Much of clinical and hospital epidemiology involves the identification andenumeration of cases or the comparison of case frequencies between two ormore groups of interest. Because both of these activities involve the use ofstatistics, it is important to pay careful attention to biostatisticalissues involved in the collection and the analysis of such data.

In this article, the first in a series of biostatistical papers, we discusssome general issues that are important in the design, analysis, andinterpretation of clinical epidemiologic data. Subsequent papers in thisseries will deal with specific methods of analysis, examples of thesemethods of analysis, and limitations and interpretations of the methods.

During three winter seasons prior to 1984-1985 the special care nursery at New England Medical Center experienced respiratory syncytial virus (RSV) epidemics that required closure of the unit. Prior to and during the 1984-1985 winter season, several measures were taken to prevent recurrent nosocomial RSV transmission. In the winters of 1984-1985 and 1985-1986 there were 26 introductions of community-acquired RSV with no transmission of nosocomial cases during 1,688 patient days at risk as compared with 1983-1984 when there were seven cases of nosocomial RSV following six introductions of RSV during 875 patient days at risk (rate = 8 per 1,000 patient days) (P = 0.0016). The institution of many infection control measures including active surveillance, cohorting infected patients, a strict winter visiting policy, and gowning, gloving, and applying mask on contact, was associated with the successful prevention of nosocomial transmission of RSV in this nursery setting.

Introduction of the diagnosis-related group (DRG)-based Medicare Prospective Payment System is one of a series of major innovations that has occurred in the payment and delivery of health care over the past ten years. Changes such as the increased prevalence of health maintenance organizations, preferred provider organizations, third-party utilization review programs, and the peer review organizations for Medicare patients have all altered the way health care is financed and delivered. The DRG-based Medicare Prospective Payment System is the most visible of these changes, given its breadth of application and its radical departure from the previous retrospective reimbursement for hospital care. The Medicare Prospective Payment System has been in effect since October, 1983. As we approach the fifth anniversary of this program, it is a good time to review its history and to make some judgments as to its future.

Conjunctivitis accounted for 5% of nosocomial infections occurring in a university-affiliated pediatric hospital between January 1984 and April 1986. Pseudomonas aeruginosa was recovered from the conjunctiva of 30 patients. The primary diseases of these patients were chronic and debilitating. Eighty percent of patients were under 18 months of age although only 30% of admissions are represented in this age group. Seventy percent of cases occurred in pediatric intensive care unit/neonatal intensive care unit patients. Seventy percent of patients who had antecedent nasopharyngeal/ endotracheal cultures obtained were colonized with P aeruginosa. All patients except one had one or more of the following interventions prior to the onset of conjunctivitis: tracheostomy, endotracheal tube, oxygen by hood, or suctioning. Two children (7.4%) have residual corneal scars. Improvements in eye care including protection of the eye during suctioning, other respiratory care, and nasogastric tube procedures are warranted.

A prospective study of 147 intermediate care ward (ICW) patients for acquisition of gentamicin-resistant Enterobacteriaceae (GRE) was carried out. Fifty (34%) were colonized or infected with one or more strains of GRE. Fifteen of these patients and one nurse were colonized with 22 strains (including ten species) of GRE bearing identical 60 kb plasmids encoding a novel β-lactamase determinant, OHIO-l and ANT (2“). Analysis of the time course of colonized patients on the ICW revealed one probable episode of cross-transmission. Five colonized patients had been residing in the ICW from one to four months prior to study initiation. Eight patients were admitted to the ICW from other hospital areas already colonized and one additional patient acquired colonization on the ICW from an unknown source. Thus, eight of ten patients admitted to the ICW during the prospective study were already colonized on admission to the ICW. To control this level of colonization it would therefore be necessary to direct efforts at limiting admission of colonized patients or attempting to eliminate the strain from persistently colonized patients, rather than trying to limit transmission within the ward.

Infusion therapy for the administration of blood products, fluids, and parenteral nutrition are essential parts of medical practice. The risks associated with such therapy are well documented but frequently unappreciated. Intravascular infusions are the single most common cause of nosocomial bacteremia.’ Many studies have focused on the epidemiology, microbiology, and pathophysiology of these infections.

In recent years, several companies have introduced a multi-lumen intravenous catheter that is placed through the subclavian or internal jugular vein. Unlike a multi-lumen pulmonary artery catheter, the multi-lumen intravenous catheter is designed solely for intravenous access. The first of these catheters was introduced in 1983, and it was soon followed by several others. These catheters have been designed to allow multiple infusions to be given simultaneously. In addition, one of the ports can be used for venous access. The advantages of these catheters are obvious. The clinician is given three ports for use instead of one. The catheter may be used to simplify infusion therapy. In some instances, cut-downs and other invasive procedures (Hickmans, Broviacs, Mediports) may be avoided. It is not surprising that the use of these catheters has grown. In many institutions, multi-lumen catheters account for a majority of all centrally placed catheters, and in some intensive care units, they are used almost exclusively. The enormous growth of the use of the multi-lumen catheter has occurred despite minimal data regarding risk, cost, and the proper procedures needed for its care.

