The concept of isolating persons with communicable diseases was in practice even in ancient times according to biblical accounts of leper colonies, although the leprosy of biblical times may have been other skin diseases (5). In modern times, hospital construction before 1850 featured crowded open wards (6). As a consequence, cross-infection was common, and mortality rates were high in urban hospitals (7). Florence Nightingale’s observations from the Crimean War (8) led her to advocate small pavilion-type wards joined by open-air corridors. Nightingale emphasized the importance of asepsis and a clean environment. Her teachings were called “fever nursing” and varied from popular concepts of disease at the time, because fever nursing implicated transmission by contact with body substances rather than the environment (9).

The germ theory of infection was accepted in US hospitals in the late 1800s, after the influence of Lister and Pasteur, and conditions began to improve as overcrowding decreased and antisepsis increased (7). Communicable disease hospitals were using individual and group isolation as early as 1889 (10). By the turn of the century, general hospitals were beginning to isolate patients with communicable diseases in an individual room with the use of separate utensils and disinfectants (9,11). Grancher in Paris promoted the theory of communicability by contact rather than airborne spread for most diseases and allowed patients with communicable diseases to be housed in general wards but with separation by wire screens (9). These screens separated the patient from other patients and served as a reminder for staff members to gown and wash their hands. Thus, began the trend in the United States from the isolation hospital to care of communicable diseases in a general hospital setting.

In the early 20th century, Charles V. Chapin of Providence City Hospital used individual isolation cubicles for patients with communicable diseases and documented that fumigation had no effect on secondary cases (12). His work was very important in emphasizing the roles of persons rather than things as spreaders of disease and helped to end the miasmatic theory of transmission (13). Richardson, the physician superintendent of Providence City Hospital, used the barrier method and the cubicle method for isolation of patients, allowing some patients with communicable diseases to be housed in the same room as other patients (9). A card outlining the barrier technique needed was placed on the patient’s bed.

The emergence of Staphylococcus aureus as a hospital pathogen in the 1950s and 1960s prompted the development of infection-control programs in US hospitals. In 1968, the first edition of the American Hospital Association’s manual (13) presented a simple barrier precautions scheme for patients with communicable diseases, listing the need for gloves, gowns, masks, and visitor screening.

While conducting healthcare-associated outbreak investigations in the 1960s, the CDC recognized that standardized
policies for isolating hospitalized patients with communicable diseases were lacking (14). A group of experts convened in 1967 to develop the first CDC isolation recommendations, and first published five categories of isolation in 1970 (15). Even in this initial manual, the philosophy behind the more recently developed Universal Precautions was expressed. By 1976, a survey showed that 93% of hospitals in the United States were using the category-specific approach to isolation (16).

Substantial changes were made in the 1983 CDC recommendations (17). The updated category-specific guidelines for isolation published in 1983 (17) included seven categories of isolation: strict isolation, respiratory isolation, enteric precautions, contact isolation, tuberculosis (acid-fast bacillus [AFB]) isolation, drainage/secretion precautions, and blood and body fluid precautions. These isolation categories grouped diseases that require the same isolation precautions. After the more recent transmissionbased guidelines from CDC, these categories are now of historical interest, but the terms are still commonly used and confused by some healthcare workers. IPs may be called on to clarify and compare these previous categories with the newer guidelines.

In addition, this 1983 guideline introduced disease-specific isolation as an alternative to category-specific isolation. Disease-specific isolation was offered as an alternative for hospitals wanting a more economic system that directed precautions at preventing transmission of a specific disease while avoiding unnecessary isolation precautions for some diseases. The guidelines stated that hospitals could choose the category-specific or the disease-specific system or design their own systems.

Disease-specific isolation was one of two isolation systems recommended by the CDC in 1983 (17). In this system, communicable diseases were considered individually with regard to mode of transmission and infective material, and accordingly, precautions are specified for each disease. The purported advantage of this system is that because precautions are specific for each disease, there are no unnecessary barriers used, and this lowers the cost of isolation. It may also enhance compliance by physicians, who more readily understand the need for specific precautions for each disease. The disadvantage of this system is that because diseases are not grouped by category, it is more difficult to train staff that are not familiar with specific diseases. Another disadvantage is that, like category-specific isolation, this system is diagnosis-driven, and isolation precautions are often important early in the patient’s hospital stay, before a diagnosis is made or even suspected.

