Table 2. Regression analysis of time trend in age-adjusted^a mesothelioma mortality rate, worldwide, 1994–2008.

Category	Regression parameterb				Annual change
	b or slope	95% CI	P		%	95% CI
Gender
Male	0.02	0.01 to 0.04	0.0003		5.85	3.25 to 8.51
Female	0.01	0.01 to 0.02	0.0008		3.48	1.74 to 5.26
All	0.02	0.01 to 0.03	0.0002		5.37	3.06 to 7.74
Anatomical disease site
Pleura	0.02	0.01 to 0.03	0.0006		5.20	2.67 to 7.78
Peritoneum	0.01	0.003 to 0.02	0.0166		2.78	0.59 to 5.03
Unspecified	0.03	0.01 to 0.05	0.0040		7.80	2.90 to 12.93
Continentc
Americas	0.03	−0.01 to 0.08	0.1148		7.89	−2.18 to 18.99
Asia	0.02	0.01 to 0.02	< 0.0001		3.67	2.64 to 4.71
Europe	0.01	0.003 to 0.03	0.0149		3.44	0.78 to 6.17
Oceania	−0.002	−0.01 to 0.01	0.6295		−0.51	−2.90 to 1.92
Country income group
High	0.02	0.01 to 0.04	0.0010		5.54	2.66 to 8.50
Middle and lowd	0.01	−0.001 to 0.02	0.0621		2.16	−0.12 to 4.51
Selected countries
United States of America	−0.004	−0.01 to −0.001	0.0076		−0.84	−1.34 to −0.34
United Kingdom of Great Britain and Northern Ireland	0.003	−0.004 to 0.01	0.3403		0.68	−0.88 to 2.27
Japan	0.01	0.01 to 0.02	< 0.0001		3.46	2.86 to 4.07
South Africa	−0.01	−0.01 to 0.001	0.0959		−1.47	−3.22 to 0.31

CI, confidence interval.

^a Ages were adjusted to the world standard population in 2000.

^b A log–linear regression model was used to analyse the variation in the logarithm of the age-adjusted mortality rate with calendar year. Weighting was applied for each country’s population.

^c Africa was excluded because the only country that satisfied the inclusion criteria for the trend analysis was South Africa, which is listed as a selected country.

^d Middle- and low-income countries were combined because only one low-income country satisfied the inclusion criteria for the trend analysis.