Risk of exposure of surgical personnel to patients' blood during surgery at San Francisco General Hospital

Occupational risks of infection in the surgical management of trauma patients

The evolution of the acquired immunodeficiency syndrome secondary to the human immunodeficiency virus (HIV) has resulted in increased concern by surgeons and other members of the health care team with respect to occupationally acquired viral infections. Hepatitis B, hepatitis C, and other viral infections of the liver remain more important than HIV as a cause of morbidity and death for surgeons. Reduction in risk of these infections for surgeons can be achieved by hepatitis B vaccination, better utilization of personal protective equipment, and by improved techniques in the performance of procedures. An overall enhanced awareness in the use of sharp instruments in the operating room is most important. Finally, serologic testing of patients prior to operative procedures, or of surgeons as a condition to performing procedures is an expensive and counterproductive exercise that cannot be recommended.

HIV risk factors and seroprevalence in surgical patients

OBJECTIVE

To determine the prevalence of risk factors for human immunodeficiency virus (HIV) infection, and HIV-1 seroprevalence, in surgical patients.

DESIGN

Prospective study in which consenting patients completed a questionnaire on factors potentially related to a higher risk of HIV infection, and underwent an HIV-1 antibody test.

PATIENTS AND SETTING

Both elective and emergency surgical admissions between July and November 1990 at St Vincent's Hospital, Sydney.

RESULTS

Of 1292 patients who were approached to participate, 27 had been previously diagnosed with HIV-1 infection and, of the remaining 1265, eight (0.63%) refused to answer the questionnaire and undergo a blood test, and 12 (0.95%) refused the blood test only. HIV-1 antibody testing was completed for 1171 study subjects. Twenty-six of the 807 men who answered the questionnaire had been previously diagnosed with HIV-1 infection. Of the remaining 781 male patients, 133 (17%) reported a history of homosexual contact, injecting drug use or blood transfusion, 132 (17%) had had a prior HIV-1 antibody test and three were newly diagnosed with HIV-1 infection as a result of the survey. Of 476 women who completed the questionnaire, one had been previously diagnosed with HIV-1 infection, and of the remaining 475, 59 (12.4%) reported injecting drug use or a blood transfusion and 72 (15%) had had a prior HIV-1 antibody test. No women were newly diagnosed with HIV-1 infection. Of patients reporting specific factors, the proportion who had had a prior HIV-1 antibody test varied from 62% for men reporting homosexual contact to 34% for recipients of a blood transfusion between 1980 and 1985.

CONCLUSION

Although an appreciable proportion of surgical patients admitted to St Vincent's Hospital reported factors associated with a higher risk of HIV infection, the prevalence of undiagnosed HIV-1 infection is very low, particularly among patients reporting no such factors.

Surgical glove perforations in otolaryngology: prevention with cut-resistant gloves

This study examines the role of cut-resistant, "orthopedic-type" gloves in prevention of exposure of surgical personnel to infectious body fluids through surgical glove perforation. The incidence of glove perforations in 12 consecutive cases of intermaxillary fixation (IMF) was studied. Surgeons were double-gloved with standard latex gloves and wore cut-resistant gloves between the outer and inner gloves. As controls, six cadaver heads were wired into IMF with surgeons double-gloving without the cut-resistant glove. The rates of latex glove perforation were then compared between the two groups. Fifty-three percent (32 of 60) of outer latex gloves, but no inner latex gloves (0 of 50) were perforated when surgeons wore cut-resistant gloves. In the control group, 45% of outer gloves (9 of 20) and 15% of inner gloves (3 of 20) were perforated. The difference of inner glove perforations--hence cutaneous exposure--between these two groups was statistically significant (p < 0.01). The use of cut-resistant gloves in addition to double-gloving with latex surgical gloves is recommended for facial plastic and reconstructive procedures with metal implants and major head and neck surgery, because these operations have a high rate of glove perforation. Because some loss of dexterity is noted when cut-resistant gloves are worn, the risk of glove perforation must be weighted against the need for optimal dexterity.