The recognition of the acquired immunodeficiency syndrome (AIDS) epidemic in the mid-1980s affected isolation policies in healthcare institutions unlike any other event in modern medicine. Before 1987, most hospitals placed patients in isolation, based on diagnosis or suspected diagnosis, according to the category-specific or disease-specific precautions as outlined by the aforementioned CDC guideline (17). As it became apparent that transmission of human immunodeficiency virus (HIV) could occur from patient to healthcare worker, new guidelines were established to minimize exposure to blood-borne pathogens from all patients, not just patients with a diagnosis or suspected diagnosis of HIV infection (18). In contrast to the 1983 CDC guideline, the 1987 CDC document (18) recommended blood and body fluid precautions for all patients, regardless of known HIV status. The belief that such precautions were unnecessary in patients not known to have blood-borne pathogens was gone. Specifically, barrier precautions were recommended to prevent contact with blood, certain body fluids, and body fluids containing blood. The application of blood and body fluid precautions to all patients was referred to as “universal blood and body fluid precautions” or “Universal Precautions.” In 1988, the CDC published an updated Universal Precautions for the prevention of transmission of HIV, hepatitis B virus (HBV), and other blood-borne pathogens to supplement the 1987 publication (19). This document made it clear that transmission of other blood-borne pathogens, such as HBV, should be prevented as well as that of HIV. In a new precedent for the healthcare industry, the Occupational Safety and Health Administration (OSHA) became involved in regulating and enforcing these guidelines (20). Now healthcare institutions were mandated to apply and enforce what was, in effect, blood and body fluid precautions as a minimum standard for protection of the healthcare worker.

Infection control programs recognized the potential benefit of this universal concept as a means of preventing cross-transmitted pathogens (blood-borne and non-bloodborne) among patients and healthcare workers. It became clear very quickly that an additional isolation system was needed to reduce the risk of transmitting non-blood-borne pathogens, because the CDC-defined Universal Precautions were primarily for preventing transmission of blood-borne pathogens. In the CDC 1988 update (19), category-specific or disease-specific isolation precautions are recommended to fill this need, as described in the 1983 CDC guidelines. IPs at Harborview Medical Center in Seattle, Washington, recognized the problem early. They implemented a BSI system at Harborview in 1984 to control cross-transmission of non-blood-borne pathogens. This system designated all body fluids and tissue as potentially infectious (21). In 1987 and 1990, Lynch et al. (22,23) described their system and its advantages in preventing the transmission of both blood-borne and non-blood-borne pathogens. This system provided an alternative to the category-specific or diseasespecific systems. Some confusion ensued because the term Universal Precautions was sometimes used to apply to barrier precautions for all body fluids, not just blood and certain body fluids as originally defined (1,2). Although the 1996 CDC guideline for isolation precautions in hospitals includes concepts of both Universal Precautions and BSI, these isolation systems are described briefly below because of their impact on current practices. Universal Precautions (now part of Standard Precautions) are also described in Chapters 73 and 74.

In 1985 and 1986, the CDC published recommendations to prevent the transmission of HIV in the workplace (24,25). In 1987, a more comprehensive document (18) was published in response to increasing concern from healthcare workers about occupational exposure to HIV. These guidelines recommended the application of blood and body fluid precautions to all patients and designated this policy “Universal Precautions” or “universal blood and body fluid precautions.”

Universal Precautions as presented by the CDC in 1987 (18) include the following concepts:

After the recommendations for Universal Precautions were published in 1987, hospitals scurried to write their own institutional policies and implement training for their personnel in the prevention of blood-borne diseases in the workplace. In 1988, the CDC published an update to Universal Precautions that indicated these precautions were also for the prevention of other blood-borne pathogens such as HBV and specified that only specific body fluids implicated in the transmission of blood-borne pathogens needed to be included under Universal Precautions. Many hospitals already had policies in place and employees trained by this time, which contributed to confusion in the use of the term Universal Precautions. A variety of different systems carried this term in individual institutions (29). The 1988 update also included further clarification on the use of protective barriers, the use of gloves for phlebotomy, the selection of gloves, and waste management.

