Gulf War veterans' illnesses: a case study in causal inference

The objective of this study was to determine from the published epidemiological literature whether there is evidence for a causal association between service in the Gulf War and illness in U.S. veterans. Eleven published studies were analyzed using standardized epidemiologic criteria for assessing causality. A consistent association was found between deployment to the Gulf and self-reports of symptoms. No consistency was seen in physical findings or laboratory results. Strength of association varied with different study designs. Dose-response information is limited, because of lack of quantitative data on exposures. Biological plausibility varies for different risk factors. Specificity of association is not seen. Frequency of self-reported symptoms is increased in U.S. Gulf War veterans compared to other veterans of the same era, but specific causes of illnesses cannot be ascertained. Major gaps in data that impeded this analysis include (1) lack of objective data on specific environmental exposures (2) lack of baseline health assessments, and (3) lack of objective measures of post-deployment health status. In future deployment of U.S. troops, accurate exposure and health data will be needed if the causes of subsequent illnesses are to be accurately assessed.

Changes in bacterial flora associated with skin damage on hands of health care personnel

In a prospective observational study of 40 nurses (20 with diagnosed hand irritation and 20 without), nurses with damaged hands did not have higher microbial counts (P = .63), but did have a greater number of colonizing species (means: 3.35 and 2.63, P = .03). Although numbers were small, nurses with damaged hands were significantly more likely to be colonized with Staphylococcus hominis (P = .03). Fifty-nine percent of S hominis isolates from nurses with damaged hands were resistant to methicillin compared with 27% of isolates from those with healthy skin (P = .14). Twenty percent of nurses with damaged hands were colonized with Staphylococcus aureus compared with none of the nurses with normal hands (P = .11). Nurses with damaged hands were also twice as likely to have gram-negative bacteria (P = .20), entercocci (P = .13), and Candida (P = .30) present on the hands. Antimicrobial resistance of the coagulase-negative staphylococcal flora (with the exception of S hominis) did not differ between the 2 groups, nor did a trend toward increasing resistance exist when compared with other studies during the past decade. Skin moisturizers and protectant products were used almost universally by nurses at work, primarily products brought from home. Efforts to improve hand condition are warranted because skin damage can change microbial flora. Such efforts should include assessment or monitoring of hand care practices, formal institutional policy adoption and control of use of skin protectant products or lotions, and prudent use of latex gloves or more widespread use of powder-free and nonlatex products.

Behavioral interventions to improve infection control practices

No single intervention has been successful in improving and sustaining such infection control practices as universal precautions and handwashing by health care professionals. This paper examines several behavioral theories (Health Belief Model, Theory of Reasoned Action and Theory of Planned Behavior, self-efficacy, and the Transtheoretic Model) and relates them to individual factors, also considering interpersonal and organizational factors. Further, this article includes recommendations of individual and organizational components to be addressed when planning a theoretically based intervention for improving infection control practices. A hypothetic framework to enhance handwashing practice is proposed.

A multifaceted approach to changing handwashing behavior

BACKGROUND

Few interventions to influence handwashing have had measurable effects. This prospective quasi-experimental study was designed to address predisposing, enabling, and reinforcing factors to improve frequency of handwashing.

METHODS

Over a 12-month time period, a multifaceted intervention including focus group sessions, installation of automated sinks, and feedback to staff on handwashing frequency was implemented in one intensive care unit; a second unit served as a control. Dependent variables observed were handwashing frequency and self-reported practices and opinions about handwashing. Study phases included baseline, three phases of about 2 months each in duration in which sink automation was incrementally increased, and follow-up 2 months after intervention.

RESULTS

During 301 hours of observation, 2624 handwashings were recorded. Proportion of times hands were washed varied by indication, ranging from 38% before invasive procedures to 86% for dirty-to-clean procedures (p < 0.00001). Although there were some significant differences between experimental and control units in handwashing during the study, these differences had returned to baseline by the 2-month follow-up. There were no significant differences in self-reported practices and opinions from before to after intervention nor between units.

CONCLUSIONS

Intensive intervention, including feedback, education, and increased sink automation, had minimal long-term effect on handwashing frequency.

