Looking at both existing and new global surveillance models, the review intends to point out any structural issues and recommend actions to make the architecture fairer, easier to monitor and more stable.

Surveillance is often divided into indicator-based surveillance (IBS) and event-based surveillance (EBS). Laboratory-confirmed cases, hospital records, and morbidity reports form the basis of IBS. It is very important to keep an eye on endemic diseases and see how their trends develop progressively.⁷ However, IBS frequently takes a long time to gather facts and confirm events which hinders rapid response (Figure 3).⁸

EBS looks at nonstructured data such as news and social media topics and notes from researchers, to identify potential outbreaks. EBS is particularly good at spotting rare or unexpected events ahead of confirmation which helps IBS a lot. A mixed model that includes both systems is often accepted as being necessary to find outbreaks quickly.⁹

In 2000, the GOARN started serving as the main response team for the WHO.¹⁰ There are over 250 technical institutions and organizations in this network which join together fast to aid those who are struck by disasters. GOARN increases the global ability to respond during emergencies, but relies on member countries reporting and requests for support.⁵

Since its establishment in 1952, GISRS has been recognized as one of the oldest and best-working disease monitoring systems. Over 150 laboratories in a wide range of countries contribute to the network, enabling the instant sharing of influenza virus samples and data.⁶ The system watches seasonal flu trends and every 2 years determines the formulation for next year’s flu vaccine based on expert-driven reviews. It is an inspiration for other surveillance systems because of how it uses standardized protocols and helps data travel internationally. However, since it focuses just on influenza, it is not very helpful for dealing with other kinds of pandemic threats (Figure 4).¹¹

After the West African Ebola outbreak in 2014, the GHSA was founded as a multilateral initiative to improve global health security more quickly. It helps provide 11 key technical services—such as surveillance, laboratory tasks, staff education and emergency management—mostly in low- and middle-income countries (LMICs). Emphasizing certain activities (such as the Joint External Evaluation (JEE)) helps GHSA tackle common problems faced by countries most prone to undetected outbreaks. On the other hand, its need for donor money and the fact that groups join by choice threaten its sustainability (Figure 5).¹²

The GDD program is managed by the CDC and has regional centers set up in areas like Southeast Asia, Africa, and Latin America.¹³ GDD provides help and solutions during outbreak investigations, works on making labs and epidemiology stronger and improves training offered in the field. Many agree with its evidence-based methods, although their effectiveness can be influenced by changing global interests and policies of the United States (Figure 6).¹⁴

Because of advancements in technology these platforms now provide early warnings about novel outbreaks before they spread widely. Started by the International Society for Infectious Diseases in 1994, ProMED-mail shares expertly compiled information from many sources and has issued important alerts on SARS and COVID-19.¹⁵ The Canadian government created GPHIN which uses computer algorithms to review information across multiple languages and identify new public health dangers. HealthMap which is led by Harvard, uses immediate data mining from digital sources to map out the spread of outbreaks (Figure 7).¹⁶

Table 1 below presents a comparative summary of these surveillance systems across key operational dimensions, including coverage, early warning lead-time, estimated costs, and data-sharing policies.

At the heart of IBS systems are the equipment and setup in laboratories that help with identifying and verifying pathogens. However, capacity is not the same everywhere in the world. Advanced health systems in rich countries allow them to do multiple tests with real-time PCR and next-generation sequencing, but LMICs continue to face challenges such as having basic laboratories, skilled staff and a reliable supply of necessary reagents and equipment.¹⁸ This challenge delays pathogen detection and identification, leading to increased reliance on larger laboratories, which further prolongs outbreak response.¹⁹

The WHO’s Strengthening Laboratory Services initiative and funding via the GHSA are trying to reduce the gap.¹⁸ Through the Regional Integrated Surveillance and Laboratory Network, the Africa CDC is demonstrating that more localized diagnostic systems are possible, although worldwide testing is still limited.¹⁹ Even though RDTs and point-of-care tools provide quick help, they differ in their ability to detect diseases and are often not enough to deal with new infectious pathogens.²⁰

