Abstract
Background
We already showed the superiority of imputation of missing data (via Multivariable Imputation via Chained Equations (MICE) method) over exclusion of them; however, the methodology of MICE is complicated. Furthermore, easier imputation methods are available. The aim of this study was to compare them in terms of model composition and performance.
Methods
Three hundreds and ten breast cancer patients were recruited. Four approaches were applied to impute missing data. First we adopted an ad hoc method in which missing data for each variable was replaced by the median of observed values. Then 3 likelihood-based approaches were used. In the regression imputation, a regression model compared the variable with missing data to the rest of the variables. The regression equation was used to fill the missing data. The Expectation Maximum (E-M) algorithm was implemented in which missing data and regression parameters were estimated iteratively until convergence of regression parameters. Finally, the MICE method was applied. Models developed were compared in terms of variables significantly contributed to the multifactorial analysis, sensitivity and specificity.
Results
All candidate variables significantly contributed to the MICE model. However, grade of disease lost its effect in other three models. The MICE model showed the best performance followed by E-M model.
Conclusion
Among imputation methods, final models were not the same, in terms of composition and performance. Therefore, modern imputation methods are recommended to recover the information.
Keywords: Data, Multivariable imputation via chained equations, Expectation maximum algorithm, Breast cancer
Introduction
The problem of missing data arises in majority of medical data sets[1] An ad hoc method was shown to substitute missing data by a fixed value such as the mean (in the case of normally distributed data) or median of observed values (in the case of skewed data). This approach might artificially reduce the variance and affect the strength of relationships with other variables since all missing data are replaced by a single value.[2][3][4] Furthermore, in case-control studies, replacement of missing data with a fixed value increases the overlap between cases and controls and tends to underestimate the true association.[5]
Recent developments in the field of analysis of missing data, including Expectation Maximum (EM) algorithm and Multiple Imputation via Chained Equations (MICE), provided methods to deal with missing data adequately. These methods are likelihood based and use partially observed data to impute incomplete data. Although such methods provide better estimates, computer software skills are required and communication of methods with clinical audiences might not be simple. It has been noted that 'the optimal method should balance validity, ease of interpretability for readers and analysis expertise of the research team.[6]
We already used a breast cancer data set to illustrate different methodological issues.[7][8][9] In a recent work, we addressed the process of the MICE method and its superiority over the complete case analysis.[10]The philosophy behind the MICE model is not simple especially to clinical audience. On the other hand, easier imputation methods are available. There is evidence that, although easier methods involve some limitations, they might provide similar model to that of the MICE under special circumstances.
In this article, we first reviewed studies published in the literature to understand the situations in which easier methods worked as a good approximation for sophisticated methods. Then main aim of this study was to apply imputation methods and to compare them in terms of variables remain in the final model, sensitivity, and specificity. In addition, results were compared with that of the MICE.[10]
Materials and Methods
A total of 310 breast cancer patients formed the study population. Data were collected from Hospital-based cancer registry of Nemazee Hospital, affiliated to Shiraz University of Medical Sciences. The main outcome of study was Breast Cancer Specific Death (BCSD). Candidate variables for the multifactorial models were tumor stage with 3 levels (early, locally advanced, and advanced), tumor grade with 3 levels (1, 2 and 3), history of benign breast disease (positive versus negative), and age at diagnosis.[7][9][11][12]
Totally, 4 models were developed. The only difference between them was the approach utilised to deal with missing data (see below). In development of all models, Cox regression model was fitted to the data. Since none of the variables under study was continuous, no attempt was made to optimise the form of association between variables and BCSD.[13]In other words, no form of polynomial regression was needed. Models developed were as follows.
In the median model, for each variable, missing data were replaced by the median of observed values. For example stage variable had three categories 1, 2 and 3. Patients with an unknown stage were ignored, and the median of stage variables for the rest of patients was calculated. Unknown stages were then simply filled by the median.
