ACP medicine, 3rd Edition

Clinical Essentials

Reducing Risk of Injury and Disease

Harold C. Sox Jr. M.D., M.A.C.P.1

1Editor, Annals of Internal Medicine

The author has no commercial relationships with manufacturers of products or providers of services discussed in this subsection.

July 2003

Prevention: A Brief Overview

During the past 2 decades, disease and injury prevention has occupied an expanding share of medical practice. Public interest in prevention is very high, driven by a steady accumulation of high-quality evidence that preventive interventions do reduce cause-specific death rates. The purpose of these interventions is to eliminate the root causes of diseases that precede death (e.g., heart disease, cancer, and stroke), which in the United States in 1990 were tobacco use (400,000 deaths), poor diet and inadequate physical activity (300,000 deaths), alcohol consumption (100,000 deaths), microbial agents (90,000 deaths), toxic agents (60,000 deaths), firearms (35,000 deaths), unprotected sexual intercourse (30,000 deaths), motor vehicle accidents (25,000 deaths), and use of illicit drugs (20,000 deaths).1 These causes of death are the targets of disease and injury prevention. They contribute to 50% of the deaths in the United States. Most are simply bad habits, and changing those habits reduces the risk of dying.

Physicians have two principal roles in prevention: they identify risk factors for disease and injury, and they act as teachers and counselors. Physicians must expand their routine questioning beyond diet, exercise, and substance abuse to include recreational activities that increase the risk of death (e.g., boating, bicycling, and riding motorcycles), gun ownership, use of swimming pools, smoke detectors in the home, and domestic violence. In counseling patients about a healthy diet (including vitamin supplements), exercise, and other elements of a healthy lifestyle, physicians must often help patients adopt healthy living habits. Some patients simply require reinforcement of a chosen lifestyle. Others need help in changing harmful habitual behaviors to healthy behaviors.

The report of the second United States Preventive Services Task Force (USPSTF)2 contains evidence-based guidelines on 70 prevention topics [see Table 1]. As they become available, the third USPSTF reports will appear on its Web site ( Other literature is also helpful [see CE:IV Diet and Exercise, CE:V Adult Preventive Health Care,7:XXXIII HIV and AIDS, 13:III Alcohol Abuse and Dependency, and 13:VI Drug Abuse and Addiction].

Table 1 Recommendations of the United States Preventive Services Task Force2

Tobacco Use

Provide tobacco cessation counseling to patients who use tobacco products. Counsel pregnant women and parents about the potentially harmful effects of smoking on fetal and child health. Prescribe nicotine patches or gum to selected patients as an adjunct to counseling. Give antitobacco messages to young people as part of health promotion counseling.

Alcohol Abuse

Screen all adults and adolescents for problem drinking, using a careful history or a standardized screening questionnaire.

  Advise pregnant women to limit or abstain from drinking.

  Counsel all persons who use alcohol about the dangers of operating a motor vehicle or engaging in other potentially dangerous activities while drinking.

Drug Abuse*

Although the evidence is insufficient for a strong recommendation to be made, it is reasonable to ask adolescents and adults about drug use and drug-related problems. All pregnant women should be advised of the potential adverse effects of drug use on fetal development. Refer drug-abusing patients to specialized treatment facilities where available.

Preventing Motor Vehicle Injuries

Counsel all patients and the parents of young people to use occupant restraints (lab/shoulder safety belts and child safety seats), to wear helmets when riding motorcycles, and to refrain from driving while under the influence of alcohol or other drugs.

Preventing Falls

Counsel elderly patients on specific measures to reduce the risk of falls. Effective measures include exercise, balance training, environmental hazard reduction, and monitoring and adjusting medications. Provide multifactorial individualized interventions to elderly patients at especially high risk for falls.


Advise homeowners to install smoke detection devices and test them periodically. Infants and children should wear flame-resistant nightclothes. Smokers should cease or reduce smoking.


Families with swimming pools should install four-sided 4-ft isolation fences with self-latching gates.

Firearm Injuries

Remove firearms from the home or store them unloaded in a locked compartment.

*There is insufficient evidence to recommend routine screening for drug abuse with standardized questionnaires or biologic assays.
There is insufficient evidence to recommend for or against counseling patients to avoid pedestrian injuries.
There is insufficient evidence of effectiveness of external hip protectors.


Although disease and injury prevention can have a significant effect on the health of the public, physicians should observe several caveats. First, the baseline risk of most diseases is very low in the average person. Each year, colon cancer occurs in 165 per 100,000 men 60 to 64 years of age. The low baseline risk means that the number needed to screen or treat to prevent one death is often very high. Annual fecal occult blood testing must be performed on more than 300 people for 12 years to prevent one death from colon cancer. Whether this inefficiency is important depends partly on the cost of the intervention. Fecal occult blood testing can be costly because it must be performed annually and because abnormal results trigger costly diagnostic tests. Seat belts and smoke alarms are very cost-effective because they incur a onetime cost.

Second, disease prevention does not prevent death. At best, it postpones death by shifting the cause of death from the targeted disease to another disease that strikes later in life. In the Minnesota Colon Cancer Control Study, a randomized trial of fecal occult blood testing, annual testing reduced deaths from colon cancer during 18 years of surveillance. However, the total mortality was the same in the control group and the intervention groups. Our preventive efforts may reduce the likelihood of death from the target disease, whose identity and natural history we know. Inevitably, we raise the lifetime probability of dying from another disease.