In many LMICs, weak health systems, constant political problems, and insufficient methods for reporting cause difficulty in surveillance. Since many zoonotic diseases start in these less-studied parts of the world, the absence of their data is very concerning. With GHSA and multilateral organizations using tools like the JEE to assess countries, the problem is that funding is often unstable and support from top officials is not always guaranteed.²¹

Fear of economic or political problems could deter a country from transparent reporting, which delays and could reduce the effectiveness of global early warning. For surveillance to improve in LMICs, it needs better technology, legal security, ways to build trust and incentives to be more open.²²

Using AI, machine learning and natural language processing (NLP) in disease surveillance has changed how outbreaks are discovered and followed.²³ BlueDot, Metabiota, and HealthMap take big data from all sorts of open sources and use algorithms to discover signs of a possible disease outbreak.¹⁸

Such tools have been shown to predict and notify people about hazards ahead of official notifications. There are reports that BlueDot spotted signs of COVID-19 much earlier in December 2019 than any international organization.¹⁹ Even so, these systems deal with issues related to the bias found in data, the interpretability of their algorithms, and how alerts should be confirmed. If a model depends too much on theories that have not been properly verified, chances for mistakes or overlooking something are higher.²¹

AI surveillance can lead to collateral issues, such as false positives, mass panic, and algorithmic bias. Overreliance on opaque models has a chance of missing marginalized areas with insufficient digital data.

Although there is limited information on the cost-effectiveness values, the research indicates that ProMED and BlueDot issued alerts about the potential risk of COVID-19 infection up to 7–10 days before the official warnings. During H1N1 and Zika, HealthMap appeared equally efficient in detecting them at an early stage. Models designed by AI like BlueDot were said to cost less than 1 million dollars per year, which is a small fraction in terms of pandemic response expenditures that ran by the billions. Comparative, systematic efficacy cost-per-case averted, lead-time benchmarking, and false positive rates, remain scarce but necessary for future prioritization.

With reduced traditional monitoring, mobile tools have greatly boosted how much data can be collected. With SMS reporting, mobile health apps and community involvement in surveillance, health professionals get real-time information about disease in remote places.¹³ Outbreaks Near Me and AfyaData from sub-Saharan Africa show how communities can use their health surveillance data to help overall efforts.²²

Yet, these ways of teaching rely on being familiar with technology, good internet or cellular coverage and users taking part regularly. However, getting mobile-collected data into national and world systems is still hard due to incompatibility among platforms and differences in standards in data validation.²¹

Under the IHR (2005), adopted by 196 countries, healthcare systems internationally must respond to shared illnesses.²³ Countries must develop main public health capacities for noticing threats, judging risks and dealing with them and must inform the WHO of any PHEIC within a day of assessment.²⁵ Even though the IHR are designed to emphasize cooperation instead of rule-breaking, they have no real enforcement measures.

Even though the pact is ratified by most nations, many countries still fail to comply with the regulations. Many low- and middle-income countries are unable to meet the IHR core standards mainly due to money, infrastructure, and policy issues. Countries that are rich and powerful have in some instances reported delays or understated risk for economic or strategic reasons.²⁶ The fact that the IHR depends on countries to report their data and does not have strong tools to verify compliance or penalize noncompliance, shows a main gap in its governance. During the first weeks of the COVID-19 outbreak in Wuhan, unclear information from China slowed the world’s response, making the weaknesses of the world’s system clearer.²⁷

The voluntary nature of the IHR establishes poor incentives toward timely reporting because countries could be afraid of being hit by trade, political, or economic sanctions. Without sanctions or effective enforcement mechanisms, underreporting becomes institutionalized.

