
'Attacks on staff are common': Inside hospital saving Haiti's malnourished children
In a simple breezeblock and cement building, cholera patients are attached to drips as they lie sprawled on hard, wooden beds.
In one section, two young boys stare into the distance through listless eyes. They are very poorly, the staff tell us, but now they are here, they will survive.
Medical staff check on their patients in the relatively cool interior of the wards, while outside the sun beats down on the grounds of the rough and ready interconnected buildings of the Fontaine Hospital in Port-au-Prince.
The hospital is built amid the slums in an area of Haiti's capital known as Cite Soleil - or Sun City.
This suburb is widely regarded to be the birthplace of the gangs of Port-au-Prince, and this section of the city has been violent and dangerous for decades.
Civil society doesn't function here. Indeed, the Fontaine Hospital is the only medical facility still operating in the gang-controlled areas of Cite Soleil.
Without it, the people who live here would have no access to doctors or medical care.
I'm standing in the cholera ward with Jose Ulysse, the hospital's founder. He opened the hospital 32 years ago. It's a charity, run purely on donations.
Mr Ulysse explained that the increasing gang violence across the whole of Port-au-Prince, and the chaos it is causing, means people are herded into displacement camps, which in turn means that cholera outbreaks are getting worse.
"Cholera is always present, but there's a time when it's more," he told me.
"Lately because of all the displacement camps there is a great deal of promiscuity and rape, and we have an increase in cases."
As we spoke, I asked him about the two young boys, and a small group of women on drips in the ward.
"Now they are here, they will be okay, but if they weren't here and this hospital wasn't here, they would be dead by now," he replied when I asked him about their condition.
We left the cholera ward, cleaning our hands and shoes with disinfectant, before moving on to the next part of the hospital under pressure - the malnutrition ward.
"Malnutrition and cholera go hand-in-hand," Mr Ulysse explained as we walked.
In the clinic, we meet parents and their little ones - all the infants are malnourished.
The mothers - and important to note - one father, are given food to feed their babies.
Those who are in the worst condition are also fed by a drip. One of the giveaway signs of malnutrition is a distended tummy, and most of these babies have that.
Poverty and insecurity combine to cause this, Mr Ulysse tells me. And like cholera, malnutrition is getting worse.
He explained that when the violence increases, parents can't go to work because it is too dangerous, so they end up not being able to make a living, which means that they can't feed their children properly.
The medics and hospital workers risk their lives every day, crossing gang lines and territories to get to the hospital and care for their patients.
The reason why this hospital is so popular is because staff show up, even when the fighting is at its worst.
Despite their meagre resources, the Fontaine Hospital's intensive care unit for premature babies is busy - it is widely regarded as one of the best facilities of its kind in the country.
A team of nurses, masked and in scrubs, tenderly care for these tiny children, some of whom are only hours old.
They are some of the most incredibly vulnerable.
I asked Mr Ulysse what would happen if his hospital wasn't there.
"Just imagine, there isn't a place where they can go, everyone comes here, normally the poorest people in the country", he told me.
But he stressed that the only way the hospital can keep going is through donations, and the cuts to the US government's USAID programme has had a direct impact on the hospital's donors.
Attacks on hospitals and staff working in the toughest areas across Port-au-Prince have become common.
We filmed outside one of the two Médecins Sans Frontières facilities in the centre of the capital, where work has been suspended because their staff were threatened or attacked.
Medical personnel from the health ministry in Port-au-Prince tell us over 70 per cent of all medical facilities in Port-au-Prince have been shut. Only one major public hospital, the Le Paix Hospital, is open.
The Le Paix Hospital's executive director, Dr Paul Junior Fontilus, says he is perplexed by the gang's targeting of medical facilities.
"It makes no sense, it's crazy, we don't know what it is they want," he said as we walked through the hospital.
The hospital is orderly and functioning well, considering the pressure it is under. They are dealing with more and more cases of cholera, an increase in gunshot wounds and sexual violence.
