INTRODUCTION
Each year in the United States, millions of people visit hospitals, physicians, and other caregivers and receive medical care of superb quality. But that’s not the whole story. Some patients’ interactions with the health care system fall short (Institute of Medicine, 1999, 2001) and widespread quality improvement has proven elusive (Chassin & Loeb, 2011).
At the beginning of the 21st century, an estimated 32,000 people died in US hospitals each year as a result of preventable medical errors (Zahn & Miller, 2003). A 2013 study suggests that the actual number is higher (James, 2013). The rate of patient harms caused by care delivered in hospitals did not change from 2002 through 2007 (Landrigan et al., 2010).
In addition, an estimated 57,000 people in the United States died because they were not receiving appropriate health care—in most cases, because common medical conditions such as high blood pressure or elevated cholesterol are not adequately controlled (National Committee for Quality Assurance, 2010). Hospitals vary greatly in their risk-adjusted mortality rates for Medicare patients; during 2009 to 2012, risk-adjusted deaths from heart failure and pneumonia were three times higher for lower-quality compared with higher-quality hospitals (Medicare Hospital Quality Chartbook, 2013). A previous study showed that if low-quality hospitals reduced mortality rates to the level of high-quality hospitals, 17,000 to 21,000 fewer deaths per year would have occurred (Schoen et al., 2006).
Fatal medication errors among outpatients doubled between 1983 and 1993 (Phillips et al., 1998). Prescribing errors occur in 7.6% of outpatient prescriptions (Gandhi et al., 2005), which amounts to 228 million errors in 2004. In 2007, about 25% of elderly patients received high-risk medications (Zhang et al., 2010). Diagnostic error rates are around 10% for a variety of medical conditions (Wachter, 2010). In some primary care practices, patients are not informed about abnormal laboratory results more than 20% of the time (Casalino et al., 2009).
Two million lives would have been saved in 2006 if preventive services had been regularly delivered to the entire population (Maciosek et al., 2010). Only 50% of people with hypertension are adequately treated (Egan, 2010) and 43% of people with diabetes are inadequately controlled (Cheung et al., 2009). Racial and ethnic minority patients experience an inferior quality of care compared with white patients (Agency for Healthcare Research and Quality, 2015).
A recent article (Chassin and Loeb, 2011) summarized that “Health care quality and safety today are best characterized as showing pockets of excellence on specific measures or in particular services at individual health care facilities … The pockets of excellence mentioned above coexist with enormously variable performance across the delivery system. Along with some progress, we are experiencing an epidemic of serious and preventable adverse events … The risk of harmful error in health care may be increasing. As new devices, equipment, procedures, and drugs are added to our therapeutic arsenal, the complexity of delivering effective care increases. Complexity greatly increases the likelihood of error, especially in systems that perform at low levels of reliability.”
A prominent Institute of Medicine report (2001) concluded that between what we know and what we do lies not just a gap, but a chasm. Quality problems have been categorized as overuse, underuse, and misuse (Chassin et al., 1998). We will first examine the factors contributing to poor quality and then explore what can be done to elevate all health care to the highest possible level.
THE COMPONENTS OF HIGH-QUALITY CARE
What is high-quality health care? It is care that assists healthy people to stay healthy, cures acute illnesses, and allows chronically ill people to live as long and fulfilling a life as possible. What are the components of high-quality health care? (Table 10-1)
Lydia and Laura were friends at a rural high school; both became pregnant. Lydia’s middle-class parents took her to a nearby obstetrician, while Laura, from a family on welfare, could not find a physician who would take Medicaid. Lydia became the mother of a healthy infant, but Laura, going without prenatal care, delivered a low–birth-weight baby with severe lung problems.
To receive quality care, people must have access to care. People with reduced access to care suffer worse health outcomes in comparison to those enjoying full access—the quality problem of underuse (see Chapter 3). Quality requires equality (Schiff et al., 1994).