Third, we know little about the age at which we should stop our efforts to prevent disease. Our studies provide good information on effectiveness in the study population, which is usually in middle age, but we do not know how the results apply to older people, whose care will occupy an increasing amount of the primary care physician's time.3 Interventions that take years to show their impact may be ill suited to people whose life expectancy is measured in years rather than decades.

The decision to do a screening test on an older person should depend on the person's general health, which may be quantified as the person's physiologic age. The physician can determine a patient's physiologic age by asking the patient to rate his or her health as excellent, good, fair, or poor.3 The most likely age at death is the sum of the patient's actual age and the life expectancy that corresponds to the patient's physiologic age. For example, a 75-year-old man in excellent health has a physiologic age of 67 years, which corresponds to a 14.7-year life expectancy [see Figure 1]. The most likely age at death is 75 years plus 14.7 years, or 90 years. This information can be very helpful in deciding how hard to press preventive efforts. With a life expectancy of almost 15 years, a 75-year-old man has plenty of time in which to experience gains from preventive efforts.


Figure 1. Life expectancy in the U.S.

Life expectancy of men and women in the United States.


Many interventions of proven efficacy are not completely effective because patients are reluctant to change long-established risky behaviors. The USPSTF recommends the following steps for helping patients use their ability to change (self-efficacy)4:

  1. Match teaching to the patient. Identify a patient's beliefs about a behavior, and adjust advice to the patient's lifestyle. Building the patient's confidence in his or her ability to change requires recognizable successes; define success in terms of goals the patient can achieve.
  2. Tell why, what, and when. Patients need to know the reason for a recommendation and the results of following the recommendation. They must also know the time scale for the results so that they do not become discouraged when results do not occur immediately.
  3. Small changes succeed. As the patient achieves small successes, propose larger but achievable goals.
  4. Be specific. Couch suggestions in terms of current behavior, and give precise instructions in writing.
  5. Add new behaviors. Adopting good habits is often easier than discarding bad ones.
  6. Link positive behaviors with the daily routine. For example, patients can be encouraged to exercise before lunch or to take medication immediately after brushing teeth.
  7. Do not mince words. Tell the patient directly, simply, and specifically what you want and why.
  8. Extract promises. Get explicit commitments from the patient. Have the patient tell you exactly how he or she will achieve a goal. Assess the patient's self-confidence and address concerns about succeeding.
  9. Use combination strategies. An approach that combines several strategies is more likely to succeed than a single strategy.
  10. Involve others. Members of the physician's office staff can become educators. Anyone can offer encouragement to patients.
  11. Refer. Subspecialists in many chronic diseases have trained teams that can educate patients far more effectively than individual physicians can. Another form of referral is sending novice patients to talk with successful patients.
  12. Stay interested. According to research findings, a call from a health professional to inquire about progress is very effective in changing a behavior. A well-organized office will have a protocol for making these calls a matter of routine.

Health Risks from Substance Abuse

Substance abuse exacts a large toll on the health of the American people, accounting for at least 520,000 deaths in 1990, or 24% of all deaths.1 Unfortunately, many physicians do not place diseases resulting from self-abuse in the same category as diseases that strike seemingly by chance. Two factors make helping patients shed their habitual use of tobacco, alcohol, and illicit drugs a very efficient way for physicians to add healthy years to patients' lives. First, there is a high baseline risk that substance abuse will lead to serious disease. Therefore, the absolute reduction in risk from a successful intervention is high. This principle is especially true of substance abuse during pregnancy. Second, substance abuse is primarily a problem of youth and middle age, so a successful intervention can add many years of healthy life.


Tobacco contains an addictive drug, nicotine, as well as other substances that contribute to death from cardiovascular disease, cancer, and chronic lung disease. Smoking also contributes to 10% of infant deaths and 20% to 30% of low-birth-weight infants.5 Tobacco use contributed to one in every five deaths in the United States in 1990 (420,000 deaths a year). In one 40-year cohort study of male physicians in the United Kingdom, half of the deaths after age 35 were smoking related, and smoking caused 25% of the residential fires that resulted in death.5 Tobacco use is less common than it was several decades ago, but 25% of adults smoke, and an increasing number of smokers are women. Women are starting to smoke earlier—many as high-school students—and they are heavier smokers; nearly twice as many women smoked at least 25 cigarettes a day in 1985 as in 1965.6

Cigarettes are highly addictive. Fewer than 10% of people who quit smoking for a day are still abstinent 1 year later. Nicotine, like other highly addictive substances, acts on the dopaminergic mesolimbic pathway, the brain reward pathway that controls motivated behaviors [see13:VI Drug Abuse and Addiction]. The use of nicotine is self-reinforcing, leading to compulsive use. Nicotine produces a withdrawal syndrome that begins within a few hours of abstinence, peaks within the first week, and continues for several weeks. The withdrawal syndrome includes dysphoria, insomnia, irritability, anxiety, difficulty in concentrating, restlessness, slowed heart rate, and increased appetite.7

Detection of cigarette smoking is easy; most smokers are truthful when asked about their habit and its extent, and the odor of tobacco is an additional diagnostic clue.

Smoking cessation reduces mortality dramatically. The risk of some diseases (e.g., myocardial infarction and stroke) declines rapidly within a few years after quitting [see Table 2].8 This information is important when one is trying to convince long-term smokers to quit.

