Top

BMC Public Health

Published in:

Open Access 01-12-2015 | Research Article

Application of the varying coefficient model to the behaviour risk factor surveillance data in Italy: a study of changing smoking prevalence among sub-populations

Authors: Shireen Assaf, Stefano Campostrini

Published in: BMC Public Health | Issue 1/2015

Abstract

Background

Behaviour risk factor surveillance (BRFS) data can be an important source of information for studying changes in various health outcomes and risk factors. Results obtained from surveillance data analysis are vital for informing health policy interventions, particularly with regards to evolutionary aspects. The objective of this analysis was to recommend a method that can be used for analysing trends in the association among variables from large public health data sets. This was demonstrated by examining the changing effects of various covariates, representing different sub-populations, on smoking status over time.

Methods

In our work, we propose the use of varying coefficient models (VCM) with non-parametric techniques to catch the dynamics of the evolutionary processes under study. This is a useful method, which allows coefficients to vary with time using smooth functions. Italian BRFS data from 2008-2012 was used with a sample size of 185,619 observations. In the application, a time VCM is fit for a smoking status binary outcome variable using the P-spline estimation method. The model includes ten independent variables comprising socio-demographic, health risk and behaviour variables.

Results

The VCM fit for the data indicates that the coefficients for some of the categories for the age and the alcohol consumption variables varied with time. The main results show that Italians aged 18-29 and 40-49 had higher odds of being smokers compared to those aged 60-69; however, these odds significantly decreased in the period 2008-2012. In addition, those who do not drink had lower odds for being a smoker compared to high risk drinkers and these odds decreased further during the observation period.

Conclusion

The application of the VCM to the BRFS data in Italy has shown that this method can be useful in detecting which sub-populations require interventions. Although the results have shown a decrease in the odds of being a smoker for certain age groups and non-drinkers, other sub-populations have not decreased their odds and health inequalities remain. This observation indicates that efforts and interventions are still required to target these non-changing sub-populations in order to modify their smoking behaviour.

Lee L, Thacker S. Principles and Practice of Public Health Surveillance. USA: Oxford University Press; 2010.CrossRef

Campostrini S, McQueen D, Taylor A, Daly A. World alliance for risk factor surveillance white paper on surveillance and health promotion. Aims Public Health. 2015; 2:10–26.CrossRef

Mokdad A. The behavioral risk factors surveillance system: Past, present, and future. Annu Rev Public Health. 2009; 30:43–54.CrossRefPubMed

Campostrini S. Surveillance systems and data analysis: continuously collected behavioural data In: McQueen D, Puska P, editors. Global Behavioral Risk Factor Surveillance. New York: Springer: 2003. p. 47–56.

Campostrini S. Surveillance for NCDs and Health promotion: an issue of theory and method In: McQueen D, editor. Global Handbook on Non-communicable Diseases and Health Promotion. New York: Springer: 2013.

Huang J, Wu C, Zhou L. Polynomial spline estimation and inference for varying coefficient models with longitudinal data. Statistica Sinica. 2004; 14(3):763–88.

Hoover D, Rice J, Wu C, Yang L. Nonparametric smoothing estimates of time-varying coefficient models with longitudinal data. Biometrika. 1998; 85(4):809–22.CrossRef

Wang L, Li H, Huang J. Variable selection in nonparametric varying-coefficient models for analysis of repeated measurements. J Am Stat Assoc. 2008; 103(484):1556–69.CrossRefPubMedPubMedCentral

Wu C, Yu K, Chiang C. A two-step smoothing method for varying-coefficient models with repeated measurements. Ann Inst Stat Mathematics. 2000; 52(3):519–43.CrossRef

10.

Wu C, Chiang C. Kernel smoothing on varying coefficient models with longitudinal dependent variable. Statistica Sinica. 2000; 10(2):433–56.

11.

Zhang D. Generalized linear mixed models with varying coefficients for longitudinal data. Biometrics. 2004; 60(1):8–15.CrossRefPubMed

12.

