Johnson et al¹ review different aspects of acute psychiatric care, with the aim to identify evidence-based practices in order to increase the range of services and improve access and quality of care. They acknowledge the assortment of services involved as well as the divergent settings across health systems and countries.

Crises are multidimensional phenom­ena and result from complex interactions be­­­­tween mental illness, substance use, emo­tional reserves and social supports. They present complex challenges for assessment of their multiple dimensions and require a multifaceted response.

The quality of evidence for current crisis interventions and models for acute psychiatric care is, at best, moderate. The availa­bility of only few studies, many of which marked by small samples, selective inclusion criteria, narrow focus of assessment of outcomes, and the lack of a comprehensive map of caregiver inputs and medication compliance, argues for the lack of robust evidence base for many interventions^{2, 3}.

Different fidelity scores for implementation of the various intervention models and programs across regions suggest variations in the translation of crisis care packages⁴. The unpredictability of crisis presentations and the need for urgent care complicate the evaluation of interventions. Randomization of participants in crisis raise difficult ethical issues.

Most appraisals have examined issues from health provider perspectives, with lim­ited user involvement in the evaluation of health care delivery. Consumer-led movements rooted in civil rights, social justice and cultural responsiveness appear promising in crisis resolution and even in prevention, and need to be included in future evaluations. The voluntary sector’s involvement in providing peer support, particularly for marginalized communities, while invaluable, needs to be systematically investigated.

The delivery of acute psychiatric care has more recently focused on telepsychiatry and substitutes to in-person interactions. While telephone, videoconferencing facilities and smartphone apps have increased resources, reduced wait times, decreased cost and improved access to care, they have not resolved issues related to digital exclusion, privacy in users’ homes, therapeutic relationships, quality of care and renumeration models. These technologies await evidence for their use in routine clinical practice.

Much of the evidence for acute psychiatric care is from high-income countries. Mental health care in low- and middle-in­come countries, with their financial and human resource constraints, urban-rural divide, and diverse mental illness perspectives (e.g., religious and traditional healer explanatory models, complementary remedies, stigma, taboo) is often marked by inadequate provision of health services, lack of evidence-based intervention guidelines and large treatment gaps. The absence of a rights-based approach, recovery-oriented responses and inclusive community practices in addressing mental health crises, and the high cost, inaccessibility and non-acceptability of specialist mental health services complicate the scenario.

Notwithstanding the success of some programs, the issues related to efficacy, effectiveness and cost-benefit of interventions in acute psychiatric care need to be examined⁵. While randomized controlled trials are the cornerstone of evidence-based medicine, the results of a single trial or a systematic review of a few such investigations, while providing evidence about the efficacy of a treatment (i.e., “The treatment works somewhere”), do not necessarily provide evidence of effectiveness in cli­nical practice (i.e., “The treatment works widely”).

Extrapolating knowledge gained from randomized controlled trials to other pa­tient populations is problematic. The evi­dence for efficacy (“Can it work?”), effectiveness (“Does it work in practice?”) and efficiency (“Is it worth it?”) will need to be addressed before widespread implementation of models and programs⁶. The Hawthorne effect also confounds comparisons between innovative interventions with “standard care control arms”. The motivational response of the subjects may be secondary to the interest, care and attention received through observation and assessment rather than due to the specific intervention.

Changes in clinical practice patterns o­ver time, differences between health sys­tems, and variations in patient demograph­ic and clinical characteristics and in social determinants of health⁷ and mental health⁸, also impact generalizability of clin­ical research. Many crisis presentations are shaped to a great extent by the social, economic and physical environments in which people live. While targeted mental health interventions will help people in crisis, structural, public health and population-wide interventions are needed to level the social gradient in health outcom­es⁸.

Divergent disciplinary perspectives (e.g., crisis intervention theory, psychiatric points of view), different levels of commu­nity supports (e.g., caregiver, peer, professional), task splitting (e.g., triage, assessment and treatment), dissimilar modes of assessments (e.g., face to face, telephone, videoconferencing), varied pathways to care (e.g., health, police), multiagency integration (e.g., police, ambulance, health professionals), distinctive legal status (e.g., voluntary, compulsory, arrest), diverse location of crisis services (e.g., provision at home, within emergency departments, colocation within mental health facilities), wide spectrum of presentations (e.g., situational crisis, personality disorder, substance use/intoxication, psychosis) and the range of harm (e.g., suicidal ideation, deliberate self-harm, suicidal attempt, violence) make comparisons across services and regions difficult. Similarly, diverse ther­apeutic interventions (e.g., psychological, pharmacological, physical restrictive practices) and differences between stepped care models make generalizations problematic.

In addition, variation in population prevalence of crisis presentations, differences in help-seeking behavior, and variation in thresholds for different types of clinical interventions further complicate gen­eralizability. Disparities in budgets, com­munity and hospital infrastructure, and hu­man resources add complexity to comparisons. Despite the success of some mo­dels, and calls for innovative approaches, the dissimilar reality across regions makes the task of identifying universally applicable models challenging.

While the evaluation of interventions is mandatory, their success will not automatically imply their generalizability to other settings. In fact, many complex programs, which often operate in project mode, succeed due to their high levels of financial, ad­ministrative and political support, but are difficult to scale up even across similar settings. Their implementation across different regions, health systems and countries can be extremely challenging.

The heterogeneity of acute psychiatric presentations, variety of interventions and diversity of settings demand the need to understand contexts. The reality of local environments and their distinctive issues demand tailored solutions. Transplanting knowledge structures, formations and prac­tice across different contexts may result in the lack of goodness of fit⁹. Standardized protocols may not recognize locally rel­evant issues, demanding contextual analysis and interpretations grounded in region­al reality. This is particularly true for multi­faceted and multi-disciplinary intervention packages for acute psychiatric and crisis presentations.