Table 2 Years of Smoking Abstinence Needed to Reduce Risk of Disease2,8


Years until Risk Is Half of a Current Smoker's Risk

Years until Risk Is Equal to a Never-smoker's Risk

Recurrent myocardial infarction or death from coronary artery disease






Oral and esophageal cancer


Lung cancer



Research has shown that a strong message from a personal physician is the most important factor in successful quitting. The elements of successful quitting are consistent, repeated, and strong advice to stop smoking; setting a specific quit date; and follow-up visits to reinforce behavior [see Table 3]. However, not all physicians counsel smokers to quit. In one study, only 78% of cigarette smokers reported that their physician had advised them to quit.9

Table 3 Elements of a Successful Smoking Cessation Strategy

Direct, face-to-face advice and suggestions

Reinforcement, especially in first 2 weeks

Office reminders: a sticker on chart of smokers may stimulate physician to deliver antitobacco message at each visit

Self-help materials

Community programs for additional help

Drug therapy

A current theory in behavioral psychology suggests that changes in behavior reflect predictable stages in the readiness to change, ranging from no intention to change, to definite plans to change in the near future, to active attempts at change. A stage of readiness to change is predictive of quitting,10 and this behavioral model may help clinicians shape smoking-cessation efforts to the patient's state of mind.

School-based prevention has received extensive study, and it is effective for at least 2 to 4 years. Clinicians, especially those caring for adolescents, must reinforce the messages of school-based programs.

Nicotine products are an important adjunct to counseling.11 Drugs are most reinforcing when the level in the brain rises very rapidly, as with inhaled nicotine. Nicotine in medication form, especially transdermal products, appears in the blood much more slowly than inhaled nicotine and produces much less of the reinforcing effect that leads to craving for cigarettes. Plasma nicotine concentrations after transdermal administration reach stable levels in 2 to 3 days. Nicotine medications reduce the symptoms of withdrawal, so that symptoms in the first week are reduced to the level of symptoms at 5 to 10 weeks. Nicotine medications may also provide some nicotinelike effects, such as helping patients to sustain concentration and deal with stress.

Nicotine medications improve abstinence rates, but abstinence at 1 year is still the exception. A meta-analysis summarized the results of 46 trials of nicotine gum and 20 trials of nicotine patches.12 At 12 months, 19% of patients who received nicotine gum were abstinent, as compared with 11% of patients who did not receive gum. The number needed to treat to achieve one success at 1 year (NNT) was 17 (P < 0.001). Transdermal nicotine led to similar rates. At 12 months, 16% of patients who received a transdermal nicotine patch had quit smoking, as compared with 9% who did not receive the patch (NNT, 16). Clonidine also increases the rate of abstinence at 12 months. Weight gain is a common occurrence in patients who have stopped smoking. The average weight gain in a national sample of adults who had stopped smoking was 4.4 kg for men and 5.0 kg for women.13

The dose of nicotine medications should depend on the degree of nicotine dependence.11 The score on the Fagerstrom questionnaire14 and the number of cigarettes a day are measures of dependence. Follow-up calls at prearranged times will help the patient to maintain abstinence. It is best to designate a specific member of the physician's office staff to be the smoking-cessation coordinator.

The starting dose of nicotine gum (nicotine polacrilex) is 2 mg per two cigarettes; in patients who smoke more than 20 cigarettes a day, the dose should be 4 mg per three or four cigarettes. Patients may take additional unit doses if their withdrawal symptoms are unpleasant. The medication should be taken at regular intervals throughout the waking hours. The patient should compress the gum a few times with the teeth and then hold it in the mouth, repeating the cycle every minute or so for 15 to 30 minutes for each dose. After 1 to 2 months, weaning can begin with a reduction of 1 unit dose a week.

With transdermal nicotine, patients who smoke more than 10 cigarettes a day should use the largest patch (21 mg). After 1 to 2 months, weaning can begin with each of the lower doses (usually 14 mg and 7 mg, respectively), prescribed for 2 to 4 weeks. Patients who smoke fewer than 10 cigarettes a day can start with the 14 mg dose. A hairless site allows the best absorption. The patient should rotate sites to avoid skin irritation.

Nicotine medications are quite safe11—certainly safer than cigarette smoking—even for patients with cardiovascular disease. The only contraindication is hypersensitivity to nicotine or to a component of the delivery system. Twenty-four-hour application of transdermal medication can result in sleep disturbance, which subsides if the medication is removed before sleep. Nicotine medication during pregnancy is of concern but probably of less concern than heavy smoking during pregnancy. Medication during pregnancy should be reserved for women who have failed to quit without medication and who smoke more than 10 to 15 cigarettes a day. Dependence on nicotine medications is most likely with delivery systems such as a nasal spray, which causes a rapid rise in the plasma nicotine concentration, and a small number of patients will still be using nicotine medication 1 year after starting treatment.

Nicotine therapy is not the only pharmacologic approach to smoking cessation. Another approach focuses on dopamine. Nicotine releases norepinephrine in the brain and increases dopamine in areas of the brain associated with reinforcing the effects of addictive substances, such as opioids. Bupropion potentiates the effect of norepinephrine and dopamine by acting as a weak inhibitor of their neuronal uptake. Thus, bupropion can mimic some of the central nervous system effects of nicotine and act as a substitute for nicotine in people who are trying to quit cigarettes. Randomized clinical trials have shown that abstinence rates are approximately twice as high with bupropion as with placebo.15 In one trial, for example, the rates of abstinence from tobacco at 1 year were 12% for those taking placebo and 23% for those taking 150 or 300 mg of bupropion.16 Side effects of bupropion include agitation and insomnia. Seizures are very uncommon when the daily dose of bupropion is 300 mg or less.