Jackson and Lynch (21) responded early on to the concern that unrecognized or undiagnosed cases resulted in unsafe exposures for healthcare workers. As early as 1984, these authors pointed out that many infectious agents are transmitted from patients who have only mild symptoms or no symptoms, and they recommended barrier precautions for anticipated contact with blood or any body fluids from all patients and reemphasized the important role of hand washing. Recognizing the limitations of the diagnosis-driven, category-specific, and disease-specific isolation systems, they systematically outlined an alternative system called BSI (22). This approach was similar to the CDC Universal Precautions in that it presumed that all patients were potentially infectious, but it differed in that barrier precautions are used to prevent contact with all body fluids and tissue, not just certain body fluids and blood-tinged body fluids, as recommended in the 1988 CDC update. The term body substance rather than body fluid is used to emphasize that barrier precautions should be used to prevent contact with solids, such as tissue and feces, and body fluids. BSI contains six major components (22):

At the time of the publication of the 1988 CDC update, there were no published data on the preference of latex versus vinyl gloves. Since that time, there have been numerous studies addressing the integrity of gloves in general and latex versus vinyl gloves in particular. The standards for testing the integrity of latex gloves were established by the American Society for Testing and Materials (ASTM) of the FDA, and compliance with them is voluntary. In 1977, the standard allowed no more than 15 defects per 1,000 (1.5%) sterile unused latex surgical gloves, as determined by the watertight method of testing (35), and 25 defects per 1,000 (2.5%) latex examination gloves (35). In 1989, the FDA method for testing gloves improved, and the standards changed to an allowable defect rate of 2.5% for surgeon’s gloves and 4.0% for examination gloves (36). Due to the continued concern of potential transmission of blood-borne pathogens HIV, HBV, and hepatitis C, the FDA issued a final rule effective December 19, 2008, which further reduced the allowable defect rate for examination gloves to 2.5%. The FDA, using calculated projections based on available CDC data on HIV, HBV, and HCV infections, estimated that this modification in acceptable quality levels could potentially avoid seven cases of HIV infection and seven cases of chronic HBV infection transmitted to healthcare workers over a 10-year period (37).

There are no standards for vinyl gloves. Concern regarding occupational exposure to HIV raised the issue of glove integrity in the clinical setting. In addition, some cases of herpetic whitlow in intensive care unit (ICU) nurses who used gloves focused more attention on this issue (38). Scanning of gloves by electron microscopy has documented inapparent pits from 30 to 50 µm in size, suggesting the possibility that viruses could penetrate this barrier (39).

In addition, several studies have documented leakage rates higher than the ASTM standard of 1.5% to 2.5%. DeGroot-Kosolcharoen and Jones (40) showed that although several brands of sterile latex surgical gloves were impermeable to water and blood, some brands showed leakage rates of up to 8%. Nonsterile latex and vinyl gloves showed leakage rates of 0% to 52%. Nonsterile packaging or packaging in suction kits increased leakage rates. Korniewicz et al. (41) studied gloves stressed by conditions mimicking those encountered in patient care and found that 63% of vinyl gloves leaked a stock solution of bacteriophage compared with 7% of latex gloves. Korniewicz et al. (42) also documented the penetration of 20% of latex gloves and 34% of vinyl gloves by S marcescens. These studies indicate that gloves reduce the risk of gross soilage from blood or body fluids but that they are not 100% effective.

Latex hypersensitivity due to repeated exposure and sensitization of the healthcare worker to latex antigens has led to the adoption of nonlatex gloves made out of various synthetic materials such as neoprene, polyurethane, and nitrile. Korniewicz et al. (43) examined both latex and nonlatex surgical gloves for defects after use by surgeons during surgical procedures and found overall glove defect rates of 5.6% and 7.5% for latex and nonlatex surgical gloves, respectively. Based on the data obtained during this review of surgical gloves, the authors recommended that surgeons change gloves within 2 to 3 hours to avoid exceeding defect rates >5%.

Doebbeling et al. (33) showed that washing gloved hands was not effective for decontamination, and in fact, 5% to 50% of hands were contaminated after gloves were removed. Washing gloves has also been shown to decrease their integrity (44). Thus, gloves should not be washed and reused between patients. These studies affirm that although gloves can be used as a barrier to reduce gross contamination from blood and body fluids, antisepsis after glove removal remains very important because occult breaks in gloves can and do occur.

The surgical literature has long been concerned with perforations in gloves during surgical procedures. In 1899, Bovie (45) stated that careful hand washing was needed, because gloves could be punctured accidentally during an operation.
More recent studies have quantitated the number and location of inapparent perforations that may occur in gloves during surgical or dental procedures. Albin et al. (46) showed a 33% leak rate of latex gloves randomly studied after surgical procedures. These authors also documented a leak rate of up to 5.5% in unused gloves. Gloves studied sequentially showed a leak rate of 58.5% at the end of surgical procedures and 32% at the end of dental procedures. Double gloving decreased the leak rate to 25%. In the sequential surgical study, 52% of the leaks occurred in the first 75 minutes; in the sequential dental study, 75% of the leaks occurred in the first 30 minutes. Gloves used in cardiovascular, orthopedic, abdominal, and oral surgical procedures had leak rates of more than 50%. The frequency of occult glove perforation has been noted to be as high as 10% after interventional radiologic procedures (47). In the Albin et al. study evaluating surgical and dental procedures, leak rates for gloves were evaluated for various members of the surgical team and were found to be highest for the surgeon (52%), followed by the first assistant (29%), and then the scrub nurse (25%) (39). Most perforations (60%) occurred in the thumb or index finger of the glove. Other studies have also documented that the largest number of perforations occur in the thumb, index finger, and middle finger (48,49).

