But such a high fever by itself is not enough to have forced him to be kept in the hospital, according to several experts in emergency room practices.
"I have certainly discharged many people with a 103-degree fever," said Dr. Kristi L. Koenig, professor of emergency medicine and director of the Center of Disaster Medical Services at UC Irvine.
There are other considerations, such as, "are they able to walk, or talk or eat? It's not just the number," she said.
In Duncan's case, the key issue was not the degree of his fever, but that he had just arrived from Liberia, a center of the
Duncan was the first person to die from Ebola in the United States. He had gone to Texas Health Presbyterian Hospital on Sept. 25 complaining of symptoms.
In a statement issued Friday, Texas Health Presbyterian said it had made procedural changes and continues to "review and evaluate" the decisions surrounding Duncan's care.
It previously defended the quality of care it offered Duncan, after several people close to him complained that the Liberian was treated less effectively than American missionaries who returned from Africa to the United States for treatment.
Duncan was infected in Africa, though officials said he did not have any symptoms when he boarded a series of flights that brought him to Dallas on Sept. 20.
He stayed with his fiancee and others in a Dallas apartment. None of them has developed symptoms, and all are still in quarantine.
His fiancee and her relatives provided Duncan's medical records to the Associated Press — more than 1,400 pages in all. The records include Duncan's time in the emergency room and his return to the hospital Sept. 28 when his condition worsened.
According to the AP, Duncan initially complained of abdominal pain, dizziness, a headache and decreased urination when he arrived at the emergency room.
He reported severe pain — rating it an 8 on a scale of 10. Doctors gave him CT scans to rule out appendicitis, stroke and numerous other serious ailments. Ultimately, he was prescribed antibiotics and told to take Tylenol, then returned to the Dallas apartment where he was staying.
"I have given patient instructions regarding their diagnosis, expectations for the next couple of days, and specific return precautions," according to the emergency room physician's note. "The condition of the patient at this time is stable."
Doctors would not discuss the specifics of Duncan's case, but agreed that a 103-degree temperature alone wasn't enough to have him admitted.
"In general, the presence of the fever doesn't mean that patient X needs to be admitted," said Dr. David Pigott of the University of Alabama at Birmingham. "It is a mark of infection.
"There are other causes of fevers," he continued. "It is not unusual to see someone with strep throat and a fever of 101, 103. Almost all of the time they are released with antibiotics. It is not necessarily linked to the need for admission."
"There is no fever that is the sole determinant of whether a patient has a particular disease or whether [he or she] needs to be hospitalized," said Dr. Tom Sugarman, the past president of the California chapter of the American College of Emergency Physicians. Sugarman practices in Berkeley and Antioch.
"Typically, we see people all of the time, you can have a fever from a cold, kidney infection, cancer, running a marathon," he said. "Fever alone is not whether you are hospitalized."