Investigating the mechanisms behind behavioural strategies to improve medicines adherence

不遵守药物治疗是一项重大的卫生政策挑战，影响到近50%的慢性病患者，导致大量浪费和患者预后不佳（世卫组织，2003年）。因此，需要大量举措和工具来支持和促进依从性行为，特别是对依赖定期长期药物治疗来维持或改善其健康的个人。甚至有人提出，提高新的和当前的依从性干预措施的有效性的潜在影响可能比任何具体的医疗改进更大（Haynes等人，2002年）。增加严格按照处方持续正确自我用药的人数需要长期、广泛的行为改变和习惯形成。因此，行为科学的应用有很大的空间，可以为当前和新的干预措施提供信息，以支持患者按规定服药。低成本行为干预提高健康预防和保护性健康行为的有效性已得到广泛调查，并取得了不同程度的成功。如果一项干预有效地利用了经济学和心理学关于人类选择、决策和行为的研究、理论和见解，那么它就可以被认为是“行为”的。最近的证据表明，在首次配发新药时，签署承诺服用新药的患者更有可能坚持服药。然而，这一结果是有条件的，强调了坚持的重要性，这与承诺是相辅相成的（Jachimowicz等人，2019）。这些结果表明，行为干预（或行为“轻推”）可能是一种有用的新工具，可以促进公共卫生倡议，从药学实践的角度出发，以最小的成本或努力加强药学实践并改善患者的结果。然而，更广泛和有效地使用这些方法的前景目前受到限制，因为人们对为什么一些推动有效而另一些不起作用的理解不足；为什么一个推动可能在某些情况下起作用，而在其他情况下不起作用；或者轻推会如何影响不同的患者群体。因此，为了利用基于轻推的工具的潜在力量来加强药房实践，需要进一步的工作来了解成功轻推背后的工作机制。我的博士研究调查了成功推动提高药物依从性背后的机制。我们在60家Boots社区药店进行了一项随机对照试验，以“一切照旧”的对照组为对象，测试四种不同的行为推动策略。我们的策略使我们能够调查是否有可能复制和改进Jachimowicz等人（2019）的结果，并揭示推动患者行为变化的潜在机制。除了这个项目，我们还进行了一系列在线实验，以评估标准化自我报告依从性问卷（Thompson等人（2000）开发的5项药物依从性评定量表）的可靠性，以准确评估行为干预对自我报告依从性的影响。在线实验使我们能够检测某些行为策略是否会促使受试者过度报告他们的依从性，并估计任何额外报告偏差的大小。

Medicines non-adherence is a major health policy challenge affecting almost 50% of patients with chronic disease, leading to massive wastage and sub-optimal patient outcomes (WHO, 2003). As such, there is significant demand for initiatives and tools to support and foster adherent behaviours, particularly for individuals dependent on regular long-term medication regimes to maintain or improve their health. It has even been suggested that the potential impact of improving the effectiveness of new and current adherence interventions, may be greater than any specific improvement in medical treatment (Haynes et al., 2002).Increasing the number of individuals who consistently and correctly self-administer medicines, exactly as prescribed, requires long-term, widespread behaviour change and habit formation. As a result, there is significant scope for the application of behavioural science to inform current and new interventions designed to support patients to take their medication as prescribed.The effectiveness of low-cost behavioural interventions to increase healthy preventative and protective health behaviours has been widely investigated to varying degrees of success. An intervention can be considered "behavioural" if it instrumentally utilises research, theory, and insights from economics and psychology on human choice, decision-making and behaviour. Recent evidence suggests that patients who sign a commitment to take a new medication, when the new medicine is first dispensed, are significantly more likely to be adherent to their medicines. However, this result was conditional on the message, highlighting the importance of adherence, that was paired with the commitment (Jachimowicz et al., 2019).These results suggest that behavioural interventions (or behavioural "nudges") may be a useful new tool for promoting public health initiatives, to enhance pharmacy practice and improve patient outcomes at minimal cost or effort from the perspective of pharmacy practice. However, the prospects for using such approaches more widely and effectively is currently limited by poor understanding of why some nudges work and others don't; why one nudge might work in some contexts and not others; or how nudges might impact different patient groups. Hence, in order to exploit the potential power of nudge-based tools to enhance pharmacy practice, further work is required to understand the mechanisms at work behind successful nudges. My PhD research investigates the mechanisms behind successful nudges to improve medicines adherence. We employ a randomised controlled trial within 60 Boots community pharmacies to test four different behavioural nudge strategies against a "business as usual" control group. Our strategy allows us to investigate whether it is possible to replicate and improve upon Jachimowicz et al.'s (2019) results, and unpick potential mechanisms driving changes in patient behaviour in response to the nudges. Alongside this project we have conducted a series of online experiments to evaluate the reliability of a standardised self-reported adherence questionnaire (the 5-item Medicines Adherence Rating Scale developed by Thompson et al. (2000)) to accurately evaluate the effect of a behavioural intervention on self-reported adherence. The online experiments allow us to detect whether certain behavioural strategies prompt subjects to over-report their adherence and estimate the size of any additional reporting bias.