By Dr. Ifeoma Oriaku, Founder of Udara Health

This reflection describes a prior clinical experience. Identifying details, including gender, relationships, and specific circumstances, have been changed to protect patient privacy.

 

You walk into the room to see your first patient of the day, and they look at you with terror in their eyes. Before they speak, you can hear what they are trying to say:

I am drowning. Can you help me?

In that moment, more than one action might be reasonable, but only a narrow range of decisions will safely carry this particular patient through what comes next. How do you decide which path offers them the best chance?

We returned them to BiPAP, a tightly fitted mask that uses pressure to support breathing, and gave a diuretic to help clear the fluid that was drowning them. Gradually, their breathing became less labored. The terror in the room softened.

But they still required nearly all the respiratory support we could provide without placing a breathing tube. Stabilizing them was only the beginning of the difficult work.

Once they could speak, another responsibility emerged. I needed to understand what mattered most to them if their condition worsened.

Would they accept intubation followed by invasive mechanical ventilation? If they survived the immediate crisis but could not be liberated from the ventilator, would prolonged life support or a tracheostomy be acceptable? Would they want cardiopulmonary resuscitation if their heart stopped? Were there outcomes they would consider worse than death? Who should speak for them if they could no longer speak for themselves?

Goals-of-care conversations help us understand the life a person values, what they hope for, what they fear, and which burdens they would accept for a meaningful possibility of recovery.

When I asked whether this conversation had happened, the patient told me it had not occurred clearly. I needed them to recover, but it was equally important that their family understand what might lie ahead.

Although I am not family, I sometimes have the privilege, and burden, of recognizing the possibility of death before a family is ready to see it. I have learned to brace my own heart while helping them prepare theirs. I find it devastating when families are not given an opportunity to use what may be their final moments to share memories, reconcile, express love, and prepare for death.

Preparing for the worst does not mean surrendering hope. My approach is to prepare honestly for the worst, hope and pray for the best, and work relentlessly alongside the team toward the outcome we all hope is still possible.

We asked their spouse to come in.

By the time they arrived, the patient looked more comfortable than earlier, which made the conversation understandably difficult. The patient did not appear to be dying in that moment, and the possibility of death had not previously been discussed so directly. Yet the patient remembered how close they had come to the edge that morning and how frightening it had been.

We did not ask the patient or their spouse to give up hope. We explained that despite everything we were doing, death remained a real possibility. The patient engaged in the conversation, asked thoughtful questions, and asked for time to speak privately with their spouse and another physician they trusted. Ultimately, they clarified that they wanted us to continue full life-sustaining treatment, including intubation and tracheostomy if necessary.

Their decision did not remove the uncertainty. It gave us a values-based direction.

During my time caring for them, a reasonable and increasingly urgent question surfaced each day:

"When are we going to intubate?"

Intubation can be lifesaving. When a person can no longer sustain the work of breathing, protect the airway, remain alert enough to cooperate with treatment, or maintain adequate oxygenation, intubation followed by mechanical ventilation may provide the support necessary to survive. Waiting too long can make an already dangerous procedure even more hazardous.

But intubation is not a neutral transition. Given the severity of this patient's lung injury, body habitus, and limited cardiopulmonary reserve, I worried it could begin a cascade: sedation, delirium, loss of mobility, further deconditioning, ventilator-associated complications, prolonged dependence on mechanical ventilation, tracheostomy, and possibly progression toward death.

The medications required for intubation suppress consciousness and may temporarily stop spontaneous breathing. In someone with severe oxygen failure and little reserve, the oxygen level can fall precipitously during those moments, even when the airway itself is not technically difficult. If the procedure were successful, a prolonged and uncertain course could still follow.

Yet there was also a small but meaningful possibility that their lungs would respond to treatment without intubation. Remaining awake would give them more time to participate in decisions and communicate with their family, even if intubation eventually became necessary.

