區間設限資料經常出現在當存活時間無法直接地被觀察到的縱貫型研究或臨床試驗中，我們唯一能得知的只有時間是落在哪個特定區間內。事實上，區間設限資料在近年來越來越常出現於大型的世代研究中，但世代研究所需的花費在有限的經費預算下，已經變成研究者們沉重的負擔。研究人員希望能找出一個比簡單隨機抽樣更具成本效益的抽樣方法來減少研究的支出。近年來，結果依賴採樣(ODS) 已被視為一個具成本效益的抽樣設計並應用在許多流行病學或大型資料庫的研究中，此設計的核心價值是希望能選取更具資訊量的樣本。在本次研究中，我們發展出針對區間設限資料在加速失效(AFT) 模型下兩種不同的結果依賴採樣設計並在相同樣本數下比較簡單隨機抽樣(SRS) 以及結果依賴採樣(ODS) 的估計表現。從模擬結果顯示，在不同配置下的相同樣本數進行抽樣，結果依賴採樣比簡單隨機抽樣的估計結果表現更佳。最後，我們也將此設計應用於Signal Tandmobiel 研究中。

Interval-censored data arise in survival analysis when the failure time can not be observed directly, and the only thing we know is the particular interval where the time is located. As the cost of a large cohort study has become an unbearable burden for researchers because of the limited budget, researchers want to search for a more cost-effective design other than just simple random sampling to lower the cost of the study. In recent years, an outcome-dependent sampling (ODS) design is regarded as a cost-effective sampling scheme and has been widely applied in many biomedical and epidemiological studies. The core value of this design is to include more informative failure subjects from the supplemental components particularly interested. In our study, we develop two ODS designs for the interval-censored failure time data under the accelerated failure time (AFT) model and compare the performance of the estimates from simple random sampling (SRS) and ODS designs. Under different settings in the simulation studies, the results show that the estimator from the ODS design is more efficient than that under the SRS design of the same sample size. We then apply the proposed designs to the Signal Tandmobiel study.

[1] J. A. Wellner. (1995). Interval censoring case 2: alternative hypotheses. in analysis
of censored data, proceedings of the workshop on analysis of censored data. IMS
Lecture Notes, Monograph Series, 27:271–291.
[2] J. Huang and J. A. Wellner. (1997). Interval censored survival data: a review
of recent progress. In Proceedings of the first seattle symposium in biostatistics,
pages 123–169.
[3] A. Schick and Q. Yu. (2000). Consistency of the gmle with mixed case intervalcensored
data. Scandinavian Journal of Statistics, 27:45–55.
[4] J. Vanobbergen, L. Martens, E. Lesaffre, and D. Declerck. (2000). The signaltandmobiel
project a longitudinal intervention health promotion study in flanders
(belgium): baseline and first year results. European Journal of Paediatric Dentistry,
2:87–96.
[5] J. Cornfield. (1951). A method of estimating comparative rates from clinical data.
applications to cancer of the lung, breast, and cervix. Journal of the National
Cancer Institute, 11:1269–1275.
[6] R. L. Prentice. (1986). A case-cohort design for epidemiologic studies and disease
prevention trials. Biometrika, 73:1–11, 1986.
[7] H. Zhou, M. A. Weaver, J. Qin, M. P. Longnecker, and M. C. Wang. (2002). A
semiparametric empirical likelihood method for data from an outcome-dependent
sampling scheme with a continuous outcome. Biometrics, 58:413–421.
[8] N. Chatterjee, Y. H. Chen, and N. E. Breslow. (2003). A pseudoscore estimator
for regression problems with two-phase sampling. Journal of the National Cancer
Institute, 98:158–168.
[9] M. A. Weaver and H. Zhou. (2005). An estimated likelihood method for continuous
outcome regression models with outcome-dependent sampling. Journal of the
American Statistical Association, 100:459–469.
[10] H. Zhou, J. Chen, T. H. Rissanen, S. A. Korrick, H. Hu, J. T. Salonen, and M.
P. Longnecker. (2007). Outcome-dependent sampling: an efficient sampling and
inference procedure for studies with a continuous outcome. Epidemiology, 18:461–
468.
[11] R. Song, H. Zhou, and M. R. Kosorok. (2009). A note on semiparametric efficient
inference for two-stage outcome-dependent sampling with a continuous outcome.
Biometrika, 96:221–228.
[12] H. Zhou, R. Song, Y. Wu, and J. Qin. (2011). Statistical inference for a twostage
outcome-dependent sampling design with a continuous outcome. Biometrics,
67:194–202.
[13] J. Ding, H. Zhou, Y. Liu, J. Cai, and M. P. Longnecker. (2014). Estimating effect
of environmental contaminants on women＇s subfecundity for the moba study data
with an outcome-dependent sampling scheme. Biostatistics, 15(4):636–650.
[14] J. Yu, Y. Liu, D. P. Sandler, and H. Zhou. (2015). Statistical inference for the
additive hazards model under outcome-dependent sampling. Canadian Journal of
Statistics, 43:436–453.
[15] Q. Zhou, J. Cai, and H. Zhou. (2018). Outcome-dependent sampling with intervalcensored
failure time data. Biometrics, 74(1):58–67.
[16] L. J. Wei. (1992). The accelerated failure time model: a useful alternative to the
cox regression model in survival analysis. Statistics in medicine, 11(14-15):1871–
1879.
[17] J. D. Kalbfleisch and R. L. Prentice. (2011). The statistical analysis of failure time
data. John Wiley & Sons.
[18] K. Kiani and J. Arasan. (2012). Simulation of interval censored data in medical and
biological studies. International Journal of Morden Physics: Conference Series,
09:112–118.
[19] Z. Zhang and J. Sun. (2010). Interval censoring. Statistical methods in medical
research, 19(1):53–70.