The recommended dose of bupropion is 150 mg/day for the first 3 days and then 150 mg twice daily. The patient should wait to stop smoking until he or she has been on bupropion for 1 week. There are no peer-reviewed reports comparing nicotine-replacement products with bupropion or detailing possible synergy between the two drugs.


Habitual excessive alcohol consumption causes 100,000 deaths annually in the United States.1 Although more than one million adults are under treatment for alcoholism, a far greater number engage in drinking that injures their health or has social consequences. The Institute of Medicine estimates that 20% of the population of the United States are problem drinkers, but only 5% are alcohol dependent.17Therefore, it is important to distinguish alcohol dependence from problem drinking. Alcohol dependence is associated with major withdrawal symptoms, tolerance, complete loss of self-control, and preoccupation with drinking. Problem drinking, on the other hand, is a less severe condition. Problem drinkers are younger, have a shorter drinking history, have fewer alcohol-related job problems, and have better social resources. In community surveys, the prevalence of problem drinking has been shown to be highest in young men (17% to 24% in 18- to 29-year-olds) and lowest in men and women older than 65 years (1% to 3% and less than 1%, respectively). Women are more frequently problem drinkers than alcohol dependent.

Problem drinking has consequences that affect others, such as motor vehicle accidents, fetal-alcohol syndrome, unsafe sex, domestic violence, and psychological damage to children of problem drinkers. Binge drinking, which is especially prevalent in young adults, leads to violence, unsafe sex, and drunk driving.

Screening for problem drinking and alcohol dependence can be time consuming, and most methods are inaccurate. The gold-standard test for alcoholism is the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) criteria, which require a detailed interview and do not constitute a suitable screening instrument. Results of physical examination and laboratory tests are often normal in problem drinkers. Screening questionnaires such as the modified Michigan Alcoholism Screening Test (MAST), the Alcohol Use Disorders Identification Test (AUDIT), and the CAGE test are the most accurate instruments for detecting problem drinking2 [see Tables 4 and 5] and 13:III Alcohol Abuse and Dependency]. The questions in the MAST and CAGE instruments focus on alcohol dependence and are much less sensitive or specific for binge drinking. The AUDIT screening instrument2 may be more generally useful because it also asks about quantity of alcohol imbibed, frequency of drinking, and binge behavior. The CAGE and MAST questionnaires may also fail to detect a level of alcohol use that is dangerous during pregnancy.

Table 4 Test Performance of Screening Questionnaires for Alcohol Abuse2

Screening Instrument

Number of Items

Sensitivity (%)

Specificity (%)

Likelihood Ratio Positive

Likelihood Ratio Negative

























*In inner-city clinic population.
In rural clinic.
AUDIT—Alcoholism Use Disorders Identification Test   CAGE—see Table 5   MAST—Michigan Alcoholism Screening Test

Table 5 The CAGE Questionnaire

C: Have you ever felt you ought to Cut down on drinking?

A: Have people Annoyed you by criticizing your drinking?

G: Have you ever felt bad or Guilty about your drinking?

E: Have you ever had a drink in the morning to steady your nerves or get rid of a hangover (Eye-opener)?

The treatment of problem drinking depends on the severity of the problem. Problem drinkers often respond to brief office interventions (as short as 10 or 15 minutes), which use motivational techniques such as goal setting, contracts, and enhancing self-efficacy. In most instances, the goal is usually controlled moderate drinking rather than abstinence. The first step toward successful counseling is to get the patient to recognize that there is a problem. It is important to help the patient see a relationship between drinking and current medical or psychosocial problems. Strong advice to reduce consumption is also important. Regular follow-up visits to monitor progress are just as important for problem drinkers as they are for patients with high blood pressure.18 One meta-analysis of brief intervention trials for nondependent problem drinkers showed a reduction of 24% in average alcohol consumption.19 Another meta-analysis of nine studies showed somewhat smaller effects. In five of nine studies in men, the number of drinks a week decreased (range of decrease, five to 20 a week). The effects in women were smaller.20 A third meta-analysis of nine randomized trials found that brief interventions were associated with alcohol moderation 6 to 12 months later (pooled odds ratio, 1.91; 95% confidence interval = 1.61 to 2.27).21 An excellent guide to managing problem drinking is available from the National Institute on Alcohol Abuse and Alcoholism (

In contrast to the success of brief office-based interventions for problem drinking, successful treatment of alcohol dependence requires intensive therapy from specialists in substance abuse. A randomized trial of employees with alcohol dependence showed the importance of intervening intensively. Participants were randomly allocated to compulsory 3-week hospitalization followed by 1 year of attendance at Alcoholics Anonymous (AA) meetings, mandatory attendance at AA meetings at least three times a week, or a choice of treatment. Rates of being fired from work were similar in all groups, but rates for hospitalization for additional alcohol treatment were much lower in the mandatory-hospitalization group.22

The personal physician does have an important role to play in the management of patients with alcohol dependence. In addition to managing medical complications of alcoholism, physicians should be able to use adjunctive therapy for alcohol dependence, such as naltrexone. Another key role for the personal physician is encouragement. Patients need to know that the abstinence rate can be as high as 60% at 10 years after intensive treatment. Finally, the personal physician can lead efforts to help patients solve life problems that are contributing to alcohol dependence.


