Our bodies run on predictable daily schedules. As morning approaches, our body temperature begins to rise, and our cortisol levels climb. Our heart rate and blood pressure spike upon awakening. By mid-morning, we tend to be at our most alert, while many components of athletic performance—including strength, coordination, and flexibility—peak in the afternoon. As darkness descends, the pineal gland dials up its release of melatonin; while we sleep, gastric-acid secretion surges. Similar 24-hour cycles, or circadian rhythms, govern nearly every cell, tissue, and organ in the human body, and almost all of its physiological functions.
Hospitals, it turns out, have daily rhythms too, and they don’t always sync with our own. According to a recent Proceedings of the National Academy of Sciences study, conducted before hospitals were handling surges in coronavirus cases, hospital clinicians prescribe medications in predictable patterns, dispensing the majority in the morning. This schedule—what the researchers call a “systemic bias in the timing of medicine”—may be convenient for clinicians, but it’s not necessarily best for patients or consistent with their biological clock.
Hospitals run all day and night, and in theory medications could be doled out whenever they’re needed. “Our findings challenge this notion,” the study’s authors write, “and reveal a potential operational barrier to best clinical care.” For busy doctors, time is a scarce resource—and hospitals are generally designed to deliver care as quickly and efficiently as possible. But the human body keeps its own time, so optimal care may be less of a race than a chronologically choreographed dance.
The research team, led by the pediatric otolaryngologist David Smith and the circadian biologist John Hogenesch, analyzed approximately 500,000 doses of a dozen drugs—including painkillers, antibiotics, and corticosteroids—given to inpatients at the Cincinnati Children’s Hospital Medical Center from 2010 to 2017. The researchers found that the times at which drugs were ordered and administered weren’t uniformly distributed—instead, there were “distinct morning-time surges and overnight lulls.” (The researchers, who are based at the Cincinnati hospital, declined an interview request.) Roughly a third of all orders were submitted in the four-hour period from 8 a.m. to noon. And most of those were concentrated between 8 and 10:30 a.m., when clinicians were conducting their morning rounds.
The finding makes sense, says Garret FitzGerald, who directs the Institute for Translational Medicine and Therapeutics at the University of Pennsylvania, and was not involved in the research. “Any of us that are physicians know that a lot of hospital behavior is driven by rounding times, and those rounding times are chosen for all sorts of reasons that don’t necessarily have any particular scientific basis,” he told me.
That doesn’t make the synchrony between rounding and prescribing medically sound, however. Patients should get drugs when they need them, not just when it’s convenient for clinicians, and there’s no evidence to suggest that, say, infections or inflammation or pain are confined to the morning. (In fact, patient pain tends to worsen at night.) “A lot of the times the schedules that are in place have nothing to do with the biological requirements of the patient that may be driven by circadian rhythms,” says Michael Smolensky, a medical chronobiologist at the University of Texas at Austin. The flood of morning orders also seemed to create an operational bottleneck; the delay between when drugs were prescribed and when they were administered was longest for medications ordered in that 8-to-noon window, the researchers found. It’s not clear whether these delays affected patient outcomes, but it’s certainly conceivable that they could have.
In addition to these potential problems, there's the fact that our bodies are not stable systems. Our own internal rhythms mean that time of day can influence how well a given drug works or how severe its side effects are. Many ailments wax and wane in predictable ways over the course of the day—cold and allergy symptoms tend to be worst in the morning, while heartburn often strikes overnight—as do blood flow, liver function, and other physiological processes that can affect drug metabolism. A compound that’s quickly eliminated from the body during the day can linger at night. “We know that giving drugs at different times of the day can often result in different blood levels of those drugs or different adverse effects,” FitzGerald said. Our responses to various drugs, he added, are “conditioned by the clock.”
Many adults, for example, take a daily dose of aspirin to help stave off the blood clots that can cause heart attacks and strokes. These cardiovascular events are especially common in the early-morning hours, and some research suggests that aspirin is most likely to prevent them when taken before bed. The drug also seems to cause less damage to the lining of the stomach when ingested at night. “Treating at the wrong biological time may not get you the results you want,” Smolensky told me, “and it may potentially aggravate the medical condition even further.”
Although these time-based effects have been documented for a wide variety of drugs, many medications “are taken without any thought about the concept of when,” Smolensky said. In fact, the Cincinnati researchers saw this play out at their hospital. Previous research has established that many blood-pressure medications are most effective when given at night. But at the Cincinnati Children’s Hospital Medical Center, almost twice as many doses of hydralazine, a drug commonly used to treat high blood pressure, were given during the day as at night. The researchers found no obvious clinical rationale for this timing—as a whole, patients’ blood pressures did not seem to be spiking significantly higher during the hours when the drug was most commonly ordered. And when the drug was dispensed at night, it prompted a bigger drop in blood pressure than when it was taken in the morning. “Paradoxically, patients were more responsive to therapy during the window of time (nighttime) when relatively few orders were placed,” the researchers write.
In an ideal world, clinicians’ prescribing behavior would be calibrated to when different drugs work best. Antipsychotics, for instance, may cause fewer metabolic problems when they’re administered in the morning. The point is that there’s no optimal time that holds for all drugs—and no medical reason to front-load all medications in the morning.
Unfortunately, correcting for this bias won’t be easy. Hospital staffing schedules are carefully designed, and rounding typically happens at the same time each morning for a reason. But simply making clinicians aware of the problem could prompt them to be a bit more thoughtful about when they’re ordering certain drugs. Hospital staff certainly have their hands full at the moment, but in the wake of this crisis, as they recover and retool, it may also be worth considering more serious changes, such as asking nurses to round more frequently or instituting regular nighttime check-ins. “Perhaps,” FitzGerald said, “we should get a little bit more sophisticated about the way that we give drugs according to time.”