Decisions in clinical practice should con­­sider the broader biopsychosocial con­­text, including clinical, psychological, social and economic problems, medical morbidity and risks, and patient and caregiver perspectives. The challenge, while attempting to replicate successful projects, is the need to understand local contexts, incorporate provincial knowledge and attempt to implement regionally tailored solutions.

Johnson 等人¹回顾了急性精神科护理的不同方面，旨在确定基于证据的实践，以扩大服务范围并提高护理的可及性和质量。他们承认所涉及的服务种类繁多，以及卫生系统和国家之间的不同环境。

危机是多维现象，是精神疾病、物质使用、情感储备和社会支持之间复杂相互作用的结果。它们对评估其多个维度提出了复杂的挑战，需要多方面的回应。

当前危机干预和急性精神病护理模型的证据质量充其量是中等的。只有少数研究的可用性，其中许多以小样本、选择性纳入标准、结果评估的狭窄焦点以及缺乏护理人员投入和药物依从性的综合地图为特征，这表明许多研究缺乏强有力的证据基础干预^{2, 3} .

不同地区实施各种干预模式和项目的忠实度得分不同，表明危机护理包的翻译存在差异⁴。危机表现的不可预测性和对紧急护理的需求使干预措施的评估变得复杂。危机参与者的随机化引发了棘手的伦理问题。

大多数评估都是从医疗服务提供者的角度审视问题，而用户对医疗服务评估的参与有限。植根于公民权利、社会正义和文化响应的消费者主导的运动在危机解决甚至预防方面似乎很有希望，需要纳入未来的评估中。志愿部门参与提供同伴支持，特别是为边缘化社区提供支持，虽然非常宝贵，但需要进行系统调查。

急性精神病护理的提供最近集中在远程精神病学和面对面互动的替代品上。虽然电话、视频会议设施和智能手机应用程序增加了资源、减少了等待时间、降低了成本并改善了获得护理的机会，但它们并没有解决与数字排斥、用户家庭隐私、治疗关系、护理质量和报酬模式相关的问题。这些技术正在等待用于常规临床实践的证据。

急性精神科护理的大部分证据来自高收入国家。低收入和中等收入国家的精神卫生保健，由于其财政和人力资源限制、城乡鸿沟和不同的精神疾病观点（例如，宗教和传统治疗师的解释模式、补充疗法、耻辱、禁忌）经常被标记由于卫生服务提供不足、缺乏循证干预指南和巨大的治疗差距。在解决精神卫生危机方面缺乏基于权利的方法、面向康复的反应和包容性的社区实践，以及专业精神卫生服务的高成本、不可及性和不可接受性，使情况变得复杂。

尽管一些项目取得了成功，但与急性精神病护理干预措施的功效、效果和成本效益相关的问题仍需审查⁵。虽然随机对照试验是循证医学的基石，但单个试验的结果或对一些此类调查的系统评价，同时提供有关治疗效果的证据（即“治疗在某处有效”），不一定提供临床实践中有效性的证据（即“治疗广泛有效”）。

将从随机对照试验中获得的知识外推到其他患者群体是有问题的。^{在广泛实施模型和计划6}之前，需要解决有效性（“它可以工作吗？”）、有效性（“它在实践中是否有效？”）和效率（“它值得吗？”）的证据。霍桑效应也混淆了创新干预与“标准护理控制臂”之间的比较。受试者的动机反应可能次要于通过观察和评估获得的兴趣、关心和关注，而不是由于特定的干预。

随着时间的推移，临床实践模式的变化、卫生系统之间的差异、患者人口统计和临床特征以及健康的社会决定因素⁷和心理健康⁸的变化，也会影响临床研究的普遍性。许多危机陈述在很大程度上受到人们生活的社会、经济和物质环境的影响。虽然有针对性的心理健康干预措施将帮助处于危机中的人们，但需要结构性、公共卫生和全人群干预措施来平衡健康结果的社会梯度 8 。

暴力）使跨服务和地区的比较变得困难。同样，不同的治疗干预（例如，心理、药理学、身体限制性实践）和阶梯式护理模式之间的差异使概括成为问题。

此外，危机表现的人群患病率差异、求助行为的差异以及不同类型临床干预的阈值差异进一步使普遍性复杂化。预算、社区和医院基础设施以及人力资源方面的差异增加了比较的复杂性。尽管一些模型取得了成功，并且需要创新方法，但不同地区的不同现实使得确定普遍适用的模型的任务具有挑战性。

虽然干预措施的评估是强制性的，但它们的成功并不自动意味着它们可推广到其他环境。事实上，许多通常以项目模式运作的复杂项目，由于其高水平的财政、行政和政治支持而取得成功，但即使在类似的环境中也难以扩大规模。它们在不同地区、卫生系统和国家的实施可能极具挑战性。

急性精神病表现的异质性、干预措施的多样性和环境的多样性要求需要了解背景。当地环境的现实及其独特的问题需要量身定制的解决方案。跨不同背景移植知识结构、形成和实践可能会导致缺乏拟合优度⁹。标准化协议可能无法识别与当地相关的问题，需要基于区域现实的背景分析和解释。对于急性精神病和危机表现的多方面和多学科干预包尤其如此。

临床实践中的决策应考虑更广泛的生物心理社会背景，包括临床、心理、社会和经济问题、医学发病率和风险，以及患者和护理人员的观点。在尝试复制成功项目的同时，面临的挑战是需要了解当地情况，整合省级知识并尝试实施适合区域的解决方案。