Background: The American Cancer Society (ACS) and the NCI collaborate every 5–8 years to update the methods for estimating numbers of new cancer cases and deaths in the current year in the United States and in every state and the District of Columbia. In this article, we reevaluate the statistical method for estimating unavailable historical incident cases which are needed for projecting the current year counts. Methods: We compared the current county-level model developed in 2012 (M0) with three new models, including a state-level mixed effect model (M1) and two state-level hierarchical Bayes models with varying random effects (M2 and M3). We used 1996–2014 incidence data for 16 sex-specific cancer sites to fit the models. An average absolute relative deviation (AARD) comparing the observed with the model-specific predicted counts was calculated for each site. Models were also cross-validated for six selected sex-specific cancer sites. Results: For the cross-validation, the AARD ranged from 2.8% to 33.0% for M0, 3.3% to 31.1% for M1, 6.6% to 30.5% for M2, and 10.4% to 393.2% for M3. M1 encountered the least technical issues in terms of model convergence and running time. Conclusions: The state-level mixed effect model (M1) was overall superior in accuracy and computational efficiency and will be the new model for the ACS current year projection project. Impact: In addition to predicting the unavailable state-level historical incidence counts for cancer surveillance, the updated algorithms have broad applicability for disease mapping and other activities of public health planning, advocacy, and research. This article is featured in Highlights of This Issue, [p. 1599][1] [1]: /lookup/volpage/30/1599?iss=9

中文翻译：

---

预测当前日历年美国和州级癌症计数的更新方法：第一部分：癌症发病率的时空建模

背景：美国癌症协会 (ACS) 和 NCI 每 5-8 年合作一次，以更新估计美国以及每个州和哥伦比亚特区当年新发癌症病例和死亡人数的方法。在本文中，我们重新评估了用于估算不可用历史事件案例的统计方法，这些案例是预测当年计数所需的。方法：我们将 2012 年开发的当前县级模型 (M0) 与三个新模型进行了比较，包括一个州级混合效应模型 (M1) 和两个具有不同随机效应的州级分层贝叶斯模型 (M2 和 M3)。我们使用了 16 个特定性别癌症部位的 1996-2014 年发病率数据来拟合模型。计算每个位点的平均绝对相对偏差 (AARD)，将观察到的计数与特定于模型的预测计数进行比较。模型还针对六个选定的性别特异性癌症部位进行了交叉验证。结果：对于交叉验证，M0 的 AARD 范围为 2.8% 至 33.0%，M1 为 3.3% 至 31.1%，M2 为 6.6% 至 30.5%，M3 为 10.4% 至 393.2%。M1 在模型收敛和运行时间方面遇到的技术问题最少。结论：州级混合效应模型（M1）在准确性和计算效率方面总体上优越，将成为 ACS 当年预测项目的新模型。影响：除了预测不可用的州级癌症监测历史发病率之外，更新后的算法广泛适用于疾病绘图和公共卫生规划、宣传和研究的其他活动。这篇文章被收录在本期的亮点中，[p. 1599][1][1]：/lookup/volpage/30/1599?iss=9