The research team took an unusual methodological route: a rapid review, sometimes called an expedited systematic review. Rather than starting from scratch, they combined the InvaCost database, a global living review of invasive species cost estimates, with an independent literature search modelled on InvaCost’s own systematic methods. InvaCost version 4.1 contained 2,597 cost estimates specific to Australia at the time of the study, but the researchers suspected that a state-level assessment would need finer, more recent data. Their targeted searches of Web of Science and Scopus, conducted in January 2024, plus grey literature and recent agricultural cost surveys, uncovered records the global database had not yet captured. In total, 217 references met the review criteria, and after screening out public expenditure, potential or avoided costs, duplicates, and low-reliability estimates, the final analytical dataset contained 374 individual cost estimates from 50 records.

One of the study’s central technical challenges was spatial. Most cost estimates in the literature are reported at national or regional scales, not for individual states. To address this, the team partitioned 90 national or regional estimates down to NSW using the Biosecurity Commons platform, a cloud-based decision-support tool. They overlaid species distribution layers built from Atlas of Living Australia occurrence records with Australian Land Use and Management Classification layers at one-square-kilometre resolution, calculating the proportion of each species’ range and impacted industry that fell within NSW borders. This spatial partitioning accounted for roughly half of the final total cost estimate, a key reason the study’s figure is nearly four times higher, in equivalent currency, than the NSW estimate in a previous Australia-wide review by Bradshaw and colleagues.

The temporal picture is stark. Average annual reported costs climbed from AU$25.51 million per year in the 1970s to AU$299.21 million in the 1980s, AU$456.22 million in the 1990s, and AU$661.22 million in the 2000s, before surging past AU$3 billion in multiple years of the 2010s and peaking at AU$3.822 billion in 2019. The apparent dip in the early 2020s, with 2022 costs reported at just AU$0.424 billion, is almost certainly an artefact of reporting lag rather than genuine relief: the dataset contains 255 distinct cost estimates from the 2010s but only 25 from the 2020s, reflecting the slow pace of academic publication and technical report release.

Using robust regression models designed to tolerate high variability and outliers, the researchers modelled the trend in annual costs from 1970 to 2022. A robust linear model projects costs of AU$2.076 billion for 2023, while a more conservative quadratic model yields AU$1.101 billion, both with extremely wide prediction intervals. The authors are emphatic that these figures should not be treated as forecasts, since the models lack the underlying covariates, such as climate, land use change, and management effort, that actually drive invasion costs. The models capture the general upward trajectory without overfitting to a tight pattern, and the researchers deliberately chose the simplest appropriate approach given the incompleteness of the data.

By taxonomy, the burden is overwhelmingly botanical. Terrestrial plants account for 82.9 percent of cumulative reported costs, with terrestrial vertebrates a distant second at roughly one-fifth of that share. The costliest plants are serrated tussock at AU$322 million, blackberry at AU$305 million, ryegrass at AU$153 million, fleabane at AU$130 million, and barnyard grass at AU$119 million. Among vertebrates, feral cats dominate at a staggering AU$2.291 billion, followed by European rabbits, wild dogs, feral pigs, and red foxes, each in the AU$390 to 443 million range. Invertebrate costs are led by oat aphids, blue oat mites, and lucerne fleas, while common carp, at AU$30 million, is the only aquatic species with species-specific estimates in the entire dataset.

The sectoral breakdown is even more lopsided. Industry and agricultural losses, dominated by production losses and pest control costs across pastoral, cropping, and forestry operations, make up 92.2 percent of all reported costs. Health, public, and social welfare costs account for 7.4 percent, almost entirely attributable to terrestrial vertebrates, including medical costs and damage to community assets. Everything else, including environmental costs and private research expenditure, falls below one percent. Environmental damage in particular is nearly invisible in the ledger: only about 0.2 percent of reported costs in this review relate to the environment sector, compared with roughly 4.1 percent in the earlier national review, a gap partly explained by this study’s exclusion of public expenditure on environmental management programmes.

This near-total absence of environmental and ecosystem service costs is the study’s most consequential finding. Non-market valuation methods such as contingent valuation, choice modelling, and benefit transfer are well developed and increasingly applied in biosecurity contexts, yet they remain rare in invasive species cost accounting. The authors point to work by Stoeckl and colleagues showing that ecosystem regulating services, including erosion control, flood control, carbon sequestration, and gene pool protection, capture around 44 percent of the benefits generated by Australia’s biosecurity system, far outweighing the roughly 25 percent contributed by agriculture, forestry, aquaculture, and fishing combined. When policymakers see cost estimates that omit these values, resources may be misallocated toward the most visible market losses rather than the greatest total damages.