The WHO, OIE and FAO have adopted the One Health approach to unite human, animal and environmental health surveillance. Since over two-thirds of EIDs come from animals, this cooperation helps spot these events early.²²

However, fully adopting One Health has been difficult. Problems with coordinating information, individual data systems and uneven approaches to funding keep sectors apart. Even though initiatives such as the Tripartite Zoonoses Guide create frameworks, their use is often blocked by the lack of coordination between health, agriculture and environment ministries and the bureaucratic hurdles.²⁸

There are several bilateral and multilateral agreements that back global surveillance and each agreement offers its own set of pros and cons. Through the GDD program, the CDC partners with single countries quickly but these steps might not always consider global health equity over the interests of the donor.²⁹

Besides, multilateral initiatives such as the GHSA or projects led by the WHO prefer to work with different countries, but they seldom succeed in building agreements, ensuring funds are present or putting their efforts together to avoid duplication. Having many groups involved without clear rules for accountability often leads to rivalry instead of teamwork which is evident in moments of crisis.¹⁸

A major problem in disease surveillance management is when outbreak information is influenced by political factors. Trying to protect country sovereignty, the risk of being banned from travels, or risks to their reputation can lead to deliberately not reporting information or not doing it immediately.¹³ Early in the outbreak of Ebola in West Africa and COVID-19 in China, important information was not shared which limited efforts to stop both viruses.¹⁹

Because no set dispute resolution exists in the IHR, countries do not have many reasons to comply with the guidelines.¹⁵ Political factors usually take priority over health-related decisions and since there is no official accountability, the world system relies a lot on good relationships and gentle diplomacy.

It is very difficult to achieve global surveillance because it must balance protecting national independence with the need for openness. Some countries might avoid transmitting full information on an outbreak since they worry it could lead to sanctions, lower their reputation, or influence political advantage.¹⁹ They can seriously get in the way of early detection and response efforts. Examples like the SARS outbreak in 2003 and the early part of the COVID-19 pandemic reveal that not reporting quickly enough given political reasons can expand containable local issues into problems that affect the whole world.⁴

There is also a lack of trust from social groups and individuals toward data-sharing initiatives. Most LMICs feel that providing data, mainly genomic, might not result in receiving fair access to diagnostics, vaccines, or treatments. Since the benefits are not equal, this makes it less likely for people to trust each other and openly share data.²¹

It is difficult for different computer systems and groups to exchange, organize and use data which is an important technological problem. There are many platforms for disease surveillance, but most work apart from one another as they do not use the same formats, systems, or protocols.²² Custom platforms that are closed-source and do not support API use are common among authorities in public health, research centers and NGOs which negatively affects the analysis of data gathered in different areas.²⁹

Lack of agreed digital networks during the pandemic resulted in important epidemiological data being used inefficiently and understanding the situation coming more slowly. Projects such as the Digital Square initiative and the Global Digital Health Partnership are working to encourage the use of open standards and similar data governance, though take-up is not spread evenly around the world. Where there are no laws or rewards for cooperation, setting up digital disease surveillance is still a haphazard project.³⁰

People are very passionate about the topic of who owns and accesses genomic surveillance data. Because of platforms such as GISAID, virus genome data from many countries could be shared quickly, making it easier to follow the different SARS-CoV-2 variants.³¹ This, however, is not without controversy, as it puts limits on what can be done with the shared data, which some scientists believe compromises transparency and repeatability.²⁵

On the other hand, general-access resources such as GenBank and Nextstrain make their data available for all but run the risk of data being misused by some, not being properly acknowledged and having too little data protection for the countries donating the data. There is a broader issue in science worldwide, since different systems may create tension between patent right and the medical needs of the many.²²

Apart from needing access, using outbreak data can cause serious ethical issues, especially in terms of who gets the benefits. Many LMICs that give genomic data and pathogen samples are usually prevented from playing a role in the development of the science and property surrounding their donations. Global South countries have provided valuable genomic data, yet most of the vaccines are still not going to them which highlights the injustice.²⁰

Besides, since not everyone has the same access to digital tools and knowledge, many nations are excluded from taking part in or profiting from, global surveillance systems. While ongoing efforts like courses and workshops have been made, they have not managed to equalize access.¹²

