The strength of statistical evidence is measured by the likelihood ratio. Two key performance properties of this measure are the probability of observing strong misleading evidence and the probability of observing weak evidence. For the likelihood function associated with a parametric statistical model, these probabilities have a simple large sample structure when the model is correct. Here we examine how that structure changes when the model fails. This leads to criteria for determining whether a given likelihood function is robust (continuing to perform satisfactorily when the model fails), and to a simple technique for adjusting both likelihoods and profile likelihoods to make them robust. We prove that the expected information in the robust adjusted likelihood cannot exceed the expected information in the likelihood function from a true model. We note that the robust adjusted likelihood is asymptotically fully efficient when the working model is correct, and we show that in some important examples this efficiency is retained even when the working model fails. In such cases the Bayes posterior probability distribution based on the adjusted likelihood is robust, remaining correct asymptotically even when the model for the observable random variable does not include the true distribution. Finally we note a link to standard frequentist methodology-in large samples the adjusted likelihood functions provide robust likelihood-based confidence intervals. Copyright 2003 Royal Statistical Society.

The impact of cancer pain on the quality of life of lung cancer patients is obvious, but the relationship of cancer pain to uncertainty and level of hope in cancer patients is not clear and has been the subject of only a few studies. The purpose of this study is to look at the relationship of pain to uncertainty and hope in Taiwanese lung cancer patients. A crosssectional and descriptive correlational design was used in this study. A convenience sample of lung cancer patients was recruited from chest medicine and oncology inpatient units at three teaching hospitals in the Taipei area of Taiwan. The research instruments included the Brief Pain Inventory-Chinese version (BPI-C), Mishel's Uncertainty Illness Scale (MUIS), and the Herth Hope Index (HHI). Data were analyzed using descriptive statistics, Pearson's correlation, and multiple regression. A total of 164 subjects were recruited, including 79 patients with cancer pain and 85 patients without cancer pain. The major findings were: 1) there were significant differences in level of uncertainty and level of hope between patients with cancer pain and those without. Patients with cancer pain reported higher levels of uncertainty and lower levels of hope than did patients without cancer pain; 2) pain severity was not significantly related to level of uncertainty; however, pain interference with daily life was positively correlated to level of uncertainty; 3) both pain severity and pain interference were negatively correlated with level of hope; and 4) after controlling for pain severity and pain interference, uncertainty was a significant predictor of level of hope. Important implications for future studies are discussed.

The purpose of this pilot study was to explore the effectiveness of a pain education program to overcome patient-related barriers in managing cancer pain for Taiwanese home care patients with cancer. The pain education program was developed based on previous studies of Taiwanese patient-related barriers to cancer pain management. The Barriers Questionnaire-Taiwan form, the Brief Pain Inventory, the Medication Adherence Questionnaire, and a demographic questionnaire were used for data collection. The sample consisted of 18 patients in the experimental group and 19 patients in the control group. Descriptive statistics, tests, and paired tests were used to analyze the data. Results of this study revealed that patients who received the pain educational program had significantly greater reduction in Barriers Questionnaire-Taiwan form scores and more improvement in medication adherence compared with patients who did not participate in the program. When compared to pretest scores, patients scores after receiving the pain education intervention showed significant improvement on the Barriers Questionnaire-Taiwan form, medication adherence, pain intensity, and pain interference. The results of this study support the effectiveness of the pain education program on overcoming the barriers to cancer pain management for Taiwanese home care patients with cancer.