The review also delivers a candid assessment of InvaCost itself, both praising its value and documenting its limitations. The independent searches found only a subset of the data already in the database, confirming that InvaCost captures substantial material, particularly grey literature comprising nearly half of its relevant records, that traditional systematic searches would miss. Yet the team also encountered apparent data duplication, difficulty tracing entries to original sources, and unextracted estimates hiding within already-cited references. Around 22 percent of the records used in the analysis came from outside the database, and these additional sources contributed a large share of the total reported costs, meaning the estimate would have been considerably lower without them.

Ultimately, the study offers a reproducible roadmap for other states, provinces, and subnational jurisdictions: filter InvaCost to the region of interest, partition broader estimates using spatial tools, supplement with independent searches and local expert engagement, and feed results back through the invacost R package. The authors stress that every figure they report is an underestimate, since costs are only counted in years where someone happened to publish a number, a bias they illustrate with feral pigs, whose costs appear in just 21 of the roughly five decades over which their impacts were ongoing. As globalisation continues to erode the geographic barriers that once kept Australia relatively pest-free, the message for biosecurity policymakers is unambiguous: the economic damage of invasive species is mounting, it is concentrated in agriculture but extends far beyond what markets can see, and the tools to measure it properly now exist and are waiting to be used.

Subject of Research: Economic costs of invasive species in New South Wales, Australia, assessed through a rapid review combining the InvaCost database with independent literature searches

Article Title: An invasive species cost review for New South Wales, Australia, highlights key drivers and limitations of economic cost estimates

Article References: Moran, N. P., Wang, L.-Y., Hanea, A. M., & Kompas, T. (2025). An invasive species cost review for New South Wales, Australia, highlights key drivers and limitations of economic cost estimates. NeoBiota, 104, 319-337. https://doi.org/10.3897/neobiota.104.157434

Image Credits: AI Generated

DOI: 10.3897/neobiota.104.157434

Keywords: invasive species, InvaCost, New South Wales, biosecurity, economic costs, agriculture, ecosystem services, non-market valuation, feral cats, weeds, rapid review, Australia

The research team conducted a rapid review of global case studies, a method designed to synthesize a broad and scattered evidence base within a structured analytical frame. Rather than cataloguing individual examples of unfairness, the authors interrogated the accumulated evidence to identify where and how climate action is engaging with the core pillars of justice that political philosophers and environmental justice scholars have developed over decades. Their conclusion is both encouraging and cautionary. The three classical dimensions of justice—distributive, procedural, and recognitional—are indeed being engaged with in urban climate initiatives worldwide, but the depth and simultaneity of that engagement determines whether an intervention becomes genuinely transformative or merely cosmetic.

Distributive justice, the first pillar, concerns who receives the benefits and burdens of climate policy. In practical terms, it asks whether a new flood barrier protects wealthy waterfront districts while informal settlements remain exposed, or whether urban tree-planting programs cool already leafy neighborhoods while heat-vulnerable districts stay bare. The review finds that cities increasingly deploy joint mitigation and adaptation interventions—projects that reduce emissions while building resilience—and that the distributional question is now explicitly on the agenda in many of them. Yet distribution alone is a blunt instrument. A solar subsidy program can be perfectly equal in formal terms and still be inaccessible to renters, low-income households, or residents without legal tenure, which is why the authors insist that distributional analysis must be paired with the other pillars.

Procedural justice addresses the second dimension: who participates in deciding. It examines whether affected communities have genuine voice in the design, implementation, and evaluation of climate interventions, or whether participation is reduced to token consultation after the key decisions have been made. The synthesized case studies show that exclusionary governance remains one of the most persistent barriers to equitable urban climate action. When planning processes are technically opaque, conducted in dominant languages only, or scheduled in ways that exclude working residents, the resulting policies tend to reproduce the very vulnerabilities they claim to address. The review identifies procedural reforms—co-production of plans with communities, transparent decision criteria, and accountability mechanisms—as a recurring feature of the more successful pathways.