Brigitte Levy, a professor of family law, was started on estrogen replacement in 1960 when she reached menopause. Her physician prescribed the hormone pills for 10 years. In 1979, she was diagnosed with invasive cancer of the uterus, which spread to her entire abdominal cavity in spite of surgical treatment and radiation. She died in 1980 at age 68, at the height of her career.
A body of knowledge must exist that informs physicians what to do for the patient’s problem. If clear scientific knowledge fails to distinguish between effective and ineffective or harmful care, quality may be compromised. During the 1960s, medical science taught that estrogen replacement, without the administration of progestins, was safe. Sadly, cases of uterine cancer caused by estrogen replacement did not show up until many years later. Brigitte Levy’s physician followed the standard of care for his day, but the medical profession as a whole was relying on inadequate scientific knowledge. Treatments of uncertain safety and efficacy may cause harm and cost billions of dollars each year.
Ceci Yu, age 77, was waking up at night with shortness of breath and wheezing. Her physician told her she had asthma and prescribed albuterol, a bronchodilator. Two days later, Ms. Yu was admitted to the coronary care unit with a heart attack. Writing to the chief of medicine, the cardiologist charged that Ms. Yu’s physician had misdiagnosed the wheezing of congestive heart failure and had treated Ms. Yu incorrectly for asthma. The cardiologist charged that the treatment might have precipitated the heart attack.
The provider must have the skills to diagnose problems and choose appropriate treatments. An inadequate level of competence resulted in poor quality care for Ms. Yu.
The Harvard Medical Practice study reviewed 30,000 medical records in 51 hospitals in New York State in 1984 (Studdert et al., 2004). The study found that in approximately 4% of hospital admissions, the patient experienced a medical injury (i.e., a medical problem caused by the management of a disease rather than by the disease itself); this is the quality problem of misuse. A more recent study placed the percent of hospital patients experiencing a medical injury at 13.8% (Meurer et al., 2006). Medical injuries can be classified as negligent or not negligent.
Jack was given a prescription for a sulfa drug. When he took the first pill, he turned beet red, began to wheeze, and fell to the floor. His friend called 911, and Jack was treated in the emergency department for anaphylactic shock, a potentially fatal allergic reaction. The emergency medicine physician learned that Jack had developed a rash the last time he took sulfa. Jack’s physician had never asked him if he was allergic to sulfa, and Jack did not realize that the prescription contained sulfa.
Mack was prescribed a sulfa drug, following which he developed anaphylactic shock. Before writing the prescription, Mack’s physician asked whether he had a sulfa allergy. Mack had said “No.”
Medical negligence is defined as failure to meet the standard of practice of an average qualified physician practicing in the same specialty. Jack’s drug reaction must be considered negligence, while Mack’s was not. Of the medical injuries discovered in the Harvard study, 28% were because of negligence. In those injuries that led to death, 51% involved negligence. The most common injuries were drug reactions (19%) and wound infections (14%). Eight percent of injuries involved failure to diagnose a condition, of which 75% were negligent. Seventy percent of patients suffering all forms of medical injury recovered completely in 6 months or less, but 47% of patients in whom a diagnosis was missed suffered serious disabilities (Brennan et al., 1991; Leape et al., 1991).
Negligence cannot be equated with incompetence. Any good health care professional may have a mental lapse, may be overtired after a long night in the intensive care unit, or may have failed to learn an important new research finding.
Nina Brown, a 56-year-old woman with diabetes, arrived at her primary care physician’s office complaining of chest pain over the past month. Her physician examined Ms. Brown, performed an electrocardiogram (ECG), which showed no abnormalities, diagnosed musculoskeletal pain, and recommended ibuprofen. Five minutes later in the parking lot, Ms. Brown collapsed of a heart attack. The health plan insuring Ms. Brown had an incentive arrangement with primary care physicians whereby the physicians receive a bonus payment if the physicians reduce use of emergency department and referral services below the community average.
