Executive Summary

Table of Contents: Background | Purpose | Methodology | Results | Conclusions |

Background

Concern about the extent of accidental injury to patients undergoing medical treatment has increased substantially in the past several years. This concern is partly in response to publicity surrounding egregious cases of mistreatment and partly in response to evidence indicating that the problem may be far more serious than previously assumed. A report issued by the Institute of Medicine in November 1999, which estimated that between 44,000 and 98,000 patients die each year as a result of preventable medical errors, and that costs associated with medical error total between $17 billion and $29 billion, has greatly increased public concerns about patient safety. This concern has stimulated a number of initiatives designed to improve patient safety. It is likely that these will lead to improvements in patient safety.

There is substantial evidence that older patients (defined as those age 65 and older) are at substantially greater risk for iatrogenic (treatment or procedure-related) medical injury than other age groups are. Because older patients have special problems and may require special measures to achieve acceptable levels of safety in health care, there is concern that they may not fully benefit from the safety improvements being designed.

Purpose

This research was initiated to identify the nature and extent of preventable medical injury among patients age 65 and older, to determine how and why their patterns of injury differ from those of younger patients, and to suggest some ways that iatrogenic injury among older patients can be addressed.

Methodology

This report synthesizes what is known about the extent and causes of iatrogenic injury in adults age 65 and older. It is based on a review of the literature and analysis of unpublished data from several databases. In addition, the Harvard Medical Practice database (previously published and referenced elsewhere in this report) was used for subanalysis within the older population.

Results

1. The Extent of Medical Injury in Older Patients

Medical injuries are unexpected and unfortunate consequences of well-intentioned health care. All ages are at risk, but older patients are at particular risk because they have more severe illnesses and more illnesses simultaneously than younger populations do. The extent of iatrogenic medical injury in patients age 65 and older is substantial. At least 6 percent of hospitalized patients in that age group suffer an adverse event, defined as a treatment-caused injury that is serious enough to result in a measurable disability or to prolong a hospital stay. That rate is substantially higher than for younger age groups.

For example, in the Harvard Medical Practice Study, patients age 65 and older suffered twice as many diagnostic mishaps, four times as many therapeutic mishaps, two and a half times as many drug complications, and nine times as many falls as those under age 65. The risk of accidental injury increases with advancing age, particularly for falls and surgical complications. Two-thirds of iatrogenic injuries are due to errors and are, therefore, potentially preventable.

The reported iatrogenic injury rate among nursing home residents is staggering: several traumatic injuries per person per year in one comprehensive study, more than half due to falls. One-quarter of nursing home residents also suffer a medication complication each year. Approximately 2 million older Americans are patients in nursing homes. Estimates are that, at any time, 200,000 of these patients are recovering from an injury caused by a fall that occurred in the nursing home. Moreover, 160,000 have a urinary tract infection, and 200,000 have pressure sores.

The extent of injury among older patients outside of the hospital or nursing home has not been well studied, but the information available suggests that the risk of falls and medication complications among confined older patients receiving home care is at least as great as among those in nursing homes.

Older patients are particularly susceptible to several forms of iatrogenic complications: adverse drug events, falls, nosocomial infections, pressure sores, delirium, and surgical complications.

Adverse drug events (ADEs) are the most common type of treatment-caused injury in hospitalized patients, including patients age 65 or greater. In addition, ADEs are frequent complications among older persons in nursing homes and the ambulatory setting. In teaching hospitals, 6-7 percent of all patients suffer ADEs; the rate is higher still in older patients who are sicker and receive more drugs. The rate is probably somewhat lower in community hospitals, where fewer medications are used.

Approximately one-third of ADEs in hospitals are due to errors and are, therefore, preventable. The remaining two-thirds are adverse drug reactions (ADR), unpreventable, and often unpredictable, side-effects that occur even when a drug is used properly.

ADEs are a particular problem among older patients in nursing homes. One prospective study showed that in a four-year period, two-thirds of nursing home residents suffered an ADE. They are also common in the noninstitutionalized older population, primarily as a result of inappropriate prescribing. Surveys have shown that a substantial percentage of older patients receive at least one inappropriate drug. These patients are also at increased risk because they take a large number of medications.

