Systematic handling of missing data in complex study designs – experiences from the Health 2000 and 2011 Surveys

Journal of Applied Statistics

Volumn 43, Issue 15, 2016, Pages 2772-2790

  Härkänen, Tommi ^a   Karvanen, Juha ^b   Tolonen, Hanna ^a   Lehtonen, Risto ^c   Djerf, Kari ^d   Juntunen, Teppo ^a   Koskinen, Seppo ^a  

^a NATIONAL PUBLIC HEALTH INSTITUTE   (Finland)

^b UNIVERSITY OF JYVÄSKYLÄ   (Finland)

^c UNIVERSITY OF HELSINKI   (Finland)

^d Statistics Finland   (Finland)

Author keywords

causal model with design; doubly robust methods; inverse probability weighting; multiple imputation; Non participation; non response

Indexed keywords

EID: 84958767739     PISSN: 02664763     EISSN: 13600532     Source Type: Journal    
DOI: 10.1080/02664763.2016.1144725     Document Type: Article

References (66)

1. H.Akaike, Maximum likelihood identification of gaussian autoregressive moving average models, Biometrika 60 (1973), pp. 255–265. doi:10.1093/biomet/60.2.255
2. A.Alkerwi, N.Sauvageot, S.Couffignal, A.Albert, M.L.Lair, and M.Guillaume, Comparison of participants and non-participants to the ORISCAV-LUX population-based study on cardiovascular risk factors in Luxembourg, BMC Med. Res. Methodol. 10 (2010), p. 80. doi:10.1186/1471-2288-10-80
3. A.Aromaa and S.Koskinen, (eds.), Health and Functional Capacity in Finland:Baseline Results of the Health 2000 Health Examination Survey, No. 12 in B, National Public Health Institute, 2004, Available at http://terveys2000.fi/julkaisut/baseline.pdf.
4. P.C.Austin, The performance of different propensity-score methods for estimating relative risks, J. Clin. Epidemiol. 61 (2008), pp. 537–545. doi:10.1016/j.jclinepi.2007.07.011
5. H.Bang and J.M.Robins, Doubly robust estimation in missing data and causal inference models, Biometrics 61 (2005), pp. 962–973, Available at http://www.jstor.org/stable/3695907. doi:10.1111/j.1541-0420.2005.00377.x
6. M.A.Brookhart, S.Schneeweiss, K.J.Rothman, R.J.Glynn, J.Avorn, and T.Stürmer, Variable selection for propensity score models, Amer. J. Epidemiol. 163 (2006), pp. 1149–1156. doi:10.1093/aje/kwj149
7. C.M.Cassel, C.E.Sarndal, and J.H.Wretman, Some uses of statistical models in connection with the nonresponse problem, in Incomplete Data in Sample Surveys, Vol. 3, W.G. Madow and I. Olkin, eds., Academic Press, New York, 1983, pp. 143–160.
8. P.Chou, H.S.Kuo, C.H.Chen, and H.C.Lin, Characteristics of non-participants and reasons for non-participation in a population survey in Kin-Hu, Kinmen, European Journal of Epidemiology 13 (1997), pp. 195–200. doi:10.1023/A:1007384525568
9. G.Cohen and J.C.Duffy, Are nonrespondents to health surveys less healthy than respondents?, Journal of Official Statistics 18 (2002), pp. 13–23.
10. L.M.Collins, J.L.Schafer, and C.M.Kam, A comparison of inclusive and restrictive strategies in modern missing data procedures, Psychol. Methods 6 (2001), pp. 330–351. doi:10.1037/1082-989X.6.4.330
11. M.H.Criqui, M.Austin, and E.Barrett-Connor, The effect of non-response on risk ratios in a cardiovascular disease study, J. Chronic Diseases 32 (1979), pp. 633–638. doi:10.1016/0021-9681(79)90093-6
12. K.Djerf, J.Laiho, R.Lehtonen, T.Härkänen, and P.Knekt, Methodology report Health 2000 Survey, Weighting and Statistical Analysis, no. 26 in B, KTL, 2008, pp. 182–200, Available at http://www.terveys2000.fi/doc/methodologyrep.pdf.