Recognitional justice, the third classical pillar, goes deeper still. It asks whether the identities, histories, and knowledge systems of marginalized groups are respected within climate governance. Recognition failures can be subtle: Indigenous worldviews dismissed as irrelevant to technical flood modeling, gendered patterns of climate vulnerability ignored in mobility planning, or the specific exposures of migrants and ethnic minorities rendered invisible in aggregate city statistics. The authors find that recognition is increasingly invoked in climate action, but that it is often the thinnest of the three pillars in practice, acknowledged in policy language without changing who holds power or whose knowledge counts.

This is where the paper’s conceptual contribution becomes significant. The team extends the standard triad by integrating two additional dimensions: epistemic justice and restorative justice. Epistemic justice concerns the credibility afforded to different kinds of knowledge. In urban climate contexts, this means treating residents’ lived experience of flooding or heat as legitimate evidence alongside satellite data and hydrological models, and creating institutional channels through which local and Indigenous knowledge can genuinely inform technical decisions. Restorative justice, meanwhile, addresses deeply rooted systemic exclusions by confronting historical wrongs—colonial dispossession, discriminatory housing policy, and the extraction that built urban wealth in some places at the expense of others. A city cannot simply plan forward from an unjust baseline, the framework suggests; it must also account for how that baseline was produced.

The central analytical claim of the review is that climate initiatives are more likely to become transformative when they address multiple dimensions of justice simultaneously. This is a testable proposition with real design implications. A cooling center program that distributes resources fairly (distributive), is sited through community deliberation (procedural), respects the needs of elderly residents and informal workers (recognitional), incorporates their knowledge of neighborhood microclimates (epistemic), and acknowledges past disinvestment in their districts (restorative) is structurally different from a program that scores well on only one axis. The authors argue that single-pillar interventions tend to stall or backslide, because unresolved injustices in other dimensions eventually undermine them—participation without recognition becomes tokenism, and recognition without redistribution becomes symbolism.

From the synthesized evidence, the researchers distill key principles for enabling just and equitable action. While the review is careful not to prescribe a single template—urban contexts vary enormously between, say, a rapidly growing South Asian megacity and a shrinking post-industrial city in Europe—the principles converge on a common logic. Justice must be embedded from the earliest stages of intervention design rather than retrofitted after opposition emerges. Power must be genuinely shared with the communities who bear climate risks. Knowledge hierarchies that privilege formal expertise must be opened to plural ways of knowing. And institutions must be built to sustain these practices over time, because justice is not a one-time consultation but an ongoing condition of governance.

The practical payoff of the framework is a shift in what the field asks of research. As the authors put it in their abstract, the work is intended to move researchers and practitioners beyond identifying climate injustices toward an actionable roadmap for co-producing equitable climate solutions. Co-production is the operative term: the pathways the review maps are not top-down blueprints delivered to passive beneficiaries, but collaborative processes in which residents, municipal agencies, researchers, and civil society organizations jointly define problems and solutions. This framing reflects a broader movement in sustainability science over the past decade, but the review gives it sharper normative teeth by insisting that the quality of co-production be judged against all five justice dimensions.

The timing of this synthesis matters. Urban populations continue to grow fastest precisely in regions with the highest climate exposure and the weakest adaptive capacity, and national governments remain slow to deliver the finance and institutional support cities need. In that gap, municipal governments, community organizations, and hybrid actors are improvising climate action on their own—and the review’s evidence suggests that those improvisations succeed most often when justice is treated as constitutive of effectiveness rather than as a constraint on it. The authors declare no competing financial interests, and the paper is published open access, making the framework available to the practitioners it is meant to serve. What the study ultimately offers is a vocabulary and a checklist for a question that every city will face this century: not simply whether climate action happens, but who it is for, who decides, and whose past it repairs.

Subject of Research: Urban climate justice frameworks for equitable mitigation and adaptation in cities

Article Title: Overcoming urban climate injustice: synthesising global pathways and principles

Article References: Prakash, A., Bulkeley, H. A., Singh, C., McGregor, D., & George, R. (2026). Overcoming urban climate injustice: synthesising global pathways and principles. npj Urban Sustainability. https://doi.org/10.1038/s42949-026-00477-z

Image Credits: AI Generated

DOI: 10.1038/s42949-026-00477-z

Keywords: urban climate justice, climate adaptation, climate mitigation, distributive justice, procedural justice, recognitional justice, epistemic justice, restorative justice, co-production, urban sustainability, climate governance, rapid review

The review arrives at a politically charged moment. The authors explicitly note that rapid policy shifts in several jurisdictions have favored treatment-focused approaches to addressing homelessness, often framing addiction treatment as a gateway to housing. Against that backdrop, the team set out to determine what the published evidence actually says about the key components and outcomes of housing-based support models. Their answer, drawn from more than 13,000 screened records, is nuanced but firm: the strongest and most consistent evidence favors low-barrier Housing First models, while evidence for treatment-first models remains more limited and heterogeneous.

