Getting someone better more quickly is not a bad thing. While family physicians who care for inpatients may be tempted to resent insurers who press for early hospital discharges, we need to remember that the aim of making inpatient care more efficient isn't — or certainly shouldn't be — simply to improve the bottom line. The point is to provide the best possible care for the patient, and that can include eliminating wasteful procedures, unnecessary admissions and unnecessary hospital days.
What follow are a number of practical suggestions for improving inpatient management and reducing the costs of hospital care. These tips — based on the experience of medical directors and utilization nurses in several managed care organizations (MCOs) as well as the lessons learned by successful hospitalists — don't relate to specific diagnoses. Instead, they reflect a belief in the imperative to do nothing that doesn't need to be done. After all, quality isn't a function of the volume of the services we provide. It's a function of delivering the right service at the right time in the right place.
Providing that kind of care benefits the patient, but it also benefits us. A family physician who can provide efficient inpatient care may be more marketable in a managed care environment — better able to win good contracts with MCOs, negotiate a higher salary or compete against encroaching hospitalists. When the patient's interest intersects with the physician's interest, you have an example of “aligned incentives” at their very best.
Improving efficiency can make inpatient care more cost-effective, but primarily it helps improve the quality of that care.
Clinical policies and protocols give physicians standards by which to assess the efficiency of the care they provide.
For a complex case, one physician should act as “commander,” and doctors covering on weekends should be empowered to discharge.
Other tips include using consultants and allied services efficiently as well as evaluating whether traditional care approaches are justified.
Prevent unnecessary admissions
Clearly, the most direct way to make inpatient care more efficient is to prevent unnecessary admissions in the first place, particularly those that stem from emergency department (ED) visits. You can prevent many of these admissions by preventing the ED visit: Ensure that patients have access to your office during office hours and to you or a colleague after hours. For those who do come to the ED, you can reduce the chance that they will expect to be admitted by educating them about good self-management of their conditions. Helpful tools include pamphlets, classes, local support groups and World Wide Web sites. [For more information on using education to help reduce unnecessary utilization, see “Demand Management: The Patient Education Connection,” September 1998.]
At some hospitals, many of the decisions to admit are made by emergency physicians, and the attending physician sees the patient during rounds the next day. Some of these admissions can be avoided if emergency physicians are aware of alternatives, such as observation beds at the hospital, direct admissions to skilled-care facilities or stabilization followed by an office visit the next day with the physician who knows the patient best. Learn the hospital's policies on options such as observation beds, and work with the emergency physicians to use them.
Another way to reduce unnecessary hospitalizations is by preventing readmissions. Modern clinical management can't be passive; no longer can we wait for an ambulatory patient with a known high-risk illness to come to us with symptoms indicating readmission. The key to reducing readmissions is good outpatient management, the strength of primary care. Periodically, your office staff can contact the patients most likely to be readmitted and check their status, encourage them to adhere to treatment regimens and urge them to seek interventions early if they suspect problems. [For more information on practicing this kind of proactive population-based medicine, see “It's Time to Start Practicing Population-Based Health Care,” June 1998.]
You can also turn to MCO case managers and various community resources for help in preventing readmissions. For example, a case manager may be able to arrange for home delivery of medications or a special waiver of costly drug co-payments for selected patients so that adherence can improve. Home care might be arranged to help spot clinical problems early, especially for patients who have a hard time getting to your office.
Use clinical policies and protocols
When patients do need to be admitted, you can make their care more efficient by basing their treatment on inpatient clinical policies such as those developed and sold by Milliman & Robertson Inc. or InterQual. These policies are flexible tools that can be used effectively in actual practice. [For more information on clinical policies, see “Where to Look for Good Clinical Policies,” February 1999, and “How to Evaluate and Implement Clinical Policies” in this issue.]
Many physicians dismiss published clinical policies without more than casual consideration. Their negative perceptions may be a result of the rigid way clinical policies have been applied by some MCOs. Some policies are voluminous and seem arcane on a first reading. But it's worth taking another look at clinical policies with an understanding that they refer to the ideal case, with no comorbidities and perfect responsiveness to every therapy.
