Most hospitalists think about the nutrition of their patients in one of two situations, according to Jeff Greenwald, MD, a hospitalist at Massachusetts General Hospital in Boston and a researcher on inpatient nutrition. The first is when they encounter an incredibly sick patient who clearly needs to be fed enterally or parenterally. The second is when they notice a patient who is frail, thin, and picking at meals, with lackluster labs.
“In the first case, we get very methodical about it, and bring in the real experts right away. In the second, we feel, at our gut level, this person doesn't look so good, and think maybe they could do better if their nutrition were better. . . but we aren't really convinced it'll make a difference. Maybe we'll order a dietary consult—maybe,” Dr. Greenwald said.
And yet, a third of patients enter the hospital malnourished, and another 38% become malnourished during their stay, noted Lisa Kirkland, MD, FACP, a hospitalist at Mayo Clinic Rochester, Minn., and critical care specialist at Abbott Northwestern Hospital, Allina Health in Minneapolis, who published a call to action for physicians to step up nutrition therapy in the hospital on the American Journal of Medicine blog in 2017.
“We should be looking at every single patient who comes in the door for nutritional status,” Dr. Kirkland said. “And not just by looking at their albumin, or saying, ‘I know malnutrition when I see it,’ or by suspecting it in nursing home patients and alcoholics but not those who are obese. Every patient.”
One barrier to this may be that hospitalists aren't taught a lot about nutrition in medical school. In addition, until recently, the evidence for nutritional therapy—that is, using oral, enteral, or parenteral nutrition to maintain or support optimal nutrition status—in general medical inpatients wasn't strong.
A systematic review and meta-analysis in the January 2016 JAMA Internal Medicine found that oral or enteral nutritional therapy for medical inpatients who were malnourished or at risk of malnutrition didn't improve clinical outcomes like mortality, length of stay, and hospital-acquired infections. That analysis was later featured in the same journal in a systematic review of articles focusing on medical overuse. “Nutritional support [in medical inpatients] should generally be avoided,” the review concluded.
What's more, some research on nutrition therapy in ICU patients has suggested that it can cause harm in certain situations, which may contribute to caution about it generally, experts said.
But ICU patients are a “totally different story” from general medical inpatients, noted Nicolaas Deutz, MD, PhD, a nutrition researcher, internist, and director of the Center for Translational Research in Aging and Longevity at Texas A&M in College Station.
“ICU patients are so much sicker, and the role of nutrition for them is way more complicated,” Dr. Deutz said. “The studies on ICU patients typically have been with a specific patient group in a specific situation with a specific design, and they aren't representative for all [patients]. You can't let those studies demotivate you” when it comes to treating malnutrition in non-ICU patients, he said.
Evidence of benefit
Indeed, more recent clinical trials have shown evidence of benefit from nutritional therapy for medical inpatients. As these studies accumulate, researchers believe, the tide may be turning in how hospitals and physicians screen, assess, and treat patients for malnutrition or nutritional risk.
“We are in the beginning of a journey here, and the more positive results physicians see with patients who get nutritional support, the more they will realize that nutrition is an important component in getting patients home and back on track,” said Philipp Schuetz, MD, an internist at Kantonsspital Aarau in Switzerland and coauthor of the 2016 meta-analysis that didn't find benefit.
For example, a more recent meta-analysis, published by JAMA Network Open on Nov. 20, 2019, found a clear mortality benefit for malnourished medical inpatients who received nutritional therapy: 8.3% with therapy versus 11% without. It also found lower rates of readmissions with nutritional therapy (14.7% vs. 18%), while energy intake, protein intake, and weight gain were higher.
Dr. Schuetz was the lead author of the former, which was published in the June 8, 2019, The Lancet. “A lot of the [earlier studies] were small and not powered for clinical endpoints, and the protocols were not well followed, compared to the newer ones,” he said. “Also, the nutritional support strategies have changed—the newer products have higher-quality proteins.”
The EFFORT trial (and the 2019 meta-analysis, also coauthored by Dr. Schuetz) didn't include ICU or surgical patients, who are more likely to have specific nutritional needs, or patients who received parenteral nutrition, which can be associated with a higher risk for adverse outcomes, he added.