Hand washing: a ritual revisited

This article summarizes a critical analysis of 18 studies that examined the link between hand washing and infection. Although several factors made it difficult to evaluate the effect of hand washing, the authors conclude that (1) hand washing can add incremental value to infection-control strategies in acute care settings, (2) patient hand hygiene may influence infection rates, and (3) the effect of "ideal" hand washing on nosocomial infection rates is unlikely to be quantifiable.

Consistency and cost of home wound management by contract nurses

The purposes of this descriptive study were to determine the consistency of wound care provided by home health care nurses with the physician's order and the agency's written wound protocol; and the type and cost of wound-management products and nursing services associated with home care. Eleven registered nurses were observed providing wound care in 117 home visits to 31 patients. In addition to the observational component, data were also collected by chart review to facilitate comparison of observed vs. documented care. There were statistically significant differences between observed care given and that which was documented in patients' records. Consistency of care for individual patients was high when care was delivered by the same nurse, but lower when different nurses were involved. The mean dollar value of supplies used for all visits was $9.40, and the average nursing charge per visit was $89.

In vitro survival of skin flora in heparin locks and needleless valve infusion devices

OBJECTIVE

To determine the extent to which two types of infusion devices, the heparin lock and a needleless valve device, allowed the persistence of inoculated microorganisms.

DESIGN

Experiment.

SETTING

Clinical microbiology laboratory.

OUTCOME MEASURES

Bacterial counts in infusion devices.

INTERVENTION

Five of each type of device were inoculated with approximately 10(5) colony-forming units of a strain of Staphylococcus epidermidis and Enterobacter aerogenes and tested for growth at seven time intervals, from 10 minutes to 72 hours after inoculation.

RESULTS

Both strains of bacteria were present in each device at every time interval tested, including 72 hours after inoculation.

CONCLUSIONS

Bacteria introduced into a heparin lock or valve device may be isolated for prolonged periods of time. This suggests that if such devices are contaminated during use, they may be a potential source of infection. The risk of infection should be one major consideration in risk and benefit deliberations when choosing new products.

Wound management in home care: an assessment

The purpose of this study was to determine the status of wound care management among home health agencies (HHAs) in the U.S. The survey was mailed to 562 randomly selected HHAs within all 10 HCFA regions. Two hundred and ninety-six agencies (53%) responded to the survey. The majority (56%) of the respondents did not have written policies and procedures specific to wound management. The investigators were able to analyze the documents of those who had indicated that they had written policies and sent copies of them (n = 87, 67%). The majority (76%) of these had separate wound assessment protocols, but only 14% had separate preventive skin assessment protocols. Sixty-four percent had treatment protocols for more than one type of wound, while the remaining 46% used no specific treatment protocol or used treatments based on physicians' orders only. Most of the HHAs with written wound management policies required documentation of wound status; however, only half had patient/caregiver teaching as part of the policy. Although there is a need for clinically useful, well-tested, standardized approaches to wound care, the results of this study indicated that there is inconsistency regarding wound care management among HHAs within the U.S.

Influence of human T-cell leukemia virus type I tax and rex on interleukin-2 gene expression

The X region of human T-cell leukemia virus type I (HTLV-I) encodes two proteins that regulate viral gene expression. The tax protein is the product of the transactivator gene and has been shown to up-regulate the expression of some cellular genes controlling T-cell replication, including that of the interleukin-2 (IL-2) T-cell growth hormone and the alpha chain of its receptor (IL-2R). Several studies have shown that tax transactivation of the IL-2R alpha-chain promoter is mediated by binding sites for the transcriptional activator NF-kappa B, and this mechanism has also been implicated in the tax activation of IL-2 promoter activity. The rex gene product of HTLV-I regulates viral protein production by influencing mRNA expression and has been implicated in the stabilization of IL-2R alpha-chain mRNA. In the present studies, the ability of the tax and rex proteins to transactivate IL-2 gene expression has been reinvestigated. The ability of the tax protein to transactivate IL-2 promoter activity appears, at least in part, to be mediated by the recognition sequence for a DNA-binding complex known as CD28RC. Consistent with this hypothesis is the observation that tax-mediated activation of IL-2 gene expression is resistant to the immunosuppressive affects of cyclosporin A, a property postulated for the CD28RC binding complex. Unexpectedly, this tax-mediated up-regulation of IL-2 expression is synergized by the presence of the rex protein. These findings demonstrate that transactivation of IL-2 gene expression by tax is augmented by mechanisms distinct from NF-kappa B and raise the possibility that rex, as well as tax, contributes to the oncogenic capability of HTLV-I by altering the expression of the IL-2 gene in T cells infected with this retrovirus.

