Abstract
Data obtained from calibration exercises are used to assess the level of agreement between examiners (and the benchmark examiner) and/or between repeated examinations by the same examiner in epidemiological surveys or large-scale clinical studies. Agreement can be measured using different techniques: kappa statistic, percentage agreement, dice coefficient, sensitivity and specificity. Each of these methods shows specific characteristics and has its own shortcomings. The aim of this contribution is to critically review techniques for the measurement and analysis of examiner agreement and to illustrate this using data from a recent survey in young children, the Smile for Life project. The above-mentioned agreement measures are influenced (in differing ways and extents) by the unit of analysis (subject, tooth, surface level) and the disease level in the validation sample. These effects are more pronounced for percentage agreement and kappa than for sensitivity and specificity. It is, therefore, important to include information on unit of analysis and disease level (in validation sample) when reporting agreement measures. Also, confidence intervals need to be included since they indicate the reliability of the estimate. When dependency among observations is present [as is the case in caries experience data sets with typical hierarchical structure (surface–tooth–subject)], this will influence the width of the confidence interval and should therefore not be ignored. In this situation, the use of multilevel modelling is necessary. This review clearly shows that there is a need for the development of guidelines for the measurement, interpretation and reporting of examiner reliability in caries experience surveys.
Similar content being viewed by others
References
Pitts NB, Evans DJ, Pine CM (1997) British Association for the Study of Community Dentistry (BASCD) diagnostic criteria for caries prevalence surveys-1996/97. Community Dent Health 14(Suppl 1):6–9
Ismail AI, Sohn W, Tellez M, Amaya A, Sen A, Hasson H, Pitts NB (2007) The International Caries Detection and Assessment System (ICDAS): an integrated system for measuring dental caries. Community Dent Oral Epidemiol 35:170–178
World Health Organization (1997) Oral health surveys. Basic methods. World Health Organization, Geneva
Pine CM, Pitts NB, Nugent ZJ (1997) British Association for the Study of Community Dentistry (BASCD) guidance on the statistical aspects of training and calibration of examiners for surveys of child dental health. A BASCD coordinated dental epidemiology programme quality standard 2. Community Dent Health 14(Suppl 1):18–29
International Caries Detection and Assessment System Coordinating Committee (2005) Criteria manual - International Caries Detection and Assessment System (ICDAS II)
Maxwell AE (1970) Comparing the classification of subjects by two independent judges. Br J Psychiatry 116:651–655
McNemar Q (1947) Note on the sampling error of the difference between correlated proportions or percentages. Psychometrika 12:153–157
Stuart AA (1955) A test for homogeneity of the marginal distributions in a two-way classification. Biometrika 42:412–416
Cicchetti DV, Allison T (1971) A new procedure for assessing reliability of scoring EEG sleep recordings. Am J EEG Technol 11:101–109
Cohen J (1968) Weighted kappa: Nominal scale agreement provision for scaled disagreement or partial credit. Psychol Bull 70:213–220
Kingman A (1986) A procedure for evaluating the reliability of a gingivitis index. J Clin Periodontol 13:385–391
Uebersax JS (1993) Statistical modeling of expert ratings on medical treatment appropriateness. J Am Stat Assoc 88:421–427
Uebersax JS (1987) Diversity of decision-making models and the measurement of interrater agreement. Psychol Bull 101:140–146
Stamm JW, Stewart PW, Bohannan HM, Disney JA, Graves RC, Abernathy JR (1991) Risk assessment for oral diseases. Adv Dent Res 5:4–17
Dice LR (1945) Measures of the amount of ecologic association between species. Ecology 26:297–302
Cohen J (1960) A coefficient of agreement for nominal scales. Educ Psychol Meas 20:37–46
Landis JR, Koch GG (1977) The measurement of observer agreement for categorical data. Biometrics 33:159–174
Tooth LR, Ottenbacher KJ (2004) The kappa statistic in rehabilitation research: an examination. Arch Phys Med Rehabil 85:1371–1376
Gwet KL (2008) Computing inter-rater reliability and its variance in the presence of high agreement. Br J Math Stat Psychol 61:29–48
Uebersax JS (2005) Statistical methods for rater agreement: the tetrachoric and polychoric correlation coefficient. http://www.john-uebersax.com/stat/tetra.htm. Accessed 10 May 2010
Hutchinson TP (1993) Focus on psychometrics. Kappa muddles together two sources of disagreement: tetrachoric correlation is preferable. Res Nurs Health 16:313–316