The efficacy of total body showering and incision site scrub with disinfectant agents was evaluated in a randomized, prospective study of 575 patients undergoing selected surgical procedures. Patients who showered twice with 4% chlorhexidine gluconate had lower mean colony counts of skin bacteria at the surgical incision site in the operating room prior to the final scrub than patients who showered twice with povidone-iodine solution or medicated bar soap. Patients in the chlor-hexidine group had no growth on 43% of the incision site skin cultures compared with 16% in the povidone-iodine group and 6% in the soap and water group. Patients who showered and who were scrubbed with chlorhexidine also had lower rates of intraoperative wound contamination. Bacteria were recovered from the wounds of 4% of patients using this regimen compared with 9% for patients who used povidone-iodine and 15% for patients who showered with medicated soap and water and were scrubbed with povidone-iodine. We noted no difference in surgery-specific infection rates among patients in the three treatment groups; however, our sample sizes were too small to evaluate this outcome parameter adequately.

These data suggest that preoperative showering and scrubbing with chlorhexidine is an effective regimen to reduce extrinsic intraoperative contamination of the surgical wound from skin bacteria. The efficacy of this regimen to prevent postoperative wound infection needs to be evaluated in a well-designed, carefully controlled prospective trial with adequate numbers of patients to achieve statistically valid conclusions.

A prospective study of 147 intermediate care ward (ICW) patients for acquisition of gentamicin-resistant Enterobacteriaceae (GRE) was carried out. Fifty (34%) were colonized or infected with one or more strains of GRE. Fifteen of these patients and one nurse were colonized with 22 strains (including ten species) of GRE bearing identical 60 kb plasmids encoding a novel β-lactamase determinant, OHIO-l and ANT (2“). Analysis of the time course of colonized patients on the ICW revealed one probable episode of cross-transmission. Five colonized patients had been residing in the ICW from one to four months prior to study initiation. Eight patients were admitted to the ICW from other hospital areas already colonized and one additional patient acquired colonization on the ICW from an unknown source. Thus, eight of ten patients admitted to the ICW during the prospective study were already colonized on admission to the ICW. To control this level of colonization it would therefore be necessary to direct efforts at limiting admission of colonized patients or attempting to eliminate the strain from persistently colonized patients, rather than trying to limit transmission within the ward.

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.

Gram-positive bacterial infections remain a serious challenge for infectious disease practitioners. Many gram-positive organisms are developing new methods of antibiotic resistance, thus rendering our standard antimicrobial agents ineffective. Penicillin G resistance, methicillin-resistant Staphylococcus aureus and recently, a laboratory strain of vancomycin-resistant S aureus have become catalysts in the search for effective alternatives in the treatment of gram-positive bacterial infections. This review will briefly describe an investigational agent (investigational in the US, although used widely overseas) that has been very successful in treating gram-positive bacterial infections. The antibiotic, named teicoplanin. (Merrell-Dow Pharmaceuticals) may be approved for release in the United States in the future.

Originally known as teichomycin A, teicoplanin is a glycopeptide antibiotic, chemically similar to the vancomycin-ristocetin antibiotic group. Teicoplanin was discovered following the fermentation process of Actinoplanes teichomyceticus The antibiotic has a linear heptapeptide structure and weighs approximately 2000 daltons.

Beyond a potential role in infection surveillance, computers offer epidemiologists several important and unique potentials. Epidemiologists deal with complex interrelationships when interpreting or contributing information in the biomedical literature. Annals of Internal Medicine recently published a series dealing with the general problem of information management. The commentary that follows examines one area in which computers can be applied to our advantage-electronic exchange of information-and presents a practical example of using database searches.

We surveyed 53 randomly chosen Maryland nursing homes for infection control policies and practices. The majority had written infection control policies, an infection control committee, and a designated practitioner for infection control; in most facilities, however, the infection control practitioner had other major duties, spent little time on infection control, and had no specific training in the field. Thirty-four percent of homes in the survey performed routine environmental cultures, and more than half had insufficient or no isolation policies for infected decubiti and acute diarrhea. In general, the intensity of infection surveillance and the extent of infection control measures increased with the level of care provided, from domiciliary homes to homes providing chronic care. Employee health care fared generally well: 60% of homes offered influenza vaccine to employees and 66% had restriction policies for employees with upper respiratory infections. While the majority of homes offered the influenza vaccine to residents, acceptance of other vaccines recommended for the elderly was less widespread. We conclude that infection control efforts are made in most Maryland nursing homes; however, appropriate guidelines and more effort to educate nursing home personnel in proper infection control practices are badly needed.