“Percolating.” That’s the word I’ve heard people use to describe the current status of Ebola here in Liberia. That’s what Dr. Peter Piot meant when he said he expects the outbreak to last through the end of 2015. Containing Ebola has been hard; ending Ebola will be harder.

For responding clinicians, “percolating” means monotony broken up by moments of emergency. It means maintaining fully staffed ETUs across the region even if their beds are empty, because we’re not sure when and where Ebola will appear next. The contact tracers and community surveillance officers are working hard to track all the suspects, but when every person with fever, weakness, vomiting and diarrhea is a suspect, it becomes a monumental task. And if a hotspot arises in an area without the infrastructure to manage it, we risk letting all of Liberia slip back to what it was in August – dying of Ebola.

But it’s no longer August, and it seems we’re entering a new phase of this crisis, one marked by tentative recovery under the shadow of constant readiness. While certainly a cause for celebration, this phase carries it’s own challenges. As the country tries to balance crisis and recuperation, the relationship between national health system and the internationally-sponsored ETUs grows more complicated.

The local clinics and hospitals have begun to reopen, and they’re calling their staff back to work, staff who were hired by foreign NGOs to man ETUs when the response was scaling up. So while efforts to return to normalcy are a positive sign, the general shortage of healthcare workers means that for now, the national hospitals and the ETUs have to share.

Those health facilities that are accepting patients remain hindered by the constant threat of a new outbreak. Memories of fellow clinicians that fell to the virus are still so fresh. If someone arrives at a clinic exhibiting any Ebola symptoms, the clinic stops all operations and waits for an ambulance to remove the patient.

The vigilance required to prevent Ebola’s rebound slows a health system that, after being closed for months, is desperately trying to catch up. All those people that were sick before the Ebola crisis now need treatment more than ever. Meeting that demand without compromising safety is difficult, to say the least.

For the ETUs, different challenges exist. Every suspect requires proof of a negative Ebola blood test before a local clinic or hospital will treat them. Signed by a doctor, the certificates we give out act as a ticket to care. To receive one, your symptoms must have started 72 hours before your blood tests negative for Ebola (research suggests an Ebola-positive individual can test negative in the first three days because their viral load is so low). This means that Ebola-negative patients in acute distress have to wait for up to 72 hours before being transferred to a facility properly equipped to care for them. The delay is potentially life-threatening, but with Ebola still percolating in the region, there are few alternatives that keep the national health infrastructure safe.

As an anecdote, a week ago we picked a patient up from a local hospital for Ebola-like symptoms. The patient denied recent contact with a sick or dead person, but complained of new onset nausea/vomiting and the hospital reported a fever. She was also pregnant, early in her third trimester.

Once admitted to the “Suspected” ward of our ETU, we found her to be afebrile. The nurse on duty drew her blood, but since her symptoms had just started, she’d have to wait 72 hours for a second test to confirm she was Ebola-negative.

Overnight, the woman’s blood pressure spiked. When the next round of nurses donned PPE and entered the ward, they found her on the ground, confused and lethargic. The doctor on shift suspected the woman had suffered a seizure due to eclampsia or pre-eclampsia, a potential life-threatening complication of pregnancy.

Our treatment unit has none of the equipment to diagnose or treat this problem, nor should we. The growing network of ETUs across Liberia is not intended to replace the country’s normal health infrastructure. And building up our capacity to treat anything other than Ebola would only confuse a population that’s just beginning to understand and not fear the ETUs’ existence. “You have symptoms of Ebola? Go the the ETU, we don’t kill you, we’ll do our best to make you better. You have any other health problem? Go to your local hospital and stay away from the ETU, coming here is an unnecessary risk.” For very good reasons, we stick to treating Ebola and Ebola only.

While necessary, these protocols worked against our pregnant patient. By morning she had improved, but she still had 48 hours before we could retest and discharge her to the hospital for more definitive care. She made it, but not without placing significant stress on our clinicians, who could only watch, hope and wait.

As we enter the “percolation” stage of Ebola in Liberia, it’s important to remember that these are good problems, ones that stem from the flood of international support, the declining transmission rate and the slow reboot of the country’s health system. And for an Ebola outbreak of unprecedented scale, these problems will continue to evolve. The best we can do is our best, and attempt to solve them as they come.