Risks to surgeons and patients from HIV and hepatitis: guidelines on precautions and management of exposure to blood or body fluids. Joint Working Party of the Hospital Infection Society and the Surgical Infection Study Group

Images

Knowledge, attitudes and practices of Greek health professionals, in relation to AIDS

In order to improve educational programmes directed at health care workers we investigated their knowledge, attitudes and practices in relation to HIV/AIDS. An anonymous self-administered questionnaires was distributed to 609 health care workers. Of these, 59.6% agreed to participate (42.4% of the medical doctors, 74.3% of the nurses and 79.6% of the laboratory technicians, health visitors and other health care workers). All studied groups believed that their knowledge of HIV modes of transmission (84.3%) was sufficient. In contrast, a relatively small percentage reported knowledge of the clinical spectrum of HIV infection (48.8%) and the diagnostic assays (57.6%). Nearly all the study participants believe (92.8%) that there is a risk of acquiring HIV infection during the hospitalization of HIV/AIDS patients. Obligatory screening of all patients was reported by nearly all participants (90.6%) as a chance to minimize their occupational risk. Although health care workers reported satisfactory knowledge of safety measures (87.0%), only 56.7% used gloves and 38.8% accept the hospitalization of HIV/AIDS patients. In spite of the educational programmes for AIDS in Greece, this study demonstrates that health professionals' knowledge and precautionary measures are not sufficient. As a result, a small percentage of them treat AIDS patients without discrimination. There is an urgent need to implement specific educational programmes for health professionals so that they will safely provide high quality care to people affected by HIV/AIDS.

Glove perforations in ophthalmic surgery

To determine the potential rate of intraoperative parenteral exposure of physicians and patients caused by glove perforation during ophthalmic surgery, gloves were analyzed after 125 procedures. Gloves were collected from all surgical team members (surgeon, assistants, scrub nurse, and circulating nurse). The rate of glove perforation was significantly lower for the surgeon, 0.3% (one of 303 gloves) than for the assistants, 5% (ten of 202 gloves [P = .001]), scrub nurses, 16% (52 of 326 gloves [P = .0001]), and circulating nurses, 15% (43 of 293 gloves [P = .0001]); similarly, assistants had a significantly lower rate than did scrub nurses (P = .0001) and circulating nurses (P = .001). There was no statistically significant difference in number of perforations on the basis of surgery duration or type of ophthalmic procedure. These findings suggest that the risk of parenteral exposure during ophthalmic surgery is low for the surgeon, and higher for other surgical personnel. We also analyzed additional safety precautions. Further study is warranted to determine the effectiveness of precautions and to guide policy formulation.

Efficacy of double gloving as a protection against blood exposure in dermatologic surgery

Double gloving has been widely suggested as a means of decreasing exposure to blood and other body fluids. However, most dermatologists do not routinely follow this practice when performing surgical procedures because it reduces manual dexterity and increases costs. To determine whether double gloving should be routinely practiced in dermatologic surgery, pairs of sterile surgical gloves worn during weekly UCLA dermatologic surgery clinics were collected over a period of 3 months and examined for perforations. The procedures gloves were used for ranged from excision of benign lesions, lasting 15 minutes, to hair transplants, lasting 3 hours. Upon completion of the procedures, 8 (5.5%) of 144 pairs of single gloves, 2 (3.7%) of 54 pairs of double inner gloves, and 3 (5.5%) of 54 pairs of double outer gloves were found to have perforations. There were no instances of both the double outer and the double inner gloves having perforations when worn on the same hand. This suggests that double gloving offers a protective advantage by providing extra protection for both the surgeon and the patient during dermatologic surgery.

Blood and amniotic fluid contact sustained by obstetric personnel during deliveries

OBJECTIVE

The objective of this study was to characterize blood and amniotic fluid contact sustained by obstetric personnel during deliveries.

STUDY DESIGN

Trained observers collected data on 1376 person procedures during 230 deliveries at Grady Memorial Hospital from May to October 1989. Rates of contact were compared by means of the chi 2 test.

RESULTS

At least one blood or amniotic fluid contact occurred during 79 (39.1%) of 202 vaginal and 14 (50.0%) of 28 cesarean deliveries; a needle stick occurred in 4 (2.0%) of the vaginal deliveries. Obstetricians and midwives had the highest rates of blood and amniotic fluid contact (18.7% and 28.8% of person procedures, respectively). Half of the contacts sustained by midwives might have been prevented by the use of gowns. Most contacts sustained by obstetricians might have been prevented by face shields, impervious gowns, and impervious shoe covers.

CONCLUSIONS

Obstetricians and midwives had substantial risk of blood and amniotic fluid contact during delivery; many of their contacts were potentially preventable.

