An ad hoc method for dual adjusting for measurement errors and nonresponse bias for estimating prevalence in survey data: Application to Iranian mental health survey on any illicit drug use

Abstract

Purpose

The prevalence estimates of binary variables in sample surveys are often subject to two systematic errors: measurement error and nonresponse bias. A multiple-bias analysis is essential to adjust for both biases.

Methods

In this paper, we linked the latent class log-linear and proxy pattern-mixture models to adjust jointly for measurement errors and nonresponse bias with missing not at random mechanism. These methods were employed to estimate the prevalence of any illicit drug use based on Iranian Mental Health Survey data.

Results

After jointly adjusting for measurement errors and nonresponse bias in this data, the prevalence (95% confidence interval) estimate of any illicit drug use changed from 3.41 (3.00, 3.81)% to 27.03 (9.02, 38.76)%, 27.42 (9.04, 38.91)%, and 27.18 (9.03, 38.82)% under “missing at random,” “missing not at random,” and an intermediate mode, respectively.

Conclusions

Under certain assumptions, a combination of the latent class log-linear and binary-outcome proxy pattern-mixture models can be used to jointly adjust for both measurement errors and nonresponse bias in the prevalence estimation of binary variables in surveys.

Keywords

Bias adjustment binary-outcome proxy pattern-mixture model binary variable illicit drug use Iranian Mental Health Survey latent class analysis latent class log-linear model measurement error nonresponse bias survey data

Get full access to this article

View all access options for this article.

References

Lash

Fox

Fink

. Applying quantitative bias analysis to epidemiologic data, Springer Science & Business Media, 2011.

Rothman

Greenland

Lash

. Modern epidemiology. 3rd ed, Lippincott Williams & Wilkins, 2008.

Lash

Fox

MacLehose

et al.

Good practices for quantitative bias analysis. Int J Epidemiol 2014; 43: 1969–1985.

Blackwell

Honaker

King

. Multiple overimputation: a unified approach to measurement error and missing data, Working Paper, 2011., http://j.mp/jqdj72 (accessed 31 December 1969)

Berzofsky

Biemer

Kalsbeek

. A brief history of classification error models. Section Survey Res Methodes-JSM 2008, pp. 3667–3673, http://ww2.amstat.org/sections/srms/Proceedings/y2008/Files/302416.pdf.

Kasprzyk

. Measurement error in household surveys: sources and measurement. Household sample surveys in developing and transition countries, New York: United Nations Publication, 2005, pp. 171–198.

Biemer

Trewin

A review of measurement error effects on the analysis of survey data. In: Lyberg

Biemer

Collins

et al. (eds). Survey Measurement and Process Quality, John Wiley & Sons, 2012, pp. 601–632.

Gustafson

Greenland

. Misclassification. Handbook of epidemiology, New York: Springer, 2014, pp. 639–658.

Biemer

Wiesen

. Measurement error evaluation of self-reported drug use: a latent class analysis of the US National Household Survey on Drug Abuse. J R Statist Soc A 2002; 165: 97–119.

10.

Walter

Irwig

. Estimation of test error rates, disease prevalence and relative risk from misclassified data: a review. J Clin Epidemiol 1988; 41: 923–937.

11.

Brenner

. Use and limitations of dual measurements in correcting for nondifferential exposure misclassification. Epidemiology 1992; 3: 216–222.

12.

Hui

Walter

. Estimating the error rates of diagnostic tests. Biometrics 1980; 36: 167–171.

13.

Goodman

. The analysis of multidimensional contingency tables when some variables are posterior to others: a modified path analysis approach. Biometrika 1973; 60: 179–192.

14.

Haberman

. Analysis of qualitative data: new developments, 1979; Vol. 2, New York: Academic Press.

15.

Vermunt

. Log-linear models for event histories, advanced quantitative techniques in the social sciences series, Thousand Oaks, CA: SAGE Publications, Inc, 1997.

16.

Bose J (ed.). Nonresponse bias analyses at the National Center for Education Statistics. Proceedings of Statistics Canada Symposium 2001, http://www.fcsm.gov/committees/ihsng/StatsCan2_JB.pdf.

17.

Andridge

. Statistical methods for missing data in complex sample surveys, Doctoral dissertation, The University of Michigan, 2009.

18.

Andridge RR and Little RJ (eds). Extensions of proxy pattern-mixture analysis for survey nonresponse. Proceedings of the Survey Research Methods Section, American Statistical Association, 2009, pp.2468–2482.

19.

Andridge

Little

RJA

. Proxy pattern-mixture analysis for survey nonresponse. J Off Stat 2011; 27: 153–180.

20.

Nandram

Choi

. A Bayesian analysis of a proportion under non‐ignorable non‐response. Stat Med 2002; 21: 1189–1212.

21.

