Jane, a 27-year-old woman, arrives to your busy clinic at the end of the day. She looks miserable and describes a lingering cold for the last few days, including sore throat, cough, congestion, and sneezing. Before you are even able to do an exam, she asks you for antibiotics. You examine her and decide she most likely has a virus, not a bacterial infection that would require antibiotics. In fact, you do not really think she will need any further diagnostic testing at all.
So you ask the patient why she wants antibiotics. She tells you that she is getting on an airplane the next day to go to a series of important meetings. She is worried about strep throat. “How will I get antibiotics if I get sicker?” she asks. You look at your watch and notice you are running 30 minutes behind schedule. You worry Jane will be upset if you do not give her what she wants. With a deep sigh, you quickly write her a prescription for azithromycin.
Antibiotics are among the most remarkable advances in modern medicine and have saved countless lives over the better part of the last century. However, when used incorrectly, antibiotics pose serious risks to both individual patients and the public health at large (we often wonder if the drug-resistant infections that catalyze the apocalypse in every modern Zombie movie are really that far off). Using antibiotics when they are not needed increases drug-resistance, leaving populations of patients vulnerable to resistant infections. Antibiotic overuse can also place individual patients at risk for allergic reactions, antibiotic-associated diarrhea, and other highly unpleasant and dangerous side effects.1 These risks are well known to clinicians—as are the decades of clear evidence, guidelines, and quality measures that argue against prescribing antibiotics in certain conditions, such as routine sore throats and acute bronchitis.2-4 Sir Alexander Fleming, the discoverer of penicillin, even warned in a 1945 New York Times article: “… the microbes are educated to resist penicillin and a host of penicillin-fast organisms is bred out…. In such cases the thoughtless person playing with penicillin is morally responsible for the death of the man who finally succumbs to infection with the penicillin-resistant organism. I hope this evil can be averted.”5,6
And yet, currently when someone is diagnosed with a viral sore throat, like our patient Jane, antibiotics are prescribed approximately 60% of the time.7 For acute bronchitis, patients are prescribed an antibiotic more than 70% of the time—a rate that has remarkably been increasing over time.8 Why is this happening?
In many cases, clinicians overprescribe antibiotics knowingly and are “caving” to perceived patient demand. In a world that often forces clinicians to see more patients in less time (see Chapter 9), it may be easier to just write out a requested prescription rather than spend the time educating a patient about why she does not need the antibiotic that she specifically came to get. Now there is the added pressure that a patient may walk out of your office and immediately leave a negative review on one of the many public-facing doctor-rating websites (see Chapter 12). Nobody wants unsatisfied patients. However, as we discuss later in this chapter, perceived patient demand and actual patient demand do not always line up. It turns out that physician-patient communication—not antibiotic prescribing—is the most important factor related to patient satisfaction in acute bronchitis.3
Many different organizations have called for mandated “antimicrobial stewardship” programs across healthcare settings.1,6 Antimicrobial stewardship “refers to coordinated interventions designed to improve and measure the appropriate use of antimicrobial agents by promoting the selection of the optimal antimicrobial drug regimen including dosing, duration of therapy, and route of administration.”6 The major goals of these programs are to achieve the best clinical outcomes while minimizing toxicity and other adverse events, protecting populations from emerging threats to public health, and reducing excessive costs.1,6 In other words, improving healthcare value, as we have defined it throughout this book.
Antibiotic stewardship is but one example that highlights both the many real cultural, operational, and financial barriers that currently hinder the delivery of high-value care, as well as the extraordinary potential for improvements (Table 10-1). Addressing these barriers will require a combination of solutions, including delivery system reform as well as improved bedside communication, feedback, and clinical decision-support. We will introduce and examine the many barriers here and then will build upon proposed solutions throughout Part III.