Occupational exposure to human immunodeficiency virus and hepatitis B virus: a comparative analysis of risk

OBJECTIVE

To estimate the occupational risk from infection with the human immunodeficiency virus (HIV) in terms of loss of (quality-adjusted) life expectancy, and to compare that risk to those posed by other hazards faced by health care workers.

DESIGN

Decision-analytic model.

RESULTS

For a 30-year-old female health care worker (unvaccinated for hepatitis B virus [HBV]), the loss of life expectancy from a needlestick from a symptomatic HIV-positive (HIV+) patient is 39 days (range, 17 to 93 days), as compared with a loss of 17 days from a needlestick from a patient who is hepatitis-B-surface-antigen-positive (HBsAg+), and 38 days from a needlestick from a patient who is hepatitis-B-e-antigen-positive (HBeAg+). When morbidity is included in the analysis of risk (through calculation of the quality-adjusted loss of life expectancy), the risk from both HBV and HIV increases. The quality-adjusted loss of life expectancy due to a needlestick exposure from a symptomatic HIV+ patient is 45 days (range, 20 to 108 days), as compared with a quality-adjusted loss of life expectancy of 48 days from a needlestick from an HBsAg+ patient, and 109 days from a needlestick from a patient who is known to be HBeAg+. By comparison, a cross-country automobile trip is associated with a loss of life expectancy of approximately 1 day. The 45- to 50-day loss of quality-adjusted life expectancy from percutaneous exposures to HIV and HBV is approximately the same magnitude as the gain in life expectancy from 10 years of annual screening for breast cancer with mammography and physical examination.

CONCLUSIONS

The risk associated with percutaneous exposures to symptomatic HIV+ patients is comparable to other risks that health care workers have faced knowingly and have accepted in the recent past. However, the loss of quality-adjusted life expectancy associated with a needlestick exposure is significant. Identification of cost-effective methods that increase the safety of medical personnel but also ensure full access to high-quality care for HIV+ patients should be a high priority.

Attitudes of cardiothoracic surgeons in the UK to human immunodeficiency virus

A survey was carried out into attitudes of cardiothoracic surgeons in the UK to human immunodeficiency virus type 1 (HIV-1) infection associated with clinical situations that would normally have been managed surgically with low operative mortality rates and long median survival times. The survey response rate was 72.4 per cent. In patients with acute valvular insufficiency or with continuing angina despite maximal medical therapy (unstable angina) who were HIV-1 antibody positive, 75.8 and 80.8 per cent, respectively, of surgeons would operate. If the patient had end-stage infection, acquired immune deficiency syndrome (AIDS), 29.7 per cent and 34.7 per cent, respectively, would consider surgical intervention. When asked to perform simple procedures such as open lung biopsy or pleurectomy on a patient with AIDS, more than half of surgeons would operate (52.2 and 65.6 per cent respectively). In patients with operable carcinoma of the lung and asymptomatic HIV-1 infection 52.3 per cent would operate. This fell to 15.0 per cent if the patient had a diagnosis of AIDS. The majority of surgeons (77.2 per cent) felt patients should have an HIV-1 antibody test before operation and this rose to 95.6 per cent if patients were in a high-risk group; 60.2 per cent of surgeons had changed their surgical practice to reduce the risks of blood-borne infection.

Infection prevention guidelines for cardiac catheterization laboratories. Society for Cardiac Angiography and Interventions Laboratory Performance Standards Committee

The following guidelines for prevention of catheterization laboratory infections are based on standard precautions for infection prevention in surgical wounds. Specific recommendations for patient preparation include proper methods for hair removal, skin cleaning and draping, antibiotic prophylaxis, wound irrigation and dressing, and sheath removal. Sterile precautions should be more vigorous for cutdown procedures compared to percutaneous. Caps, masks, gowns, and gloves help to protect both the patient and operator. Handwashing is the most important procedure for preventing infections. Maintenance of the catheterization laboratory environment includes appropriate cleaning, limitation of traffic, and maintenance of adequate ventilation. Proper catheterization technique and appropriate use of sterile equipment will decrease the wound infection rate. Protection of personnel may be accomplished by proper gowning and gloving, disposal of contaminated equipment, and care of puncture wounds and lacerations. All personnel should receive vaccination for hepatitis B.