In regression model, each variable including missing data, in turn, was considered as dependant variable, using other variables as independent ones. A linear regression model was then fitted only using cases with available data on all variables. For example, a regression model linked stage to grade, age, status, and history of benign disease. Missing data were estimated from model derived, following a rounding approach to the nearest plausible value. For example, for stage and grade, which had 3 levels, imputed values <1.5 were rounded to 1. Furthermore, values higher than 2 but lower than 2.5 were rounded to 2. For family history of benign form of disease and age at diagnosis, which had 2 levels, imputations <1.5 were rounded to 1 and values >1.5 were rounded to 2.
The E-M model was the extension of the regression model. Here, using cases with available data, regression models would be fitted. This model was used to impute the missing data. Then regression parameters would be re-estimated using new sets of data (i.e. using cases with available data plus imputed data). Again, new parameters of regression model (i.e. coefficients) would be estimated. Using ew regression coefficients, imputed values would be updated. These two steps would be continued iteratively until convergence. The whole process would be stopped when difference between regression coefficients in two successive steps became less than 10-6. Although it is necessary that data follow a multivariate normal distribution, it was suggested that for binary and categorical data, a rounding approach to the nearest possible value might work well in practice.[14]
Regression and E-M imputations methods replace each missing data by one single value. On the other hand, the MICE method was taken into account for the imputation uncertainty. Methodological issues behind the MICE model were illustrated elsewhere.[10]The key concept of the MICE method was to use the distribution of the observed data to estimate plausible values for the missing data and to incorporate random components into the estimated values so as to reflect their uncertainty. Here, each missing data was replaced by 10 values, thus creating 10 imputed data sets.[15] Each data set was analysed individually but e.13stimates derived from imputed data sets were combined applying Rubin's rule, to get a single Hazard Ratio (HR) and Confidence Interval (CI).[16]
Models developed were compared in terms of variables contributed significantly to them and estimated HR's. To compare the sensitivity and specificity of models, in each model, linear combination of variables multiplied into the estimated regression coefficients was calculated (i.e. risk score). Risk scores were dichotomised at median to divide patients into good and bad prognosis groups. Two by two tables were constructed using prognostic groups and patients' status. Sensitivity and specificity were calculated with respect to patients' status.
A series of packages which worked under R software(version 2.5.1) were used.[17] Missing data were imputed using MICE package.[18] Estimated regression coefficients and standard errors were combined across imputed data sets using Mitools library.[19] The E-M and regression imputations were done using SPSS software.
Results
Total number of BCSD was 56 (out of 310 patients). The numbers (percentages) of patients with missing value were as follows: Node status 63 (20.3%), grade 64 (20.6%), history of benign disease 47 (15.2%), and age at diagnosis 0 (0%).
In terms of variables contributed o the models, all four variables were significant in the MICE model (Table 1). However, in the E-M and median models, grade of disease lost its significant impact on BCSD. This was the case for age of diagnosis as well. Regression model produced poorest result. In the MICE model, the risk of death for patients with high stage was about 3 times higher than that of patients with low stage. This variable was of marginal significance in the regression model. On the other hand, the age variable was retained in the regression model.
Table 1. Comparison between imputation models in terms of composition and performance.
