The contribution of smoking to regional mortality differences in the Netherlands

Sigma and beta convergence in regional mortality: A case study of the Netherlands

“…Previous research has shown that the larger differences in regional mortality among males as compared to females can largely be explained by smoking (Janssen and Spriensma 2012). Actually, for non-smoking-related mortality, regional differences in 2004-2008 were slightly larger for females as compared to males.…”

“…Even though the Netherlands is a small country, there have been significant differences in the level of mortality across regions. For 2004-2008, for example, 12 out of 40 NUTS-2 regions had statistically significant higher age-standardised mortality than the average level of 81.0 deaths per 10,000 population for males and 84.4 for females, and 15 regions had lower age-standardised mortality than average (Janssen and Spriensma 2012). Furthermore, a renewed increase in life expectancy at the national level has occurred since 2002 (Mackenbach and Garssen 2011), which could potentially affect the existence and level of convergence.…”

Role of smoking in regional variation in mortality in Poland

“…We derived smoking‐attributable mortality using the simplified indirect Peto–Lopez method , as proposed by Mackenbach et al . and applied, for example, by Janssen & Spriensma . The simplified Peto–Lopez method was chosen in this study over the original method because it applies smoking‐attributable risks to all‐cause mortality instead of cause‐specific mortality rates.…”

“…As the high share of ill‐defined causes of death in Poland means that the cause‐of‐death data are of low quality , smoking‐attributable mortality rates obtained through methods using cause‐specific data would be questionable. Janssen & Spriensma showed for the Netherlands that estimates of smoking‐attributable mortality were approximately the same using both the original and simplified Peto–Lopez methods. The regression‐based method proposed recently by Preston, Glei & Wilmoth , which uses the relationship between lung cancer mortality and all other causes combined, also resulted—according to the authors—in estimates very similar to the original Peto–Lopez method.…”

“…) and mainly applied to explain differences in mortality between countries . To a lesser extent, smoking‐attributable mortality is used to study the significance of smoking for within‐country mortality differences , but to our knowledge no earlier study has formally examined the role of smoking in changes over time in regional variation. Nevertheless, analyses of regional mortality levels and trends due to smoking provide useful information for local and national policymakers aiming to conduct effective anti‐tobacco campaigns and improve access to early diagnosis and treatment of people with diseases caused by tobacco use.…”