We were holding two real dangers at the same time: intubating before it was necessary, and waiting beyond the point when it could be performed safely.

The team's concern about missing the safest window was appropriate, and I shared it. So was my concern that intubation itself could alter the patient's trajectory. Our task was not to eliminate uncertainty, since we could not, but to build enough safety around it to keep reassessing which risk had become greater.

Their lungs were not failing in isolation. Their heart and kidneys were also struggling, creating a difficult physiological loop.

When pressure rises within the circulation, fluid can move into the lung interstitium and airspaces, and oxygen must then cross a thicker, wetter barrier to reach the bloodstream. Removing excess fluid can reduce the burden on the lungs and heart, but aggressive diuresis can sometimes worsen kidney function, particularly when renal perfusion is already fragile.

Reducing diuresis might protect the kidneys from further injury while allowing pulmonary congestion to worsen. Continuing it might improve oxygenation while accelerating the need for dialysis. Dialysis would require placing a large catheter, not a small undertaking in someone who could not safely lie flat without risking respiratory collapse.

Neither the pulmonary nor the kidney concern was wrong. We were not choosing to care about one organ while disregarding another. We were repeatedly asking which imbalance posed the most immediate threat to the whole person.

I explained to the patient that my intention was never to sacrifice one organ while caring for another. Although diuresis could worsen laboratory measures of kidney function, persistent venous congestion could also impair the kidneys while worsening pulmonary edema. If their oxygenation deteriorated further, every organ, including the kidneys, would suffer. In that moment, I believed careful fluid removal offered the more favorable overall balance of benefit and risk, while accepting that dialysis might still become necessary.

Whole-person care does not mean every organ can always be protected equally. Sometimes it means recognizing how the organs are connected, making the least harmful tradeoff available, and revisiting that decision as the physiology changes.

I believed their lungs might still respond to disease-directed treatment. Our plan included corticosteroids, continued fluid management, intensive oxygen support, antimicrobial treatment, and awake proning.

At the same time, we kept revisiting the diagnosis. We considered infection, inflammatory lung injury, autoimmune disease, inhalational or chemical injury, pulmonary edema, organizing pneumonia, etc. Additional consultants created safety guardrails as we intensified anti-inflammatory therapy and continued diuresis.

None of us could know with certainty whether the plan would work, or whether it would work quickly enough, so we did not simply wait.

We created a closely monitored therapeutic trial. We communicated repeatedly with the patient and their spouse, consultants, nurses, respiratory therapists, and the rest of the care team. We monitored not only oxygen levels but also work of breathing, mental status, hemodynamics, ability to manage secretions, kidney function, treatment tolerance, and overall trajectory.

We remained prepared to change course if the physiology told us the window for noninvasive support was closing. Worsening respiratory distress, exhaustion, unsafe agitation, aspiration risk, hemodynamic deterioration, worsening gas exchange, or evidence that our working hypothesis was wrong could all shift the balance toward intubation.

In medicine, the most visible intervention is not always the most active treatment. Supporting someone without intubation can require extraordinary vigilance, coordination, labor, and resource use.

Awake proning became one of the few remaining interventions we could attempt without intubation. Helping a patient lie on their abdomen can improve how air and blood flow are distributed through injured lungs, allowing different regions to participate more effectively in gas exchange.

It sounds simple. It was not.

Staff had to protect oxygen devices, intravenous lines, skin, pressure points, and overall safety while closely monitoring the patient. The patient had to tolerate an uncomfortable position while critically ill, frightened, and short of breath.

With their spouse present, they tolerated three to five hours of proning during the day. Their spouse rubbed their back, cracked jokes, offered encouragement, and helped them stay oriented while the team monitored closely. That night, with their spouse still beside them, they remained on their abdomen for another remarkable five hours.

Their spouse's presence created familiarity that helped the patient tolerate an intervention, and it became part of the environment that allowed them to participate in the care their body needed.