Nandram

Choi

. Hierarchical Bayesian nonresponse models for binary data from small areas with uncertainty about ignorability. J Am Stat Assoc 2002; 97: 381–388.

22.

Nandram

Han

Choi

. A hierarchical bayesian nonignorable nonresponse model for multinomial data from small areas. Survey Methodol 2002; 28: 145–156.

23.

Nandram

Liu

Choi

et al.

Bayesian non‐response models for categorical data from small areas: an application to BMD and age. Stat Med 2005; 24: 1047–1074.

24.

Stasny

. Hierarchical models for the probabilities of a survey classification and nonresponse: an example from the National Crime Survey. J Am Stat Assoc 1991; 86: 296–303.

25.

Little

Rubin

. Statistical analysis with missing data, 2nd ed. New York: Wiley, 2002.

26.

Little

. A class of pattern-mixture models for normal incomplete data. Biometrika 1994; 81: 471–483.

27.

Goodman

. Exploratory latent structure analysis using both identifiable and unidentifiable models. Biometrika 1974; 61: 215–231.

28.

Agresti

. Categorical data analysis, 2nd ed. Hoboken, NJ: John Wiley, 2002, pp. 721–721.

29.

Glymour

M. GS

. Causal diagrams. Modern epidemiology, 3rd ed. Philadelphia: Lippincott Williams & Wilkins, 2008.

30.

Jewell

. Statistics for epidemiology, London: CRC Press, 2003.

31.

Mansournia

Hernán

Greenland

. Matched designs and causal diagrams. Int J Epidemiol 2013; 42: 860–869.

32.

Mansournia

Higgins

Sterne

et al.

Biases in randomized trials: a conversation between trialists and epidemiologists. Epidemiology 2017; 28: 54–59.

33.

Pearl

. Causality, Cambridge: Cambridge University Press, 2009.

34.

Casella

Berger

. Statistical inference, Pacific Grove, CA: Duxbury, 2002.

35.

R Core Team.

. a language and environment for statistical computing, Vienna, Austria: R Foundation for Statistical Computing, 2015, https://www.R-project.org/.

36.

Duffy D. Package ‘gllm’: generalised log-linear model. 0.35 ed. CRAN, 2014, https://cran.r-project.org/.

37.

Rahimi-Movaghar

Amin-Esmaeili

Sharifi

et al.

Iranian Mental Health Survey: design and field procedures. Iran J Psychiat 2014; 9: 96–109.

38.

Turner CF, Lessler JT and Gfroerer JC. Survey measurement of drug use: methodological studies. United States of America, National Institute on Drug Abuse, Division of Epidemiology and Prevention Research, US Department of Health and Human Services, Public Health Service, Alcohol, Drug Abuse, and Mental Health Administration, 1992, p. 421.

39.

Harrison

. The validity of self-reported drug use in survey research: an overview and critique of research methods. NIDA Res Monogr 1997; 167: 17–36.

40.

Lin

Dayton

. Model selection information criteria for non-nested latent class models. J Educat Behavior Stat 1997; 22: 249–264.

41.

Dayton

Macready

. A scaling model with response errors and intrinsically unscalable respondents. Psychometrika 1980; 45: 343–356.

42.

Mansournia

Altman

. Inverse probability weighting. BMJ 2016; 352: i189–i189.

43.

Kachitvichyanukul

Schmeiser

. Binomial random variate generation. Commun ACM 1988; 31: 216–222.

44.

Toft

Jørgensen

Højsgaard

. Diagnosing diagnostic tests: evaluating the assumptions underlying the estimation of sensitivity and specificity in the absence of a gold standard. Prevent Veterin Med 2005; 68: 19–33.

45.

Berzofsky

Biemer

Kalsbeek

. Local dependence in latent class analysis of rare and sensitive events. Sociol Methods Res 2014; 43: 137–170.

46.

West

Little

RJA

. Non-response adjustment of survey estimates based on auxiliary variables subject to error. Appl Statist 2013; 62: 213–231.

47.

Tourangeau

Yan

. Sensitive questions in surveys. Psychol Bull 2007; 133: 859–859.

48.

Zarrabi

Najafi

Kafi

et al.

Substance use among students of Guilan University of Medical Sciences in Iran in 2005-2006. Acta Medica Iranica 2009; 47: 473–478.

49.

Khosravi

Mousavi

Chaman

et al.

Crosswise model to assess sensitive issues: a study on prevalence of drug abuse among university students of Iran. Int J High Risk Behav Addict 2015, pp. 4: e24388. DOI: 10.5812/ijhrba.24388v2.

50.

Shamsipour

Yunesian

Fotouhi

et al.

Estimating the prevalence of illicit drug use among students using the crosswise model. Substance Use Misuse 2014; 49: 1303–1310.

51.

Ziaaddini

. The household survey of drug abuse in Kerman, Iran. J Appl Sci 2005; 5: 380–382.