全国队列中采用分子检测是否导致甲状腺切除术率下降?高采用率与低采用率州之间的准实验研究
该研究基于Merative™ MarketScan®数据库回顾性分析2011至2021年接受甲状腺细针穿刺的471,364例美国成人患者,采用中断时间序列和双重差分分析,发现2015年后分子检测采用加速,甲状腺切除术率与分子检测使用呈负相关(r=-0.98),但双重差分调整率为-3.3%[CI -5.6%至-0.8%],作者认为该相关性并非因果。
要点
- 分子检测使用率从0.01%增至2021年的10.1%,2015年后立即(β2=1.61,p=0.002)且更深(β3=0.6,p<0.001)增加。
- 黑人、老年人、农村地区及医疗补助患者接受分子检测较少(p<0.05)。
- 甲状腺切除术率与分子检测使用呈负相关(r=-0.98,p<0.0001),但作者认为此相关性并非因果。
核心结论
- 2015年后分子检测采用的加速可能与美国甲状腺协会指南的发布同时发生。
- 黑人、老年和农村患者较少接受分子检测。
- 尽管甲状腺切除术率与分子检测使用呈负相关,但本研究提示该相关性并非因果。
- 分子检测对甲状腺切除术率的影响可能被甲状腺结节评估侵袭性的降低所掩盖。
研究设计与样本
- 回顾性分析美国成人患者,数据来自Merative™ MarketScan®研究数据库,纳入2011至2021年接受甲状腺细针穿刺(FNA)者。
- 分子检测包括市售DNA和RNA平台及传统靶向突变分析。
- 采用中断时间序列分析评估2015年后分子检测采用和甲状腺切除术率的拐点。
- 采用双重差分(DID)分析因果评估2015至2021年高采用州(分子检测使用增加至少10%)与低采用州(增加不超过5%)分子检测采用对甲状腺切除术率的影响。
主要发现
- 共识别471,364例接受甲状腺FNA的患者。
- 分子检测使用率从0.01%[CI 0.00%至0.02%]增至2021年的10.1%[CI 9.7%至10.5%],2015年后立即(β2=1.61,p=0.002)且更深(β3=0.6,p<0.001)增加。
- 黑人患者、老年人、农村地区及医疗补助患者中分子检测使用较低(p<0.05)。
- 甲状腺切除术率与分子检测使用呈负相关(r=-0.98,p<0.0001)。
- 2015至2021年,高采用州平均分子检测使用率从2.4%增至15.3%,低采用州从1.6%增至5.6%;低采用州甲状腺切除术率下降更多但降至相似水平(18.5-13.2%),高采用州为(15.9-13.4%),调整后DID率为-3.3%[CI -5.6%至-0.8%]。
关键数值与推荐
- 高采用州定义为分子检测使用增加至少10%,低采用州定义为增加不超过5%。
- 分子检测使用率从0.01%[CI 0.00%至0.02%]增至2021年的10.1%[CI 9.7%至10.5%]。
- 甲状腺切除术率与分子检测使用呈负相关(r=-0.98,p<0.0001)。
- 调整后DID率为-3.3%[CI -5.6%至-0.8%]。
临床怎么用
- 原文未据此给出推荐。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
Background: Over the last decade, the utilization of molecular testing (MT) for the evaluation of thyroid nodules has increased. Rates and patterns of adoption of MT and its effect on thyroidectomy rates nationally are unknown. Varying rates of MT adoption at the state level provide an opportunity to study the effects of MT on thyroidectomy rates using a quasiexperimental study design. Methods: We performed a retrospective analysis of American adult patients in the Merative™ MarketScan® Research Databases who underwent thyroid fine-needle aspiration (FNA) from 2011 to 2021. MT included commercially available DNA and RNA platforms and traditional targeted mutational analysis. Interrupted time series analysis was used to evaluate the inflection of MT adoption and thyroidectomy rates after 2015. Difference-in-differences (DID) analysis was used to causally analyze the effect of MT adoption on thyroidectomy rates in high-adoption (at least a 10% increase in MT utilization) versus low-adoption states (no more than 5% increase in MT utilization) from 2015 to 2021. Results: We identified 471,364 patients who underwent thyroid FNA. The utilization of MT increased over the study period from 0.01% [confidence interval, CI: 0.00% to 0.02%] to 10.1% [CI: 9.7% to 10.5%], in 2021, with an immediate (β2 = 1.61, p = 0.002) and deeper (β3 = 0.6, p < 0.001) increase in MT adoption after 2015. Utilization of MT was lower in black patients, the elderly, rural areas, and patients with Medicaid (p < 0.05). Thyroidectomy rates were inversely correlated with MT utilization (r = -0.98, p < 0.0001). From 2015 to 2021, the average MT utilization rate increased from 2.4% to 15.3% in high-adoption states and 1.6% to 5.6% in low-adoption states. In low-adoption states, thyroidectomy rates decreased more but to similar levels (18.5-13.2%) compared with high-adoption states (15.9-13.4%) with an adjusted DID rate of -3.3% [CI -5.6% to -0.8%]. Conclusions: The acceleration in adoption of MT after 2015 likely coincides with the publication of American Thyroid Association guidelines. Black, elderly, and rural patients are less likely to receive MT. Although thyroidectomy rates were inversely correlated with MT utilization, our study suggests that this correlation is not causal. The effect of MT on thyroidectomy rates may be overshadowed by decreasing aggressiveness of thyroid nodule evaluation.
引用
- Huang Y, Chan SJ, Wright JD, et al. Does the Adoption of Molecular Testing Cause Decreased Thyroidectomy Rates in a National Cohort? A Quasiexperimental Study of High- Versus Low-Adoption States. Thyroid. 2024. PMID: 38251649. doi:10.1089/thy.2023.0651
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。