From the Center for Stroke Research Berlin (I.N., I.L.M, A.H., M.E., E.F., P.H., M.E., B.S., H.J.A.), Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin and Humboldt-Universität zu Berlin, Germany; Klinik für Neurologie (M.E.), Medical Park Berlin Humboldtmühle, Germany; Klinik und Hochschulambulanz für Neurologie (E.F., P.H., M.E.), Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin and Humboldt-Universität zu Berlin, Germany; German Centre for Cardiovascular Research (DZHK) (M.E.), Partner Site Berlin, Germany; German Center for Neurodegenerative Diseases (DZNE) (M.E.), Partner Site Berlin, Germany; NeuroCure Cluster of Excellence (M.E.), Charité-Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin and Humboldt-Universität zu Berlin, Germany; Helios Klinikum Berlin Buch (G.H.), Germany; Klinik für Neurologie Mit Stroke Unit (H.-C.K.); Vivantes Klinikum Im Friedrichshain, Berlin, Germany; Klinik für Neurologie Mit Stroke Unit (B.-M.M.), Vivantes Auguste-Viktoria-Klinikum, Berlin, Germany; and Department of Clinical Epidemiology (B.S.), Leiden University Medical Center, Leiden University, The Netherlands.
Neurology. 2022 Sep 27;99(13):e1335-e1344. doi: 10.1212/WNL.0000000000200916. Epub 2022 Aug 2.
Restricting follow-up assessment of both interventional and observational studies to patients who provide informed consent introduces relevant selection bias-particularly by underrepresenting patients with neurologic communication deficits and impaired capacity to consent. Many patients who are initially unable to give consent may be willing to do so after recovery. Informing patients on study purposes and procedures with offering them the option of nonparticipation but not requesting explicit consent is called "opt-out" approach. We investigated whether an opt-out strategy yields meaningful follow-up rates in an acute stroke registry with an embedded controlled study.
The citywide Berlin-SPecific Acute Treatment in Ischemic or hAemorrhagic Stroke With Long Term Follow-up (B-SPATIAL) registry was designed to provide reliable information on process indicators and outcomes of specific acute stroke treatments to inform health care providers about quality of care and best practice strategies including the effects of a mobile stroke unit implementation. Because this information was regarded of high public interest, Berlin data protection authorities permitted data sampling without prior informed consent, using instead follow-up assessment on an "opt-out" basis. Patients were included if they had neurologic symptoms at ambulance or hospital arrival within 6 hours of onset and had a final diagnosis of stroke or TIA. Information on data collection and outcome assessment was sent by letter to patients 1 month before follow-up.
From February 1, 2017, to January 31, 2020, a total of 10,597 patients were assessed. Thirty-one (0.3%) patients declined any data use, whereas 578 (5.5%) opted out of follow-up assessment. Of those not opting out (n = 9,988), functional outcome (modified Rankin Scale) was collected in 8,330 patients (83.4%) and vital status in 9,741 patients (97.5%). We received no complaints regarding data collection procedures.
Opt-out-based follow-up collection offers a way to achieve high follow-up rates along with respecting patients' preferences.
限制对介入性和观察性研究的随访评估仅限于提供知情同意的患者，这会引入相关的选择偏倚——尤其是会低估那些存在神经沟通障碍和同意能力受损的患者。许多最初无法同意的患者在康复后可能愿意参与研究。通过向患者告知研究目的和程序，并为其提供不参与的选择但不要求明确同意的方式，被称为“选择退出”方法。我们研究了在一个嵌入对照研究的急性脑卒中登记处中，选择退出策略是否能产生有意义的随访率。
全市范围的柏林特定急性缺血性或出血性卒中治疗的长期随访（B-SPATIAL）登记处旨在为特定急性卒中治疗的过程指标和结果提供可靠信息，以便向医疗保健提供者提供有关护理质量和最佳实践策略的信息，包括移动卒中单元实施的效果。由于这些信息具有高度的公共利益，柏林数据保护机构允许在没有事先知情同意的情况下进行数据抽样，而是使用“选择退出”的方式进行随访评估。如果患者在发病后 6 小时内到达救护车或医院时出现神经症状，并最终诊断为卒中或 TIA，则纳入该研究。在随访前 1 个月，通过信件向患者发送有关数据收集和结果评估的信息。
从 2017 年 2 月 1 日至 2020 年 1 月 31 日，共评估了 10597 例患者。有 31 例（0.3%）患者拒绝任何数据使用，而 578 例（5.5%）选择退出随访评估。在未选择退出的患者中（n=9988），有 8330 例（83.4%）收集了功能结局（改良 Rankin 量表），有 9741 例（97.5%）收集了生命状态。我们没有收到任何关于数据收集程序的投诉。
基于选择退出的随访收集方法提供了一种既能实现高随访率，又能尊重患者意愿的方式。