"We are overrun with demand, and this surpasses our capacity to respond," he explained to me.
"But we are obliged to meet the challenge and offer services to the population."
7:17
Gang violence is crushing the life out of Port-au-Prince, affecting all of society. And, as is often the case, the most vulnerable in society suffer the most.
Hashtags

Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles


Geeky Gadgets
33 minutes ago
- Geeky Gadgets
What Is a Triage AI Agent Automation & Multi-Agent Systems Explained
What if the key to solving complex, high-stakes challenges in industries like healthcare, cybersecurity, or customer service wasn't a human expert, but an intelligent system that never sleeps? Enter the world of triage AI agents, a new evolution in automation that's reshaping how organizations prioritize and manage tasks. Inspired by the life-or-death urgency of medical triage, these systems don't just organize workflows—they make data-driven decisions in real-time, making sure that critical issues are addressed immediately while routine tasks are handled with precision. In a world where delays can cost lives, revenue, or reputations, triage AI agents are proving to be indispensable allies. IBM Technology explores the mechanics and fantastic potential of triage AI agents, revealing how they combine multi-agent systems, large language models (LLMs), and domain-specific knowledge to transform task management. You'll uncover how these systems operate behind the scenes, from collecting and assessing data to routing tasks with unparalleled efficiency. Whether you're curious about their role in reducing patient wait times, thwarting cyber threats, or enhancing customer satisfaction, this deep dive will illuminate why triage AI agents are more than just another automation tool—they're a glimpse into the future of intelligent decision-making. Could this be the breakthrough your industry has been waiting for? Triage AI Agents Overview How Triage AI Agents Work Triage AI agents streamline workflows by automating the intake, assessment, and routing of tasks. This structured process is composed of three primary components: Intake Agent: This component collects data from users or systems through conversational interfaces or APIs. It connects to knowledge sources such as historical records, templates, or client-specific data to ensure the information gathered is both accurate and relevant. This component collects data from users or systems through conversational interfaces or APIs. It connects to knowledge sources such as historical records, templates, or client-specific data to ensure the information gathered is both accurate and relevant. Assessment Agent: Once data is collected, this agent evaluates the information using domain-specific knowledge, search APIs, or LLMs. It identifies the task's nature, diagnoses potential issues, and determines the most appropriate course of action. Once data is collected, this agent evaluates the information using domain-specific knowledge, search APIs, or LLMs. It identifies the task's nature, diagnoses potential issues, and determines the most appropriate course of action. Routing Agent: In the final step, the task is either executed or directed to the appropriate resource. This is achieved through automation tools, communication platforms, or priority management systems. By integrating these components, triage AI agents reduce delays, optimize task management, and enhance overall operational efficiency. Their ability to handle complex workflows with minimal human intervention makes them a valuable asset for organizations aiming to improve productivity. The Evolution of Triage: From Medicine to AI The concept of triage has its roots in military medicine, where patients were prioritized based on the urgency of their conditions rather than their rank or status. This principle of prioritization has since been adopted across various industries where efficient task management is critical. In the digital age, AI agents have taken this concept to new heights by replicating and enhancing human decision-making through automation and artificial intelligence. These systems are designed to handle the growing complexity of modern workflows, offering precision and speed that surpass traditional methods. By automating the triage process, organizations can allocate resources more effectively, making sure that critical tasks receive immediate attention while routine tasks are managed systematically. What are Triage AI Agents? Watch this video on YouTube. Below are more guides on AI agents from our extensive range of articles. Applications Across Key Industries Triage AI agents are versatile tools with applications that span multiple sectors. Their ability to intelligently prioritize and manage tasks is transforming key industries in the following ways: Healthcare: Triage AI agents streamline patient intake processes, assess symptoms, and route cases to the appropriate medical professionals. This reduces wait times, enhances diagnostic accuracy, and improves overall patient care. Triage AI agents streamline patient intake processes, assess symptoms, and route cases to