Completely healthy at age 45, Henry Fung reluctantly submitted to a treadmill exercise test at the local YMCA. The study was inconclusive and Mr. Fung, who had fee-for-service insurance, sought the advice of a cardiologist. The cardiologist knew that treadmill tests are sometimes positive in healthy people. Yet he ordered a coronary angiogram, which was perfectly normal. Three hours after the study, a clot formed in the femoral artery at the site of the catheter insertion, and emergency surgery was required to save Mr. Fung’s leg.
No one can know what motivated the physician to send Ms. Brown home instead of to an emergency department when unstable coronary heart disease was one possible diagnosis (underuse); nor can one guess what led the fee-for-service cardiologist to perform an invasive coronary angiogram of questionable appropriateness on Mr. Fung (overuse). One factor that bears close attention is the impact of financial considerations on the quantity (and thus the quality) of medical care (Relman, 2007). As noted in Chapter 4, fee-for-service reimbursement encourages physicians to perform more services, whereas capitation payment rewards those who perform fewer services.
More than 40 years ago, Bunker (1970) found that the United States performed twice the number of surgical procedures per capita than Great Britain. He postulated that this difference could be accounted for by the greater number of surgeons per capita in the United States and concluded that “the method of payment appears to play an important, if unmeasured, part.” Most surgeons in the United States are compensated by fee-for-service, whereas most in Great Britain are paid a salary. From 8% to 86% of surgeries—depending on the type—have been found to be unnecessary and have caused substantial avoidable death and disability (Leape, 1992). An analysis of the National Practitioner Data Base suggests that 10% to 20% of all surgeries in several specialties are unnecessary (Eisler & Hansen, 2013). As an example, spinal fusion surgery increased by 77% from 1996 to 2001, though little evidence supports this procedure in many cases. From 2002 to 2007, Medicare patients undergoing surgery for lumbar spinal stenosis experienced a doubling of complex rather than simple operations resulting in a major increase in surgical complications, rehospitalizations, and costs. Rates of reoperation (because of worsening pain) are high. Payment for this procedure is greater than that provided for most other procedures performed by orthopedists and neurosurgeons (Deyo et al., 2004 and 2010).
It was a nice dinner, hosted by the hospital radiologist and paid for by the company manufacturing magnetic resonance imaging (MRI) scanners. After the meal came the pitch: “If you physicians invest money, we can get an MRI scanner near our hospital; if the MRI makes money, you all share in the profits.” One internist explained later, “After I put in my $10,000, it was hard to resist ordering MRI scans. With headaches, back pain, and knee problems, the indications for MRIs are kind of fuzzy. You might order one or you might not. Now, I do.”
Relman (2007) writes about the commercialization of medicine: “The introduction of new technology in the hands of specialists, expanded insurance coverage, and unregulated fee-for-service payments all combined to rapidly increase the flow of money into the health care system, and thus sowed the seeds of a new, profit-driven industry.”
During the 1980s, many physicians formed partnerships and joint ventures, giving them part ownership in laboratories, MRI scanners, and outpatient surgicenters. By 1990, 93% of diagnostic imaging facilities, 76% of ambulatory surgery centers, and 60% of clinical laboratories in Florida were owned wholly or in part by physicians. The rates of use for MRI and CT scans were higher for physician-owned compared with nonphysician-owned facilities (Mitchell & Scott, 1992). In a national study, physicians who received payment for performing x-rays and sonograms within their own offices obtained these examinations four times as often as physicians who referred the examinations to radiologists and received no payment for the studies (Hillman et al., 1990).
Profitable diagnostic, imaging, and surgical procedures have rapidly migrated from the hospital to free-standing physician-owned ambulatory surgery centers, endoscopy centers, and imaging centers, with rapid increases in the number of tests and procedures performed (Berenson et al., 2006). The number of CT scans performed for Medicare patients increased by 65% from 2000 to 2005; during those years, the number of MRI scans jumped by 94% (Bodenheimer et al., 2007). The number of CT scans was growing by more than 10% per year, increasing patients’ risk of radiation-related cancer (Smith-Bindman, 2010). A significant association exists between surgeon ownership of ambulatory surgery centers and a higher volume of surgeries; surgery volume increases immediately following surgeons’ acquisition of the surgicenter (Hollingsworth et al., 2010).
