a vs XX_b) are significantly different at p < .05 after adjusting for the Bonferroni correction. If they share the same subscript (e.g. XX_a vs XX_a or XX_a,_b vs XX_a) then they are not significantly different at p < .05.

Table A4. Annual indirect costs (Japanese Yen in thousands), taking age and gender into account, for those in the labor force by smoking status, after controlling for demographic and health characteristics.

	Never Smoked (A) (N = 7,766)						Former S moker (B) (N = 3,575)				Current Smoker (C) (N = 4,198)
			RR	p-value	RR	p-value			RR	p-value
	Mean (SE)	95% LCI, UCI	B vs. A	B vs. A	C vs. A	C vs. A	Mean (SE)	95%LCI, UCI	C vs. B	C vs.B	Mean (SE)	95% LCI, UCI
Absenteeism Costs	103.49 (5.78)	92.77, 115.45	1.07	.516	1.17	.107	110.44 (8.93)	94.25, 129.41	1.10	.403	121.09 (9.23)	104.28, 140.61
Presenteeism Costs	498.82 (8.84)	481.80, 516.45	1.03	.350	1.13	<.001	513.92 (13.30)	488.50, 540.65	1.10	.007	564.61 (13.49)	538.78, 591.67
Indirect Costs	602.50 (10.98)	581.37, 624.40	1.04	.271	1.14	<.001	624.64 (16.60)	592.95, 658.03	1.10	.010	685.10 (16.85)	652.86, 718.93

The calculation of costs takes respondent age and gender into consideration, and the calculation varies based on these respondent characteristics. Rate ratio indicates the increase in the outcome associated with a change in the first group to the second group, as indicated in smoking status.

Abbreviations. LCI, lower confidence interval; UCI, upper confidence interval; SE, standard error, RR, rate ratio.