13. T.Drivsholm, L.F.Eplov, M.Davidsen, T.Jörgensen, H.Ibsen, H.Hollnagel, and K.Borch-Johnsen, Representativeness in population-based studies:A detailed description of non-response in a Danish cohort study, Scand. J. Public Health 34 (2006), pp. 623–631. doi:10.1080/14034940600607616
14. A.Gelman, Struggles with survey weighting and regression modeling, Statist. Sci. 22 (2007), pp. 153–164. doi:10.1214/088342306000000691
15. K.M.Gorey and M.Trevisan, Secular trends in the United States black/white hypertension prevalence ratio:Potential impact of diminishing response rates, Amer. J. Epidemiol. 147 (1998), pp. 95–99. doi:10.1093/oxfordjournals.aje.a009434
16. J.Gundgaard, O.Ekholm, E.H.Hansen, and N.K.Rasmussen, The effect of non-response on estimates of health care utilisation:Linking health surveys and registers, Eur. J. Public Health 18 (2008), pp. 189–194. doi:10.1093/eurpub/ckm103
17. M.Hara, S.Sasaki, T.Sobue, S.Yamamoto, and S.Tsugane, Comparison of cause-specific mortality between respondents and nonrespondents in a population-based prospective study:Ten-year follow-up of JPHC study cohort i, J. Clin. Epidemiol. 55 (2002), pp. 150–156. doi:10.1016/S0895-4356(01)00431-0
18. T.Härkänen, J.Karvanen, H.Tolonen, R.Lehtonen, K.Djerf, T.Juntunen, and S.Koskinen, Supplement to ‘systematic handling of missing data in complex study designs – experiences from the health 2000 and 2011 surveys’.
19. Y.He, A.M.Zaslavsky, M.B.Landrum, D.P.Harrington, and P.Catalano, Multiple imputation in a large-scale complex survey:A practical guide, Stat. Methods Med. Res. 19 (2010), pp. 653–670, Available at http://dx.doi.org/10.1177/0962280208101273.
20. J.J.Heckman, Sample selection bias as a specification error, Econometrica:J. Econometric Soc. (1979), pp. 153–161. doi:10.2307/1912352
21. D.W.Hosmer, T.Hosmer, S.Le Cessie, and S.Lemeshow, A comparison of goodness-of-fit tests for the logistic regression model, Stat. Med. 16 (1997), pp. 965–980. doi:10.1002/(SICI)1097-0258(19970515)16:9<965::AID-SIM509>3.0.CO;2-O
22. R.Jackson, L.E.Chambless, K.Yang, T.Byrne, R.Watson, A.Folsom, E.Shahar, and W.Kalsbeek, Differences between respondents and nonrespondents in a multicenter community-based study vary by gender and ethnicity, J. Clin. Epidemiol. 49 (1996), pp. 1441–1446. doi:10.1016/0895-4356(95)00047-X
23. P.Jousilahti, V.Salomaa, K.Kuulasmaa, M.Niemela, and E.Vartiainen, Total and cause specific mortality among participants and non-participants of population based health surveys:A comprehensive follow up of 54 372 Finnish men and women, J. Epidemiol. Community Health 59 (2005), pp. 310–315. doi:10.1136/jech.2004.024349
24. J.Karvanen, Study design in causal models, Scand. J. Stat. 42 (2015), pp. 361–377, Available at http://arxiv.org/abs/1211.2958. doi:10.1111/sjos.12110
25. Kela, Statistics on disability benefits and services provided by Kela (2015), Available at http://www.kela.fi/web/en/statistics-by-topic_statistics-on-disability-benefits-provided-by-kela.
26. Kela, Statistics on reimbursement entitlements in respect of medicines (2015), Available at http://www.kela.fi/web/en/statistics-by-topic_statistics-on-reimbursement-entitlements-in-respect-of-medicines
27. J.K.Kim, J.Michael Brick, W.A.Fuller, and G.Kalton, On the bias of the multiple-imputation variance estimator in survey sampling, J. R. Statist. Soc:Ser. B (Statist. Methodol.) 68 (2006), pp. 509–521. doi:10.1111/j.1467-9868.2006.00546.x
28. M.Kjoller and H.Thoning, Characteristics of non-response in the Danish health interview surveys, 1987–1994, Eur. J. Public Health 15 (2005), pp. 528–535. doi:10.1093/eurpub/cki023