A study of older veterans showed that they were taking an average of eight drugs each, and 35 percent reported having had a drug reaction in the prior year--28 percent of which required an emergency room visit or hospitalization. Overall, untoward reactions to medications account for 15 percent of admissions to the hospital for patients over 60 (compared to 6 percent for younger patients).

Falls are a major public health problem among older persons. It has been estimated that, annually, falls are responsible for over 2,000,000 injuries, 369,000 admissions to hospital, and 9,000 deaths, at a cost of $8 billion a year. The risk of falling and the likelihood that the resultant injuries (mostly hip fractures and brain injuries) will result in death increase substantially with advancing age. Over 250,000 hip fractures occur annually, almost all in older persons. In the health care setting, falls are most common in nursing homes; they occur less frequently in hospitalized patients. On average, half of older patients residing in nursing homes suffer falls each year.

Nosocomial infections occur in 6-17 percent of hospitalized patients, and may be equally frequent among nursing home residents. Like other iatrogenic complications, the risk of infection increases sharply with advancing age. Among nursing home residents, pneumonia is the most common form of infection and most likely to be fatal.

Pressure sores are a preventable form of iatrogenic injury that occurs in approximately 1.7 million patients each year. The vast majority develop in older patients in nursing homes. Up to 20 percent of nursing home patients have been found to have pressure sores.

Delirium is a common affliction of older patients, complicating the course of 2.3 million hospitalized patients annually, at a cost of $4 billion. The underlying causes are related to aging of the brain and are unpreventable, but surgery and drug therapy are frequently precipitating events.

Surgical complications are twice as likely to occur in patients over the age of 65 as in younger patients, and the rates of complications and death increase sharply with age. Surgery is riskier in older patients because they have reduced physiologic reserve and are more likely to have multiple medical conditions. Patients age 65 and older account for half of all surgical emergencies and three-fourths of operative deaths.

2. Why Older Patients Are More at Risk of Injury

The greater risk of harm to older patients from medical interventions results both from increased exposure to opportunities for medical mistakes and from the likelihood that those mistakes will then lead to actual injury. Older patients are more susceptible to injury because of reduced compensatory mechanisms related to declining organ function. Functional decline (confusion, incontinence, loss of appetite, and tendency to fall) is particularly likely to occur in older patients when they are hospitalized. These problems lead to additional treatments with their attendant risk of complications ("cascade iatrogenesis"). Patients who are already impaired (such as those who are unable to walk or take care of themselves) are at special risk.

Unfortunately, hospitalization itself--involving a change in environment and routines, often accompanied by enforced bed rest--can be a major cause of functional decline and lead to a progressive downhill course in patients who were previously functioning well. Physician errors that are well tolerated in younger patients can be devastating in older ones. For example, excessive fluid administration that would be relatively harmless in a young patient can lead to congestive heart failure in an older person with reduced cardiac reserve. In addition, dehydration or inappropriate drug usage that would be tolerated in the young can lead to renal failure in older patients with marginal kidney function.

Doctors are also more likely to miss diagnoses in older patients. The reasons are multiple and complex. Patients may give them inadequate clues, for example. Older patients are more likely to deny symptoms, in part because they have become accustomed to living with aches and pains of one sort or another, or because they interpret a new symptom as part of a chronic disorder they have. Older patients are also less likely than their children to be aggressive or assertive with doctors, often accepting inadequate explanations and not voicing concerns.

But most missed diagnoses result from physicians' lack of sufficient training or experience to recognize that older patients often present with symptoms (e.g., delirium or fainting caused by an infection such as pneumonia) that are not directly related to a new disease process. Atypical presentation of diseases is a major pitfall for the nongeriatric specialist. In addition, some physicians have biases against older patients and conflicting attitudes about dying that compromise their objectivity.

Finally, the recent pressure by managed care organizations to shorten hospital stays has led to more care of the chronically ill being given outside the hospital setting, where supervision and expertise of caregivers are greatly reduced. This is particularly the case in nursing homes, where inexperienced caregivers confuse symptoms of disease or complications with the normal course of patients in decline.

Older persons are more susceptible to adverse drug events because they absorb and metabolize drugs differently than younger patients do, and they have more diseases requiring treatment, resulting in multiple drug use. In addition, they face the challenge of complying with complex medication dosing schedules. Also contributing to adverse events from medication is inadequate caregiver-to-patient instruction about how to take each medication and what side effects to look for-plus the fact that patients may get prescriptions from multiple physicians who may not be aware of other medications the patient is taking.

