FORT SILL, Okla. -- Jessica Zeppa, five months pregnant, the wife of a soldier, showed up four times at Reynolds Army Community Hospital here in pain, weak, barely able to swallow and fighting a fever. The last time, she declared that she was not leaving until she could get warm.
Without reviewing her file, nurses sent her home anyway, in a wheelchair, with an appointment to see an oral surgeon to extract her wisdom teeth.
Zeppa returned the next day, in an ambulance. She was airlifted to a civilian hospital, where despite relentless efforts to save her and her baby, she suffered a miscarriage and died Oct. 22, 2010, of complications from severe sepsis, a bodywide infection. Medical experts hired by her family said later that because she was young and otherwise healthy, she most likely would have survived had the medical staff at Reynolds properly diagnosed and treated her.
"She was 21 years old," her mother, Shelley Amonett, said. "They let this happen. This is what I want to know: Why did they let it slip? Why?"
The hospital doesn't know, either.
Since 2001, the Defense Department has required military hospitals to conduct safety investigations when patients unexpectedly die or suffer severe injury. The object is to expose and fix systemic errors, often in the most routine procedures, that can have disastrous consequences for the quality of care. Yet there is no evidence of such an inquiry into Zeppa's death.
The Zeppa case is emblematic of persistent lapses in protecting patients that emerged from an examination by The New York Times of the nation's military hospitals, the hub of a sprawling medical network - entirely separate from the scandal-plagued veterans system - that cares for the 1.6 million active-duty service members and their families.
Internal documents obtained by The Times depict a system in which scrutiny is sporadic and avoidable errors are chronic.
As in the Zeppa case, records indicate that the mandated safety investigations often go undone: From 2011 to 2013, medical workers reported 239 unexpected deaths, but only 100 inquiries were forwarded to the Pentagon's patient-safety center, where analysts recommend how to improve care. Cases involving permanent harm often remained unexamined as well.
At the same time, by several measures considered crucial barometers of patient safety, the military system has consistently had higher than expected rates of harm and complications in two central parts of its business - maternity care and surgery.
More than 50,000 babies are born at military hospitals each year, and they are twice as likely to be injured during delivery as newborns nationwide, the most recent statistics show. And their mothers were more likely to hemorrhage after childbirth than mothers at civilian hospitals, according to a 2012 analysis conducted for the Pentagon.
In surgery, half of the system's 16 largest hospitals had higher than expected rates of complications over a recent 12-month period, the American College of Surgeons found last year. Four of the busiest hospitals have performed poorly on that metric year after year.
Little known beyond the confines of the military community, the Pentagon's medical system has recently been pushed into the spotlight. In late May, Defense Secretary Chuck Hagel ordered a review of all military hospitals, saying he wanted to determine if they had the same problems that have shaken the veterans system.
Hagel said the review would study not just access to treatment, the focus of investigations at the veterans hospitals, but also quality of care and patient safety -- issues that The Times has been looking at, and asking the Pentagon about, for months.
Defense Department health officials say their hospitals deliver treatment that is as good as or better than civilian care, while giving military doctors and nurses the experience they may one day need on the battlefield. In interviews, they described their patient-safety system as evolving but robust, even if regulations are not always followed to the letter.
"We strive to be a perfect system, but we are not a perfect system, and we know it," said Dr. Jonathan Woodson, assistant secretary of defense for health affairs. He added, "We must learn from our mistakes and take corrective actions to prevent them from reoccurring."
The Times' examination, based on Pentagon studies, court records, analyses of thousands of pages of data, and interviews with current and former military health officials and workers, indicates that the military lags behind many civilian hospital systems in protecting patients from harm. The reasons, military doctors and nurses said, are rooted in a compartmentalized system of leadership, a culture of interservice secrecy and an overall failure to make patient safety a top priority.
The investigations of unforeseen deaths or permanent harm, called root-cause analyses, are widely regarded as a centerpiece of efforts to make care safer. Asked about the military's missing inquiries, Dr. James P. Bagian, director of the University of Michigan's Center for Healthcare Engineering and Patient Safety, said, "If in fact unexpected deaths were reported and ignored, there would appear to be no good answer for that except that someone is sleeping at the switch."
Avoidable errors can and do occur at the best of hospitals. But the military's reports show a steady stream of the sort of mistakes that patient-safety programs are designed to prevent.