Variable | Level | Median model | Regression model | E-M model | MICE model | ||||
HRa (95% CIb) | Pvalue | HRa (95% CIb) | Pvalue | HRa(95% CIb) | Pvalue | ||||
Stage | 1 | 1 | 1 | 1 | 1 | ||||
2 | 3.79 (1.96, 7.33) | <0.001 | 2.57 (1.39, 4.75) | 0.003 | 3.84 (1.94, 7.22) | <0.001 | 3.13 (1.64, 5.97) | <0.001 | |
3 | 2.99 (1.24, 7.13) | 0.014 | 2.17 (0.94, 4.99) | 0.07 | 3.21 (1.35, 7.65) | 0.01 | 2.53 (1.05, 6.12) | 0.03 | |
Grade | 1 | 1 | 1 | 1 | 1 | ||||
2 | 1.69 (0.82, 3.49) | 0.16 | 2.03 (1, 4.10) | 0.05 | 1.56 (0.76, 3.19) | 0.22 | 2.46 (1.15, 5.24) | 0.02 | |
3 | 1.25 (0.56, 2.80) | 0.59 | 1.51 (0.67, 3.37) | 0.32 | 1.27 (0.56, 2.84) | 0.57 | 1.52 (0.65, 3.60) | 0.34 | |
Age | <48 | 1 | 1 | 1 | 1 | ||||
≥48 | 1.80 (0.95, 3.45) | 0.07 | 2.12 (1.41, 3.95) | 0.02 | 1.72 (0.89, 3.32) | 0.11 | 1.92 (1.01, 3.65) | 0.04 | |
Benign | No | 1 | 1 | 1 | 1 | ||||
Yes | 2.26 (1.25, 4.11) | 0.01 | 2.29 (1.27, 4.13) | 0.01 | 2.13 (1.15, 3.94) | 0.02 | 2.32 (1.24, 4.33) | 0.01 | |
Performance of models | |||||||||
Sensitivity | 23% | 36% | 88% | 88% | |||||
Specificity | 59% | 48% | 56% | 60% |
a HR: Hazard ratio
b CI: Confidence interval, SD: Standard deviation
In terms of performance, the MICE model was able to classify 49 died patients (out of 56) into bad prognosis group giving sensitivity of 0.88. Furthermore, out of 254 live patients, 151 ones were allocated to good prognosis group. This was corresponded to a specificity of 0.60. Results of he E-M model were fairly the same as MICE. However, regression and median substitution models produced poorest results. Sensitivity and specificity of these two models were 0.36 and 0.23 respectively. Corresponding figure for specificity was 0.48 and 0.59 respectively.
Discussion
Using an empirical data set, our results showed that different imputation methods led to different models in terms of composition and performance. It should be emphasized that we did not aim to perform complicated simulation studies, so as to study behaviour of imputation methods under different circumstances. We simply tried to show the impact of imputation method on modelling. This provided an excellent stage to explain to clinicians why different methods to handle missing data yielded different results.
In our data set, the MICE model gave highest sensitivity and specificity and retained all variables as being significant (Table 1). However, the communication of results with clinical audiences was not simple and especial software was required (Table 2).
Table 2. Advantages and disadvantages of methods to tackle missing data.
Feature | Complete-case | Median substitution | EM | MICE |
No special software is needed | Yes | Yes | Yes | No |
Easy to communicate with clinical audience | Yes | Yes | Yes | No |
Do not require distributional assumption | Yes | Yes | No | Yes |
Preserve data characteristics | No | No | Yes | Yes |
Convergence of imputation model is not an issue | Yes | Yes | No | No |
Takes imputation uncertainty into account | No | No | No | Yes |
Any particular problem | Diminishes the power Gives biased estimated if not MCAR | Artificially reduces the variance | Might give out of range estimates | Requires aggregation of estimates |
We have seen that results of the median model were not comparable with that of MICE. Neither composition of model nor performance was satisfying. It has been argued that although replacement of missing data with a single value does not hold characteristics of data, this ad hoc method is reasonable when missing rate is low.[20][21][22][23][24] Substitution of missing data by median of observed values was straightforward. However, in our data set, this model was only able to identify less than 40% of died patients. The specificity rate was less than 60%.
Results of the regression model were disappointing as well. Stage which was one of the most important prognostic factors of breast cancer was of marginal significance in this model. Furthermore, sensitivity and specificity of the model was as poor as that of median replacement model. This is because the effect of stage, which is known as one of the most important risk factor[20][21] did not reach a significant level.
The main disadvantage of EM method was that uncertainty in imputation of missing data was not taken into account. Furthermore, although EM preserves characteristics of data set[25] but might fail to converge when sample size relative to number of variables was low. In our application, performance of models derived applying EM and MICE methods were the same in terms of performance but not in terms of contribution of variables in the model. Age at diagnosis was not retained in the E-M model but it did not affect the performance of the model. This might indicate low contribution of this variable in terms of classification of patients.