HIV disease: present status and future directions

This review of the human immunodeficiency virus (HIV) epidemic shows that HIV has had and will have a major impact on dentistry, just as it has had on so many other aspects of medical practice and society. These areas include the prevention of HIV transmission in the dental care workplace, the early and safe care and treatment of those who are infected, and the protection of those who are vulnerable or made more vulnerable because of HIV infection. To do this, the dental professional must be educated about HIV and its diseases, their treatment, and what must be done to prevent HIV transmission. Early recognition and treatment of HIV-related oral diseases have become the norms of practice today. Although more and more dentists face potential exposure to HIV, excellent dental care can be provided while minimizing this risk.

Frequency of glove puncture in the post mortem room

The post mortem room is a potential source of danger from contamination caused by glove punctures. The objective of this study was to assess the incidence of hand injury and hand contamination during post mortem examinations. A prospective study of injury and glove puncture rate was conducted during post mortem examinations performed in 1990 on adults in Northwick Park Hospital mortuary. Four pathologists and two assistants performed a total of 44 post mortem examinations on adults. All participants completed a questionnaire after each examination, recording the total number of gloves used and the number of hand injuries and glove perforations sustained. Just over 8.3% of gloves were punctured; 31.8% of these punctures went unnoticed. Evisceration was the procedure most likely to result in hand contamination. The study highlights a significant risk which could be reduced by more care, frequent glove changes, and hand washing during post mortem examinations. In particular, gloves should be changed and hand washed when evisceration has been completed.

Surgical glove perforation and maxillofacial trauma: to plate or wire?

The technique of interdental wiring was compared with a small-plate osteosynthesis technique (SPO) in a prospective study of surgical glove perforations acquired during the treatment of mandibular fractures. Using the SPO technique there was a significant reduction in the incidence of skin penetrating injuries in the surgeon (P less than 0.005) and assistant surgeon groups (P less than 0.05). The reduction in the incidence of glove perforation in the assistant surgeon group was very highly significant in the SPO series (P less than 0.001). The reduction in the surgeon group was not significant. No difference was noted in the scrub nurse group. The small-plate osteosynthesis technique has the advantage of reducing the risk of intraoperative cross-infection transmitted by hand contamination or penetrating injury. The recommended precautions for preventing the transmission of blood-borne pathogens are reviewed.

A look-back investigation of patients of an HIV-infected physician. Public health implications

BACKGROUND

Transmission of the human immunodeficiency virus (HIV) to five patients receiving care from an HIV-infected dentist in Florida has recently been reported. Current data indicate that the risk of HIV transmission from health care workers to patients is low. Despite this low risk, programs to notify patients of past exposure to an HIV-infected health care worker are being conducted with increasing frequency.

METHODS

We recently conducted an investigation of all the patients cared for by an HIV-infected family physician during a period when he had severe dermatitis caused by Mycobacterium marinum on his hands and forearms. After reviewing the patients' records, we notified 336 patients who had undergone one or more procedures (digital examination of a body cavity or vaginal delivery) placing them at potentially increased risk of HIV infection. The patients were offered tests for HIV infection and counseling.

RESULTS

Of the 336 patients, 325 (97 percent) had negative tests for HIV antibody, 3 (1 percent) refused testing, 1 (less than 1 percent) died of a cause unrelated to HIV infection before notification, and the HIV-antibody status of 7 (2 percent) remained unknown. The direct and indirect public health costs of this investigation were approximately $130,000.

CONCLUSIONS

The results of this investigation raise important questions about the risk of HIV transmission from health care workers to patients and the usefulness of HIV look-back programs, particularly in the light of recently published recommendations from the Centers for Disease Control. We propose that before a look-back investigation is undertaken, there should be a clearly identifiable risk of transmission of the infection, substantially higher than the risk requiring limitation of an HIV-infected health care worker's practice prospectively.

Risk of blood contamination and injury to operating room personnel

The potential for transmission of deadly viral diseases to health care workers exists when contaminated blood is inoculated through injury or when blood comes in contact with nonintact skin. Operating room personnel are at particularly high risk for injury and blood contamination, but data on the specifics of which personnel are at greater risk and which practices change risk in this environment are almost nonexistent. To define these risk factors, experienced operating room nurses were employed solely to observe and record the injuries and blood contaminations that occurred during 234 operations involving 1763 personnel. Overall 118 of the operations (50%) resulted in at least one person becoming contaminated with blood. Cuts or needlestick injuries occurred in 15% of the operations. Several factors were found to significantly alter the risk of blood contamination or injury: surgical specialty, role of each person, duration of the procedure, amount of blood loss, number of needles used, and volume of irrigation fluid used. Risk calculations that use average values to include all personnel in the operating room or all operations performed substantially underestimate risk for surgeons and first assistants, who accounted for 81% of all body contamination and 65% of the injuries. The area of the body contaminated also changed with the surgical specialty. These data should help define more appropriate protection for individuals in the operating room and should allow refinements of practices and techniques to decrease injury.