Barriers to high-value care
|Barriers to High-Value Care||Examples|
|Misaligned financial incentives||A patient with viral pharyngitis is seen in the office because telephone care is not reimbursed.|
|Time pressure||A patient with a viral upper respiratory tract infection who asks for antibiotics is given a prescription because it takes less time than explaining why the patient does not need antibiotics.|
|Imprecise measurements||Insurance claims data do not account for clinical decision making based on individual patient characteristics, nor do they assess the quality of the patient experience.|
|Lack of education and training||Clinicians do not incorporate costs into decision making because they were not taught where to find costs of common tests and treatments.|
|Healthcare system fragmentation||A test done at another institution is repeated because the electronic medical records are not interoperable and the results are not available.|
|Local culture and hidden curriculum||The attending physician commends the medical student for working up a rare but unlikely diagnosis on his/her patient.|
|Discomfort with diagnostic uncertainty||Ordering the additional testing when the patient has a straightforward clinical diagnosis “just to be sure.”|
|Fear of malpractice||Increased hospital admissions for atypical chest pain after a clinician was sued for a bad outcome when he/she sent a patient with chest pain home from the ED.|
|Patient expectations||A desire to please the patient by ordering advanced imaging for low back pain because the patient requests the study.|
Clinicians and policymakers do not always see eye to eye, but there is one thing that nearly everyone can agree on: a major barrier to improving value is a healthcare system that is replete with misaligned financial incentives. As we discussed in Chapter 3, the stakeholders (patients, providers, and payers) each consider the costs of care from their own perspective.
In most cases, US providers are paid more for doing more. Conversely, doing less can mean decreased revenue for a delivery system and lower salary for a clinician.9 The fee-for-service payment model rewards clinicians for maximizing the number of patient encounters and procedures but often does not reimburse for important, value-added activities, like telephone follow-up and care coordination, that improve the quality of care.10 As discussed in the last chapter (Chapter 9), compensation models for clinicians are typically based on relative value units (RVUs). Thus, even in delivery organizations that have risk-sharing contracts at the delivery system level (see Chapter 15), their clinicians are still strongly incentivized to provide as many services as possible.
If providers were reimbursed based on patient outcomes as opposed to the number of services they deliver, they would have stronger incentives to provide value-based care. But it is not only the clinicians – the vast majority of hospital executives, including those at nonprofit hospitals, are also compensated based on metrics related to productivity, rather than patient outcomes or community benefit.11 There is substantial resistance to changing financial models at this point.
The cost structure of a typical healthcare setting is relatively rigid due to a high proportion of fixed costs (Chapter 4) that are minimally sensitive to small changes in patient volume, resource use, or the severity of patient illness.12 At the same time, the operations of the hospital are designed to optimize revenue by delivering as many services as possible. In addition, more clinicians are employed by hospital systems than ever before and may be financially motivated to use their facilities and advanced technology.9,13,14 Many health professionals are well-intentioned and trying to do all they can to take the best care possible for their patients, but it is hard to always do the right thing when placed in a system that floats on a river moving in the wrong direction.
As examined in Chapter 3, the situation becomes even more messy when considering the misaligned financial incentives of other stakeholders in the system, including payers, patients, pharmaceutical companies, and technology vendors. Each party is encouraged to charge different amounts to different customers, all creating “chaos behind a veil of secrecy.”15 Understanding the limitations of our current reimbursement models will help guide our work redesigning care delivery and realigning financial incentives with improved patient outcomes (see Chapter 15 on “Shifting Incentives”).
Story From the Frontlines: “I’ll Do It for $1100”
Where I am from, you can have someone killed for $5000. I will do it for $1100. I am a hand surgeon.
I practice (or practiced, by the time you read this) in an area that is what we often refer to as “underserved.” Rather, the area isn’t, but the people I treat are. I work in a large urban referral center that has a very high proportion of Medicaid, as well as unfunded patients. No one else in town will touch them. I am not blaming them—they are in private practice and they cannot cover their expenses if they are paid nothing or close to nothing for their time and supplies. While there may be an element of greed, it is not all greed. I know my colleagues in private practice and, almost without exception, I respect them all as physicians and people.
In my referral center, the hospital has favorable contracts with Medicaid that yield good revenue for the center from Medicaid patients. But, as a consequence, all of my procedures are “hospital-based” as opposed to “clinic-based.” It is a semantic billing distinction that I do not completely understand myself (another part of the problem) and it allows the hospital to generate enough revenue to cover costs for an unquestionably needy population. For the patients with Medicaid, it allows us to care for them with little, if any out-of-pocket cost to them and keep the lights on. However, for patients with private insurance that is anything less than a top-of-the-line plan, procedures done in outpatient “hospital-based” clinics are not covered and are billed at very high rates that come out-of-pocket. A steroid injection for tendonitis can yield a bill in excess of $1000. Doing a simple wound “clean-up” or debridement can be north of $1100. I am on salary and do not make an extra nickel either way.
Which is why when Mr. Jones, an overweight diabetic with private insurance presented with a small local infection that I probably could have addressed in the office, I took him to the OR. His insurance company would only cover the costs if it were done upstairs, but would pay nothing if I did it in my clinic.