With our experience, the main advantages and disadvantages of imputation methods were gathered in Table 2. Each technique had its own limitations. For example, median replacement was simple to be implemented but underestimated the true association. Our results suggested that the MICE method resulted into the best composition and performance. However, our findings can be considered only as a case study. This is because several items (such as sample size, nature of variables, and rate of missing data) affect performance of imputation models (see limitations of our work in the rest of the text). Therefore, here we reviewed the literature to compare our findings with them.
In a comprehensive study, it has been shown that the MICE method was the best technique to deal with missing data (see rest of the text).[26] However, there is controversy in the appropriateness of replacement of missing data by a fixed value. Usefulness of ad hoc methods, such as replacement with a fixed value, depends to a great extent to missing rate and content. Here, some examples were presented.
As an example, performance of complete-case analysis (C-C), mean replacement, and MICE imputation methods were compared.[23] Models were compared in terms of magnitude of estimated coefficients and standard error (SE), direction of association, and discrimination ability. Data for 398 cases with suspected pulmonary embolism were available of which 246 participants (62%) had complete information on all 26 variables studied. Rate of missing values were as follows: 0% for12 variables, <10% for 11 variables 14% for 1 variable and 21% for 2 variables. Variables selected in the complete-case analysis differed with other methods. Results of MICE were comparable with mean replacement because of low overall number of missing values. As expected, replacement of missing data by mean yielded smaller SE's.
On the other had, to address replacement of missing data by a single value was a bad practice, a total of 1000 samples of 500 subjects were generated.[5] True odds ratio (OR) between diagnostic test and disease status was 2.7. Omitting data under MCAR mechanism, diagnostic test for 20% of diseased and non-diseased subjects were omitted. When missing data were then replaced with overall mean, estimated OR was 1.73.
In another study, using a data set of a study on substance use among American Indian adolescents, artificial reduction in SE in the case of replacement of missing data by a single value was addressed.[4] In the original paper, 76% of cases with available data were analysed.[27] This gave mean at first use (SE) of 14.66 (0.19). Use of E-M or substitution of missing data by mean reduced SE to 0.16.
A sample of 492 patients with complete data, from a longitudinal study on the stress and health of elder adults was used to address the ability of list-wise deletion, mean substitution, and EM algorithm to handle missing data.[25] True mean (SD) for all cases (n=492) was 7.34 (7.28). Data on a single variable for 96 cases were dropped out (missing rate of 20%) under MAR assumption. Estimates corresponding to C-C and mean substitution were 6.47 (6.82) and 6.38 (6.12). EM gave best estimate of 6.79 (6.23). Furthermore, after application of EM method, correlation between variable with missing value and rest of variables were fairly similar to that of original data.
In a comprehensive study with binary outcomes, the ability of mean replacement, MI techniques, and complete case analysis were compared.[26] The data set of Society of Cardiothoracic Surgeons of Great Britain and Ireland was used. The original data included 20378 cases of which 1404 patients had died. Nine variables, out of 14 variables studied involved less than 15% missing values. The missing rate for 3 variables was higher than 20%. Actual missing rates were 20.2%, 42.7%, and 53.4%. In total, 32% had complete data on all variables studied. Authors found that, under MAR and MCAR mechanisms, performance of the MICE and conditional mean substitution were comparable and much better than complete-case analysis in terms of the proportion of patients classified into the correct risk group and the estimated spearman rank correlation between true and fitted probabilities. Estimated root mean square error (which quantified difference between fitted and true probabilities) for the mean substitution method was marginally higher than that of the MICE but better than complete-case analysis. Comparing estimated regression coefficients, it has been shown that the MICE produced the lowest level of bias.[26]
Application of ad hoc methods such as replacement with mean was criticised.[4] This might lead to artificially narrow confidence intervals. However, it has been suggested that when missing rate is low (about 10%), replacement of missing data by median or mean or exclusion of missing data is a reasonable approximation for the MICE, in terms of variables that contribute to the multifactorial models. However, such ad hoc methods affect estimated HR's and model performance.