Does knowledge of human immunodeficiency virus infection decrease the frequency of occupational exposure to blood?

Implementation of universal precautions is advocated to prevent exposure to human immunodeficiency virus (HIV) and other bloodborne pathogens. Critics of this approach argue that additional benefit can be derived by identifying and labeling patients known to be infected. Knowledge of HIV infection status could result in a reduced exposure rate by two mechanisms: (a) by motivating improved compliance with universal precautions, or (b) by allowing changes in procedure or technique not feasible for all patients. Compliance with universal precautions may reduce the frequency of some types of exposure but has not been associated with a reduction in the frequency of needlestick exposures in several studies. Despite the perception by some health care workers that awareness of HIV status will result in improved safety, no objective data have demonstrated a direct benefit from testing or identifying infected patients. Health care workers who recognize the presence of occupational HIV risk are apt to be motivated to practice universal infection control precautions and experience little additional benefit from testing individual patients. Health care workers in low prevalence areas may experience less incentive to comply routinely with universal precautions and selectively may be motivated when HIV infection is diagnosed in individual patients. However, routine testing in areas of low HIV prevalence is not likely to be cost effective. Until further research clarifies the efficacy and costs of universal precautions or HIV testing, infection control standards should maximize local autonomy in developing rational policies consistent with institutional needs.

Human immunodeficiency virus transmission in health care settings: risk and risk reduction

Surveillance data and case reports document that health care workers (HCWs) risk occupationally acquired human immunodeficiency virus (HIV) infection. Transmission of HIV to patients of an infected HCW during invasive procedures has also been reported. The risk to a susceptible HCW depends on the prevalence of HIV infection among patients, the nature and frequency of occupational blood exposures, and the risk of transmission per exposure. Blood exposure rates vary by occupation, by procedure, and by compliance with preventive measures. Future efforts to protect both HCWs and patients must include improved surveillance, risk assessment, study of postexposure prophylaxis, and an emphasis on exposure prevention, including development of safer medical devices, work practices, and personal protective equipment that are acceptable to HCWs and do not adversely affect patient care.

Epidemiology of hospital sharps injuries: a 14-year prospective study in the pre-AIDS and AIDS eras

The world pandemic of acquired immunodeficiency syndrome (AIDS) has focused enormous attention on the problem of accidental sharps injuries sustained by health care workers (HCWs) and the risk of occupationally acquired infection by human immunodeficiency virus (HIV). At the 1980 Conference, we reported a 4-year epidemiologic study (1975-1979) of sharps injuries in HCWs at our hospital. Using the same reporting system and analyses, we now report the epidemiology of sharps injuries in our center during the current AIDS era (1987-1988) and assess trends over the 14-year period. Despite greatly increased institutional efforts to prevent sharps injuries, the annual incidence has increased more than threefold (60.4 to 187.0/1,000 HCWs), reflecting better reporting and increased exposure. Reported injuries by house officers have increased ninefold. Adjusting for inflation, the direct costs of sharps injuries has increased sevenfold ($5,354 to $37,271/year). Environmental service HCWs (305.8 sharps injuries per 1,000 employees) now have the highest incidence in our center, followed by nursing personnel (196.5/1,000) and laboratory personnel (169.9/1,000), but as in 1975-1979, two thirds of all injuries occur in nursing personnel. Although phlebotomy team members have a very low risk per procedure (1/26,871 draws), their annual incidence is extraordinarily high, 407.0/1,000. Injuries continue to occur mainly during disposal of waste, linen, or used procedure trays (19.7% of all injuries), administration of parenteral injections or infusion therapy (15.7%), surgery (16.0%), blood drawing (13.3%), or recapping of used needles (10.1%). Making disposal units available at every bedside has reduced injuries from needle disposal two-fold since 1975-1979. With consistent application of a stringent postexposure protocol, and wide acceptance of the hepatitis B vaccine, we have had no sharps injury-related infections during the past 3 years. These data indicate the increasing risk, complexity and cost of sharps injuries in HCWs and the need for more innovative--ideally, technology-based--approaches to prevention. Certain groups of HCWs are at very high risk. Comprehensive postexposure protocols that are uniformly applied can provide substantial protection to exposed HCWs.