29. A.K.Knudsen, M.Hotopf, J.C.Skogen, S.Overland, and A.Mykletun, The health status of nonparticipants in a population-based health study:The Hordaland health study, Amer. J. Epidemiol. 172 (2010), pp. 1306–1314. doi:10.1093/aje/kwq257
30. K.Korkeila, S.Suominen, J.Ahvenainen, A.Ojanlatva, P.Rautava, H.Helenius, and M.Koskenvuo, Non-response and related factors in a nation-wide health survey, Eur. J. Epidemiol. 17 (2001), pp. 991–999. doi:10.1023/A:1020016922473
31. S.Koskinen, A.Lundqvist, and N.Ristiluoma, (eds.), Terveys, toimintakyky ja hyvinvointi Suomessa 2011, no. 68 in Raportti, National Institute for Health and Welfare, 2012, Available at http://urn.fi/URN:ISBN:978-952-245-769-1.
32. J.Laiho, K.Djerf, and R.Lehtonen, Methodology report:Health 2000 Survey, Sampling Design, no. 26 in B, National Public Health Institute, 2008, pp. 13–15, Available at http://www.terveys2000.fi/doc/methodologyrep.pdf.
33. S.B.Larsen, S.O.Dalton, J.Schüz, J.Christensen, K.Overvad, A.Tönneland, C.Johansen, and A.Olsen, Mortality among participants and non-participants in a prospective cohort study, Eur. J. Epidemiol. 27 (2012), pp. 837–845. doi:10.1007/s10654-012-9739-x
34. J.M.Last, A Dictionary of Epidemiology, 4th ed., Oxford University Press, New York, 2001.
35. R.Lehtonen and E.Pahkinen, Practical Methods for Design and Analysis of Complex Surveys, Wiley, Chichester, 2004.
36. R.J.Little, Missing-data adjustments in large surveys, J. Bus. Econ. Stat. 6 (1988), pp. 287–296.
37. R.J.Little and N.Zhang, Subsample ignorable likelihood for regression analysis with missing data, J. R. Statist. Soc:Ser. C (Appl. Stat.) 60 (2011), pp. 591–605. doi:10.1111/j.1467-9876.2011.00763.x
38. A.J.V.Loon, M.Tijhuis, H.S.Picavet, P.G.Surtees, and J.Ormel, Survey non-response in the netherlands:Effects on prevalence estimates and associations, Ann. Epidemiol. 13 (2003), pp. 105–110. doi:10.1016/S1047-2797(02)00257-0
39. I.Lundberg, K.D.Thakker, T.Hallstrom, and Y.Forsell, Determinants of non-participation, and the effects of non-participation on potential cause-effect relationships, in the part study on mental disorders, Social Psychiatry Psychiatric Epidemiol. 40 (2005), pp. 475–483. doi:10.1007/s00127-005-0911-4
40. P.Lynn, Methodology of Longitudinal Surveys, John Wiley & Sons, Chichester, 2009.
41. G.Molenberghs and M.G.Kenward, Missing Data in Medical Research, Wiley, Chichester, 2007.
42. J.Pearl, Causal inference in statistics:An overview, Stat. Surv. 3 (2009), pp. 96–146, Available at http://projecteuclid.org/euclid.ssu/1255440554. doi:10.1214/09-SS057
43. J.Pearl, Causality:Models, Reasoning, and Inference, 2nd ed., Cambridge University Press, New York, 2009.
44. C.P.Robert and G.Casella, Monte Carlo Statistical Methods, 2nd ed., Springer, New York, 2004.
45. J.M.Robins, A.Rotnitzky, and L.P.Zhao, Estimation of regression coefficients when some regressors are not always observed, J. Amer. Statist. Assoc. 89 (1994), pp. 846–866, Available at http://amstat.tandfonline.com/doi/full/10.1080/01621459.1994.10476818. doi:10.1080/01621459.1994.10476818
46. D.B.Rubin, Inference and missing data, Biometrika 63 (1976), pp. 581–592. doi:10.1093/biomet/63.3.581
47. D.B.Rubin, Statistical matching using file concatenation with adjusted weights and multiple imputations, J. Bus. Econ. Stat. 4 (1986), pp. 87–94.
48. D.B.Rubin, Multiple Imputation for Nonresponse in Surveys, Wiley, New York, 1987.
49. J.L.Schafer, Multiple imputation:A primer, Statist. Methods Med. Res. 8 (1999), pp. 3–15. doi:10.1191/096228099671525676