Handwashing practices and resistance and density of bacterial hand flora on two pediatric units in Lima, Peru

The handwashing practices and bacterial hand flora of 62 pediatric staff members of a teaching hospital in Lima, Peru, were studied. Handwashing followed patient contact 29.3% of the time (204/697 contacts). Mean duration was 14.5 seconds, and significant differences in practices were found by unit (rehydration or neonatal intensive care), type of staff member (nurses or physicians), and type and duration of patient contact. Mean count of colony-forming units was log10 5.87 +/- 0.41, with significant differences in density of flora found between patient care and kitchen staffs. There was no significant effect of handwashing on counts of colony-forming units. Significant differences were also found by unit and by staff position with regard to species isolated and antimicrobial resistance of isolates. A more efficacious and cost-effective form of hand hygiene and a more prudent use of antimicrobial agents are indicated.

A comparison of the auscultated acceleration test and the nonstress test as predictors of perinatal outcomes

In this prospective study, the predictive ability of the nonstress test (NST), the most widely used antepartum screening test to assess fetal well-being, was compared with that of the auscultated acceleration test (AAT) in predicting perinatal outcomes. The AAT is a more easily administered test than the NST, and, unlike the NST, does not use electronic fetal monitors. Study subjects were 205 women with singleton pregnancies greater than 34 weeks' gestation, whose delivery occurred within 7 days of receiving antepartum testing by NST at Johns Hopkins Hospital. The AAT yielded better prediction of poor perinatal outcomes than the NST. The NST, however, was a significantly better predictor of favorable outcomes than the AAT. The AAT has the potential to affect perinatal care if false positive results can be decreased through further research.

Indigenous multiresistant bacteria from flowers in hospital and nonhospital environments

The microbial flora of 60 vase water samples from cut flowers obtained from several environments, including a hospital, were examined in this study. Forty-one different bacterial species were identified, including 12 species of Pseudomonas. The mean total aerobic bacterial count per 500 ml of vase water was 4.5 x 10(8) organisms, and high levels of antibiotic resistance were found. To ascertain the origin of the bacteria found on the flowers as well as their growth patterns, natural cut flowers were compared with sterilized cut flowers in tap water over time. Although the density of organisms was similar, the flora in vase water of sterilized flowers consisted almost entirely of aerobic spore formers while mixed flora of gram-negative bacteria, staphylococci, aerobic spore formers, and fungi were isolated from natural flowers. Our results indicate that the multiply-resistant microbial flora found in vase water is indigenous to flowers, rather than originating from the environment in which they are placed, and that such water is a reservoir of large numbers of multiresistant organisms.

Alcohol-impregnated wipes as an alternative in hand hygiene

The antimicrobial effectiveness of four hand-wash products for health care personnel included three liquid soaps that contained 4% chlorhexidine gluconate, 1% triclosan, or no antiseptic ingredient, respectively, and a 30% w/w ethyl alcohol-impregnated hand wipe. These products were evaluated for reduction in bacterial counts on hands after extended use of 15 handwashes per day for 5 consecutive days. The order of greatest to least log reduction among products at the end of the 5-day test period was chlorhexidine gluconate (2.01), triclosan (1.52), alcohol wipe (0.04), and control soap (0.03). Skin condition before and after handwash was assessed for each treatment group. Subjects reported less skin irritation with alcohol wipes than with the two antiseptic products. Repeated washing with alcohol wipes results in reductions in bacterial colony counts comparable with nonmedicated soap, sufficient to prevent transmission of pathogens by the hands in most situations that arise in nonacute health care settings. This evidence, in addition to increased user acceptability reported by the subjects who used alcohol wipes, suggests that alcohol wipes are an acceptable alternative to soap-and-water handwashing in nonacute health care settings.