Methodologically, the review was designed for speed without sacrificing rigor. The researchers searched five major databases — CINAHL, Embase, Medline, APA PsycInfo, and Web of Science — for peer-reviewed studies published between January 1, 2000 and April 30, 2025. After screening 13,393 records, they included 57 studies: eight randomized controlled trials, thirty quasi-experimental studies, six qualitative studies, five mixed-methods studies, and eight other designs. Study quality was appraised using Joanna Briggs Institute tools, and the team extracted data on intervention characteristics, model features, comparison conditions, and outcomes spanning housing stability, substance use, mental health, and service utilization. That breadth of study designs, from gold-standard randomized trials to real-world quasi-experiments, strengthens the ecological validity of the findings while also revealing where the evidence base thins out.

From this body of work, four distinct program types emerged. Scattered-site models place participants in independent apartments dispersed across a community, with support services delivered where people live. Single-site models concentrate housing and on-site services within one building, offering concentrated support but sometimes recreating institutional environments. Transitional housing provides time-limited accommodation intended as a bridge toward permanent housing. Finally, outreach support-to-housing programs begin with street-level engagement and gradually connect individuals to housing. Each architecture carries different assumptions about what people experiencing homelessness need first — and each produces different patterns of outcomes in the literature.

On the headline question of housing stability, the verdict was unambiguous. Housing First models consistently improved housing stability compared with treatment-first, linear treatment-contingent models, or treatment-as-usual conditions. This finding held across randomized trials and observational studies alike, reinforcing what has become a robust international evidence pattern: when the immediate goal is ending homelessness, removing preconditions works. The mechanism is conceptually straightforward. Requiring people to demonstrate sobriety or complete treatment before accessing housing creates a barrier that many with active substance use disorders cannot clear, leaving them cycling through shelters, hospitals, and the street. Housing First inverts that sequence, treating stable housing as the platform from which health improvements become possible.

The picture grows more complicated when the review turns to substance use outcomes. Effects of Housing First on substance use were heterogeneous across studies, with alcohol-related outcomes improving more consistently than illicit drug use. Some trials found meaningful reductions in alcohol consumption and alcohol-related harms among participants in scattered-site Housing First, while effects on illicit drug use varied widely, with some studies showing no significant differences compared with control conditions. This heterogeneity matters for policy. It suggests that housing alone, while transformative for stability, does not automatically resolve the full spectrum of substance use health needs — particularly for stimulant and opioid use, where the evidence for spontaneous improvement in the absence of integrated treatment is weakest.

Mental health outcomes followed a similarly mixed pattern. Some studies reported no differences between Housing First and comparison conditions on measures of psychiatric symptoms, while others documented improvements in psychological distress. The reviewers caution against reading these mixed results as evidence of failure. Mental health trajectories among this population are shaped by many factors beyond housing, including trauma history, physical health, social connection, and access to psychiatric care. What the data do support is that Housing First does not worsen mental health outcomes — a persistent concern among critics — and in several studies actively reduces distress, all while delivering far superior housing outcomes.

Beyond housing and behavioral health, the review examined acute healthcare utilization and costs. Evidence for reductions in psychiatric hospitalization and overall costs was described as promising but less definitive than the housing stability findings. Several studies suggested that Housing First participants spend fewer days in hospitals and emergency departments, generating offsets that partially or fully compensate for the cost of housing and support services. However, the reviewers found this evidence less consistent across studies, likely reflecting differences in health system contexts, measurement approaches, and follow-up durations. For policymakers weighing fiscal arguments, the review suggests the cost case is plausible but not yet ironclad — an area where longer-term, more rigorous economic evaluations are needed.