A policy for management of congestive heart failure, for example, might say that the rales and pedal edema will clear on the day following admission, or one for management of pneumonia may indicate that a patient's fever will abate after he or she takes antibiotics for 24 hours. All of us, including the authors of the clinical policies, know this is not true in every case. But the policies state the expected sequence of events, the various therapies a prudent physician would consider and when to move to a lower level of care. Therefore, you can use them to see whether you're ordering the sorts of interventions that efficient physicians would order and whether you're moving the case along as quickly as you might.
Clinical policies also can help you gauge whether surgeries are taking place in a timely manner at your hospital. One significant factor in shortening lengths of stay has been performing surgeries on the first admission day instead of the second. By comparing current practice among your surgeons with published policies, you may find additional opportunities to shorten the length of stay for certain surgeries.
Finally, clinical policies are valuable for helping you make an argument for a continued stay. If a policy says a certain clinical event needs to happen before discharge and your patient hasn't reached that point, a delayed discharge can be justified to the MCO that uses the policy.
You can often obtain portions of vendors' clinical policies (perhaps a page or two) from MCOs, although copyright restrictions, the size of the policies, and their cost prevent MCOs from distributing them in complete form. In fact, according to a standard from the National Committee for Quality Assurance, MCOs should provide their utilization criteria to physicians on request. If possible, read the introductory material that accompanies the policies, which will make their interpretation far easier. Since it's unlikely that a practicing physician will get a complete set of a vendor's clinical policies, try to obtain those that relate to the limited number of diagnoses that make up the bulk of your inpatient case load.
In addition to vendors' policies, you may find that hospital protocols for certain diagnoses or surgeries are useful in making your care more efficient. Often they are designed to correct utilization problems at the facility in question, and they can serve as helpful reminders when writing orders.
For example, one hospital committee I encountered had created a set of standard admission orders for patients who had had strokes. These orders included early consultation with physiotherapists because some physicians weren't returning stroke patients to mobility quickly, a practice that can have an adverse effect on the patients' ultimate level of functioning. Physicians who forgot to request physiotherapy weren't just delaying discharge; they were probably delivering suboptimal care. The protocols served as a reminder of high-quality care as well as a means of standardizing care appropriately.
In another example, staff at a small urban hospital reduced the length of stay and lowered the cost of care for patients with cardiac chest pain by promoting a clinical protocol. At the same time, they demonstrably improved care by increasing the use of aspirin in these patients by 45.7 percent.1 On the other hand, hospital protocols can suffer from not being as aggressive in moving patients along as they could be. Sometimes protocols fail to take advantage of other resources — for example, not fully enlisting the help of pharmacists in monitoring aminoglycoside administration or failing to consider patients' post-discharge needs for care at home. Some end up codifying traditional local inefficiencies.
Perform work-ups in the right setting
Some of the greatest length-of-stay reductions in recent years have been the result of doing preadmission and preoperative testing on an outpatient basis. Sometimes the logic that underlies these improvements also applies to inpatients: Even if the problem that prompted a patient's admission remains unresolved and you determine that further studies are necessary, discharge may still be appropriate.
For example, a patient is admitted appropriately because of chest pain. In a day or so, a work-up shows that cardiac causes were not responsible for the pain, but the patient's discomfort continues. You suspect a gastrointestinal cause, and you could initiate a full work-up, including gastroscopy and colonoscopy. But at this point, it's important to ask whether immediately dangerous conditions have been ruled out and whether the patient could be discharged to have the studies done as an outpatient. The question to ask is this: If the patient came to you today in the office with these signs and symptoms, would you admit him or her, knowing what you know now?
Know who's in charge
Providing the right care in the hospital is only part of the struggle for efficiency. Someone also needs to coordinate that care. An inpatient may have admitting, attending, consulting and primary care physicians, but sometimes no one's really in charge. Patients often express frustration that no single doctor appears to take responsibility for the whole case.
Multiple physicians sometimes defer to one another to the point that no one seems willing to write a discharge order. Utilization nurses gaze with frustration at orders such as “Discharge when OK with Dr. X” written by several consultants, creating a situation in which perhaps five specialists must agree before a patient can be discharged. The process is made far more efficient if one physician is designated as the attending — in a real, not nominal, sense.