50. G.Schwarz, Estimating the dimension of a model, Ann. Stat. 6 (1978), pp. 461–464, Available at http://www.jstor.org/stable/2958889. doi:10.1214/aos/1176344136
51. S.Seaman, J.Galati, D.Jackson, and J.Carlin, What is meant by ‘missing at random’, Statist. Sci. 28 (2013), pp. 257–268. doi:10.1214/13-STS415
52. S.R.Seaman and I.R.White, Review of inverse probability weighting for dealing with missing data, Statist. Methods Med. Res. 22 (2013), pp. 278–295. doi:10.1177/0962280210395740
53. S.R.Seaman, I.R.White, A.J.Copas, and L.Li, Combining multiple imputation and inverse-probability weighting, Biometrics 68 (2012), pp. 129–137. doi:10.1111/j.1541-0420.2011.01666.x
54. A.J.Sögaard, R.Selmer, E.Bjertness, and D.Thelle, The Oslo health study:The impact of self-selection in a large, population-based survey, Int. J. Equity Health 3 (2004), p. 3. doi:10.1186/1475-9276-3-3
55. E.Shahar, A.R.Folsom, and R.Jackson, The effect of nonresponse on prevalence estimates for a referent population:Insights from a population-based cohort study, Ann. Epidemiol. 6 (1996), pp. 498–506. doi:10.1016/S1047-2797(96)00104-4
56. M.Spratt, J.Carpenter, J.A.Sterne, J.B.Carlin, J.Heron, J.Henderson, and K.Tilling, Strategies for multiple imputation in longitudinal studies, Amer. J. Epidemiol. 172 (2010), pp. 478–487. doi:10.1093/aje/kwq137
57. M.A.Tanner and W.H.Wong, The calculation of posterior distributions by data augmentation, J. Amer. Statist. Assoc. 82 (1987), pp. 528–540. doi:10.1080/01621459.1987.10478458
58. T.Therneau, B.Atkinson, and B.Ripley, rpart:Recursive Partitioning and Regression Trees (2015), Available at http://CRAN.R-project.org/package=rpart, R package version 4.1-10.
59. THL, Care register for health care (2015), Available at http://www.thl.fi/en_US/web/en/statistics/information/register_descriptions/careregister_healthcare
60. A.Thomas, B.O'Hara, U.Ligges, and S.Sturtz, Making BUGS open, R News 6 (2006), pp. 12–17.
61. H.Tolonen, A.Dobson, and S.Kulathinal, Effect on trend estimates of the difference between survey respondents and non-respondents:Results from 27 populations in the who monica project, Eur. J. Epidemiol. 20 (2005), pp. 887–898. doi:10.1007/s10654-005-2672-5
62. A.M.Tolppanen, H.Taipale, M.Koponen, P.Lavikainen, A.Tanskanen, J.Tiihonen, and S.Hartikainen, Use of existing data sources in clinical epidemiology:Finnish health care registers in Alzheimer's disease research–the medication use among persons with Alzheimer's disease (MEDALZ-2005) study, Clin. Epidemiol. 5 (2013), p. 277. doi:10.2147/CLEP.S46622
63. S.Van Buuren, H.C.Boshuizen, and D.L.Knook, et al., Multiple imputation of missing blood pressure covariates in survival analysis, Stat. Med. 18 (1999), pp. 681–694. doi:10.1002/(SICI)1097-0258(19990330)18:6<681::AID-SIM71>3.0.CO;2-R
64. S.Van Buuren and K.Groothuis-Oudshoorn, mice:Multivariate imputation by chained equations in R, J. Statist. Softw. 45 (2011), pp. 1–67, Available at http://www.jstatsoft.org/v45/i03/.
65. E.von Elm, D.G.Altman, M.Egger, S.J.Pocock, P.C.Gøtzsche and J.P.Vandenbroucke, and for the STROBE Initiative, The strengthening the reporting of observational studies in epidemiology (STROBE) statement:Guidelines for reporting observational studies, Epidemiology 18 (2007), pp. 800–804. doi:10.1097/EDE.0b013e3181577654
66. I.R.White, P.Royston, and A.M.Wood, Multiple imputation using chained equations:Issues and guidance for practice, Stat. Med. 30 (2011), pp. 377–399. doi:10.1002/sim.4067