Young L, Gotway C, Yang J, Kearney G, DuClos C. Assessing the association between environmental impacts and health outcomes: A case study from florida. Stat Med. 2008; 27(20):3998–4015.CrossRefPubMed

13.

PASSI - Coordinating technical group of the behavioural risk factor system. PASSI (Progressi Delle Aziende Sanitarie per la Salute in Italia) - The Italian behavioral risk factor surveillance system. 2013. Available at: http://www.epicentro.iss.it/passi/en/english.asp. Rome. Retrieved on: 27 June 2013.

14.

Baldissera S, Campostrini S, Binkin N, Minardi V, Minelli G, Ferrante G, et al. Features and initial assessment of the Italian Behavioral Risk Factor Surveillance System (PASSI), 2007-2008. Preventing Chronic Disease. 2011; 8(1):A24. http://www.cdc.gov/pcd/issues/2011/jan/10_0030.htm.PubMed

15.

Minardi V, Campostrini S, Carrozzi G, Minelli G, Salmaso S. Social determinants effects from the italian risk factor surveillance system passi. Int J Public Health. 2011; 56(4):359–66.CrossRefPubMed

16.

Binkin N, Gigantesco A, Ferrante G, Baldissera S. Depressive symptoms among adults 18–69 years in italy: results from the italian behavioural risk factor surveillance system, 2007. Int J Public Health. 2010; 55(5):479–88.CrossRefPubMed

17.

Chandola T, Head J, Bartley M. Socio-demographic predictors of quitting smoking: how important are household factors?. Addiction. 2004; 99(6):770–7.CrossRefPubMed

18.

Hiscock R, Bauld L, Amos A, Fidler J, Munafò M. Socioeconomic status and smoking: a review. Ann New York Acad Sci. 2012; 1248(1):107–23.CrossRef

19.

Levy D, Romano E, Mumford E. The relationship of smoking cessation to sociodemographic characteristics, smoking intensity, and tobacco control policies. Nicotine Tobacco Res. 2005; 7(3):387–96.CrossRef

20.

Siahpush M, Borland R. Socio-demographic variations in smoking status among australians aged ≥18: multivariate results from the 1995 national health survey. Aust New Zealand J Public Health. 2001; 25(5):438–42.CrossRef

21.

Tramacere I, Gallus S, Zuccaro P, Colombo P, Rossi S, Boffetta P, et al. Socio-demographic variation in smoking habits: Italy, 2008. Preventive Med. 2009; 48(3):213–7.CrossRef

22.

Cook J, Fucito L, Piasecki T, Piper M, Schlam T, Berg K, et al. Relations of alcohol consumption with smoking cessation milestones and tobacco dependence. J Consulting Clin Psychology. 2012; 80(6):1075.CrossRef

23.

Boden J, Fergusson D, Horwood L. Cigarette smoking and depression: tests of causal linkages using a longitudinal birth cohort. Br J Psychiatry. 2010; 196(6):440–6.CrossRefPubMed

24.

Pratt L, Brody D. Depression and smoking in the US household population aged 20 and over, 2005-2008. NCHS Data Brief. 2010; 34:1.

25.

Trost S, Owen N, Bauman A, Sallis J, Brown W. Correlates of adults’ participation in physical activity: review and update. Med Sci Sports Exercise. 2002; 34(12):1996–2001.CrossRef

26.

Eilers P, Marx B. Generalized linear additive smooth structures. J Comput Graphical Stat. 2002; 11(4):758–83.CrossRef

27.

Hastie T, Tibshirani R. Varying-coefficient models. J R Stat Soc. Ser B (Methodological). 1993; 55(5):757–96.

28.

Cheng M, Zhang W, Chen L. Statistical estimation in generalized multiparameter likelihood models. J Am Stat Assoc. 2009; 104(487):1179–91.CrossRef

29.

Cai Z, Fan J, Li R. Generalized Varying-coefficient models. UC Los Angeles: Department of Statistics, UCLA. 1999. Available at: http://escholarship.org/uc/item/7n0494s4, Department of Statistics, UCLA, Los Angeles.