The most common errors are strikingly prosaic - the unread file, the unheeded distress call, the doctor on one floor not talking to the doctor on another. But there are also these, sprinkled through the Pentagon's 2011 and 2012 patient-safety reports:
A viable fetus died after a surgeon operated on the wrong part of the mother's body.
A 41-year-old woman's healthy thyroid gland was removed because someone else's biopsy result had been recorded on her chart.
A 54-year-old retired officer suffered acute kidney failure and permanent hearing loss after an incorrect dose of chemotherapy.
Such treatment failures are known as "never events," because they are potentially so grave -- and so preventable. They do not happen frequently. But a persistent rate of such mistakes can indicate broader patterns of slipshod care.
Malpractice suits can also be a rough indicator of risk. From 2006 to 2010, the government paid an annual average of more than $100 million in malpractice claims from surgical, maternity and neonatal care, records show. It would be paying far more if not for one salient reality of military health care: Active-duty service members are required to use military hospitals and clinics, but unlike the other patients, they may not sue. If they could, the Congressional Budget Office estimated in 2010, the military's paid claims would triple.
Experience in civilian hospitals, and in the veterans system, has shown that stricter procedures and more sophisticated surveillance can limit errors, sometimes markedly. Among some in the military network, concerns about patient safety are longstanding, if rarely acknowledged in public. But calls for change have consistently foundered in the convoluted bureaucracy.
The military heath system is split into three major branches, with the Army, Navy and Air Force each controlling its own hospitals and clinics. The Pentagon's Defense Health Agency also runs the Walter Reed National Military Medical Center and Fort Belvoir Community Hospital, both outside Washington. Any systemwide change involves a carefully calibrated consensus of three equally ranked surgeons general, as well as the Defense Health Agency. Woodson, who oversees the system, cannot order the surgeons general to act. He can only recommend.
Progress can be glacial: In 2007, for example, the military started rewriting regulations for handling events that harmed or endangered patients. It finished only last October. Several former Pentagon officials said embarrassing statistics were often filtered out, glossed over or buried amid larger data sets before they reached senior health leaders. Two measures used in major civilian hospitals to monitor quality of care - rates of death and readmission, adjusted for seriousness of illness - are simply not tracked. "The patient-safety system is broken," Dr. Mary Lopez, a former staff officer for health policy and services under the Army surgeon general, said in an interview.
"It has no teeth," she added. "Reports are submitted, but patient-safety offices have no authority. People rarely talk to each other. It's 'I have my territory, and nobody is going to encroach on my territory.'"
In an internal report in 2011, the Pentagon's patient-safety analysts offered this succinct conclusion about military health care: "Harm rate -- unknown."
Lethal medical errors
Katie Guill checked into the hospital at Fort Leonard Wood, Missouri, on Christmas morning 2008, expecting to give birth to a healthy baby boy. She left with an infant so severely brain-damaged that at age 5, he cannot crawl, speak or swallow. He must be fed through a pump.
In the three hours before a doctor finally delivered their son, Justen, by cesarean section, the Guills said in a lawsuit, a monitor sounded 32 alerts that the baby's heart rate had slowed. The suit also said the nurse had warned the doctor on duty four times that the baby was in distress before he arrived at Katie Guill's bedside. The family's lawyer, Laurie Higginbotham, said she believed the outcome might have been different had the nurse alerted the doctor's superiors.
The government settled the case for $10 million, but Pentagon records give no indication that a safety investigation was conducted. Nor is there a record of any action taken against the doctors and nurses involved. A spokeswoman said the Defense Department was legally prohibited from discussing how any specific case had been handled.
"We don't know what went wrong because no one has ever told us," said Justen's father, Jon Guill, a former Army mechanic who served 18 months in Iraq.
The Pentagon had promised to look harder at such mistakes.
In 1999, the Institute of Medicine estimated that medical errors killed between 44,000 and 98,000 patients at hospitals nationwide every year. Those numbers -- which most experts now consider an undercount -- stunned the medical community and kick-started an aggressive effort to protect patients from accidental harm.
Simply penalizing doctors and nurses for malpractice had failed, the institute concluded, because most mistakes arose from weak procedures, not reckless individuals. It called for new strategies, including mandating that medical workers report mistakes and hospitals investigate and correct the lapses that allowed them to occur.