The main limitations of our work were as follows. We used a data set contained only four categorical variables. Therefore questions whether number and nature of variables affected our conclusions remains to be addressed. In regression imputation, E-M algorithm, and the MICE methods regression modelling has been used to draw the imputation. It is clear that, generally speaking, the more the number of variables the better the prediction. In addition, in regression and E-M imputations, it is assumed that the data follows a normal distribution. This was not the case in our data set. This might partly explain poorer performance of these two methods. Another limitation of our work was that we did not compare performance of the imputation methods under different missing rates. One last issue was the process of model building. We developed all four models using ENTER method. This method fits a model including all variables offered to. Therefore, the behavior of imputation models under different variable selection methods (i.e. Backward and Forward) should be investigated. All issues noted affect the conclusions[5][25][26][27][28] and their influence should be explored in future studies.
Although the MICE method does not provide unique estimates,[29] and communication of results with clinical audiences is not simple, evidence from literature suggested that the MICE method is the best approach to impute missing data. However, our literature review showed that under special circumstances, easier methods might provide comparable estimates. It should be noted that even when easier imputation methods work, results should be compared to that of MICE, to enrich the body of the literature and enhance the understanding of the value of the statistical methods.
Acknowledgments
We should thank staff of Motahhari Paraclinic and Shahid Faghihi Hospital who facilitated our access to patients' folder and information.
Footnotes
Conflict of interest: None declared.
References
- 1.Burton A, Altman DG. Missing covariate data within cancer prognostic studies: a review of current reporting and proposed guidelines. Br J Cancer. 2004;91:4–8. doi: 10.1038/sj.bjc.6601 907. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Baneshi MR. Statistical Models in Prognostic Modelling of Many Skewed Variables and Missing Data: A Case Study in Breast Cancer. (PhD thesis submitted at Edinburgh University) 2009 [Google Scholar]
- 3.Donner A. The relative effectiveness of procedures commonly used in multiple regression analysis for dealing with missing values. American Statisticians. 1982;36:378–81. doi: 10.2307/2683092. [DOI] [Google Scholar]
- 4.Croy CD, Novins DK. Methods for addressing missing data in psychiatric and developmental research. J Am Acad Child Adolesc Psychiatry. 2005;44:1230–40. doi: 10.1097/01.chi.0000181044.06337.6f. [DOI] [PubMed] [Google Scholar]
- 5.Donders AR, van der Heijden GJ, Stijnen T, Moons KG. Review: a gentle introduction to imputation of missing values. J Clin Epidemiol. 2006;59:1087–91. doi: 10.1016/j.jclinepi.2006.01.014. [DOI] [PubMed] [Google Scholar]
- 6.Shrive FM, Stuart H, Quan H, Ghali WA. Dealing with missing data in a multi-question depression scale: a comparison of imputation methods. BMC Med Res Methodol. 2006;6:57. doi: 10.1186/1471-2288-6-57. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Rajaeefard AR, Baneshi MR, Talei AR, Mehrabani D. Survival models in breast cancer. Iran Red Crescent Med J. 2009;11:295–300. [Google Scholar]
- 8.Baneshi MR, Talei AR. Impact of imputation of missing data on estimation of survival rates: an example in breast cancer. Iranian Journal of Cancer Prevention . 2010;3:127–31. [Google Scholar]
- 9.Baneshi MR, Talei AR. Dichotomisation of continuous data: review of methods, advantages, and disadvantages. Iran J Can Prev . 2010;4:26–32. [Google Scholar]