The abuse of illicit and legal drugs is a large problem in the United States. A 2001 household survey showed that use of illicit drugs within the previous month peaked among 18 to 20 year olds, at 22.4%, and declined steadily with increasing age. Casual use of marijuana accounts for most of these reports, but as many as 1.7 million Americans use cocaine weekly and 130,000 use heroin. The drug abuser is at risk for many medical complications, but the social cost of drug abuse far outweighs the personal costs. Illicit drug use plays a major role in spreading HIV infection and in homicide, suicide, and motor vehicle accidents. The health care costs of drug abuse are estimated to be $3.2 billion annually, and the cost of federal and state government efforts to stem the flow of illicit drugs is several times higher [see 13:VI Drug Abuse and Addiction].

Many professional organizations recommend that physicians ask about drug abuse as part of a periodic health examination of a well person. However, learning about drug abuse may be difficult in the office setting. Patients may be unwilling to acknowledge drug abuse until presented with incontrovertible evidence or after persistent questioning by an alerted physician. There is little information about the accuracy of the history or questionnaires in detecting drug abuse.

Toxicologic testing is the best way to detect illicit drug use. Compared with reference tests, current tests can detect drugs in the urine with 99% sensitivity. However, detection depends on when the patient supplies the specimen relative to the last drug exposure. Marijuana is detectable up to 14 days after use, whereas cocaine, opiates, amphetamines, and barbiturates are present for only 2 to 4 days after use.

Although physicians sometimes test for illicit drugs without obtaining the patient's consent, they do so in the context of trying to determine the cause of a clinical problem that could be caused by an illicit drug. Whether it is ethical to test for illicit drugs in an apparently healthy person who is at high risk for drug abuse is an open question. Regardless of the circumstances leading to testing, abnormal results deserve the physician's best efforts to maintain confidentiality, because they may affect the patient's employability, insurability, and personal relationships. At present, no professional organization recommends drug testing in apparently healthy people.

Physicians must learn to think of drug abuse as a chronic disease. Recidivism after intensive treatment programs is very common, no doubt in part because psychiatric disorders, unemployment, and homelessness often coexist with drug abuse. On the other hand, treating heroin abuse with maintenance methadone, an opioid agonist, can dramatically reduce the social effects of abusing the drug. Heroin addicts in methadone maintenance programs report less use of heroin and reduced rates of HIV infection, criminal behavior, and unemployment. There is no similarly effective treatment for cocaine addiction. Changes in the law now encourage methadone maintenance in office practice. Physicians who wish to treat their regular patients will need to learn about the new regulations.23

Health Risks from Accidents and Violence

A person's environment contains many threats to health: motor vehicle accidents, accidents in the home, recreation-related accidents, and domestic violence. Passive strategies, which change the environment in which accidents can occur, are generally more successful at accident prevention than active strategies, which require people to change their behavior. Improving roads saves more lives than exhorting people to drive carefully. For a fuller account, refer to the report of the USPSTF2 and to a comprehensive review published in 1997.24


Motor vehicle accidents are the leading cause of loss of potential years of life before age 65. Alcohol-related accidents account for 44% of all motor vehicle deaths. One can experience a motor vehicle accident as an occupant, as a pedestrian, or as a bicycle or motorcycle rider.

Injuries to Motor Vehicle Occupants

In 2000, 37,409 people died of injuries sustained in motor vehicle accidents in the United States.25 The two greatest risk factors for death while one is driving a motor vehicle are driving while intoxicated and failing to use a seat belt. The physician's role is to identify patients with alcoholism [see Alcohol Abuse, above], to inquire about seat-belt use, and to counsel people to use seat belts and child car seats routinely. In one study, 53.5% of patients in a university internal medicine practice did not use seat belts. Problem drinking, physical inactivity, obesity, and low income were indicators of nonuse. The prevalence of nonuse was 91% in people with all four indicators and only 25% in those with no indicators.25 Seat belts confer considerable protection, yet in one survey, only 3.9% of university clinic patients reported that a physician had counseled them about using seat belts.25

Three-point restraints reduce the risk of death or serious injury by 45%.26 Air bags reduce the risk of death by an additional 6% in drivers using seat belts.27 Because air bags reduce the risk of death by only 14% in unbelted drivers,27 physicians must tell their patients not to rely on air bags.

Injuries to Motorcyclists

Motorcycle deaths in the United States have been rising since 1997, reaching 3,181 in 2001.28 The chance of death per mile when one is riding a motorcycle is 35 times higher than when one is riding in an automobile. Most deaths are caused by head injuries. Helmets reduce the risk of a fatal head injury by 27%, but only 50% of riders use helmets. Laws mandating helmet use are quite effective; the rate of helmet use rose to 95% in California after passage of a law, and the rate of head injuries dropped by 34%. Substance abuse is very common among injured motorcyclists.

Physicians should inquire about motorcycle use. They should redouble their efforts to screen for substance abuse in motorcyclists and should recommend using helmets.

Injuries to Pedestrians

Pedestrian injuries caused by motor vehicles accounted for 4,739 deaths in 2000, a 27% decrease from 1990. Children are at greatest risk for injury. Among adults, the elderly are at greatest risk, principally because of sensory deficits, locomotor disability, and inability to process simultaneous stimuli.

Injuries to Cyclists

Each year, there are approximately 900 deaths from bicycle injuries in the United States.29 Children are at greatest risk. Head injuries account for two thirds of hospitalizations and three quarters of deaths related to bicycling. A meta-analysis of case-control studies showed that use of safety helmets reduced the risk of head injuries by 63% to 88%.29 Helmets are effective for all ages and provide protection even in collisions with motor vehicles. Community-based education efforts have raised the rate of helmet use to 50%. Physicians should ask about bicycle use and counsel riders to use safety helmets.