Such a commanding role is perfect for family physicians. You may well be the one most skilled in dealing with the patient's and the family's concerns, and you may be the best at coordinating the work of the health care team, including the consultants. Performing this role may require you to use your communication skills as much as your clinical skills, but orchestration of a complex case can be rewarding, too.
Let stand-in physicians discharge
Even if a patient does have a strong attending physician, that doctor can't be there all the time. MCO medical directors and their staffs commonly observe that the progress of inpatient cases seems to stall as the weekend approaches or if the attending physician takes a few days off. Relatively unimportant orders often are written over the weekend, only to be followed by diagnostic decisiveness on Monday. The “stand-in” physician responsible for the patient while the “real” doctor is unavailable may feel that active management should wait. Sometimes this is because the hospital doesn't make certain diagnostic services and procedures available on weekends, but too often it's because the covering physician is simply temporizing.
If you're the attending physician handing over care to a stand-in, you can avoid this inefficiency by discussing the clinical indications you expect to see before discharge and explicitly giving the covering physician your permission to discharge during the weekend if it's indicated. You can make this far easier by telling the patient and family that you trust the stand-in doctor to make the discharge decision in your absence.
If you're the stand-in physician, don't passively accept a colleague's advice that discharge can wait until Monday. Agree on specific clinical endpoints, not a date on the calendar.
Turn to efficient consultants
Another way to affect inpatient-care efficiency is through your choice of consulting physicians. Some doctors appear to approach diagnostic dilemmas sequentially: Consider disorder A, and order test A; then, when test A fails to confirm disorder A, consider disorder B and test B, which can eventually lead to consideration of disorder C — and so on. At times this approach is appropriate, but some cases move faster toward resolution if several diagnoses are considered, and tested for, at once.
You know which of the available consultants are the most efficient, so use them preferentially. If a given MCO doesn't include a particularly efficient consultant on its panel, discuss the matter with the plan's medical director, who may be unaware of the issues involved. If the MCO ignores your advice, it may be because there is some overriding business reason from the MCO's point of view (such as confidential contracting problems) for not including that physician in its network. Don't take it personally.
Take advantage of allied services and resources
It's also important for family physicians to know the resources available during and after an inpatient stay that can speed a case toward resolution. A call to the MCO for preauthorization will alert its concurrent review staff and get them involved at the earliest possible point. They can help with discharge planning sometimes before the patient is even admitted.
Most physicians eventually get to know the pharmacists, physiotherapists, health educators and other professionals in the hospital, but the efficient physician actively seeks these people out and asks them for their ideas and support. Many allied health professionals complain that physicians often don't know what they can do and ignore their advice. A captain disregards the advice of the crew at the ship's peril.
A study at one hospital found that by bringing various professionals together to conduct daily rounds on medicinal patients, both the lengths of stay and the cost of care were reduced compared with the traditional model of an attending physician rounding in relative isolation.2
You also can take steps to reduce “social admissions.” For example, an elderly man left at the ED at 9 p.m. by relatives who insist they can no longer care for him needs social rather than medical care. Physicians need to be prepared with an extensive list of local resources to help them deal with this and similar scenarios without automatically admitting these patients. Know whom to contact at odd hours.
Resist the technological imperative
Of course, just because resources are available doesn't mean they must be tapped, particularly if they're expensive and likely to lead to high-intensity procedures. Before seeking a consult or ordering a test, consider the likely outcome of what you're tempted to order. When expensive equipment reveals something, doctors tend to want to act on it. Imaging studies often show abnormalities that lead to follow-up, even though the findings may or may not be related to disease. By statistical definition, roughly 5 percent of laboratory test results on healthy people fall outside the laboratory's normal range, sometimes prompting further investigation of problems that patients aren't experiencing. Patients, having been admitted, expect things to be done, and physicians expect to do them. Consultations with pulmonologists tend to lead to bronchoscopy. Consultations with surgeons tend to lead to surgery. And so on.
