Generalized Additive Models applied to analysis of the relation between amount and type of alcohol and all-cause mortality

  • Ditte Johansen
  • Morten Grønbæk
  • Kim Overvad
  • Peter Schnohr
  • Per Kragh Andersen


The J-shaped relation between alcohol intake and mortality is well established, whereas the nadir of the curve is not determined. Due to non-linearity of the relation, categorical alcohol variables have been used to model the relation. In Generalized Additive Models (GAM) non-linear relations can be modelled without the disadvantages of categorization and without assumptions regarding the functional form. The aim of this study was to use GAM to evaluate the relation between alcohol intake, amount and type, and mortality. The relation was investigated using data from the Copenhagen City Heart Study (11,920 participants of whom 5552 died during 20 years follow-up). Using GAM, a smooth J-shaped relation between alcohol and mortality was found. However, if non-drinkers were categorized separately there was a positive association between alcohol and mortality even for low alcohol intake. For equal total alcohol intake, men and women drinking wine or spirits had lower mortality than beer drinkers. The nadir of the relation between alcohol and mortality was sensitive to the handling of non-drinkers. When non-drinkers were categorized separately we found no indication of a beneficial influence of low alcohol intake on mortality.


Alcohol Beverage type Generalized Additive Models Mortality Poisson analysis 


  1. Fuchs, CS, Stampfer, MJ, Colditz, GA,  et al. 1995Alcohol consumption and mortality among women.N Engl J Med33212451250CrossRefPubMedGoogle Scholar
  2. Thun, MJ, Peto, R, Lopez, AD,  et al. 1997Alcohol consumption and mortality among middle-aged and elderly US adults.N Engl J Med33717051714CrossRefPubMedGoogle Scholar
  3. Renaud, SC, Guéguen, R, Schenker, J, d ′Houtaud, A 1998Alcohol and mortality in middle-aged men from eastern France.Epidemiology9184188CrossRefPubMedGoogle Scholar
  4. Keil, U, Chambless, LE, Döring, A, Filipiak, B, Stieber, J 1997The relation of alcohol intake to coronary heart disease and all-cause mortality in a beer-drinking population.Epidemiology8150156PubMedCrossRefGoogle Scholar
  5. Hart, CL, Smith, GD, Hole, DJ, Hawthorne, VM 1999Alcohol consumption and mortality from all causes, coronary heart disease, and stroke: Results from a prospective cohort study of Scottish men with 21 years of follow-up.Br Med J31817251729Google Scholar
  6. Farchi, G, Fidanza, F, Mariotti, S, Menotti, A 1992Alcohol and mortality in the Italian rural cohorts of the Seven Countries Study.Int J Epidemiol217482PubMedCrossRefGoogle Scholar
  7. Shaper, AG, Wannamethee, SG, Walther, M 1988Alcohol and mortality in British men: Explaining the U-shaped curve.Lancet212671273CrossRefPubMedGoogle Scholar
  8. Boucher, KM, Slattery, ML, Berry, TD, Quesenberry, C, Anderson, K 1998Statistical methods in epidemiology: A comparison of statistical methods to analyze dose-response and trend analysis in epidemiologic studies.J Clin Epidemiol5112231233CrossRefPubMedGoogle Scholar
  9. Hastie, T, Tibshirani, R 1990Generalized Additive Models.Chapman & HallLondonGoogle Scholar
  10. Grønbæk, M, Becker, U, Johansen, D,  et al. 2000Type of alcohol consumed and mortality from all causes, coronary heart disease, and cancer.Ann Intern Med133411419PubMedGoogle Scholar
  11. Klatsky, AL, Armstrong, MA, Friedman, GD 1992Alcohol and mortality.Ann Intern Med117646654PubMedGoogle Scholar
  12. Renaud, SC, Guéguen, R, Siest, G, Salamon, R 1999Wine, beer, and mortality in middle-aged men from eastern France.Arch Intern Med15918651870CrossRefPubMedGoogle Scholar