Falling is a serious health risk for older persons [see Table 6]. The lifetime risk of hip fracture, perhaps the most important consequence of a fall, is 40% for a 50-year-old woman. One approach to reducing the risk of hip fracture is to prevent osteoporosis. Strategies for prevention of osteoporosis include vitamin D and supplementation of dietary calcium intake,30 drugs that increase bone mass, such as etidronate and alendronate, and estrogen replacement after menopause. In a cohort study, weight-bearing exercise, such as walking, was associated with a 40% lower risk of hip fracture in women and a 50% lower risk in men.31 Exercise works in part by increasing bone mass and in part by reducing the likelihood of a fall. Combined interventions that included home visits, modifying home hazards, and exercise and gait programs reduced the risk of falls by 31% in a randomized clinical trial.32 Physicians should identify patients who are at greatest risk for falling [see Table 6], treat osteoporosis, and link the patient to community-based programs for improving mobility and reducing hazards in the home. Hip pads are also effective; in a randomized trial, a pad worn over each hip reduced the risk of hip fracture from 46 per 1,000 patient-years to 21.3 per 1,000 patient-years (P = 0.008).33

Table 6 Risk Factors for Falls among the Elderly32

Prior falls

Low body mass index

Cognitive impairment

Female sex

Chronic illness

General frailty

Balance and gait impairment

Hazards in the home


Prevention of death from fires is an example of a successful passive strategy. Smoke detectors prevent fire injury. A study in Oklahoma City measured the effects of door-to-door distribution of smoke detectors to residents of an area that had much higher rates of burn injuries than the rest of the city. The fire-injury rate declined 80%, to the same level as that in the rest of the city. The injury rate per fire also declined dramatically.34 Physicians should inquire about smoke detectors in the home and recommend them to people who don't have them. Persons who are alcohol dependent or who smoke in bed are at high risk and need special effort.


In most instances of witnessed drowning, bystanders report that the victim becomes motionless while swimming or simply fails to surface after a dive. Struggle is unusual. This observation raises the possibility that many cases of drowning occur when something such as a seizure, an arrhythmia, or an injury occurs.35

All victims of immersion have hypoxemia. Aspirated freshwater is hypotonic and therefore rapidly absorbed by the pulmonary circulation and distributed throughout the body water compartment. Freshwater alters pulmonary surfactant and causes alveolar collapse and atelectasis. Saltwater is hypertonic and draws water into the alveoli, causing perfused but poorly ventilated alveoli and hypovolemia with concentration of electrolytes. The end result with both types of water is venous admixture and hypoxemia, often resulting in metabolic acidosis. Saltwater drowning often leads to hypovolemia as well.

The main goal of treatment is to prevent brain injury.35 The first step is to initiate cardiopulmonary resuscitation if the victim is apneic and pulseless. The American Heart Association recommends abdominal thrusts only to clear the airway in case of suspected foreign-body aspiration or failure to respond to artificial ventilation. Supplemental oxygen is indicated as long as the patient is hypoxemic. The most effective single treatment of hypoxemia is continuous positive airway pressure (CPAP), using mechanical ventilation to expand collapsed alveoli caused by freshwater immersion.35 Hypothermia, which often accompanies near-drowning, can protect the brain from injury by reducing its metabolic requirements when the patient is hypoxemic.

Most efforts to prevent drowning focus on children, for whom the passive strategy of requiring fencing around swimming pools is associated with reduced drowning rates. In adults, alcohol ingestion is a risk factor for drowning. The efficacy of personal flotation devices is not known. Relatively few boaters (14%) wear personal flotation devices, but this rate is similar to that of drowning victims. Physicians should ask patients whether they use boats recreationally and advise avoiding alcohol and using a personal flotation device while boating.


For women especially, the home is the most dangerous place. In one large study of women in a primary care clinic, one in 20 had experienced domestic violence in the previous year, one in four had experienced it as adults, and one in three had experienced it in their lifetime.36,37 A condition so prevalent in primary care practice demands the attention of the physician.

Among those abused in the previous year, approximately equal numbers had been abused once, two or three times, or four or more times. In this study, the definition of domestic violence was an affirmative answer to the question “Have you been hit, slapped, kicked, or otherwise physically hurt by someone?” or “Has anyone forced you to have sexual activities?” Generally, a husband, ex-husband, boyfriend, or relative is the abuser in domestic violence.

Most of the rapidly developing literature on domestic violence focuses on screening, diagnosis, and management, and there is little on preventing the first episode. Screening for domestic violence typically occurs in the office setting. Many authorities recommend that physicians routinely ask about domestic violence as part of the screening history.37,38 Domestic violence occurs in homosexual relationships as well, so it is best to ask both men and women. Some physicians introduce the question by saying that they are now asking all their patients about domestic abuse, in view of the growing awareness of the problem. Then they ask, “At any time [or since I last saw you] has your husband [lover, partner, boyfriend] hit, kicked, threatened, or otherwise frightened you?” If the patient replies in the affirmative, the physician should gather more information, including the name of the abuser, and record it. Because many people are ashamed of their situation and their inability to break out of it, the physician should avoid any judgmental statements other than to confirm that what is being done to the patient is wrong.

