To prospectively examine the epidemiology of blood and body fluid exposures sustained by medicine housestaff, medical school students, registered nurses (RNs), licensed practical nurses (LPNs), and nurses' aides (NAs) on general medicine wards and to define problem areas that may be amenable to change.

Daily data collection during 9 months using a self-reporting questionnaire.

General medical wards in 2 tertiary referral hospitals.

Medicine housestaff/students and nursing personnel.

Physicians reported 644 exposures, of which 98 (15.2%), 296 (46.0%), and 250 (38.8%) were sustained by medicine residents, interns, and students, respectively. Blood contact occurred with 591 (91.8%) exposures. For physicians, 575 (89.3%) exposures occurred during venipuncture, intravenous catheter manipulation, and arterial punctures. Interns and students most commonly incurred exposures during venipunctures and intravenous manipulations; residents commonly were exposed during emergent intravenous catheter placements. Five-hundred-twenty-two (8 1%) exposures occurred between 7 A.M. and 7 P.M. During 524 (81.4%) exposures, physicians were not using barrier devices. Nurses reported 235 exposures, of which 140 (59.6%), 23 (9.8%), and 72 (30.6%) were sustained by RNs, LPNs, and NAs, respectively. RN exposures commonly occurred during intravenous manipulations and glucometer fingersticks. LPNs and NAs incurred a higher percentage of exposures during nonprocedural patient care. Blood contact and wound drainage accounted for 167 (71.1%) and 31 (13.2%) exposures, respectively.

Exposures to blood and body fluids frequently are incurred by healthcare workers on general medical wards. Efforts to reduce these exposures should be directed not only at improving procedural skills of healthcare workers for venipunctures, intravenous catheter insertions, and glucometer fingersticks, but also in increasing barrier use during procedural and nonprocedural tasks.

To determine the prevalence of infections and antibiotic use at a community hospital in 1990 and to compare the results with those from previous surveys in 1971, 1979, and 1984.

Point prevalence survey.

Community teaching hospital.

All hospitalized patients, excluding those on the nursery, psychiatry, and short-stay wards.

Medical records were reviewed and a pertinent physical examination was performed on each patient.

Overall, 46 (20.1%) of 229 patients had infection in 1990. Community-acquired and hospital-acquired infections were present in 26 (11.4%) and 24 (10.5%) of the patients, respectively. The most common site of community-acquired infection was the deep abdomen in 8 (25.8%) of 32 sites, while hospital-acquired infection most frequently involved the lower respiratory tract in 8 (25.0%) of 32 sites. Antimicrobial agents were being administered to 88 (38.4%) of the patients. Cephalosporins (particularly newer, broad-spectrum agents) were the most frequently used antibiotics, comprising 53 (45.3%) of the 117 total antimicrobial prescriptions. seven (18.4%) of the 38 antibiotics used for surgical prophylaxis were given for more than 48 hours. Trend analysis revealed no significant changes in the proportions of patients with community-acquired, hospital-acquired, or total infection over the last 20 years (p=.18, p=.12, and p=.O7, respectively). While the overall use of antibiotics increased p<.ool), the administration of prophylactic perioperative antibiotics for greater than 48 hours decreased p=.009).

Despite major shifts in factors associated with an increased risk of infection, the overall prevalences of hospital-acquired and community-acquired infections remained remarkably constant during the last 20 years. Although total antibiotic use increased, the duration of surgical prophylaxis decreased and the pattern of use of therapeutic antibiotics also changed with increased use of broad-spectrum agents

To know the incidence, etiology, risk factors, morbidity, and mortality of nosocomial diarrhea in adults.

Nested case-control study, matched by service, length of stay, date of admission, and presence of leukopenia and/or the acquired immunodeficiency syndrome (AIDS). Cases were those who developed nosocomial diarrhea. Controls were those who did not develop nosocomial diarrhea during a comparative period nor during the next ten days. Stool samples were processed in search for parasites, yeasts, bacteria, and rotavirus.

Third-level referral center, in Mexico City, Mexico, for general internal medicine and surgical problems.

Eligible subjects were all new admissions to the hospital from November 1987 to September 1988. Reasons for exclusion were presence of chronic diarrheal disease or melena. There were 115 cases and 111 controls.

Overall risk of acquiring nosocomial diarrhea was 5.5%, or 1.8 episodes per 100 patient-weeks. A potential pathogen was found in 59%. Yeasts and Entamoeba histolytica were the most frequently isolated pathogens. Mortality in cases was 18%, as compared with 5% in controls (p<.01). Multivariate analysis showed enteral feeding, recent enemas, presence of Candida species, use of antacids/Hz-blockers, and presence of nasogastric tubes as significant risk factors for nosocomial diarrhea.

Diarrhea is a common complication in hospitalized patients. It occurs more often than previously suspected and is linked with a substantial mortality, The spectrum of etiologic agents is different from that reported in pediatric hospitals. Given that nosocomial diarrhea may constitute, at least, a marker of severity of illness, it should receive more attention in general hospitals.

To define accurate wound infection rates for the cardiac surgery service based on site of infection and characterization as “deep” or “incisional” and to determine whether a correctable cause for an apparent increase in deep wound infection rates existed.

Observational.

Tertiary-care teaching hospital.

All adults undergoing open heart surgery in 1988 and 1989.

Changed from razor to clipper preoperative hair removal in January 1989.

Deep sternotomy wound infections decreased significantly from 1.2% in 1988 to 0.2% in 1989 (p=.010) and deep venectomy (vein donor site) wound infections declined from 1.6% to 0.4% (p=.014) during the same time period. Incisional wound infection rates did not change. Patients with deep infections more likely required readmission or operation to treat their infection than those with incisional wound infections. The percentage of gram-negative organisms causing wound infections decreased from 56.3% in 1988 to 34.7% in 1989 (p= ,017).

Preoperative hair removal using a clipper appears to have decreased the risk of deep wound infection compared with razor preparation. The dichotomous wound classification of “deep” and “incisional” distinguished between patients who required additional interventions for treatment of wound infections.