  13. Wannamethee, SG, Shaper, AG 1999Type of alcoholic drink and risk of major coronary heart disease events and all- cause mortality.Am J Public Health89685690PubMedGoogle Scholar
  14. Rimm, EB, Giovannucci, EL, Wilett, WC 24-8-1991Prospective study of alcohol consumption and risk of coronary disease in menLancet338464468CrossRefGoogle Scholar
  15. Appleyard, M, Hansen, AT, Schnohr, P 1989The Copenhagen city heart study. A book of tables with data from the first examination (1976–78) and a 5-year follow-up (1981–83). Scand J Soc Med1701160Google Scholar
  16. Chambers J, Hastie T. Statistical models. S. Chapman & Hall/CRC, 1999.Google Scholar
  17. Dominici F, McDermott A, Hastie TJ. Issues in semi-parametric regression with applications in time series models for air pollution and mortality. Technical Report, Department of Biostatistics, John Hopkins University, Baltimore MD, 2003.Google Scholar
  18. Ramsay, TO, Burnett, RT, Krewski, D 2003The effect of concurvity in generalized additive models linking mortality to ambient particulate matter.Epidemiology141823CrossRefPubMedGoogle Scholar
  19. S-plus 6 for Windows Programmers Guide for Statistics, Insightful Corporation, Seattle, WA. 6. 2001.Google Scholar
  20. Dominici, F, McDermott, A, Zeger, SL, Samet, JM 2002On the use of generalized additive models in time-series studies of air pollution and health.Am J Epidemiol156193203CrossRefPubMedGoogle Scholar
  21. Hastie, TJ, Sleeper, L, Tibshirani, RJ 1992Flexible covariate effects in the Cox model.Breast Cancer Res Treat22241250PubMedCrossRefGoogle Scholar
  22. Dominici F, McDermott A, Hastie TJ. Improved semi-parametric time series models of air pollution and mortality. 2003. Unpublished work.Google Scholar
  23. Ambler, G, Royston, P, Head, J 2003Non-linear models for the relation between cardiovascular risk factors and intake of wine, beer and spirits.Stat Med22363383CrossRefPubMedGoogle Scholar
  24. Dorfman, A, Kimball, AW, Friedman, LA 1985Regression modeling of consumption or exposure variables classified by type.Am J Epidemiol12210961107PubMedGoogle Scholar
  25. Kimball, AW, Friedman, LA, Moore, RD 1-6-1992Nonlinear modeling of alcohol consumption for analysis of beverage type effects and beverage preference effectsAm J Epidemiol13512871292Google Scholar
  26. Wannamethee, SG, Shaper, AG 1998Alcohol, coronary heart disease and stroke: An examination of the J-shaped curve.Neuroepidemiology17288295CrossRefPubMedGoogle Scholar
  27. Shinton, R, Sagar, G, Beevers, G 1993The relation of alcohol consumption to cardiovascular risk factors and stroke. The west Birmingham stroke projectJ Neurol Neurosurg Psychiat56458462PubMedCrossRefGoogle Scholar
  28. Cryer, PC, Saunders, J, Jenkins, LM, Neale, H, Cook, AC, Peters, TJ 2001Clusters within a general adult population of alohol abstainers.Int J Epidemiol30756765CrossRefPubMedGoogle Scholar
  29. Wannamethee, SG, Shaper, AG 2003Men who do not drink: A report from the British Regional Heart Study.Int J Epidemiol17307316CrossRefGoogle Scholar
  30. Becker, U, Grønbæk, M, Johansen, D, Sørensen, TI 2002Lower risk for alcohol-induced cirrhosis in wine drinkers.Hepatology35868875CrossRefPubMedGoogle Scholar
  31. Mertens, JR, Moos, RH, Brennan, PL 1996Alcohol consumption, life context, and coping predict mortality among late-middle-aged drinkers and former drinkers.Alcohol Clin Exp Res20313319PubMedCrossRefGoogle Scholar
  32. Suadicani, P, Hein, HO, Gyntelberg, F 2001Socioeconomic status and ischaemic heart disease mortality in middle-aged men: Importance of the duration of follow-up. The Copenhagen Male StudyInt J Epidemiol30248255CrossRefPubMedGoogle Scholar