Universal Precautions are now a minimum standard in US hospitals as a result of OSHA regulations. Many hospitals also have BSI or some modification of BSI in place because of increasing emphasis on the potential infectiousness of body fluids from all patients and the increasing rate of multidrug-resistant pathogens. To review the advantages of these systems over category-specific or disease-specific isolation, the latter systems may be inconsistently or incorrectly applied, whereas precautions that are used for all patients not only are easier to implement but also protect cross-transmission from patients who may lack signs or symptoms of a disease. Furthermore, there is less psychological trauma for individual patients identified as having a microorganism transmissible by blood or body fluids because all patients are treated in a standard manner. Because of healthcare worker concern about HIV in particular, this system at least theoretically eliminates the need for routine screening of all patients and personnel for HIV at periodic intervals—a process that would prove extremely costly (50).

Some disadvantages of the Universal Precautions concept have been proposed. Because gloves were used more extensively for barrier precautions in BSI and in Universal Precautions, some healthcare workers have sometimes neglected to change gloves between patients (30,32), and such practices have been associated with cross-transmission of microorganisms (51,52). Education and reinforcement of appropriate use of gloves and changing gloves between patients can be successful in reducing such practices (23,51).

The CDC has stated that each institution may design its own system of isolation (17). Indeed, as hospitals have tailored Universal Precautions or BSI to their own institutional needs, each system has incorporated elements of the other and the terms have been used interchangeably, even though there are real and philosophical differences between the two systems (2). Consequently, confusion has ensued regarding the term Universal Precautions in particular (1). The primary purpose of Universal Precautions is to reduce healthcare worker exposure to blood-borne pathogens, whereas the primary intent of BSI is to reduce cross-transmission of microorganisms between patients by transient carriage on the hands of personnel. An additional benefit is the protection of the healthcare worker from the patient’s microorganisms (2).

The effectiveness of Universal Precautions has been evaluated using the frequency of personnel nonparenteral exposures to blood and body fluids (including sputum, urine, feces) as a monitor. Fahey et al. (53) and Wong et al. (54) documented a significant decrease in nonparenteral exposures to blood and body substances after the implementation of Universal Precautions. Saghafi et al. (55) also documented a reduction in exposure of unprotected skin to blood, but the rate of needlestick exposures remained unchanged. So it appears that although Universal Precautions or BSI may significantly reduce nonparenteral exposures to blood or body fluids, other measures such as engineering controls are needed to reduce parenteral exposures such as needlesticks.

As Universal Precautions or BSI systems were implemented throughout the country, glove use increased substantially and cost became a concern. Doebbeling and Wenzel (56) evaluated the costs of using Universal Precautions, and McPherson et al. (57) evaluated the cost of BSI. Universal Precautions increased the total annual costs for isolation materials at a large university teaching hospital by $350,900—an increase, adjusted for inflation, from $13.70 to $22.89 (67%) per admission. Although BSI theoretically could be more costly, it caused an unadjusted increase in cost of 147% for isolation materials compared with an unadjusted increase in cost of 167% for Universal Precautions (56,57). There was an approximately 80% increase in the use of gloves for BSI compared with a 64% increase in glove use for Universal Precautions. Doebbeling and Wenzel (56). estimated that Universal Precautions cost approximately $269 million annually nationwide (using the dollar value from 1989) in hospitals alone and approximately $67 million in the outpatient setting, accounting for $336 million total per year nationwide.

Although these systems are expensive, the alternatives must be considered. The alternative of testing all patients admitted to US hospitals each year is estimated to be $2.6 billion or approximately eight times the cost of Universal Precautions (58). Thus, Universal Precautions are less expensive than universal testing. In addition, a decrease in healthcare-associated infection rates has been documented after the implementation of Universal Precautions and BSI (2,56), providing further evidence for the cost-benefit of these systems in the United States.