“Our study focused on medical inpatients in acute care—those with cardiac failure, renal failure, kidney failure, etc.—but we purposely didn't include intensive care patients, with whom you may use a different feeding strategy,” Dr. Schuetz said. “Our idea was more to look at the others, to see if you should wait and see if they need support—which is what many physicians do—or if you should be proactive. The evidence is you should be proactive.”
The trial included medical inpatients with a Nutrition Risk Screening 2002 (NRS-2002) score of at least 3, randomized to either individualized nutritional therapy to reach protein and caloric goals or to standard hospital food. At 30 days, significantly fewer patients in the nutrition group had died (7% vs. 10%) or had an adverse clinical outcome (23% vs. 27%).
“We need to do more trials and understand the field of individualized nutrition—different nutrition for different patients with different illnesses,” Dr. Schuetz said. “But I am already seeing physicians who are starting to rethink the evidence, and to think it points to nutrition as a supportive measure for the fragile, elderly, polymorbid population of inpatients.”
The NOURISH trial, published in Clinical Nutrition in 2016, comprised malnourished inpatients who were at least 65 years old. The intervention group received high-protein oral nutritional supplements and showed no reduction in the composite endpoint of death or readmission at 90 days compared to those getting placebo. However, when broken out as a single endpoint, 90-day mortality was lower with the intervention group (4.8% vs. 9.7%). Intervention patients also had improved nutritional status at day 90.
“The EFFORT trial showed roughly the same idea as my paper: It is medically important to treat patients who are malnourished in the hospital,” said Dr. Deutz, who was lead author on NOURISH. “We want more results pointing in this direction. . . . the medical profession is conservative and doesn't change habits quickly.”
Evidence into action
One of the most important ways hospitalists can improve nutrition therapy at their facilities is to help establish a protocol for screening all patients within 24 hours of admission. This doesn't have to be particularly onerous, Dr. Greenwald said.
“From an operational perspective, the screening process isn't a 20-question, five-minute-long, crazy annoying thing. It can be as simple as the NRS-2002, which has just four basic questions and one or two follow-ups, and is part of the MDCalc app that a lot of physicians already use. If a patient is flagged, you get a nutrition consult, which every hospitalist knows how to do,” Dr. Greenwald said.
Nutrition screening could potentially be added to the initial nursing admissions process, Dr. Greenwald said. Ideally, a nurse would have the explicit authority to order a dietary consult if needed, Dr. Kirkland said, and dietitians could communicate with physicians directly about the results.
“Results need to be brought directly to the physician's attention, just like an abnormal lab would be,” Dr. Kirkland said. “Especially because, as of now, there is no routine place in the patient chart where a [nutrition] assessment score is recorded and red flagged and comes up on the physician dashboard.”
Having a routine system for screening is helpful. “You could develop a memorandum of understanding between the hospital medicine group and the nutrition department that the former will do an automatic screen at admissions and, if the screen is positive, there will be a standing order for a nutrition consult, for example,” Dr. Greenwald said.
Hospitalists can help think through the best screening policies for their individual services, given staffing and other resources. Elements to consider include whether a positive screen for malnutrition automatically triggers a dietary consult, whether nurses can order nutrition consults directly, and whether dietitians can accompany physicians on rounds to monitor for nutrition status.
Drs. Greenwald and Schuetz recently coauthored an article calling on hospitalists to get involved in optimizing inpatient nutrition. “Hospitalists can play a pivotal role in helping to identify and intervene in this important problem by working to implement multidisciplinary nutritional support programs in conjunction with nursing and nutrition colleagues,” they wrote in the Feb. 18 “Annals for Hospitalists” article published by Annals of Internal Medicine.
One of the big decisions to make about screenings is whether they should lead to further action only for patients whose results show they are malnourished, or whether patients who screen as at risk (but not yet malnourished) should be subject to additional protocols, such as rescreening during the hospital stay.
