OBJECTIVE: Analyze the influence of patients lost to follow-up in estimated survival rates calculated by the Kaplan-Meier method.
STUDY DESIGN: Only patients with previously untreated squamous cell carcinoma of the upper aerodigestive tract were selected. For the patients lost to follow-up anytime, the last medical evaluation date was collected to calculate the “estimate A” survival time. If the same patient returned to our outpatient clinic or further health information was obtained, the updated last information/evaluation date was also collected to calculate “estimate B” survival time. The survival curves considering “estimate A” and “B” survival rates were compared.
RESULTS: The overall 5 and 10-years survival rates for all patients calculated for “estimate A” were 54.0% and 46.0%, respectively; compared with 42.8% and 28.2% when were calculated considering “estimate B” (P < 0.001).
CONCLUSION: Close follow-up of the head and neck cancer patients is essential for an accurate estimate of survival by KaplanMeier method.
EBM RATING: C
Get full access to this article
View all access options for this article.
References
1.
de VisscherAVManniJJRoutine long-term follow-up in patients treated with curative intent for squamous cell carcinoma of the larynx, pharynx, and oral cavity: Does it make sense?Arch Otolaryngol Head Neck Surg1994;120:934–9.
2.
HaasIHauserUGanzerU.The dilemma of follow-up in head and neck cancer patients. Eur Arch Otorhinolaryngol2001;258:177–83.
3.
BlackWCHaggstromDAWelchG.All-cause mortality in randomized trials of cancer screening. J Natl Cancer Inst2002;94:167–73.
4.
KaplanELMeierP.Nonparametric estimation from incomplete observations. J Am Static Assoc1958;53:457–81.
5.
GudjonssonB.Survival statistics gone away. Pancreatic cancer, a case in point. J Clin Gastroenterol. 2002;35:180–4.
6.
ParkinDMSanghviLDCancer registration in developing countries. In: JensenOMParkinDMMaclennanR, et al eds. Cancer registration principles and methods. Lyon: IARC;1991. p. 185–98. (IARC Scientific Publications, no 95).
7.
YoungJLThe hospital-based cancer registry. In: JensenOMParkinDMMaclennanR, et al eds. Cancer registration principles and methods. Lyon: IARC;1991. p. 177–84. (IARC Scientific Publications, no 95).
8.
HubbardAEvan der LaanMJEnanoriaW, et al. Nonparametric survival estimation when death is reported with delay. Lifetime data analysis2000;6:237–50.
9.
GoldmanAIThe cure model and time confounded risk in the analysis of survival and other timed events. J Clin Epidemiol1991;44:1327–40.
10.
DuchateauLColletteLSylvesterR, et al. Estimating number of events from the Kaplan-Meier curve for incorporation in a literature-based meta-analysis: What you don't see you can't getBiometrics2000;56:886–92.
11.
SimonRWittesREMethodologic guidelines for reports of clinical trial. Cancer1986;58:212–4.