  33. Nielsen NR, Thygesen LC, Johansen D, Jensen G, Grønbæk M. The influence of duration of follow-up on the interpretation of estimates from prospective studies. Alcohol and cause-specific mortality in The Copenhagen City Heart Study. Ann Epidemiol. 2004. In press.Google Scholar
  34. Wannamethee, SG, Shaper, AG, Whincup, PH, Walker, M 1999Role of risk factors for major coronary heart disease events with increasing length of follow up.Heart81374379PubMedGoogle Scholar
  35. Gmel, G, Guthahr, E, Rehm, J 2003How stable is the risk curve between alcohol and all-cause mortality and what factors influence the shape? A precision-weighted hierarchical meta-analysisEur J Epidemiol18631642CrossRefPubMedGoogle Scholar
  36. Corrao, G, Rubbiati, L, Bagnardi, V, Zambon, A, Poikolainen, K 2000Alcohol and coronary heart disease: A meta-analysis.Addiction9515051523CrossRefPubMedGoogle Scholar
  37. White, IR, Altmann, DR, Nanchahal, K 2002Alcohol consumption and mortality: Modelling risk for men and women at different ages.Br Med J325191194CrossRefGoogle Scholar
  38. Marques-Vidal, P, Ducimetiere, P, Evans, A, Cambou, J, Arveiler, D 1996Alcohol consumption and myocardial infarction: A case-control study in France and northern Ireland.Am J Epidemiol14310891093PubMedGoogle Scholar
  39. Hertog, MG, Feskens, EJ, Hollman, PC, Katan, MK, Kromhout, D 1993Dietary antioxidant flavonoids and risk of coronary heart disease: The Zutphen elderly study.Lancet34210071011CrossRefPubMedGoogle Scholar
  40. Tjønneland, A, Grønbæk, M, Stripp, C, Overvad, K 1999Wine intake and diet in a random sample of 48,763 Danish men and women.Am J Clin Nutr694954PubMedGoogle Scholar
  41. Grønbæk, M, Mortensen, EL, Mygind, K,  et al. 1999Beer, wine, spirits and subjective health.J Epidemiol Commun Health53721724CrossRefGoogle Scholar
  42. Mortensen, EL, Jensen, HH, Sanders, SA, Reinisch, JM 2001Better psychological functioning and higher social status may largely explain the apparent health benefits of wine.Arch Intern Med16118441848CrossRefPubMedGoogle Scholar
  43. Rehm, J, Bondy, S 1998Alcohol and all-cause mortality: An overview.Novart Found Symp216223232CrossRefGoogle Scholar
  44. Mukamal, KJ, Conigrave, MB, Mittleman, MA,  et al. 2003Roles of drinking pattern and type of alcohol consumed in coronary heart disease in men.N Engl J Med348109118CrossRefPubMedGoogle Scholar
  45. Gaziano, JM, Buring, JE 1998Alcohol intake, lipids and risks of myocardial infarction.Novart Found Symp2168695CrossRefGoogle Scholar
  46. Grønbæk, M, Heitmann, BL 1996Validity of self-reported intakes of wine, beer and spirits in population studies.Eur J Clin Nutri50487490Google Scholar

Copyright information

© Springer 2005

Authors and Affiliations

  • Ditte Johansen
    • 1
    • 2
  • Morten Grønbæk
    • 1
    • 2
  • Kim Overvad
    • 3
  • Peter Schnohr
    • 4
  • Per Kragh Andersen
    • 5
    • 6
  1. 1.Danish Epidemiologic Science Centre at the Institute of Preventive MedicineCopenhagen University HospitalDenmark
  2. 2.Centre for Alcohol ResearchNational Institute of Public HealthCopenhagenDenmark
  3. 3.Department of Clinical Epidemiology, Department of Epidemiology and Social MedicineAalborg Hospital, Aarhus UniversityDenmark
  4. 4.The Copenhagen City Heart Study, Epidemiologic Research UnitCopenhagen University HospitalBispebjerg
  5. 5.Department of Biostatistics, Faculty of Health ScienceUniversity of CopenhagenDenmark
  6. 6.Danish Epidemiologic Science Centre at the Statens Serum InstitutCopenhagenDenmark

Personalised recommendations