The Pentagon embraced the report, requiring that military treatment facilities produce a written root-cause analysis within 45 days in all cases of unexpected serious injury or death. "Such events are called 'sentinel' because they signal the need for immediate investigation and response," the regulations state. Military hospitals must also report sentinel events to the Joint Commission, an independent accreditation group. Specialists at the Pentagon's patient-safety center, created in 2001, were told to review the analyses and recommend changes.
Certainly it is difficult to assess such a divided and diverse medical system, with 56 hospitals, domestic and overseas, ranging from the flagship, Walter Reed, to a hospital in the middle of the Mojave Desert with an average of three inpatients a day. They serve not only young, typically healthy active-duty families but also the longest-serving military retirees and their families. Even so, experts say safety reviews can reveal trouble spots as well as patterns of error across an entire system.
But annual patient-safety reports and other internal documents obtained by The Times show that, for years, the center's analysts have often found themselves staring into darkness.
As early as 2003, a Pentagon audit noted that medical workers had reported 80 cases of severe harm or death in the preceding 12 months, but that only 32 root-cause analyses had been forwarded to the center. Five years later, another audit concluded, "Unfortunately, RCAs are used relatively infrequently."
The most recent safety reports paint a similar -- and more detailed -- picture.
In 2011, 50 unexpected deaths were identified but only 25 analyses submitted.
The next year, the center was informed of 110 deaths but received only 44 root-cause analyses.
And in 2013, the report documented 79 deaths and 31 root-cause analyses.
The safety audits also make clear that of the root-cause analyses that are done, the cases of clearest-cut blame -- the "never events" -- make up the largest group. Even so, the reports show, those mistakes have not subsided.
In 2010, the safety center had sounded a hopeful note. For the first time in nine years, it said, "all surgeries and procedures were performed on the right person." But the next year, the center said surgeons were still performing the wrong procedure or operating on the wrong patient or part of the body at an "alarming" rate. It called for intensive hospital audits to lower it. Last year, medical workers reported virtually the same number of errors. They also reported more cases in which medical devices were inadvertently left inside patients than they had four years earlier.
In a written response to questions from The Times, the Pentagon acknowledged that it had taken a decade before the reporting system was "operational and collecting data in a uniform manner." Not until last October, for instance, were the Army, Navy and Air Force required to identify the facilities where patients were severely harmed or died.
Senior defense heath officials say the missing investigations are not a true measure of attention paid to serious harm. "There are many mechanisms for reviewing significant adverse events," the Pentagon said in a statement.
In an interview, Woodson said a different kind of inquiry that hospitals conduct -- a risk-management investigation -- might have gotten to the bottom of what went wrong. Those investigations focus on whether individual doctors or nurses provided substandard care for which the government could be liable.
"I feel confident that we capture and investigate the overwhelming majority of these adverse events," Woodson said. "The key is having a robust system and that you pick all of them up and make the changes that are appropriate."
But military regulations specifically require both types of inquiries, and for good reason, patient-safety experts say: Otherwise, even catastrophic errors - mistakes for which no one is specifically to blame, but that instead result from systemic lapses - can be easily swept under the rug.
Busy but Troubled Hospital
Womack Army Medical Center in Fort Bragg, North Carolina, is one of the system's largest, busiest hospitals. Lately it has also been one of its more troubled.
For three years, it has had a higher-than-average rate of surgical complications, and in March it suspended all elective surgery for two days after inspectors found problems with surgical infection controls. Then last month, the Army ousted the hospital's leadership after the unexpected deaths of two patients in their 20s: a mother of three who had undergone a low-risk surgical procedure and a soldier who had been sent home from the emergency room.
That same day, Hagel ordered the systemwide review.
At Womack and elsewhere, some doctors and nurses complain that no one listens to their safety warnings. One staff member interviewed by The Times recalled filing roughly 50 reports of safety problems since 2007, each time providing contact information. Only once, the worker said, did a supervisor respond, and then only to express irritation at the fusillade of filings.
"It is an exercise in futility," said the staff member, speaking on condition of anonymity for fear of job repercussions. "We can jump up and down and shake our fists, but nothing changes."
Lopez, the former Army staff officer, said some hospital officials had told her that they felt pressure from superiors to focus on budget cutting and efficiency, while patient safety got a cursory nod.