the appropriate medical professionals. This reduces wait times, enhances diagnostic accuracy, and improves overall patient care. Cybersecurity: These agents analyze potential threats, prioritize vulnerabilities, and direct incidents to the appropriate teams. By automating these processes, organizations can bolster their security posture and respond to threats more effectively. These agents analyze potential threats, prioritize vulnerabilities, and direct incidents to the appropriate teams. By automating these processes, organizations can bolster their security posture and respond to threats more effectively. Customer Service: In customer service, triage AI agents automate the prioritization of inquiries, making sure that urgent issues are addressed promptly while routine queries are resolved efficiently. This improves customer satisfaction and reduces response times. The adaptability of triage AI agents makes them suitable for any domain requiring intelligent task management. Their ability to integrate seamlessly with existing systems further enhances their utility across diverse applications. Technologies Powering Triage AI Agents The effectiveness of triage AI agents is underpinned by a combination of advanced technologies and tools. These technologies work in tandem to create intelligent, adaptable, and efficient AI solutions: Large Language Models (LLMs): Models like GPT enable triage AI agents to process and interpret natural language, enhancing their ability to assess and prioritize tasks accurately. Models like GPT enable triage AI agents to process and interpret natural language, enhancing their ability to assess and prioritize tasks accurately. Domain-Specific Knowledge: By incorporating industry-specific data, these agents make informed decisions that are tailored to the context of each task. By incorporating industry-specific data, these agents make informed decisions that are tailored to the context of each task. Search APIs: These tools allow agents to access and analyze external data sources, making sure comprehensive evaluations and informed decision-making. These tools allow agents to access and analyze external data sources, making sure comprehensive evaluations and informed decision-making. Frameworks: Platforms such as Langflow, Langchain, and Crew AI provide developers with the resources needed to build and customize triage AI systems, allowing them to meet specific organizational needs. These technologies collectively empower triage AI agents to deliver consistent and reliable performance across a variety of use cases. Why Triage AI Agents Matter The adoption of triage AI agents offers several significant benefits that address common challenges in modern workflows: Speed: By automating the prioritization and routing of tasks, these systems minimize delays and improve response times, making sure that critical issues are addressed without unnecessary lag. By automating the prioritization and routing of tasks, these systems minimize delays and improve response times, making sure that critical issues are addressed without unnecessary lag. Consistency: Automated decision-making eliminates human biases, making sure uniform and objective task prioritization across all operations. Automated decision-making eliminates human biases, making sure uniform and objective task prioritization across all operations. Scalability: Triage AI agents can handle increasing volumes of tasks without compromising performance, making them ideal for organizations experiencing growth or managing high workloads. By addressing these challenges, triage AI agents enhance operational efficiency and deliver measurable improvements across industries. Their ability to adapt to evolving demands ensures their long-term relevance and value. The Future of Triage AI Agents As digital transformation continues to accelerate, triage AI agents are poised to become an integral part of workflow automation. Their ability to integrate seamlessly with existing systems and adapt to changing requirements ensures their utility across a wide range of applications. In the future, these agents are expected to play an even more critical role in intelligent decision-making, helping organizations navigate complex challenges with greater ease. By using advancements in artificial intelligence and automation, triage AI agents will continue to redefine task management, offering scalable, consistent, and efficient solutions to meet the demands of an increasingly complex world. Media Credit: IBM Technology Filed Under: AI, Guides Latest Geeky Gadgets Deals Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, Geeky Gadgets may earn an affiliate commission. Learn about our Disclosure Policy.


Daily Mail
an hour ago
- Daily Mail
Ozempic linked to terrifying eye-bleeding disorder that turns patients blind
Ozempic has been linked to a potentially blinding condition that causes internal bleeding behind the eyes. An Israeli study found patients taking Ozempic or similar weight-loss shots were twice as likely to suffer from a serious degenerative eye disease compared to those not on the medication. Your browser does not support iframes.