The challenge, then, is to proceed aggressively with necessary testing and procedures while resisting the reflex to order more than necessary simply because it's available. Some excessive testing may be driven by worry about being sued for having missed something or having failed to investigate for serious but rare disorders. Physicians who are well-informed about medical malpractice issues understand that ordering lots of tests will not protect them from lawsuits. In fact, since overly aggressive interventions can produce untoward complications, that defensive strategy has its own risks. It's better to proceed deliberately and thoughtfully, with good documentation, than to be overly aggressive just for the sake of appearing thorough.
Find ways to avoid futile care
It has been written that “physicians frequently practice as though every available medical measure, including absurd and overzealous interventions, must be used to prolong life unless patients give definitive directions to the contrary.”3 For the most part, insurers are loath to intervene in cases of extended, apparently futile intensive care. Family members are generally confused and frightened. It's the doctor's role to look beyond the automatic reaction of doing something, anything, to prolong life.
For your own psychological comfort, if not for the benefit of your patients, familiarize yourself with recent debates about futile medical care.3,7–9 Find out whether your hospital has a policy on futile care and an ethics committee to help you with the more challenging cases. The ethical considerations are changing as rapidly as the capabilities of high-tech care.
Discuss utilization with patients and families
Families and patients, understandably, may perceive efforts to decrease lengths of stay and prevent admissions as denials of necessary care. The problem is exacerbated when patients expect their admissions to be longer than they are, perhaps after having been told something that seems like a promise. Imagine telling a patient that his or her admission will continue until a certain drug has been infused over several days, only to discover that the drug can be safely infused at home. The physician will either discharge the patient earlier than promised, at best creating confusion, or try to save face by keeping the patient in the hospital unnecessarily.
Use language carefully to avoid offering false certainties. Rather than saying, “You'll be going home by Tuesday” — thus setting a date regardless of the clinical situation and making discharge awkward even if the patient's condition improves unexpectedly — tell the patient he or she will be discharged when specific endpoints are reached. For tough, complex cases, family and staff conferences can help (and physicians would do well to attend them).
Look for recurring problems
A physician can affect inpatient-care efficiency on the macro level as well as the micro. Using a proven industrial quality-improvement technique, you could keep a log showing major categories of reasons for delayed discharges. After a period of time, you should find that one or two categories stand out as the most common problem areas. Then you can focus on eliminating the root causes. This is not a task for one lonely family physician crusader. A diverse team of professionals is best at solving challenging problems in the health care system.
Such a team might find, for example, that the most common reason for delays in discharge is difficulty in finding skilled-nursing beds. The team should resist the first solution that leaps to mind (build a new skilled-care center!) as a panacea. Then it should consider all the possible reasons why skilled-nursing beds are hard to find, digging as deeply as possible to find the underlying causes (such as unnecessarily long stays in skilled-care centers, which have their own underlying causes). Once the team has listed the underlying problems, it will be clear that some are easier to solve than others or that some will have a greater return on the effort that goes into the solution. Then the team can plan a logical set of interventions to deal with those problems.
Know your own performance
Those who seek solutions should always ensure that they aren't the ones causing problems. So be sure to take a good look at your own inpatient-care performance. MCOs and hospitals may have data they will share on the efficiency of your inpatient care.
If external data aren't available (or helpful), you can evaluate yourself. For a series of admissions over a period of time, calculate your average length of stay. If you're managing a specific population of a known size, calculate your bed days per thousand per year or your admission rate per thousand per year, and then compare yourself to benchmarks. Collecting these data could help you defend yourself against an aggressive managed care medical director or perhaps help you get a new contract. There are limits to how useful this record-keeping can be, however, if your number of inpatients is small or if you've had a few patients with unusually long lengths of stay or high frequencies of admission, which will skew your results.
At the level of individual cases, you can avoid being part of the problem by making sure others can follow your charting. Be sure to chart clear progress notes that show what you're planning and what clinical endpoints you're watching for. Not only will you be doing your part to make your patients' care more efficient, you'll receive fewer calls from the MCOs' utilization staff.
Today, an inpatient stay that is handled well is increasingly appreciated by patients, payers and providers alike. It's not just a matter of making care more cost-effective. It's a matter of helping patients get back on their feet and back into their lives more quickly — and helping you to move on to the next person who needs your help.