Infection control: the premier quality assessment program in United States hospitals

The three decennial international meetings of the Centers for Disease Control have reaffirmed the national health priority for prevention of nosocomial infections. Moreover, the past 20 years have seen the institutionalization of the infection control practitioner and hospital epidemiologist. The epidemiologic activities of these two groups have included access to medical care, identification of risk factors, and both prevention and control of adverse nosocomial events. As such, they represent experience with issues commonly referred to as quality of care. It is argued that infection control not only is now but has been from its inception the premier quality assessment program in the United States. Moreover, it is the only activity in quality assessment that has been shown to be efficacious.

Reducing occupational risk of HIV infection

At San Francisco General Hospital and many other institutions in areas of high HIV prevalence, a policy of body substance isolation has proved easier to implement than the CDC's across-the-board recommendations. Prevention of needle-stick injury and use of barrier methods of infection control are emphasized. Preoperative HIV testing is addressed.

Percutaneous injury during dermatologic surgery

An anonymous survey was conducted among 100 randomly selected fellows of the American Society for Dermatologic Surgery. Forty-one respondents provided information about their experience with percutaneous injury. Causes of injury during 6278 invasive procedures performed during a 1-month period included suture needlesticks (two injuries), injection needlestick (one), needle recapping (one), scalpel blade (one), skin hook (one), and an injury during the transport of an instrument (one). Dermatologic surgeons were more likely to injure their dominant fingers. Attitudes of dermatologic surgeons were surveyed regarding operating on patients while the physician or assistant was actively infected with human immunodeficiency virus, hepatitis B virus, herpetic whitlow, or paronychia caused by Staphylococcus aureus. Dermatologic surgeons believed that they should be allowed to operate while infected with human immunodeficiency virus (41.5%), hepatitis B virus (40%), herpetic whitlow (25%), and paronychia caused by S. aureus (20%). Few dermatologic surgeons would disclose preoperatively to patients infections present in themselves or in operating team members that were due to human immunodeficiency virus (29.4%), hepatitis B virus (27.8%), herpetic whitlow (28.6%), or paronychia caused by S. aureus (33.3%).

Compulsory HIV antibody testing, universal precautions and the perceived risk of HIV: a survey among medical students and consultant staff at a London teaching hospital

Routine screening of patients and health care staff for HIV has not been endorsed by the medical profession. Instead universal precautions have been recommended as being the most effective way of minimizing the occupational risk of HIV infection. Consultant staff and undergraduate medical students at the Royal Free Hospital and School of Medicine, London, were asked about their attitudes towards compulsory HIV antibody testing, their compliance with universal precautions and their perceived risk of HIV infection. A substantial proportion of staff and students supported compulsory HIV antibody testing for patients, health care workers and certain subgroups of the population. Most of the clinical students and about half the consultants failed to comply with universal precautions. Staff and students saw themselves at greater risk of HIV infection in the hospital than in their personal lives. Clearly, these beliefs and practices must be taken into account when introducing a policy of universal precautions.

Some notes on the impact of treating AIDS patients in hospices

As part of a larger nation-wide study of attitudes of hospice personnel, we incorporated questions about patient autonomy and economic justice in which we asked the respondents to share their beliefs about AIDS patients compared to patients terminally ill with other diagnoses. The convenience sample of 826 hospice workers, 76% of whom were clinical professionals (nurses, physicians, social workers) rated terminally ill patients and those with AIDS the same in terms of right to refuse life sustaining medical therapy. Although over one half believed that we are not currently spending enough on the care of AIDS patients, 25% thought that we spend too much on those terminally ill with other diagnoses. Issues of survival time, costs of care, and staff concerns about treating AIDS patients in hospices are briefly discussed.

HIV infection in healthcare workers. How great is the risk? What can be done before and after exposure?

Healthcare workers know that there is a risk of HIV infection through exposure to AIDS patients. In both hospital and office settings, physicians have the opportunity to set standards and promote education about the degree of risk, effective precautions, and postexposure testing, prophylaxis, and treatment. Drs Henry and Thurn share the latest findings and offer policy recommendations based on their own experience.