Determining and targeting reasons for low statin use to improve guideline-concordant statin therapy in high-risk patients

背景：最佳他汀类药物治疗与心血管事件风险降低相关， 心血管疾病（CVD）。VA外部同行评审计划（EPRP）还建议 中-高强度他汀类药物治疗CVD患者。我们已经表明，很大一部分退伍军人， CVD（41%）正在接受次优他汀类药物治疗。指南一致的他汀类药物治疗超过5年， 潜在预防20，611例血管事件，包括3，435例CVD退伍军人死亡。次优他汀类药物使用 可能是由于他汀类药物相关的副作用或由于提供者的临床惰性。我们已经证明， 和他汀类药物不耐受不能很好地捕获在结构化数据集中，通常记录在免费的 临床医生的文本注释。因此，重要的是要确定他汀类药物使用不佳的原因， 设计干预措施，以增加适当CVD患者接受指南推荐的治疗。 目的：我们的目标是从目前可用的结构化数据集中检查药物使用模式 以及使用自然语言处理（NLP）来识别指南不一致的CVD患者 他汀类药物治疗由于他汀类药物不耐受或提供者临床惰性。我们亦会进行一项试验研究， 以即时通讯辅助工具的形式向患者调整护理团队（PACTs）提供简洁的信息 针对这些差距。目的1：确定退伍军人患有以下疾病的原因（过敏、不耐受或临床惰性）： CVD未接受指南推荐的基于结构化数据的他汀类药物治疗， NLP的信息提取目的2：了解提供者和患者对他汀类药物不耐受的理解 并描述供应商的脂质管理策略，通过定性访谈解决不耐受。 根据这些访谈，我们将完善针对PACT提供者的简明沟通援助 这将使他们能够有效地启动和/或滴定他汀类药物，以证据为基础的剂量在心血管疾病患者 （包括对他汀类药物不耐受的患者）。目标3：在休斯顿对一种通信辅助设备进行试点测试， 纳什维尔VA PACTs，以确定其使用是否与指南一致的他汀类药物增加相关 与常规治疗相比，CVD患者的治疗使用。 方法：对于目标1，我们将从整个VA系统中随机识别CVD患者（500例最佳 他汀类药物，500例接受他汀类药物但低于中等强度剂量，500例未接受他汀类药物），并随机分组 将它们转化为训练集和测试集。然后，我们将训练我们的NLP系统，以实现以下灵敏度和特异性： > 90%，与手动图表审查相比。在目标2中，我们将与供应商进行定性访谈， VISN 16中的患者，以了解他们对他汀类药物相关临床惰性和不耐受的看法。这些 访谈将有助于完善交流辅助工具的内容，供试点期间使用 拟议研究的阶段。在目标3中，我们将与休斯顿和纳什维尔的VAMC PACTs进行试点试验 作为干预场所。干预部位的所有PACT将获得通信辅助， 在接受次优他汀类药物治疗的CVD患者中开始他汀类药物治疗和/或滴定。在常规护理中心 （休斯顿和纳什维尔VAMC附属的社区门诊诊所），PACT提供者将只 收到一份关于心血管病患者使用次优他汀类药物比例的季度报告。我们的主要成果是 接受最佳（至少中等强度）他汀类药物治疗的CVD患者比例的变化。 对退伍军人医疗保健的预期影响：我们的结果将是重要的，因为它们将确定巨大的 大多数高风险退伍军人由于不耐受与临床惰性而未接受最佳他汀类药物治疗。我们 沟通援助将确定战略，开始或滴定他汀类药物的高风险退伍军人谁站在派生 通过解决患者不耐受和临床惰性，最大程度地受益于这种挽救生命的治疗。最后， 这些结果对于退伍军人管理局卫生保健系统识别高风险的“真正的他汀类药物”患者具有重要意义 不耐受”，谁将是未来的候选人，非常昂贵的新药最近批准的FDA。

Background: Optimal statin therapy is associated with a lower risk of cardiovascular events in patients with established cardiovascular disease (CVD). The VA External Peer Review Program (EPRP) also recommends moderate-high intensity statins in patients with CVD. We have shown that a large proportion of Veterans with CVD (41%) are on suboptimal statin therapy. Guideline concordant statin treatment over 5 years could potentially prevent 20,611 vascular events including 3,435 deaths in Veterans with CVD. Suboptimal statin use may be due to statin associated side effects or due to provider clinical inertia. We have shown that side effects and statin intolerance are not well captured in the structured datasets, and are usually documented in the free text notes by clinicians. It is therefore important to identify the reasons for suboptimal statin utilization and to design an intervention to increase receipt of guideline recommended care in appropriate CVD patients. Objective: Our objective is to examine patterns of medication use from currently available structured datasets and the use of natural language processing (NLP) to identify the CVD patients not on a guideline concordant statin therapy due to statin intolerance or provider clinical inertia. We will also perform a pilot study providing succinct information to patient aligned care teams (PACTs) in the form of a point-of-care communication aid which targets these gaps. Aim 1: To identify reasons (allergy, intolerance, or clinical inertia) Veterans with CVD are not receiving guideline-recommended statin therapy based on structured data and automated information extraction with NLP. Aim 2: To understand provider and patient understanding of statin intolerance and to describe providers' lipid management strategies addressing intolerance through qualitative interviews. Based on these interviews, we will refine a succinct communication aid targeted towards PACT providers which will allow them to effectively initiate and/or titrate statins to evidence-based doses in CVD patients (including patients who are intolerant to statins). Aim 3: To pilot test a communication aid in Houston and Nashville VA PACTs to determine whether its use is associated with an increase in guideline-concordant statin therapy use in patients with CVD compared to usual care. Methods: For Aim 1, we will randomly identify CVD patients from throughout the VA system (500 on optimal statins, 500 on statin but on less than moderate-intensity dose, and 500 not on a statin) and randomly partition them into training and a test set. We will then train our NLP system to achieve a sensitivity and specificity of >90%, compared with manual chart review. In Aim 2, we will conduct qualitative interviews with providers and patients in VISN 16 to elicit their perspectives on clinical inertia and intolerance related to statins. These interviews will help facilitate the refinement of the content of the communication aid for use during the pilot phase of the proposed study. In Aim 3, we will conduct a pilot trial with Houston and Nashville VAMC PACTs serving as the intervention sites. All PACTs at the intervention sites will receive the communication aid to assist them with statin initiation and/or titration in patients with CVD on suboptimal statins. At usual care sites (community based outpatient clinics affiliated with the Houston and Nashville VAMC), PACT providers will only receive a quarterly report of the proportion of their CVD patients on suboptimal statins. Our primary outcome is the change in the proportion of CVD patients receiving optimal (at least a moderate intensity) statin therapy. Anticipated Impact on Veteran's Healthcare: Our results will be important as they will identify the vast majority of high-risk Veterans not on optimal statin therapy due to intolerance versus clinical inertia. Our communication aid will identify strategies to initiate or titrate statins in high-risk Veterans who stand to derive the most benefit from this life saving therapy by addressing both patient intolerance and clinical inertia. Finally, these results will be important for the VA Health Care System to identify high-risk patients with “true statin intolerance” who will be future candidates for highly expensive new drugs recently approved by the FDA.