The causes of falls among older persons are complex. Progressive instability and difficulty in walking is a common affliction of aging, and complete protection against falling is impossible, short of constant supervision or restraints. However, patients with dementia or depression, those on psychotropic drugs, those with visual or neurologic impairment, and those with mobility problems or a history of falling are at special risk and require special attention. Use of restraints is rarely a needed option, however, and the evidence is that they probably increase the risk of serious injury. Recent episodes of tragedies associated with unsupervised use of restraints have highlighted the hazard and led to calls for stricter regulation.

Nosocomial infections are a particular hazard for older patients. Pneumonia is disturbingly common in both hospitalized and nursing home patients. Decreased lung capacity, impaired cough reflex, and declining immunity all contribute to this increased susceptibility to pneumonia. Pulmonary aspiration is a major precipitating event, particularly in post-operative patients or those who are neurologically impaired. The use of urinary catheters is also a significant cause of infection. Both of these complications are common sequelae of prolonged hospitalization and multiple treatments.

The development of pressure sores is a potential hazard for anyone who is bedridden or chairbound, as many older hospitalized or nursing home patients are. A host of factors add to the risk: neurologic injury, malnutrition, fecal incontinence, orthopedic injuries, hypoalbuminemia, to mention a few.

Delirium is another iatrogenic complication with many causes, including infections, metabolic imbalances, alcohol withdrawal, insufficient social support, sleep deprivation, unfamiliar surroundings, pain, and isolation. A major problem is that doctors and nurses frequently fail to diagnose or misdiagnose delirium, mistakenly attributing mental changes to aging or dementia. As many as 25% of older patients experience delirium following surgery. Patients undergoing coronary-artery bypass surgery are particularly at risk. Inadequate control of pain in the post-operative period is a significant cause of delirium.

Age alone is not a risk factor for post-operative complications, but older patients tend to have reduced organ function and multiple chronic diseases that increase the risk of iatrogenic injury. In addition, the hazards of bed rest, blood clots, and nosocomial infections are real and present dangers for post-operative patients. Unfortunately, fear of increased risk of complications may also lead to risky behavior by the surgeon, who may delay surgery until it becomes necessary as an emergency, at which point the patient's reduced reserve and the ravages of the disease have been shown to increase the mortality risk as much as tenfold.

3. What Can Be Done to Reduce Accidental Medical Injury in Older Patients

Interest in improving patient safety has accelerated significantly in the past few years. The American Medical Association (AMA) founded the National Patient Safety Foundation, major private and governmental health organizations have formed the National Patient Safety Partnership, and national advisory boards have called for an increased emphasis on prevention of medical errors.

All of these efforts will help older patients as well as others. Four strategies hold special promise for reducing the risks associated with health care for older patients: (1) application of lessons in error prevention from other industries, (2) reducing variability in medical care, (3) enhancing the roles of geriatric specialists, and (4) using risk profiling and discharge planning.

(1) Application of lessons in error prevention from other industries. Design for safety using human factors principles such as standardization, simplification, improved information access, appropriate automation, and training all workers to work in teams. These principles will work in health care as well.

(2) Reducing variability in medical care. Studies show substantial variation from hospital to hospital and from doctor to doctor in the application of life-saving treatments. Eliminating this variation through the dissemination of guidelines and the enforcement of standards in health care organizations will enhance safety.

(3) Enhancing the roles of geriatric specialists. Not surprisingly, care of older patients is better when provided by those who specialize in their care (geriatricians). Enhancing their role and developing multidisciplinary teams of geriatric specialists should be an agenda item for every hospital.

(4) Using risk profiling and discharge planning. Assessing and identifying older patients at risk for iatrogenic complications when they are admitted to the hospital is the first step in a comprehensive injury prevention program. Similarly, adequate assessment and discharge planning dramatically reduce the risk of post-hospitalization complications. These principles need to be applied in the outpatient setting as well.

For the six areas where older patients are at special risk, a host of interventions have been developed that, if applied throughout the health care industry, could have a major impact on the extent of accidental medical injury.