Once in the OR, the regional block he received did not work well (which happens) and his sedation was increased. The increased sedation made it difficult for him to breathe and he had to be ventilated emergently. He lost his airway. His oxygen saturation dipped below 60% of normal, briefly, and the anesthesiologists were able to right the ship and wake him up.
Ultimately, things went well. The patient’s hand was healed and he didn’t face a medical bill that would have decimated his financial health. However, I nearly killed him to save him $1110.
—Robert Gray. “I’ll Do It for $1100.”
Costs of Care, 2013. (www.costsofcare.org)
As the story that opens this chapter illustrates, time pressure is frequently identified by practicing clinicians as a significant barrier to delivering the best possible care. This is hardly surprising. Remember from Chapter 1 that it would take 21 hours a day for a primary care provider to provide all of the care recommended to meet his patients’ acute, preventive, and chronic disease management needs.16
Often, the way clinicians deal with time pressure is by looking for ways to decrease their workload. For a busy emergency department (ED) physician with a packed waiting room, figuring out how to empty the stretcher in front of her may seem like the number one goal. Instead of ordering one test and then waiting for the result before ordering more, ordering several tests at once may be more expeditious. Pressure to reduce waiting times is one of the most significant sources of stress for ED staff.17 Similarly, hospital physicians under pressure to decrease length of stay may order multiple tests and consultations simultaneously to expedite patient disposition. In a different era of practicing medicine, diagnoses were often clarified by just relying on a “tincture of time.” In our hospitals and ambulatory offices, there is little patience for this approach despite the evidence that “watchful waiting” (sometimes referred to as “expectant management”) is a feasible plan for patients with unexplained complaints and it does not result in delayed testing.18 Indeed, as length of stay has decreased, there is some evidence that it does not actually result in overall cost savings, but instead merely shifts the increased costs to the ambulatory setting.19
In the ambulatory setting, time pressure also drives inappropriate use of tests and treatments. As our story illustrates, it can be quicker to write a prescription or order a test than it is to explain to a patient why these interventions may not be necessary. Shorter office visits, particularly those less than 15 minutes, are more likely to lead to inappropriate prescribing.20 Recent studies exploring barriers to improved outcomes in hypertensive patients highlight the association of improved therapy adherence with strong patient-provider relationships.21 Provider time might be better spent establishing and maintaining continuous healing relationships with patients than ordering tests and treatments that are unlikely to improve patient outcomes. In fact, this “time pressured” approach to care delivery may result in the creation of unnecessary additional workload from checking results, informing patients of results, and providing follow-up testing or treatment that occurs from the discovery of incidental findings. There are multiple components of care that are affected by time, including patient satisfaction, outcomes of chronic diseases, prescribing practices, physician satisfaction, and risk of malpractice claims.22
Throughout healthcare, poorly functioning work processes are created by unnecessary pauses and rework, delays, and established workarounds.23 The inefficiency in hospital care delivery has created a system where nurses spend less than one-third of their working time performing direct patient care.24 In turn, time pressure, burnout, and workload for nurses have been associated with patient safety risks.25,26
Clinicians may become accustomed to some of the pressures and flaws inherent in the current healthcare system, but the maddening fragmentation is starkly obvious to patients and their families. In healthcare provider offices, faxes—yes, faxes, in 2014—often provide vital documents such as discharge summaries or home care orders. These flimsy sheets of paper that often never make it to the provider or to the patient’s chart, symbolize our antiquated and insufficient systems of communications in healthcare. Left with increasing provider specialization, poor continuity of care, and insufficient communication among providers, patients, and families, the modern-day US healthcare system is incredibly complex and fragmented—a single patient often receives care from multiple clinicians who work in different facilities.19,27 This fragmentation may lead to duplication of tests and confusion about the care plan and often results in poor outcomes at a higher cost. In the United Kingdom, discontinuity of care was associated with decreased rates of cancer screening and prevention, and when these deficits were addressed in a national program, outcomes improved.27
Healthcare providers have been encouraged to transition to the use of electronic health records (EHR) through a Federal initiative called the “meaningful use” program in hopes that patient records could be more easily shared and communication among providers would be enhanced. However, the current EHR system has not lived up to this promise because electronic records at one health facility are rarely compatible with those at another facility.28 Even within the same hospital system, it is not uncommon to find that the inpatient and outpatient electronic records are not connected. When information is not easily accessible, providers often opt to duplicate testing rather than spend the time trying to track down previous results.29 Improving communication, continuity and access in such a way that incentivizes providers to have ongoing healing relationships with their patients would go a long way in combating the fragmented system, reducing waste and improving health outcomes.