Across the system, Pentagon officials cite some signs of progress. In 2008, for instance, the composite rate of 11 types of harm - like pressure ulcers and postoperative hemorrhages - was more than twice that of civilian hospitals with a similar patient mix. Last year, it was better than the civilian average, although the Pentagon's own analysts warned results might be skewed by reporting problems.
Even so, the most recent patient-safety report complained of a general lack of headway in building a safer system. While the number of reports of harm has varied over the last decade, and "there are certainly pockets of excellence," it noted, "The leading trends remain consistent." What was needed was "enterprisewide change."
The operating room has been one focus of concern.
The study by the American College of Surgeons found that in addition to Womack, three other major hospitals - Madigan Army Medical Center in Tacoma, Washington; San Antonio Army Medical Center in Texas; and Portsmouth Naval Hospital in Virginia - have had high rates of surgical complications for two or three years in a row. Five of the eight cited last year had also been flagged repeatedly for high rates of infection related to surgery.
Dr. Brian Lein, the Army's deputy surgeon general, said hospitals that fell below the benchmark "have dug deep into the data to find the actual issues and are addressing those issues." The Navy echoed that response.
With so many young military families, the system's maternity wards are among its busiest. Pentagon officials say maternity care is top notch, and on some leading measures of safe childbirth, the military hospitals indeed compare well with their civilian counterparts.
For example, their rate of infant mortality was equal to or lower than that of civilian hospitals in the most recent data analyzed by the National Perinatal Information Center, a private group with a Pentagon contract. In routine vaginal births, the rate of injury to the mother has consistently been below the national average.
On other measures, though, the military system lags.
In 40 percent of the military hospitals, mothers were significantly more likely to suffer hemorrhages after birth than at the civilian hospitals tracked by the perinatal center. The hemorrhages can lead to hysterectomies or even death. About 2,500 cases were recorded in military hospitals in 2012, roughly 760 more than if the military had met the civilian benchmark.
If doctors used instruments such as forceps to assist the delivery, mothers in military hospitals were about 15 percent more likely to be injured than mothers nationally, the most recent data shows.
One of the broadest measures of safe childbirth is the rate of injuries to babies, ranging from cerebral hemorrhages to small cuts on babies' scalps. From 2009 to 2011, according to a Times analysis, the rate at military hospitals was twice the national average.
In 2011, nearly 5 in every 1,000 babies born at military hospitals suffered some kind of birth trauma. Had the military met the national average, 107 newborns would have been injured instead of 239.
Woodson said the military is looking "closely at areas where we are falling short" on maternity care and measuring its hospitals against civilian ones in order to improve.
(BEGIN OPTIONAL TRIM.)
Ordeal Giving Birth
When Stephanie King felt labor pains on Easter in 2004, she drove to Reynolds Army Community Hospital. She was 34, a kindergarten teacher and mother of two. Her husband, an Army artillery officer, was serving six months in Iraq, so her 12-year-old son accompanied her to the hospital.
Her contractions were coming every three to five minutes, court records show, but a resident on the maternity ward refused to admit her, saying her cervix was not sufficiently dilated. The attending physician agreed.
King spent the next two hours in the hospital's first-floor lobby, waiting room, cafeteria and bathroom. She wanted to seek care at the civilian hospital 15 minutes away, but her military insurance would not cover it. In the midst of her ordeal, her husband called from Baghdad to say he was being airlifted to Germany after an emergency appendectomy.
Finally, fearful that she would deliver on the bathroom floor, she took the elevator back to maternity. A nurse greeted her with what seemed to be a joke: "'Oh, Mrs. King, you are back,'" she recalled the nurse saying. "'You don't look as good as you did when you first came in.'"
Any amusement vanished, however, when staff members realized not only that King was about to give birth - but that her file showed she carried a common but symptomless strain of group B streptococcus bacteria. Women in labor must be given antibiotics at least an hour before delivering to avoid transmitting the infection to their newborns, hospital workers later testified.
Dawson King was born just 42 minutes after his mother was admitted. Doctors soon determined that he had contracted an infection, and warned Stephanie King that the only effective medication could cause deafness. Four months later, when Dawson did not turn his head when his parents walked into the room, it became clear that he was profoundly deaf.