30.

Cai Z, Fan J, Li R. Efficient estimation and inferences for varying-coefficient models. J Am Stat Assoc. 2000; 95(451):888–902.CrossRef

31.

Chiang C, Rice J, Wu C. Smoothing spline estimation for varying coefficient models with repeatedly measured dependent variables. J Am Stat Assoc. 2001; 96(454):605–19.CrossRef

32.

Huang J, Wu C, Zhou L. Varying-coefficient models and basis function approximations for the analysis of repeated measurements. Biometrika. 2002; 89(1):111–28.CrossRef

33.

Marx B. P-spline varying coefficient models for complex data In: Kneib T, Tutz G, editors. Statistical Modelling and Regression Structures. Berlin Heidelberg: Springer: 2010. p. 19–43.

34.

Fan J, Zhang W. Statistical methods with varying coefficient models. Stat Its Interface. 2008; 1(1):179–95.CrossRef

35.

Eilers P, Marx B. Flexible smoothing with b-splines and penalties. Stat Sci. 1996; 11(2):89–102.CrossRef

36.

Wasserman L. All of Nonparametric Statistics. New York: Springer; 2006.

37.

Hastie T, Tibshirani R, Friedman J. The Elements of Statistical Learning, 2nd. New York: Springer; 2009.CrossRef

38.

Wood S. Fast stable restricted maximum likelihood and marginal likelihood estimation of semiparametric generalized linear models. J R Stat Soc: Ser B (Stat Methodology). 2011; 73(1):3–36.CrossRef

39.

Wood S. Generalized Additive Models: an Introduction with R. Boca Raton, FL: Chapman & Hall; 2006.

40.

Campostrini S. Measurement and effectiveness: methodological considerations, issues and possible solutions. In: Global Perspectives on Health Promotion Effectiveness. New York: Springer: 2007. p. 305–25.

41.

La Vecchia C, Bosetti C, Lucchini F, Bertuccio P, Negri E, Boyle P, et al. Cancer mortality in europe, 2000–2004, and an overview of trends since 1975. Ann Oncol. 2010; 21(6):1323–60.CrossRefPubMed

42.

R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2012. R Foundation for Statistical Computing. ISBN 3-900051-07-0. http://www.R-project.org/.

43.

Wood S. Mgcv: Mixed GAM Computation Vehicle with GCV/AIC/REML Smoothness Estimation. 2007. R package version 1.7-22.

Title: Application of the varying coefficient model to the behaviour risk factor surveillance data in Italy: a study of changing smoking prevalence among sub-populations
Authors: Shireen Assaf
Stefano Campostrini
Publication date: 01-12-2015
Publisher: BioMed Central
Published in: BMC Public Health / Issue 1/2015
Electronic ISSN: 1471-2458
DOI: https://doi.org/10.1186/s12889-015-1805-3

Keynote webinar | Spotlight on medication adherence

Springer Medicine

Application of the varying coefficient model to the behaviour risk factor surveillance data in Italy: a study of changing smoking prevalence among sub-populations

Abstract

Background

Methods

Results

Conclusion

Keynote webinar | Spotlight on medication adherence

Springer Medicine

Abstract

Background

Methods

Results

Conclusion

Please log in to get access to this content

Other articles of this Issue 1/2015

Impact of the joint association between sex, age and diabetes on long-term mortality after acute myocardial infarction

Coverage of skin cancer and recreational tanning in North American magazines before and after the landmark 2006 International Agency for Research on Cancer report

Maternal socioeconomic status and the risk of asthma among offspring

Updated prevalence rates of overweight and obesity in 11- to 17-year-old adolescents in Germany. Results from the telephone-based KiGGS Wave 1 after correction for bias in self-reports

Content analysis of press coverage during the H1N1 influenza pandemic in Germany 2009–2010

Association between smoking intensity and duration and tooth loss among Finnish middle-aged adults: The Northern Finland Birth Cohort 1966 Project