We performed a prospective randomized study on 71 patients on chronic outpatient hemodialysis to determine whether a sterile technique was better than a clean technique for preparation of the skin over the vascular access site prior to cannulation. In addition, we wanted to determine overall and site-specific infection rates, microbial etiologies of infection, and risk factors for infection. The overall infection rate was 4.7 infections per 100 dialysis months; the vascular access-site infection rate was 1.3 infections per 100 dialysis months; and the rate for bacteremia was 0.7 cases per 100 dialysis months. Staphylococcus aureus was the most common pathogen, but infections were equally divided between gram-positive cocci and gram-negative bacilli. Advanced age (P = 0.02), a low Karnofsky activity score (P = 0.05), poor hygiene (P = 0.0004) and number of hospitalizations (P = 0.0002) were risk factors for infections in general while only poor hygiene (P = 0.002) was a risk factor for vascular access-site infection. Sterile preparation of the skin over the vascular access site was no more effective at preventing infection than was clean technique (P = 0.80). Maintenance of good personal hygiene may be one of the most important measures for prevention of infections in hemodialysis patients

An outbreak of gastroenteritis in New York City's largest jail involved 145 cases over a two-month period. The outbreak was unusual in that two Salmonella strains (serogroups B and D) were involved. Management of the outbreak involved screening kitchen workers by culture of stool samples, and education regarding personal hygiene. Obstacles to investigation and management of the outbreak arose out of the special nature of the jail environment; these included jurisdictional problems and high turnover of the inmate population.

During the course of a tuberculosis skin testing program at a chronic care Veterans Administration Medical Center, we uncovered evidence of occult transmission of endemic tuberculosis. Skin test conversion of eight patients (one of whom had unsuspected progressive primary tuberculosis) and two employees was ultimately traced to a patient in whom tuberculosis was first diagnosed at autopsy three years earlier. Identification of employee skin test conversions was a key factor in recognizing and terminating disease transmission. Throughout the institution, 33% of patients were tuberculin-positive; 10.8% demonstrated the booster phenomenon after initial negative skin tests. Prevalence of tuberculin positivity among employees was 28%. Twelve percent of initially tuberculin-negative employees converted during employment. Our experience documents the value of tuberculin testing of both patients and staff in a chronic care environment, and the necessity of vigorous investigation of skin test conversions.

In a one-year retrospective review, 138 symptomatic infections were identified in 145 admissions to a rehabilitation unit. One hundred twenty-six (91%) infections were either urinary or skin and soft tissue infections. The daily incidence of infection increased with increasing numbers of infections for infected patients. Patient factors associated with infection included male sex, young age, spinal cord injury, admission for management of decubitus ulcers or urinary infection, history of urinary infection, urologic interventions or skin breakdown, and bladder and bowel incontinence. All patients with a history of chronic urinary infection or skin breakdown developed infection. In a logistic regression model, factors that were independently associated with risk of infection in the remaining patients included sex, incontinence, chronic neurologic disease, and prior urologic interventions. This review suggests rehabilitation patients are at high risk of acquiring nosocomial infection. The subgroup of patients with the highest risk of infection are identifiable by specific characteristics.

A recent article in this section reviewed some problem areas of susceptibility testing. In particular, the “myth” of susceptibility testing being predictive of clinical outcome was discussed. The realization of this myth often leaves clinicians less comfortable when selecting antimicrobial agents. The question they often ask is, “How does one select antibiotics?” This question has always been important for empiric therapy of the septic patient. It has become even more important in the cost containment era. The introduction of new antimicrobial agents that are very costly when compared with older agents has greatly complicated antimicrobial therapy. It is now possible to use antibiotics correctly in a therapeutic sense, but to misuse them in an economic sense. Nonetheless, recognition of new pathogens, increasing resistance of pathogens to older antimicrobial agents and more complicated clinical problems have increased the need for these new agents. The following discussion will focus on those factors upon which the selection of a specific antimicrobial agent (or combination of agents) is based, the least of which should be the results of susceptibility tests.

Previous papers in this series have discussed the design of epidemiologic studies and analysis for data that can be presented in a 2 × 2 table. The purpose of this paper is to explain how sample sizes are determined for unmatched prospective and retrospective studies. One of the most common questions asked in planning a project is “What sample size do I need for my study?” Determining the correct sample size is an important consideration that is best resolved prior to data collection. In this paper we assume the goal of the study is to examine the association between a dichotomous risk factor and the presence or absence of disease; therefore, the analysis will be that of a 2 × 2 table. We also assume throughout that sample sizes are to be equal for the two comparison groups (ie, equal numbers of exposed and unexposed for a prospective study, and equal numbers of cases and controls for a retrospective study). Sample size curves used to determine sample sizes will be presented. The results here are based on a paper by Schlesselman that contains sample size formulae for these study designs, and to which the reader may refer for an excellent technical discussion of their derivation.