The authors’ conclusions translate directly into policy recommendations. They call for scaling up low-barrier Housing First programs while integrating expanded voluntary substance use treatment, mental health care, harm reduction services, and wraparound supports within housing. The emphasis on voluntariness is significant: the evidence does not support coercive treatment mandates as a condition of housing, but it does support making comprehensive, choice-based services readily available to tenants who want them. The four program types identified in the review offer a menu rather than a single prescription — scattered-site models may suit people seeking independence, while single-site and transitional models may serve others — but the common thread is that housing comes first, not last.

The review also candidly maps the gaps that remain. While some treatment-first models reported positive outcomes, the evidence base for them was more limited and heterogeneous than for Housing First, making head-to-head comparisons difficult. The authors highlight the need for further comparative research and, critically, more robust longer-term studies to identify which specific program components best support substance use health among this high-needs population. As jurisdictions worldwide grapple with visible homelessness crises and intensifying debates over housing versus treatment paradigms, this review offers a rare commodity: a rigorously synthesized evidence base. Its message is neither that housing solves everything nor that treatment is irrelevant, but that the sequence matters — and the science says house people first, then wrap every effective service around them.

Subject of Research: Housing-based support models for adults experiencing homelessness with substance use and mental health needs

Article Title: Housing-based support models for individuals experiencing homelessness, substance use and mental health needs: a rapid review

Article References: Somefun, O. D., Kerman, N., Corace, K., Taha, S., Buckley, L., & Stergiopoulos, V. (2026). Housing-based support models for individuals experiencing homelessness, substance use and mental health needs: a rapid review. Addiction Science & Clinical Practice. https://doi.org/10.1186/s13722-026-00716-9

Image Credits: AI Generated

DOI: 10.1186/s13722-026-00716-9

Keywords: homelessness, Housing First, substance use, mental health, harm reduction, treatment-first models, housing stability, rapid review, addiction science, psychiatric hospitalization, supportive housing, public health policy

The research was commissioned to inform the World Health Organization’s efforts to develop a practical toolkit for human resources for health planning, often abbreviated as HRH planning. The stakes are considerable. Effective planning ensures a balance between the workforce and patient needs, optimizing both the quality of care and the job satisfaction of health and care workers. A diverse mix of approaches and methodologies tailored to different health-system contexts has been developed over the years, but the authors note that many health systems have simply not implemented them effectively. The gap between what the planning literature offers and what ministries of health actually use has persisted through pandemics, economic shocks and sweeping health reforms, and closing it requires knowing precisely what tools exist, what they demand of their users and where their blind spots lie.

To map that landscape, the team employed a rapid review methodology, a streamlined form of systematic review designed to deliver rigorous evidence under time constraints. Working in consultation with experienced library scientists, the researchers developed both a database search strategy and a targeted grey literature search strategy, the latter capturing reports, manuals and working papers that never appear in peer-reviewed journals but often document the real-world use of planning models. Only documents written in English or French were included, and each source was carefully screened against explicit inclusion criteria. Relevant articles were then extracted, and the resulting data were used to classify the tools into distinct planning approaches, such as needs-based models, and to rate them along five domains: accessibility, capacity, adaptability, usability and applicability.

The scale of the resulting inventory is striking. From 123 articles that passed screening, the team identified a total of 96 distinct planning tools that have been applied across 184 countries. That figure alone tells a story: workforce planning is not a niche technical exercise but a genuinely global undertaking, with tools deployed in high-income countries and low- and middle-income countries alike. The tools were classified into eight planning approaches and assessed against ten criteria spanning the five assessment domains. Among the well-known models captured in the review are the Workload Indicators for Staffing Need method, widely used to translate service workload into staffing requirements, and Health Labour Market Analysis, which examines the supply of and demand for health workers as an economic system rather than a static headcount.

The headline finding, however, is a negative one: no tool received a high score across all assessment criteria. Every instrument in the review, no matter how sophisticated or widely adopted, carries trade-offs. Some models that excel at projecting future requirements demand granular, disaggregated data that many countries cannot readily produce. Others are easy to use but too rigid to adapt to different epidemiological profiles, service delivery models or professional mixes. The review also found no consensus, either within the review itself or among the broader HRH community, on a single best planning tool. This absence of a clear winner is not a failure of the field so much as a reflection of the diversity of contexts in which planning takes place; a tool calibrated for a data-rich national health service may be poorly matched to a decentralized system with fragmented information systems.