Nurses and respiratory therapists repeatedly performed work that could easily be overlooked but was essential to giving the strategy a chance. Sometimes highly technological treatment receives the most attention, yet some of the most meaningful ICU care comes through human coordination, careful observation, trust, and a willingness to repeat labor-intensive interventions.

For several days, the patient's condition remained tenuous, with little sustained progress.

I questioned myself. Was I giving a biologically plausible treatment enough time to work, or was I becoming attached to my own hypothesis? Was continued noninvasive support preserving a meaningful possibility of recovery, or were we moving closer to an unsafe emergency intubation?

I revisited the differential diagnosis, examined the signals that supported or weakened our working hypothesis, consulted with colleagues, and reinforced the safety guardrails.

The team was understandably uneasy. Their concern reflected a legitimate fear that we could miss the safest opportunity to intubate, a concern I shared. My concern was that intubation itself might move this patient toward a course from which they would struggle to recover, or never recover.

Both concerns came from a desire to protect them.

A strong multidisciplinary team does not require everyone to feel equally confident. It requires people to voice their concerns, examine competing risks, define what they are watching for, and remain aligned around the patient. Discomfort within a team can be useful when it sharpens the plan rather than dividing the people responsible for carrying it out.

The patient and their spouse were also uncertain. We never asked them to manufacture confidence. We explained what we believed, what we did not know, what we were watching, and what would cause us to change course. Although unsure, they remained engaged and trusted the process.

I cherished that trust not as validation that I was right, but as a responsibility to remain attentive, honest, and willing to update the plan.

Then the trajectory changed.

After corticosteroids, continued fluid removal, intermittent awake proning, antimicrobial treatment, and intensive supportive care, the patient's oxygen needs fell dramatically. They progressed from maximal noninvasive respiratory support to low-flow oxygen. They could get out of bed and eat.

The atmosphere in the room changed. The patient and their spouse could see possibility again, and the team seemed rejuvenated, with renewed energy for the work ahead.

I cannot say that one intervention determined this improvement. I cannot separate the effects of corticosteroids from diuresis, proning, antimicrobial treatment, oxygen support, time, or the patient's own biological capacity to recover, nor can I separate their resilience from their spouse's steadfast presence or the extraordinary work of the nurses, respiratory therapists, consultants, and physicians who cared for them.

A favorable response does not prove that every part of my original hypothesis was correct, and it does not mean another patient with similar oxygen needs should be managed the same way. What it does suggest is that, in this case, a carefully monitored therapeutic trial was worth giving.

This experience reminded me that clinical courage is not the belief that we cannot be wrong. It is the ability to make a reasoned decision when certainty is unavailable: building a hypothesis from the available clinical signals, acknowledging what does not fit, establishing safeguards, defining what failure might look like, and continuing to listen to the patient's physiology.

Stubbornness protects a conclusion. Clinical courage stays committed to the patient while letting the conclusion change if it needs to.

Critical illness creates pressure for everyone. The patient is struggling to breathe. Family members are frightened. Nurses and respiratory therapists are watching minute-to-minute changes. Consultants and physicians understand how quickly apparent stability can disappear, and no one wants to realize, in retrospect, that an intervention came too late.

Under those conditions, action can relieve more than physiological danger; it can also relieve the emotional discomfort of uncertainty. That doesn't make action wrong. Sometimes immediate escalation is exactly what the patient needs. But clinicians have to be willing to ask whether they're intervening because the patient's physiology requires it now, or partly because the uncertainty has become hard to tolerate. The reverse question matters just as much: am I continuing the current plan because the patient still has a safe window, or because I've become attached to being right?

Both action bias and therapeutic inertia are human, and neither can be eliminated completely. They can, however, be made more visible through team discussion, explicit escalation thresholds, repeated reassessment, and a culture where people can voice concern without being seen as disloyal or obstructive.

This case also reminded me why clinicians need to protect their mental health.