Nutritional therapy has greater mortality and morbidity benefit among patients who are malnourished versus those only at risk, Dr. Greenwald noted, and the patients who are most likely to deteriorate from “at risk” into malnutrition are those who have been in the hospital for a while and are therefore likely to already be on the radar of several services.
“I'd rather see priorities and resources go toward making sure everyone is screened the first time, within 24 hours of admission. Certainly, there can be left turns that patients take in their clinical course, and we could build an arbitrary rescreening interval for that. But given that we are scratching a new surface here, I think we should focus on that first screening, then figure out the process of rescreening at-risk patients down the line,” Dr. Greenwald said.
Another decision for hospitals is which tool to use for assessment. The Global Leadership Initiative on Malnutrition (GLIM), which comprises major nutrition societies such as the American Society of Parenteral and Enteral Nutrition (ASPEN) and the European Society of Parenteral and Enteral Nutrition, says any validated tool can be used to screen patients for nutritional risk. That includes the NRS-2002 and the 2012 ASPEN criteria.
GLIM also recommends that patients who are “at risk” upon screening should move quickly to diagnosis (or not) of malnutrition, and if malnourished, have the severity graded using a measure GLIM created to be simple and useful for a range of clinicians. Such guidance may be difficult to follow in some U.S. hospitals, Dr. Greenwald said.
“It would be naive to expect that after the initial screen, a non-nutritionist is going to go ahead and assess the severity of nutritional insufficiency. Doctors and nurses don't have experience with that and more importantly, we already have experts in nutrition at every hospital who can do it. Our job is to identify who needs a consult, and we are able to do that just from a screening,” he said.
From assessment until near discharge, the role of hospitalists in nutritional therapy is relatively limited. They should be available for discussion with the consulting services and keep an eye on overall health, the experts said.
Ideally, hospitals have a nutrition service or team to assess and manage patients' malnutrition throughout their stay, Dr. Deutz said. “The nutrition team is a doctor [such as a hospitalist], a nurse, a dietitian, and maybe a pharmacist. Handling nutrition isn't just for one provider; it needs to be a team effort.”
Such teams are associated with better patient outcomes and shorter lengths of stay and are cost-effective, according to ASPEN's 2018 Standards for Nutrition Support for Adult Hospitalized Patients, which cite a number studies to support this.
Hospitalists should be more involved in nutrition when it comes time for discharge, the experts agreed. The hospitalist may need to specify that nutritional therapy, such as oral supplements, ordered in the hospital should be continued after discharge and ensure that outside clinicians are aware of the progress of patients' nutritional status during the hospital stay. It's also important that patients are interviewed about their access to food, with an eye toward connecting them to resources if needed.
“When the patient leaves the hospital, it's important to stratify who needs to be followed up with,” Dr. Schuetz said. “Depending on the nature of the patient's illness and situation, it may be important [for an outpatient physician] to continue to check up, or to recommend that [the] patient see a dietitian.” These next steps should be part of the discharge plan, he advised.
The hospitalist can request a formal discharge plan of nutrition therapy from the dietitian, which would include the number of calories and the amount of fluid and protein to consume once the patient is at home. The hospitalist should then review it with the patient as part of the discharge process, and make sure it gets into the hands of the patient's primary care clinician, said Dr. Kirkland.
“This should be a very specific plan so patients can take it home and use it as a guide so their nutrition doesn't deteriorate at home, where they are still recovering, and maybe don't feel good and aren't shopping regularly,” Dr. Kirkland said. “And primary care doctors should follow up on it like they would with test results and medication.”
Of course, to be successful, this approach necessitates systems change and that requires champions in leadership. “It would be nice to know under whom nutrition falls in the organizational chart,” Dr. Greenwald said. “Getting that person to buy in is going to be critical, because we are talking about increasing utilization of their services.”
Together with that champion, representatives from nursing, hospital medicine, nutrition, and information technology should work together to figure out how to routinize the screening process and embed it in the electronic health record, which can help with both standardization and preservation of the resulting data.
“This is legitimate information about a patient; it shouldn't be on a piece of paper that's thrown out later,” Dr. Greenwald said. “It needs to be somewhere where you can see the results and track the impact.”