In many cases, patients will not disclose an abusive relationship but their medical and social history contains clues to the true situation. Somatic symptoms that are particularly indicative (prevalence ratio > 2.5) of an abusive relationship include multiple symptoms (especially with no apparent physical cause), poor appetite, nightmares, eating binges, pain in the pelvic region, vaginal discharge, musculoskeletal injuries, and diarrhea.36 Even more indicative are emotional symptoms such as high anxiety, severe depression, a high level of somatization, and low self-esteem; current or past use of street drugs; positive items on the CAGE questionnaire for alcohol abuse; a current or past drinking problem; a husband or partner who abuses alcohol or uses street drugs; a history of suicide attempt; and abuse as a child (prevalence ratio > 10.0 for all of these).36,39 Pregnancy is often associated with an escalation of violence.

Some abused patients will disclose a history of abuse, but many will not. Therefore, during the physical examination, the physician should be alert to signs of injury. One expert states that a woman who presents with any injury should be considered a victim of domestic abuse until proved otherwise.38 Trauma to the face, abdomen, breasts, or genitals is especially likely to be from domestic abuse, as are bilateral or multiple injuries, injuries in different stages of healing, and injuries that occurred well before the patient sought help. Injuries to the ulnar aspect of the elbows may occur as a woman raises her hands to protect herself during an assault.

Older persons are subject to several forms of abuse: self-neglect or caregiver neglect, emotional and psychological abuse, fiduciary exploitation, and physical abuse. Signs of elder abuse include bruising and other signs of trauma, malnutrition, volume depletion, and poor hygiene.

The physician should communicate concern and validate the patient's belief that domestic abuse is wrong. The physician should not only provide medical treatment of injuries but also talk with the patient about how to avoid serious injury during an assault, review the patient's options, and facilitate referral to community and other resources for abused partners [see Sidebar Domestic Violence Information on the Internet]

Domestic Violence Information on the Internet

Medical Resources

Family Peace Project

Domestic Violence: A Practical Approach for Clinicians

Legal Resources

Women's Law Initiative

American Bar Association Commission on Domestic Violence


The rate of death by firearms in the United States peaked at 39,595 in 1993, and it has since declined, reaching 28,663 in 2000.40,41Firearms accounted for 992 accidental deaths, 13,677 homicides, and 17,767 suicides in 1997. In two large communities, 58% of suicide victims used a firearm. Seventy percent of the suicides occurred at home. Firearms kill more teenagers than all natural causes of death combined. The Bureau of Alcohol, Tobacco and Firearms estimates that there are 192 million firearms in private hands. Firearms, often bought for protection in the home, are far more dangerous to the occupants than to an intruder. After controlling for other suicide risk factors, the odds of suicide are 1.9 times greater in homes in which there is at least one gun.42,43 The odds of homicide are 2.2 times higher in homes in which there is a firearm.44

Physicians strongly support regulation of firearms and community efforts to restrict ownership.45 Physicians also have a role to play in preventing injury from firearms.46 They should inquire about firearms in the home and counsel owners about storing their firearms in a safe place. With the increased number of teenagers who own guns and commit homicide with guns, the need to educate parents is urgent. The American Academy of Pediatrics has developed an information kit for physicians to use in counseling parents and children. The kits, called Steps To Prevent Firearm Injury In The Home, are available without charge from the Brady Center to Prevent Gun Violence, 1225 Eye Street NW, Suite 1100, Washington, DC 20005, or they can be obtained on the Internet, at


Figure 1 Marcia Kammerer. Adapted from Life table from the Vital Statistics of the United States, 1999.