When our nervous systems stay chronically overwhelmed, uncertainty becomes harder to tolerate. Sleep deprivation, emotional exhaustion, accumulated grief, and prolonged hypervigilance can narrow attention toward the most immediate threat. Activation of the sympathetic nervous system and hypothalamic-pituitary-adrenal axis prepares us to respond quickly during an emergency, and that response can be protective. When it becomes chronic, though, it can impair working memory, reduce cognitive flexibility, and make it harder to hold several competing possibilities in mind at once.

In clinical care, that narrowing can look like premature closure, defensive escalation, reluctance to reconsider a plan, or action taken partly to quiet the anxiety in the room.

Rest doesn't make clinicians infallible. Personal healing doesn't guarantee the correct diagnosis. Emotional regulation doesn't remove the uncertainty that's inherent to medicine.

I don't believe my own healing made me more certain. It helped me tolerate being uncertain. It gave me more internal steadiness to hear the team's concerns without becoming defensive, question my own hypothesis, stay present with the patient and their spouse, and avoid using an intervention just to relieve the pressure in the room.

I'm deeply grateful for the community that has supported my healing as I continue practicing medicine grounded in humanity.

Caring for our mental health helps preserve what our patients actually need from us: attention, flexibility, humility, and presence. It lets us notice our own fear without asking the patient's treatment plan to resolve it, and it helps us stay regulated enough to communicate honestly and think clearly while the situation keeps evolving. Prioritizing clinician well-being isn't separate from excellent patient care. It's part of what makes that care possible.

The patient's improvement brought hope back into the room. Hope matters in critical illness. It can help a patient participate in proning, mobility, nutrition, pulmonary hygiene, and the long work of recovery. It can help a family stay present, and a care team keep doing difficult work.

But hope doesn't require certainty or a promised outcome. Sometimes it just means recognizing that possibility remains, and caring for that possibility while still preparing for what may come next.

This patient's story was still unfolding when I wrote this reflection. I remain grateful for their effort, their spouse's presence, and the vigilance of the nurses, respiratory therapists, consultants, and physicians involved in their care. The lesson I carry forward isn't that escalation should always be delayed, or that intuition should override concern. It's that good medicine sometimes asks us to act immediately, and other times asks us to create enough safety, support, and time for the right treatment to work. Wisdom is in figuring out, again and again, which moment you're actually in.

And to do that well, clinicians need care too.

Why I Founded Udara Health

I chose to work as a locum physician while building Udara Health for one reason: it removes the bureaucracy that can get in the way of whole-person care in a traditional practice or hospital setting. And as a physician, I believe better care starts with better understanding. That's the philosophy I've carried into Udara Health.

A lot of patients I see in clinic have been living with their own quieter version of exhaustion and shortness of breath, with some now believing it is just a part of getting older, and have bounced between specialists. The theme of most of these visits is that patients have been going through this process for years, and their visit with me is the first time they've understood, felt listened to, felt empowered, etc.

Patients in the ICU with severe illness range from previously healthy to patients having several chronic illnesses. Most of these patients have chronic medical illnesses, with several of these illnesses preventable or better managed with support. Thus, one reason Udara Health was created is to care for patients not in silos, but in the context of their lived experiences, risk factors, and resource availability, acknowledging the complexities of their diagnosis. By doing this, the hope is to empower patients with the tools to live a healthful life, reduce their health burden, and hopefully, lower their risk of hospitalization and critical illness.

For those who end up hospitalized, critically Ill, and in post-acute illness, feeling stuck, Udara Health also serves as that extra support that patients, clinicians, and families can tap into to feel empowered with a whole-picture recovery map they can follow toward their own personal health journey.

I started Udara Health because the fifteen-minute visit of traditional medicine rarely leaves room for root-cause thinking or the kind of honest, unhurried conversation that helped this patient and their spouse work through real uncertainty. Sleep, breathing, and lifestyle issues deserve that same kind of attention, and ideally, they get it before things turn into a crisis, not only after.