The COVID-19 infodemic demonstrated that modern information systems can both help and harm. Digital systems are quick to distribute public messages but also accelerate the spread of false information and conspiracy theories, which reduces trust in health monitoring and support. Algorithms on popular social media sites share emotional content (Figure 8).³²

Therefore, governments and public health agencies should treat strategic communication as a core part of their surveillance efforts.³⁰ Engaging with the public in real time, being transparent about uncertainties, and collaborating with trusted community members are needed to combat misinformation and build public trust.³²

The first major alert about the COVID-19 outbreak was generated by ProMED-mail and not public health departments at the state level. On the last day of December 2019, ProMED reported a group of pneumonia cases in Wuhan, China, before the WHO had shared any news.¹³ It demonstrates how EBS can contribute by uncovering critical information from clinicians, media, and similar sources.¹⁸

Likewise, the GISAID made it possible to quickly share genetic information on SARS-CoV-2 soon after the outbreak was recognized. The open-access system at GISAID, with some licenses, enabled researchers everywhere to monitor the virus and create new tools to address it, much quicker than ever.²¹

Africa CDC and its network of regional surveillance units are major steps toward letting Africa take the lead in managing epidemic information. These hubs facilitate real-time information sharing and utilization, enhance team training, and foster better cooperation among member countries.³³ The Africa CDC contributed to COVID-19 prevention and control efforts by strengthening genomic surveillance and supporting vaccine distribution which showed that building capacity is crucial for underserved communities.³¹

Next strain and Pathogenwatch have turned genetic information into useful knowledge. Seeing how pathogens spread in real time helps researchers and health agencies react faster to new variants, measure their mutation rates, and decide on updates to vaccines and restrictions for travel.³⁴ The dashboards developed for interactive diagnosis and infectious disease tracking are strong evidence of what open science and real-time information can do.³⁵

AI has demonstrated its potential in epidemic prediction. Using NLP and airline information from around the world, the company BlueDot found a possible health threat in Wuhan on December 31, 2019.²⁹ It predicted that COVID-19 would spread within the next week before authorities gave clear reports. It shows that AI-based models may increase the strength of traditional surveillance by recognizing little-known, complex trends in numerous streams of data.³¹

Integrating genomic surveillance with other forms of data is needed right away. This integration allows for rapid tracking of pathogen evolution, the emergence of antimicrobial resistance, and the unfolding of disease outbreaks. An unequal distribution of sequencing access worldwide is making it more challenging for early warning systems to work well in LMICs.³⁶ By establishing both decentralizing sequencing and cloud-based data transfer pathways, the goal of democratizing pathogen genomics can be achieved. It is very important to give training in bioinformatics and genomic epidemiology to local scientists and public health workers.³²

The guide in this flow chart (Figure 9) simplifies how public health officials can align the characteristics of an emerging threat with the most appropriate surveillance strategy given some of the dynamics are pathogen familiarity, infrastructure, and digital access.

Because AI and digital tools are being used quickly in outbreak response and prediction, formal governance systems are now needed. They ought to cover privacy issues, accountability in algorithms and ethical rules. WHO and similar organizations ought to be given the necessary powers and support to oversee agreed standards and help nations collaborate on health issues.³⁴ Digital governance principles can be added to an international treaty or updated IHR to support innovation and rights.

It is necessary for the global health community to use financial incentives to encourage sharing of data instead of stockpiling it. For example, such actions include credit systems, official agreements on data sharing and systems that guarantee developing countries equal access to important tools related to vaccines, diagnostics, and treatments.³⁷ Consistency in standards and APIs encourages systems to communicate easily among countries and organizations.

A sustainable approach involves ongoing building of surveillance in places where risks are highest. Capacity building should be redefined from development aid to a key aspect of global security.¹⁷ A sustainable financial system should offer stability, flexibility, and high accountability to support infrastructure, foster workforce development, and strengthen regional partnerships. Building trust among stakeholders and community groups can foster greater self-reliance in surveillance efforts.²⁴