SECONDARY ANALYSIS OF CRISP PILOT STUDY DATA

本研究的目的是进行二次分析， 降低老年人胆固醇计划（CRISP）数据集。至少一 手稿将被写在以下每个主题。 1.依从性预测：我们假设， 与教育程度低，年龄较大，男性，黑人，没有 心血管疾病和大量合并用药。 依从性将通过药丸计数、门诊出勤率和LDL-C来衡量 各治疗组的变化（安慰剂组除外）。 2.安慰剂导入的效用：我们假设， 安慰剂导入期将预测治疗期的不良依从性。贫困 粘附性（药丸计数小于75%）和良好的粘附性（药丸计数 大于75%）将与依从性进行比较 随机化后测量。 3.治疗对生活质量的影响：假设 洛伐他汀（两种剂量）对生活质量几乎没有影响 将通过比较两个活性治疗组和 安慰剂组的许多生活质量的措施。 4.对饮食的脂质反应：记录血脂变化（如有） 将使用贝叶斯方法调整饮食回归平均值。 饮食改变与年龄、种族、 将测试性别和教育程度。 5. CRISP参与者与CHS参与者的比较：我们 假设受血脂标准的影响，CRISP参与者将 比CHS参与者更健康。他们将被比较 关于基线特征，包括年龄、种族、BP， 教育、吸烟、性别特异性血脂水平、心血管疾病 和药物治疗。

The purpose of the study is to conduct secondary analyses of The Cholesterol Reduction in Seniors Program (CRISP) data set. At least one manuscript will be written on each of the following topics. 1. Prediction of Adherence: We hypothesize that por adherence will be associated with low education, greater age, male gender, black race, no cardiovascular disease and large number of concurrent medications. Adherence will be measured by pill count, clinic attendance and LDL-C change (except placebo group) in each treatment group. 2. Utility of a Placebo Run-in: We hypothesize that poor adherence at placebo run-in will predict poor adherence in the treatment phase. Poor adherence (pill count less than 75%) and good adherers (pill count greater than 75%) at the run-in phase will be compared to adherence measures post-randomization. 3. Effect of Treatment on Quality of Life: The hypothesis that lovastatin (at either dose) will have little effect on quality of life will be tested by comparing the two active treatment groups and the placebo group over a number of quality of life measures. 4. Lipid Response to Diet: Documentation of serum lipid change (if any) to diet will be adjusted for regression to the mean using Bayes method. The null hypothesis that diet change is not associated with age, race, gender and educational attainment will be tested. 5. Comparison of CRISP Participants with CHS Participants: We hypothesize that subject to the lipid criteria, CRISP participants will exhibit a healthier profile than CHS participants. They will be compared with respect to baseline characteristics including age, race, BP, education, smoking, gender specific lipid levels, cardiovascular disease and medication taking.