Reducing adverse drug events requires a multipronged approach. Several are related to better education of physicians who care for older patients. First, doctors need to be much better educated about drugs that carry special hazard when used in older patients. Fortunately, specific criteria have been developed to help physicians avoid inappropriate prescribing. The task is to get these criteria disseminated and used. Second, physicians need to be trained to better recognize adverse drug events so they do not treat them as "new illnesses." Third, physicians and others need to learn about and use effective nonpharmacologic alternatives for such conditions as anxiety and sleep disorders.

Fourth, pharmacists need to be much more widely involved in medication management for patients in all care settings. Fifth, regulators and private organizations can be major forces for change. The FDA is beginning to implement geriatric drug labeling that will provide critically needed information on how certain drugs should be used by older patients. The Institute for Healthcare Improvement (IHI) has launched a series of collaborative efforts in which IHI staff work with hospital teams to redesign their medication systems for safety.

Finally, information technology holds great promise for reducing medication errors. Computerized physician order entry not only eliminates paper prescriptions, but is powerfully effective at intercepting and preventing errors at the time of ordering. Its universal application could have a major impact on medication errors. "On-line" medication guidelines and information from real-time analysis of laboratory and drug data can also help physicians use medications correctly.

Falls can be prevented by implementing comprehensive programs that identify patients at risk and provide adequate safeguards. It has been shown that, when personnel are appropriately trained, the use of physical or chemical restraints can be dramatically reduced without an increase in falls.

The prevention of nosocomial infections requires more rigorous application of principles of infection control that have been well understood and taught for at least 50 years. In addition, efforts to avoid use of invasive treatments, such as endotracheal tubes and urinary catheters, should be encouraged. Similarly, the prevention of pressure sores depends on the application of long-established principles of nursing care, such as frequent turning of patients who are identified as being at risk. The ability to give such intensive "hands on" care is related to nursing-patient staffing ratios. A rising rate of pressure sores might well be an indicator of excessive staff reductions resulting from cost-cutting programs.

As with pressure sores, prevention of delirium begins with identifying patients at high risk and instituting prophylactic measures promptly. Effective pain control is an important component of prevention. Prevention of post-operative complications, perhaps more than with any other type of iatrogenic injury, requires a true multidisciplinary effort in which surgeon, geriatric specialist, and anesthesiologist work together to insure optimal preoperative preparation and attentive post-operative care. If such partnerships could be established early in the course of a surgical problem for each patient, they could help eliminate the most serious surgical risk of all: delay of needed surgery until it becomes an emergency.

Finally, we have much to learn from research about how to improve safety in caring for older patients. The development of more extensive longitudinal databases and analysis of outcomes can provide rich opportunities to better understand the limits, opportunities, and challenges in such care. Particular attention needs to be given to determining the effect of specialized training and care, staffing levels, and effective means of providing safe care outside the hospital setting. And doctors, nurses and patients will all profit from clearer information on the risks and probability of benefits for all treatments.

Conclusions

The risk of accidental injury is unacceptably high for all who undergo medical treatment but it is especially so for older patients. Some of these risks are unavoidable, the consequences of normal aging; others are brought on by the treatments themselves. In addition, many diseases are more prevalent in older patients, further reducing organ function and increasing susceptibility to injury. Finally, decreasing motor activity and dependency in the very old lead to complications such as pressure sores and falls.

However, the risk of complications from treatments is higher for older patients in large measure because they receive so many more of them. This is particularly true of medications. Specific treatments, such as surgical operations, have not been found to be intrinsically more hazardous in older patients, but they take on increased risk to the extent that patients have reduced organ function or coexistent ("comorbid") disease, or if surgery is delayed until it becomes an emergency.

The risks of medical treatment could be greatly reduced by wider application of principles and techniques that have been learned in recent years from geriatric research. The sad fact is that too many older patients are cared for by doctors and nurses with no geriatric training. This is true for patients in the hospital as well as at home or in nursing homes. With careful and expert care, older patients can weather many a medical storm. Providing that care is a major challenge to our medical care system.

Written by Jeffrey M. Rothschild, MD, Harvard School of Medicine and
Lucian L. Leape, MD, MPH, Harvard School of Public Health
Andrew H. Smith, Project Manager
September 2000

� U. S. Health Care Reform. All Rights Reserved.