- 10.Baneshi MR, Talei AR. Multiple imputation in survival models: Applied on breast cancer data. Iran J Can Prev. 2011:[In Press]. [PMC free article] [PubMed] [Google Scholar]
- 11.Mehrabani D, Tabei SZ, Heydari ST, Shamsina SJ, Shokrpour N, Amini M, Masoumi SJ, Julaee H, Farahmand M, Manafi A. Cancer occurrence in Fars province, Southern Iran. Iran Red Crescent Med J. 2008;10:314–22. [Google Scholar]
- 12.Mokarian F, Ramezani MA, Heydari K, Tabatabaeian M, Tavazohi H. Epidemiology and trend of cancer in Isfahan 2005-2010. J Res Med Sci. 2011;16:1228–33. [PMC free article] [PubMed] [Google Scholar]
- 13.Royston P, Sauerbrei W, Altman DG. Modeling the effects of continuous risk factors. J Clin Epidemiol. 2000;53:219–21. doi: 10.1016/S0895-4356(99)00163-8. [DOI] [PubMed] [Google Scholar]
- 14.Schafer JL. Analysis of Incomplete Multivariate Data. Florida: Chapman and Hall; 1997. [Google Scholar]
- 15.Van Buuren S, Boshuizen HC, Knook DL. Multiple imputation of missing blood pressure covariates in survival analysis. Stat Med. 1999;18:681–94. doi: 10.1002/(SICI)1097-0258(19990330)18:6<681::AID-SIM71>3.0.CO;2-R. [DOI] [PubMed] [Google Scholar]
- 16.Rubin DB. Multiple imputation for non response in surveys. 1978. [Google Scholar]
- 17.The R Project for Statistical Computing [computer program] 2008
- 18.Mice: Multivariate Imputation by Chained Equations [computer program] 2007
- 19.mitools: Tools for multiple imputation of missing data [computer program] 2008
- 20.Baneshi MR, Warner P, Anderson N, Bartlett JSM. Tamoxifen resistance in early breast cancer: statistical modelling of tissue markers to improve risk prediction. Br J Cancer. 2010;102:1503–10. doi: 10.1038/sj.bjc.6605627. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Baneshi MR, Warner P, Anderson N, Tovey S, Edwards J, Bartlett JM. Can biomarkers improve ability of NPI in risk prediction? a decision tree model analysis. Iran J Can Prev. 2010;2:62–74. [Google Scholar]
- 22.Barzi F, Woodward M. Imputations of missing values in practice: results from imputations of serum cholesterol in 28 cohort studies. Am J Epidemiol. 2004;160:34–45. doi: 10.1093/aje/kwh175. [DOI] [PubMed] [Google Scholar]
- 23.Van Der Heijden GJ, Donders AR, Stijnen T, Moons KG. Imputation of missing values is superior to complete case analysis and the missingindicator method in multivariable diagnostic research: a clinical example. J Clin Epidemiol. 2006;59:1102–9. doi: 10.1016/j.jclinepi.2006.01.015. [DOI] [PubMed] [Google Scholar]
- 24.Kristman VL, Manno M, Cote P. Methods to account for attrition in longitudinal data: do they work? A simulation study. Eur J Epidemiol . 2005;20:657–62. doi: 10.1007/s10654-005-7919-7. [DOI] [PubMed] [Google Scholar]
- 25.Musil CM, Warner CB, Yobas PK, Jones SL. A comparison of imputation techniques for handling missing data. West J Nurs Res. 2002;24:815–29. doi: 10.1177/019394502762477004. [DOI] [PubMed] [Google Scholar]
- 26.Ambler G, Omar RZ, Royston P. A comparison of imputation techniques for handling missing predictor values in a risk model with a binary outcome. Stat Methods Med Res. 2007;16:277–98. doi: 10.1177/0962280206074466. [DOI] [PubMed] [Google Scholar]
- 27.Novins DK, Beals J, Mitchell CM. Sequences of substance use among American Indian adolescents. J Am Acad Child Adolesc Psychiatry. 2001;40:1168–74. doi: 10.1097/00004583-200110000-00010. [DOI] [PubMed] [Google Scholar]
- 28.Bono C, Ried LD, Kimberlin C, Vogel B. Missing data on the Center for Epidemiologic Studies Depression Scale: a comparison of 4 imputation techniques. Res Social Adm Pharm. 2007;3:1–27. doi: 10.1016/j.sapharm.2006.04.001. [DOI] [PubMed] [Google Scholar]
- 29.Kneipp SM, McIntosh M. Handling missing data in nursing research with multiple imputation. Nurs Res. 2001;50:384–9. doi: 10.1097/00006199-200111000-00010. [DOI] [PubMed] [Google Scholar]