Increasing clinical use of pulse oximetry

Changing the system from measuring blood gases through invasive measures to using noninvasive pulse oximetry is a challenge in the Critical Care Unit where invasive techniques are taken for granted. The authors report a project that was successful in increasing the use of noninvasive monitoring techniques by critical care nurses in a Surgical Intensive Care Unit. This clinical project became an important aspect of incorporating the staff in a change to more extensive use of pulse oximetry. This study defines the change in nursing practice with the use of pulse oximetry. The authors discuss three areas: (1) demonstration of the correlation between O2Sat as measured by the pulse oximeter and arterial blood gas saturations; (2) introduction of the pulse oximeter as a reliable alternative to ABGs when monitoring oxygenation; and (3) the establishment of guidelines for using pulse oximetry within the clinical setting.

Alcohol for surgical scrubbing?

To test the effects of four surgical scrub products on colonizing hand flora, 60 healthy adult volunteers were assigned by block randomization (12 subjects per group) to use one of the following formulations: 70% ethyl alcohol with 0.5% chlorhexidine gluconate (ALC); a liquid detergent base containing 1% triclosan (TRI); a liquid detergent base containing 4% chlorhexidine gluconate (CHG); a liquid detergent base containing 7.5% povidone-iodine (PI); or a nonantimicrobial liquid soap (control). Using standard protocol, subjects performed a surgical scrub daily for five consecutive days. Hand cultures were obtained at baseline and on test days 1 and 5 immediately after the scrub and following four hours of gloving. After the first and last scrubs, ALC, CHG and PI resulted in significant reductions in colonizing flora when compared to the control. Additionally, by day 5 ALC was associated with an almost 3-log reduction as compared to an approximate 1.5-log reduction for CHG and PI and less than a 1-log reduction of TRI and the control (p = .009). After four hours of gloving on both days 1 and 5, microbial counts on hands of subjects using ALC, TRI and CHG were significantly lower than counts for the control (p less than .001), whereas there was no significant difference in counts between the PI and control groups (p = .41). Skin assessment by study subjects rated products from least to most harsh as follows: control, TRI, CHG, ALC and PI p = .00001). It was concluded that ALC could be an efficacious and acceptable alternative for surgical scrubbing.

Microbial flora on the hands of health care personnel: differences in composition and antibacterial resistance

The composition and antibiotic sensitivity pattern of bacteria recovered from the hands of nurses and physicians in two service units of a major teaching hospital were compared with those found in a control population. Significant differences in the composition of bacteria were found in dermatology and oncology unit personnel. Staphylococcus aureus was recovered from 31% of dermatology nurses and 37% of dermatology physicians compared with 20% of oncology nurses, 15% of oncology physicians, and 17% of controls. Oncology personnel had a significantly higher carriage of gram-negative bacteria, yeasts, and multiple antibiotic-resistant, aerobic coryneforms (group JK bacteria). Both dermatology and oncology nursing personnel were colonized by organisms resistant to multiple antibiotics. Methicillin resistance was found in 26% and 66% of the staphylococci recovered from dermatology and oncology nurses respectively. Flora from physicians on the two units had sensitivity patterns similar to controls.

Composition and density of microflora in the subungual space of the hand

There were significant quantitative differences in the composition and density of microflora in different areas of the hands of 26 adult volunteers. The subungual spaces had an average log10 CFU of 5.39, compared with a range from 2.55 to 3.53 for other hand sites. In quantitative cultures from five subungual spaces in 26 subjects, coagulase-negative staphylococci were the dominant organisms, with Staphylococcus epidermidis, S. haemolyticus and S. hominis being the most frequently isolated species. Other bacteria recovered from subungual spaces included gram-negative bacilli in 42.3% of subjects, with Pseudomonas species composing 31.3% of this group, and coryneforms in 42.3% of subjects, with multiply resistant JK group coryneforms making up 12.5%. Yeasts were isolated from 69.0% of subjects sampled, with 51.3% of the yeasts identified as Candida parapsilosis. The subungual coagulase-negative staphylococci were susceptible to most antibiotics, with resistance to penicillin, ampicillin, and erythromycin detected in 23 to 38% of isolates.