The Independent
2 hours ago
- The Independent
Consultant should have referred teenager to intensive care, tribunal rules
A senior doctor should have referred a teenager to intensive care when she displayed several 'high-risk indicators' of sepsis days before she died, a medical tribunal has found. Martha Mills, 13, had been an inpatient on the Rays of Sunshine Ward at King's College Hospital in London after she suffered a serious injury to her pancreas when she slipped while riding a bike on a family holiday in Wales in July 2021. Weeks later she experienced a fever, increased heart rate and had a catheter inserted into her vein, which was 'ultimately considered' to be a likely source of the infection that led to her death from sepsis, the Medical Practitioners Tribunal Service (MPTS) hearing was told. More spikes in her temperature followed, before consultant hepatologist Professor Richard Thompson saw Martha on his morning ward round on Sunday August 29 at the hospital, one of three locations in the UK which specialise in the treatment of paediatric pancreatic injuries. The on-call consultant left the hospital at 3pm, but was phoned at home two hours later by a trainee doctor, who gave an update on Martha's condition. Medical records showed she had deteriorated over the course of the afternoon, and into the early evening, with a drop in her blood pressure, the appearance of a new rash and increases in heart rate, respiratory rate and body temperature. Tribunal chairman Robin Ince noted that by 5pm there were 'several high-risk indicators' as set out in the Nice guidelines relating to sepsis. He said: 'These changes collectively indicated a sudden and significant deterioration for no clearly identified reason…it was evident that by around 5pm the clinical condition of Martha justified escalation to PICU (paediatric intensive care unit) and this opportunity was not taken.' The duty registrar called Prof Thompson again at 8.30pm because of ongoing concerns over Martha's fever, but she was kept on the ward despite the continued presence of moderate to high-risk indicators and the absence of meaningful clinical improvement, said the tribunal. Mr Ince said: 'The tribunal notes that Professor Thompson expressed a preference to prearrange any PICU involvement, rather than having PICU clinicians arrive unexpectedly and cause distress to Martha's parents. 'While the tribunal understands the desire to manage the family anxiety sensitively, this did not, in its view, justify withholding or delaying a clinically indicated escalation of care.' He added: 'The tribunal therefore concluded that the GMC (General Medical Council) has proved its case that Professor Thompson failed to take more aggressive intervention from 12 noon onwards in that he did not escalate Martha to the PICU team for a direct clinical review.' The tribunal also ruled that Prof Thompson should have conducted a direct in-person review and assessment of Martha, including the developing rash, from 5pm, which would have been 'beneficial'. Mr Ince said: 'He would no doubt have discovered that no observations had been taken in respect of Martha from 2pm to at least 4.45pm. 'The tribunal notes that one of the advantages of escalating a patient to PICU would have been an increased level of monitoring. 'Although no explanation has been given to the tribunal for the failure to monitor Martha between 2pm and 4.45pm, this absence of monitoring on the ward would no doubt have been a significant factor in justifying such an escalation.' The tribunal cleared Prof Thompson of the GMC's allegations that he gave 'outdated, misleading' information on Martha's condition to a consultant colleague in the intensive care unit, and that he failed to mention her rash. Martha collapsed on August 30 and was moved to intensive care, before she was transferred to London's Great Ormond Street Hospital, where she died in the early hours of August 31. At a 2022 inquest into her death a coroner ruled Martha would most likely have survived if doctors had identified the warning signs and transferred her to intensive care earlier. Martha's mother, Merope Mills, an editor at The Guardian, said she and her husband, Paul Laity, raised concerns about Martha's deteriorating health a number of times but these were not acted on. The couple later successfully campaigned for Martha's Rule to give patients, families and carers the chance to easily request a second opinion from a senior doctor in the same hospital in the event of a suspected deterioration or serious concern. The MPTS hearing in Manchester continues as the tribunal considers whether Prof Thompson's fitness to practise is impaired.