Still, the analysis did identify a small group of standouts. Four tools received high ratings on five key criteria: capacity for multi-professional and scenario-based planning, opportunity to engage stakeholders, adaptability across settings, and utility of results. These are exactly the qualities planners prize. Multi-professional capacity means a model can account for the full team, from physicians to community health aides, rather than optimizing one cadre in isolation. Scenario-based planning allows users to test how demographic change, task-shifting or new treatment guidelines might alter future requirements. Stakeholder engagement ensures that the people who must implement the resulting plans, including professional associations, educators and service managers, have a voice in the assumptions that drive the numbers. And adaptability determines whether a tool built in one country can be meaningfully applied in another with a very different disease burden and resource base.

Yet even these top performers come with caveats that matter enormously for real-world uptake. The four leading models require moderate user capacity and substantial data, necessitating disaggregate data for model operation and some prior knowledge of HRH planning from the user. In practice, this means the tools most likely to produce useful, credible results are also the tools most likely to stall in settings where workforce information systems are weak or where planning expertise is concentrated in a handful of national officials. The review’s authors are candid about this tension, and it shapes their central recommendation: the value of the project depends greatly on the WHO’s complementary capacity-building initiatives, which they describe as essential for real-world uptake by countries. A toolkit, in other words, cannot simply be published and downloaded; it must be accompanied by training, data-system investment and sustained technical support.

The technical details of the assessment framework deserve attention from anyone who plans or evaluates planning tools. By rating each instrument on ten criteria across the five domains of accessibility, capacity, adaptability, usability and applicability, the review creates a common vocabulary for comparing tools that were previously described in incompatible ways. Accessibility speaks to how easily a country can obtain and license the tool. Capacity captures the skills and data the user must bring. Adaptability measures how well the model can be reconfigured for new contexts. Usability reflects the learning curve and the clarity of documentation. Applicability, perhaps the most consequential criterion, asks whether the outputs are actually useful for decision-making. This structure allows the forthcoming WHO toolkit to guide users toward tools matched to their own capacity and data realities rather than toward a one-size-fits-all recommendation that the evidence does not support.

The timing of the review is significant. Health workforces worldwide emerged from the COVID-19 pandemic depleted and demoralized, with the WHO and its member states confronting projected shortfalls of millions of health and care workers concentrated disproportionately in low- and middle-income countries. Planning tools are among the few levers that can turn anxiety about shortages into concrete strategies for education, recruitment, retention and deployment. By cataloguing 96 tools and their applications across 184 countries, the review provides an evidence base for countries seeking to expand beyond their traditional planning processes and work towards sustained improvements in their health systems, as the authors put it. It also signals a shift in emphasis at the global level: from producing yet another planning model to curating, assessing and supporting the use of the many models that already exist.

For health planners, the practical message is to start from an honest appraisal of local data and capacity before selecting a tool, to favour models that support multi-professional, scenario-based planning and stakeholder engagement where capacity permits, and to pair any tool with the training and data investments it requires. For researchers, the review exposes persistent gaps, including the lack of consensus on best practice and the untested assumption that high-scoring tools translate into better workforce outcomes on the ground. For the WHO, the findings set a clear specification for the guidance toolkit now under development: it must help countries match tools to context, build the user capacity those tools demand, and integrate planning into the routine machinery of health-system governance. The full open-access article, published on 6 October 2026 in BMC Health Services Research, offers the detailed assessments that will underpin that work.

Subject of Research: Systematic review of human resources for health planning tools to inform WHO guidance for health workforce planning

Article Title: A comprehensive review and analysis of human resources for health planning tools to inform the development of a guidance toolkit for HRH planning

Article References: Tomblin Murphy, G., Sampalli, T., Diallo, K., Chamberland-Rowe, C., Nair, T. S., Lawrence, L., Murdoch, J., Kunjumen, T., Boniol, M., Rigby, J., MacInnis, M., & Murphy-Boyle, K. (2026). A comprehensive review and analysis of human resources for health planning tools to inform the development of a guidance toolkit for HRH planning. BMC Health Services Research. https://doi.org/10.1186/s12913-026-14326-3

Image Credits: AI Generated

DOI: 10.1186/s12913-026-14326-3

Keywords: health workforce, human resources for health, HRH planning, systematic review, World Health Organization, planning tools, health systems, workforce planning, rapid review, health policy, capacity building, BMC Health Services Research