Moving to the other side of the overuse—underuse spectrum, payment by capitation, or salaried employment by a for-profit business, may create a climate hostile to the provision of adequate services. In the 1970s, a series of HMOs called prepaid health plans (PHPs) sprang up to provide care to California Medicaid patients. The quality of care in several PHPs became a major scandal in California. At one PHP, administrators wrote a message to health care providers: “Do as little as you possibly can for the PHP patient,” and charts audited by the California Health Department revealed many instances of undertreatment. The PHPs received a lump sum for each patient enrolled, meaning that the lower the cost of the services actually provided, the greater the PHP’s profits (US Senate, 1975). More recent approaches to capitation payment have attempted to mitigate incentives for undertreatment by requiring providers to achieve quality of care targets and risk-adjusting capitation payments, with payments for patients at greater risk for needing medical services higher than payments for low-risk patients. The quantity and quality of medical care are inextricably interrelated. Too much or too little can be injurious. The research of Fisher et al. (2003) has shown that similar populations in different geographic areas have widely varying rates of surgeries and days in the hospital, with no consistent difference in clinical outcomes between those in high-use and low-use areas.
The personnel cutbacks were terrible; staffing had diminished from four RNs per shift to two, with only two aides to provide assistance. Shelley Rush, RN, was 2 hours behind in administering medications and had five insulin injections to give, with complicated dosing schedules. A family member rushed to the nursing station saying, “The lady in my mother’s room looks bad.” Shelley ran in and found the patient unconscious. She quickly checked the blood sugar, which was disastrously low at 20 mg/dL. Shelley gave 50% glucose, and the patient woke up. Then it hit her—she had injected the insulin into the wrong patient.
Health care institutions must be well organized, with an adequate, competent staff. Shelley Rush was a superb nurse, but understaffing caused her to make a serious error. The book Curing Health Care by Berwick et al. (1990) opens with a heartbreaking case:
She died, but she didn’t have to. The senior resident was sitting, near tears, in the drab office behind the nurses’ station in the intensive care unit. It was 2:00 AM, and he had been battling for 32 hours to save the life of the 23-year-old graduate student who had just suffered her final cardiac arrest.
“Routine screening chest x-ray, taken 10 months ago. The tumor is right there, and it was curable—then. By the time the second film was taken 8 months later, because she was complaining of pain, it was too late. The tumor had spread everywhere, and the odds were hopelessly against her. Everything we’ve done since then has really just been wishful thinking. We missed our chance. She missed her chance.” Exhausted, the resident put his head in his hands and cried.
Two months later, the Quality Assurance Committee completed its investigation.… “We find the inpatient care commendable in this tragic case,” concluded the brief report, “although the failure to recognize the tumor in a potentially curable stage 10 months earlier was unfortunate.… ” Nowhere in this report was it written explicitly why the results of the first chest x-ray had not been translated into action. No one knew.
One year later …. it was 2:00 AM, and the night custodian was cleaning the radiologist’s office. As he moved a filing cabinet aside to sweep behind it, he glimpsed a dusty tan envelope that had been stuck between the cabinet and the wall. The envelope contained a yellow radiology report slip, and the date on the report—nearly 2 years earlier—convinced the custodian that this was, indeed, garbage … He tossed it in with the other trash, and 4 hours later it was incinerated along with other useless things.
This patient may have had perfect access to care for an illness whose treatment is scientifically proved; she may have seen a physician who knew how to make the diagnosis and deliver the appropriate treatment; and yet the quality of her care was disastrously deficient. Dozens of people and hundreds of processes influence the care of one person with one illness. In her case, one person—perhaps a file clerk with a near-perfect record in handling thousands of radiology reports—lost control of one report, and the physician’s office had no system to monitor whether or not x-ray reports had been received. The result was the most tragic of quality failures—the unnecessary death of a young person.