Byrt T, Bishop J, Carlin JB (1993) Bias, prevalence and kappa. J Clin Epidemiol 46:423–429
Declerck D, Leroy R, Martens L, Lesaffre E, Garcia-Zattera MJ, Vanden BS, Debyser M, Hoppenbrouwers K (2008) Factors associated with prevalence and severity of caries experience in preschool children. Community Dent Oral Epidemiol 36:168–178
Assaf AV, de Castro MM, Zanin L, Tengan C, Pereira AC (2006) Effect of different diagnostic thresholds on dental caries calibration—a 12-month evaluation. Community Dent Oral Epidemiol 34:213–219
Assaf AV, Tagliaferro EP, Meneghim MC, Tengan C, Pereira AC, Ambrosano GM, Mialhe FL (2007) A new approach for interexaminer reliability data analysis on dental caries calibration. J Appl Oral Sci 15:480–485
Vanobbergen J, Lesaffre E, Garcia-Zattera MJ, Jara A, Martens L, Declerck D (2007) Caries patterns in primary dentition in 3-, 5- and 7-year-old children: spatial correlation and preventive consequences. Caries Res 41:16–25
Williamson JM, Datta S, Satten GA (2003) Marginal analyses of clustered data when cluster size is informative. Biometrics 59:36–42
Hoehler FK (2000) Bias and prevalence effects on kappa viewed in terms of sensitivity and specificity. J Clin Epidemiol 53:499–503
Spitznagel EL, Helzer JE (1985) A proposed solution to the base rate problem in the kappa statistic. Arch Gen Psychiatry 42:725–728
Maclure M, Willett WC (1987) Misinterpretation and misuse of the kappa statistic. Am J Epidemiol 126:161–169
Braga MM, Oliveira LB, Bonini GA, Bonecker M, Mendes FM (2009) Feasibility of the International Caries Detection and Assessment System (ICDAS-II) in epidemiological surveys and comparability with standard World Health Organization criteria. Caries Res 43:245–249
Fyffe HE, Deery C, Nugent ZJ, Nuttall NM, Pitts NB (2000) Effect of diagnostic threshold on the validity and reliability of epidemiological caries diagnosis using the Dundee Selectable Threshold Method for caries diagnosis (DSTM). Community Dent Oral Epidemiol 28:42–51
Gilthorpe MS, Griffiths GS, Maddick IH, Zamzuri AT (2000) The application of multilevel modelling to periodontal research data. Community Dent Health 17:227–235
Tu YK, Gilthorpe MS, Griffiths GS, Maddick IH, Eaton KA, Johnson NW (2004) The application of multilevel modeling in the analysis of longitudinal periodontal data-part I: absolute levels of disease. J Periodontol 75:127–136
Williams FM, Nan G (2006) Estimation of sensitivity and specificity of clustered binary data. Statistics and data analysis, SUGI 31 Proceedings, SAS Proceedings
Burnside G, Pine CM, Williamson PR (2007) The application of multilevel modelling to dental caries data. Stat Med 26:4139–4149
Mutsvari T, Lesaffre E, Garcia-Zattera MJ, Diya L, Declerck D (2010) Factors that influence data quality in caries experience detection: a multilevel modeling approach. Caries Res 44:438–444
Williamson JM, Lipsitz SR, Manatunga AK (2000) Modeling kappa for measuring dependent categorical agreement data. Biostatistics 1:191–202
Liang KY, Zeger SL (1986) Longitudinal data analysis using generalized linear models. Biometrika 73:13–22
Lesaffre E, Mwalili SM, Declerck D (2004) Analysis of caries experience taking inter-observer bias and variability into account. J Dent Res 83:951–955
Barnhart HX, Williamson JM (2002) Weighted least-squares approach for comparing correlated kappa. Biometrics 58:1012–1019
Lin HM, Williamson JM, Lipsitz SR (2003) Calculating power for the comparison of dependent κ-coefficients. J R Stat Soc C Appl Stat 52:391–404
Wacholder S, Armstrong B, Hartge P (1993) Validation studies using an alloyed gold standard. Am J Epidemiol 137:1251–1258
Brenner H (1996) Correcting for exposure misclassification using an alloyed gold standard. Epidemiology 7:406–410
Conflict of interest
The authors declare that they have no conflict of interest.
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
Calculation of kappa from sensitivity, specificity and disease prevalence.
Let Y* = examiner score, Y = benchmark score
j and k take values of 0 and 1 e.g. if j = 1 and k = 1we obtain \( {P_{{11}}} = P\left( {Y* = {1},\;Y = {1}} \right) = P\left( {{{{Y* = {1}}} \left/ {{Y = {1}}} \right.}} \right) \times P\left( {Y = {1}} \right) \)
if j = 0 and k = 0 we obtain \( {P_{00}} = P\left( {Y* = 0,\;Y = 0} \right) = P\left( {{{{Y* = 0}} \left/ {{Y = 0}} \right.}} \right) \times P\left( {Y = 0} \right) \)
Similarly P 01 and P 10 are obtained
Rights and permissions
About this article
Cite this article
Agbaje, J.O., Mutsvari, T., Lesaffre, E. et al. Measurement, analysis and interpretation of examiner reliability in caries experience surveys: some methodological thoughts. Clin Oral Invest 16, 117–127 (2012). https://doi.org/10.1007/s00784-010-0475-x
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00784-010-0475-x