Only three years later, after Stephanie King filed a malpractice claim, was the case discussed at Reynolds, court records indicate. The attending physician said a risk-management meeting was held to ensure that the standard of care was met. No report was written, and the doctors and the nurse emerged with spotless licenses.
That result would appear to point to a lapse in hospital procedures. But both the resident and the nurse testified that they had never attended a meeting at Reynolds to discuss what had gone wrong.
"That's the disturbing part," King said, while Dawson draped himself over the arm of his father's chair, looking at photos of himself as a 1-year-old, his head swathed in bandages from surgery for cochlear implants. "Doesn't the hospital want to know what happened?"
King believes that she knows: Her file documented her history of strep B infection and the fact that her second child had been born after a 90-minute labor. And the resident testified that he could not recall if he had read her record before declining to admit her. "It boils down to they did not even read my records," King said.
The government settled the Kings' case in 2009 for $300,000. King describes her son, now an engaging, soccer-playing 10-year-old, as "a success story." Still, she said, "it was devastating how easily it could have been prevented."
The Kings' malpractice payout was relatively modest. An examination of court records and Pentagon data from Reynolds and seven other hospitals turned up dozens of settlements, ranging from $30,000 to $10 million, but no record of a root-cause analysis.
Flying Blind on Safety
On Dec. 7, 2011, Woodson summoned senior health officials to a Pentagon conference room to discuss the safety of military hospitals and clinics. The deputy surgeons general of the Army, Navy and Air Force were present. So was Dr. Ronald Wyatt, then the director of the Pentagon's patient-safety center.
According to accounts from several people familiar with the session, Wyatt suggested that the center was flying blind. Without knowing the facilities' death and harm rates, as well as the rates of paid malpractice claims, he argued, his staff was unable to identify safety lapses - much less correct them.
And serious lapses did exist, he said, for "this system, like many systems throughout the country, hurts and kills people every day." Angered, Dr. Richard A. Stone, then the Army's deputy surgeon general, shot back: "I demand that you retract those statements." Dr. Warren Lockette, the deputy assistant secretary for clinical policy, stood by Wyatt. "What I am hearing is you are all satisfied with the status quo," he said. Woodson tried to defuse the tension with a compromise: He would recommend that the services turn over the data.
Asked about the meeting, Stone said that data should be shared but that Wyatt's statement was "inflammatory."
The standoff was typical, former Defense Department officials say, of a continual tug-of-war between health care officials in the Pentagon and in the individual armed services that has crippled efforts to improve patient safety. In such a politicized system, data can be a weapon.
"Why should the Army safety system want to play with DOD, because then I have less control over my data, less control over my kingdom, and potentially DOD is going to tell me what to do?" said Lopez, the former Army health policy officer.
To keep Pentagon overseers at bay, surgeons general have often relied on a provision in a 1986 law, known as 1102, that prohibits disclosure of medical quality assurance records. Originally adopted to ensure that medical personnel could be honestly evaluated without fear of publicity, former department officials say, it became a tool to withhold a broad range of data from the Pentagon itself.
And from the public. In response to Freedom of Information Act requests, the Pentagon provided The Times with thousands of pages of data. But much information was redacted and some reports were withheld as confidential, including all reports by the Navy's inspector general on patient safety or quality of care. By contrast, the veterans system posts the reports on a public website.
While infighting held the military's patient-safety programs in check, some civilian hospital systems cut death and harm rates.
At Ascension Health and Kaiser Permanente, two of the nation's biggest nonprofit systems, investigating workers' reports was just a first step. The companies also analyze a vast array of data, including readmission and mortality rates.
The death rate is a broad measure that cannot pinpoint where care falls short. Nonetheless, "mortality is the mother of all outcomes," said Ascension's chief medical officer, Dr. Ziad Haydar. Measured over time, a death rate, adjusted for seriousness of illness, can show if a medical system is getting more or less safe.
Officials at Ascension and Kaiser say their hospitals have gotten safer.
Ascension estimates that its safety measures have saved 1,500 lives in the last six years. Doug Bonacum, Kaiser's vice president for quality, safety and resource management, said the mortality rate at Kaiser's 38 hospitals had fallen more than 30 percent in the last four years.
The Pentagon does not routinely track the total number of deaths, and has no method yet to calculate adjusted rates. "Frankly it is not yet a helpful measure for assessing quality," Woodson said.