How health care systems and institutions are organized has a major impact on health care outcomes. For example, large multispecialty group practices in 22 metropolitan areas have better-quality measures at lower cost than dispersed physician practices in those areas (Weeks et al., 2009). Nurse understaffing is associated with higher hospital mortality rates (Needleman et al., 2011). Studies have shown that hospitals with more RN staffing have lower surgical complication rates (Kovner & Gergen, 1998) and lower mortality rates (Aiken et al., 2002).
Oliver Hart lived in a city with a population of 80,000. He was admitted to Neighborhood Hospital with congestive heart failure caused by a defective mitral valve. He was told he needed semiurgent heart surgery to replace the valve. The cardiologist said “You can go to University Hospital 30 miles away or have the surgery done here.” The cardiologist did not say that Neighborhood Hospital performed only seven cardiac surgeries last year. Mr. Hart elected to remain for the procedure. During the surgery, a key piece of equipment failed, and he died on the operating table.
Quality of care must be viewed in the context of regional systems of care (see Chapter 6), not simply within each health care institution. In one study, 27% of deaths related to coronary artery bypass graft (CABG) surgery at low-volume hospitals might have been prevented by referral of those patients to hospitals performing a higher volume of those surgeries (Dudley et al., 2000). Hospitals performing more surgical procedures have lower mortality rates for that surgical procedure (Gonzalez et al., 2014). Quality improves with the experience of those providing the care (Kizer, 2003; Peterson et al., 2004). Had Mr. Hart been told the relative surgical mortality rates at University Hospital, which performed 500 cardiac surgeries each year, and at Neighborhood Hospital, he would have chosen to be transferred 30 miles down the road. Not only does the volume of surgeries in a hospital matter; equally important is the volume of surgeries performed by the specific surgeon (Birkmeyer et al., 2003).
In the late 1980s, Dr. Donald Berwick (1989) and others realized that quality of care is not simply a question of whether or not a physician or other caregiver is competent. If poorly organized, the complex systems within and among medical institutions can thwart the best efforts of professionals to deliver high-quality care.
There are two approaches to the problem of improving quality … [One is] the Theory of Bad Apples, because those who subscribe to it believe that quality is best achieved by discovering bad apples and removing them from the lot.… The Theory of Bad Apples gives rise readily to what can be called the my-apple-is-just-fine-thank-you response … and seeks not understanding, but escape. [The other is] the Theory of Continuous Improvement.… Even when people were at the root of defects, … the problem was generally not one of motivation or effort, but rather of poor job design, failure of leadership, or unclear purpose. Quality can be improved much more when people are assumed to be trying hard already, and are not accused of sloth. Fear of the kind engendered by the disciplinary approach poisons improvement in quality, since it inevitably leads to the loss of the chance to learn.… When one is clear and constant in one’s purpose, when fear does not control the atmosphere (and thus the data), when learning is guided by accurate information … and when the hearts and talents of all workers are enlisted in the pursuit of better ways, the potential for improvement in quality is nearly boundless.
Good-quality care can be compromised at a number of steps along the way.
Angie Roth has coronary heart disease and may need CABG surgery. (1) If she is uninsured and cannot get to a physician, high-quality care is impossible to obtain. (2) If clear evidence-based guidelines do not exist regarding who benefits from CABG and who does not, Ms. Roth’s physician may make the wrong choice. (3) Even if clear guidelines exist, if Angie Roth’s physician fails to evaluate her illness correctly or sends her to a surgeon with poor operative skills, quality may suffer. (4) If indications for surgery are not clear in Ms. Roth’s case but the surgeon will benefit economically from the procedure, the surgery may be inappropriately performed. (5) Even if the surgery is appropriate and performed by an excellent surgeon, faulty equipment in the operating room or poor teamwork among the operating room surgeons, anesthesiologists, and nurses may lead to a poor outcome.
The Institute of Medicine, in its influential 2001 report Crossing the Quality Chasm, conceptualized six core dimensions of quality: safe, effective, patient-centered, timely, efficient, and equitable. These dimensions, defined in greater detail in Table 10-2, are consistent with the components of quality discussed earlier.
|
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
