Scholarlink

Background <p>Health systems are expected to reduce environmental impacts while preserving safe, continuous and equitable care. Artificial intelligence (AI) may help optimize energy, water, waste and circular-economy processes, but the evidence is dispersed across sectors and its relevance to nursing and health policy is uncertain.</p> Methods <p>We conducted a cross-sector scoping review and evidence map of studies published from January 2020 to April 2025. Searches were conducted in Scopus and Web of Science, with backward reference-list checking. Eligible studies evaluated an AI-enabled intervention, decision-support pathway or optimization process and reported a quantified environmental sustainability endpoint, sustainability-relevant operational proxy or clearly linked circular-economy process outcome. Outcomes were mapped as real-world direct environmental endpoints, simulated or modeled direct endpoints, indirect operational proxies or technical enabling metrics. Full platform-specific search strategies are provided in Supplementary file 1 (Supplementary Table <InternalRef RefID="MOESM1">S1</InternalRef>).</p> Results <p>After a tightened eligibility audit, 11 studies were included. No study evaluated AI-enabled sustainability interventions in routine healthcare delivery or nursing practice. Energy-management studies were concentrated in building heating, ventilation and air-conditioning control or microgrid scheduling. Direct environmental evidence was sparse and was either non-healthcare real-building evidence or simulated/modeled estimates, including building energy use, irrigation water use and route-derived emissions or carbon-cost indicators. Indirect evidence included demand-response flexibility, operating-cost reductions, productivity or profit improvements, operational-efficiency gains, battery reuse-pack performance and bioleaching process optimization. Technical metrics such as forecasting or classification accuracy were treated as implementation-enabling evidence and were not interpreted as environmental effects.</p> Conclusions <p>The current evidence base is better characterized as a cross-sector evidence map than as a focused healthcare effectiveness review. AI-enabled sustainability applications may offer candidate functions for health-system infrastructure and operations; however, nursing implications are interpretive, hypothesis-generating and not directly evidenced. Future healthcare studies should evaluate environmental endpoints alongside patient safety, infection prevention, nursing workload, equity and the lifecycle footprint of AI systems.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Artificial intelligence for environmental sustainability across sectors: a scoping review and evidence map for nursing and health policy

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月30日

查看全文
  • Fatma M. Ibrahim,
  • Ghada Shahrour,
  • Abdulkarem Radhi Alanazi,
  • Husain Mohammed Al Hakami,
  • Asmaa Alsuwayed

摘要

Background

Health systems are expected to reduce environmental impacts while preserving safe, continuous and equitable care. Artificial intelligence (AI) may help optimize energy, water, waste and circular-economy processes, but the evidence is dispersed across sectors and its relevance to nursing and health policy is uncertain.

Methods

We conducted a cross-sector scoping review and evidence map of studies published from January 2020 to April 2025. Searches were conducted in Scopus and Web of Science, with backward reference-list checking. Eligible studies evaluated an AI-enabled intervention, decision-support pathway or optimization process and reported a quantified environmental sustainability endpoint, sustainability-relevant operational proxy or clearly linked circular-economy process outcome. Outcomes were mapped as real-world direct environmental endpoints, simulated or modeled direct endpoints, indirect operational proxies or technical enabling metrics. Full platform-specific search strategies are provided in Supplementary file 1 (Supplementary Table S1).

Results

After a tightened eligibility audit, 11 studies were included. No study evaluated AI-enabled sustainability interventions in routine healthcare delivery or nursing practice. Energy-management studies were concentrated in building heating, ventilation and air-conditioning control or microgrid scheduling. Direct environmental evidence was sparse and was either non-healthcare real-building evidence or simulated/modeled estimates, including building energy use, irrigation water use and route-derived emissions or carbon-cost indicators. Indirect evidence included demand-response flexibility, operating-cost reductions, productivity or profit improvements, operational-efficiency gains, battery reuse-pack performance and bioleaching process optimization. Technical metrics such as forecasting or classification accuracy were treated as implementation-enabling evidence and were not interpreted as environmental effects.

Conclusions

The current evidence base is better characterized as a cross-sector evidence map than as a focused healthcare effectiveness review. AI-enabled sustainability applications may offer candidate functions for health-system infrastructure and operations; however, nursing implications are interpretive, hypothesis-generating and not directly evidenced. Future healthcare studies should evaluate environmental endpoints alongside patient safety, infection prevention, nursing workload, equity and the lifecycle footprint of AI systems.