As a result, why some military hospitals report many more deaths than others with similar numbers of patients is a mystery. The Army, Navy and Air Force each said that hospitals with above-average death rates treated older, sicker patients, but did not produce statistical evidence to verify that.
When it comes to gauging the frequency of errors, systems like the military's that rely on workers to report harm have been shown invariably to undercount. Kaiser has long used another technique, called the Global Trigger Tool, that winnows out indicators of poor care from randomly selected patient files.
Using that technique, researchers concluded in 2010 that one-third of patients at three major civilian hospitals had suffered some kind of harm. A similar pilot study by the Pentagon last year found that nearly half the patients whose files were reviewed at a major military hospital had been harmed at least once. The study suggested 99 percent of harm at that hospital was not reported by medical workers.
When patients die unexpectedly, medical workers often cite a breakdown in communications.
That appears to be the overriding explanation for the delay 3 1/2 years ago in treating Jessica Zeppa, a case that ended with a $1.25 million malpractice settlement. But that is only conjecture. Her death was apparently never subjected to a patient-safety examination.
Zeppa had been married and living at Fort Sill for only nine months when her husband, James, an air defense tactician, was deployed to the United Arab Emirates. She had their two dogs and a cat for company. And to her delight, she was pregnant. "She was just out of this world about it," said her mother, Amonett.
But five months into the pregnancy, Zeppa became so weak that she struggled to climb stairs. She complained that it hurt to drink or swallow.
At Reynolds Army Community Hospital, she was initially treated in the obstetrics and gynecology unit, where an ear infection was diagnosed and she was prescribed antibiotics and ear drops, court records show. Three days later, she arrived at the emergency room at 4:51 a.m.; she was prescribed a painkiller for erupting wisdom teeth.
She returned that same evening and was found to have a fever, a fast pulse and an elevated white blood cell count -- possible symptoms of serious infection. The emergency room physician, Dr. Raul Young-Rodriguez, treated her intravenously with fluids and a powerful antibiotic and sent her upstairs to the obstetrics and gynecology unit for possible admission. No one called the obstetrician on duty to inform her of the patient's condition, the Zeppas' lawyers, Heather Mitchell and Steven Clark, said in court papers.
Nor did the obstetrician, Dr. Debra A. Carson, call Young-Rodriguez.
As far as she knew, Carson later testified, Jessica Zeppa was there for "obstetrical clearance." She examined her and sent her home within 20 minutes. Amonett, Jessica's mother, said she protested but was told her daughter could not be admitted if the fetus was not in distress.
Carson testified that she had not checked the military's electronic record system for Zeppa's history, because all too often she had found patient records missing. In Zeppa's case, lab results were posted less than an hour after she left. Had she seen them, Carson testified, "I would more than likely have admitted her."
Two days later, Zeppa returned to the obstetrics and gynecology unit, insisting that she would not leave until she got warm.
A nurse midwife, Kelly West, treated her with intravenous fluids and again released her. West testified that she did not review Zeppa's records either.
The next afternoon, with Zeppa struggling to breathe, her mother summoned an ambulance. Zeppa was airlifted the following morning to a civilian hospital in Oklahoma City, where she miscarried 10 days later and died the next month.
Five months after that, facing a malpractice claim, Reynolds officials conducted a risk-management investigation. In an interview, West, the nurse, said she had been cleared of violating the standard of care. Nor is there any public indication that the two doctors were penalized. They did not respond to requests for comment.
Medical experts hired by the family's lawyers said that had the Fort Sill doctors recognized that Zeppa was suffering from septic shock and immediately hospitalized and aggressively treated her, she and the baby probably would have survived. The government's experts disagreed, noting that civilian doctors had been unable to save Zeppa in five weeks of treatment.
Justice Department lawyers called Zeppa's death a "unique and tragic case, but not a case of bad and actionable medicine." Beyond the risk-management assessment, they said, they knew of no other inquiry. West also said she knew of none in the roughly eight months before she left Fort Sill.
That left any missteps that contributed to her death unexplained.
"She was really pretty, and she had a really big heart," James Zeppa, Jessica Zeppa's husband, said. Now, he said, he no longer trusts military medicine.
Jessica Zeppa's father, Mike Amonett, had one thing to say about the Fort Sill hospital: "I just want that place shut down."