  1. McGinnis JM, Foege WH: Actual causes of death in the United States. JAMA 270:2207, 1993
  2. United States Preventive Services Task Force: Guide to Clinical Preventive Services. Williams & Wilkins, Baltimore, 1996
  3. Welch HG, Albertsen PC, Nease RF, et al: Estimating treatment benefits for the elderly: the effect of competing risks. Ann Intern Med 124:577, 1996
  4. Lorig K: Patient Education and Counseling. United States Preventive Services Task Force. Williams & Wilkins, Baltimore, 1996
  5. American Academy of Pediatrics: Tobacco's toll: implications for the pediatrician. Committee on Substance Abuse. Pediatrics 107:794, 2001
  6. Bartecchi CE, MacKenzie TD, Schrier RW: The human costs of tobacco use. N Engl J Med 330:907, 1994
  7. Diagnostic and Statistical Manual of Mental Disorders 4th ed.: DSM-IV. American Psychiatric Association. Washington, DC, 1994
  8. Kawachi I, Colditz GA, Stampfer MJ, et al: Smoking cessation and decreased risk of stroke in women. JAMA 269:232, 1993
  9. Dietrich AJ, O'Connor GT, Keller A, et al: Cancer: improving early detection and prevention: a community practice randomised trial. BMJ 304:687, 1992
  10. Djikstra A, Roijackers J, De Vries H: Smokers in four stages of readiness to change. Addict Behav 23:339, 1998
  11. Henningfield JE: Nicotine medications for smoking cessation. N Engl J Med 333:1196, 1995
  12. Silagy C, Lancaster T, Stead L, et al: Nicotine replacement therapy for smoking cessation. Cochrane Database Syst Rev (4):CD000146, 2002
  13. Flegal KM, Troiano RP, Pamuk ER, et al: The influence of smoking cessation on the prevalence of overweight in the United States. N Engl J Med 333:1165, 1995
  14. Heatherton TF, Kozlowski LT, Frecker RC, et al: The Fagerstrom test for nicotine dependence: a revision of the Fagerstrom Tolerance Questionnaire. Br J Addict 86:1119, 1991
  15. Holm KJ, Spencer CM: Bupropion: a review of its use in the management of smoking cessation. Drugs 59:1007, 2000
  16. Hurt RD, Sachs DPL, Glover ED, et al: A comparison of sustained-release bupropion and placebo for smoking cessation. N Engl J Med 337:1195, 1997
  17. Institute of Medicine: Broadening the base of treatment for alcohol problems. National Academy Press, Washington, DC, 1990
  18. Bien TH, Miller WR, Tonigan JS: Brief interventions for alcohol problems: a review. Addiction 88:315, 1993
  19. Brief interventions and alcohol use. Effective Health Care, Bulletin No. 7. Nuffield Institute for Health Care, University of Leeds, Leeds, England, 1993
  20. Kahan M, Wilson L, Becker L: Effectiveness of physician-based interventions with problem drinkers: a review. Can Med Assoc J 152:851, 1995
  21. Wilk AI, Jensen NM, Havighurst TC: Meta-analysis of randomized control trials addressing brief interventions in heavy alcohol drinkers. J Gen Intern Med 12:274, 1997
  22. Walsh DC, Hingson RW, Merrigan DM, et al: A randomized trial of treatment options for alcohol-abusing workers. N Engl J Med 325:775, 1991
  23. Fiellin DA, O'Connor PG: Clinical practice: office-based treatment of opioid-dependent patients. N Engl J Med 347:817, 2002
  24. Rivara FP, Grossman DC, Cummins P: Medical progress: injury prevention. N Engl J Med 337:536, 1997
  25. Traffic safety facts 2000. A compilation of motor vehicle crash data from the fatality analysis reporting system and the general estimates system (DOT HS 809 337). National Highway Traffic Safety Administration. US Department of Transportation. December, 2001
  26. Hunt DR, Lowenstein SR, Badgett RG, et al: Safety belt nonuse by internal medicine patients: a missed opportunity in preventive medicine. Am J Med 98:343, 1995
  27. Effectiveness of occupant protection systems and their use: third report to Congress. National Highway Traffic Safety Administration. US Department of Transportation. Washington, DC, December 1996
  28. Kahane CJ: Fatality reduction by airbags: analyses of accident data through early 1996 (DOT HS 808 470). National Highway Traffic Safety Administration, US Department of Transportation. Washington, DC, August, 1996
  29. The National Highway Traffic Safety Administration Motorcycle Safety Program. National Highway Traffic Safety Administration. US Department of Transportation, January 2003
  30. Thompson DC, Rivara FP, Thompson R: Helmets for preventing head and facial injuries in bicyclists. Cochrane Database Syst Rev (2):CD001855, 2000
  31. Bischoff HA, Stahelin HB, Dick W, et al: Effects of vitamin D and calcium supplementation on falls: a randomized controlled trial. J Bone Miner Res 18:343, 2003
  32. Paganini-Hill A, Chao A, Ross RK, et al: Exercise and other risk factors in the prevention of hip fracture: the Leisure World Study. Epidemiology 2:16, 1991
  33. Tinetti ME, Baker DI, McAvay G, et al: A multi-factorial intervention to reduce the risk of falling among elderly people living in the community. N Engl J Med 331:821, 1994
  34. Kannus P, Parkkari J, Niemi S, et al: Prevention of hip fracture in elderly people with use of a hip protector. N Engl J Med 343:1506, 2000
  35. Mallonee S, Istre GR, Rosenberg M, et al: Surveillance and prevention of residential-fire injuries. N Engl J Med 335:27, 1996
  36. Bierens JJ, Knape JT, Gelissen HP: Drowning. Curr Opin Crit Care 8:578, 2002
  37. McCauley J, Kern DE, Kolodner K, et al: The “battering syndrome”: prevalence and clinical characteristics of domestic violence in primary care internal medicine practices. Ann Intern Med 123:737, 1995
  38. Barrier PA: Domestic violence. Mayo Clin Proc 73:271, 1998
  39. Alpert EJ: Violence in intimate relationships and the practicing internist: new “disease” or new agenda? Ann Intern Med 123:774, 1995
  40. Kyriacou DN, McCabe F, Anglin D, et al: Emergency department-based study of risk factors for acute injury from domestic violence against women. Ann Emerg Med 31:502, 1998
  41. Nonfatal and fatal firearm-related injuries—United States, 1993–1997. MMWR Morb Mortal Wkly Rep 48:1029, 1999
  42. Deaths: Final Data for 2000. National vital statistics reports, 50, no 15. National Center for Health Statistics, Hyattsville, Maryland, 2002
  43. Miller M, Azrael D, Hemenway D: Household firearm ownership and suicide rates in the United States. Epidemiology 13:517, 2002
  44. Cummings P, Koepsell TD: Does owning a firearm increase or decrease risk of death? JAMA 280:471, 1998
  45. Kellerman AL, River FP, Rushforth NB, et al: Gun ownership as a risk factor for homicide in the home. N Engl J Med 329:1084, 1993
  46. Cassel CK, Nelson EA, Smith TW, et al: Internists' and surgeons' attitudes toward guns and firearm injury prevention. Ann Intern Med 128:224, 1998
  47. American College of Physicians: Firearm injury prevention. Ann Intern Med 128:236, 1998

Editors: Dale, David C.; Federman, Daniel D.