Standardized infection ratio: the next generation of risk-adjusted wound infection measures

Background

The scientific literature overwhelmingly supports the value of surgical site infection (SSI) surveillance with feedback of infection rates to surgeons. Recognizing that feedback has little value without some form of adjustment for patient factors, surgeons long ago developed the four-level wound classification system. In the 1980s the SENIC multivariate risk index was introduced, followed by the National Nosocomial Infection Surveillance multivariate risk index which is the currently recommended US standard.

Methods

To address recent advances in surgical technology, ambulatory surgery, and differences between the USA and other countries, newer risk-adjustment methods were explored. The ideal approach would overcome the following common practical problems encountered in generating SSI rates such as: (1) new patient-level risk factors, (2) technology-specific factors, (3) small monthly denominators and (4) postdischarge surveillance. Despite increased mathematical complexity, an acceptable risk-adjustment technique must still provide easily interpretable results.

Results

The standardized infection ratio (SIR) is proposed as the next generation in epidemiological approaches for monitoring and feeding back risk-adjusted SSI rates to surgeons. The SIR is a single monthly number for each surgeon, calculated by dividing the observed number of SSIs by the expected number predicted from a complex statistical model of patient-level and technology-specific risk factors. Like the well known standardized mortality ratio, the SIR offers the advantage of providing one generalizable quality index regardless of the mix of operations performed. In addition, the SIR is well suited to control chart techniques and for providing valid interhospital comparisons.

Conclusion

The SIR model offers simple data input and easily interpretable output suitable for application in all healthcare settings worldwide.

Occipital brain perfusion deficits in children with major depressive disorder

Occipital lobe perfusion defects have been identified on regional cerebral blood flow (rCBF) SPECT scans of adolescent children and young adults with major depressive disorder (MDD). We reinvestigated a series of rCBF SPECT scans obtained several years ago on drug-naive children with a clinical diagnosis of MDD and on healthy children.

METHODS

To test whether visually apparent abnormalities in rCBF constitute statistically significant differences between patients, given the relatively small sample sizes, we applied the technique of statistical parametric mapping (SPM).

RESULTS

Two groups of patients were identified: 8 with significant posterior flow deficits in the occipital cortex (Brodmann's areas 18 and 19), usually symmetric, and best visualized on paramedian sagittal sections, and 13 without obvious occipital perfusion deficits but with anterior rCBF deficits in a pattern often described in the literature, attaining statistical significance in the right frontal region. Other localizations in the left frontal and bilateral prefrontal regions did not attain significance, but each localization contained statistically significant maxima (z scores). The scan findings of all 18 healthy children were normal.

CONCLUSION

With the aid of SPM, 2 groups of children with significantly different rCBF behavior were identified. The reason for this difference is not known but should be investigated to determine its possible significance to patients with MDD.

Gene therapy to prevent organophosphate intoxication

The specific hydrolytic activity of PON1 paraoxonase/arylesterase enzymes in liver and blood provides a natural barrier against the entry of organophosphate toxins into the central and peripheral nervous systems. Inherited differences in PON1 enzyme concentrations may determine levels of susceptibility to organophosphate injury in humans. To test whether boosting serum levels of PON1 enzymes by gene therapy might provide increased protection, we compared the degree of inactivation of whole brain acetylcholinesterase of mice exposed to chlorpyrifos 4 days after intravenous injection of recombinant adenoviruses containing PON1-LQ or PON1-LR genes or no PON1 gene. Both recombinant viruses containing PON1 genes boosted serum arylesterase concentrations by approximately 60% and significantly prevented the inactivation of brain acetylcholinesterase. Some mice were completely protected. These findings indicate that boosting serum levels of PON1 enzymes by a gene delivery vector raises the threshold for organophosphate toxicity by hydrolytic destruction before the chemical can enter the brain.

Use of structural equation modeling to test the construct validity of a case definition of Gulf War syndrome: invariance over developmental and validation samples, service branches and publicity

To attempt to replicate the syndrome-like structure identified by exploratory factor analysis of symptom reports from 249 Gulf War veterans of a Naval reserve battalion (the developmental sample), we administered Haley's original symptom questionnaire to 335 Gulf War veterans who served primarily in active-duty US Army units living in North Texas (the validation sample). On the basis of recently validated goodness-of-fit criteria (SRMR<or=0.08, RMSEA<or=0.06, and CFI>or=0.95), a structural equation model (Model 1) with four symptom scales loading on each of three first-order latent syndrome factors fit both the developmental and validation samples well and was invariant across both samples. Additional models validated a higher-order latent factor (a single Gulf War syndrome) explaining the variances and covariances of the first-order factors, four additional symptom scales loading on the higher-order factor, and four possible secondary factor loadings that also fit the data well. All structural models were invariant across cohorts of the validation sample surveyed before and after intense publicity following publication of the case definition. These findings suggest that the apparent syndrome structure of a single Gulf War syndrome with three variants may be found widely and justify a confirmatory sample survey of Gulf War-era veterans.

Serum paraoxonase (PON1) isozymes: the quantitative analysis of isozymes affecting individual sensitivity to environmental chemicals

In a recent study on Gulf War veterans who developed delayed neurotoxicity symptoms, we found their levels of serum paraoxonase (PON1) isozyme type Q to be significantly lower than in the control, unaffected veteran group. These results were obtained in 25 ill veterans and 20 well control subjects, of which 10 were deployed and 10 were nondeployed battalion members who remained in the United States during the Gulf War. The blood samples were also assayed for serum butyrylcholinesterase in our laboratory, and more recently in Dr. C. Broomfield's laboratory for somanase and sarinase activities. The cholinesterase activities showed no significant correlation with the PON1 isozyme levels or the severity of the clinical symptoms, but the somanase and sarinase levels ran parallel to the PON1 type Q isozyme concentrations. Although there is no direct evidence that these Gulf War veterans were directly exposed to or encountered either of these nerve gases, they may have been exposed to some environmental or chemical toxin with a similar preference for hydrolysis by the PON1 type Q isozyme. The number of subjects is relatively small, but the results should encourage other investigators to examine both the individual phenotypes and the levels of PON1 isozymes in other groups exhibiting neurological symptoms.

Commercial tattooing as a potentially important source of hepatitis C infection. Clinical epidemiology of 626 consecutive patients unaware of their hepatitis C serologic status

Tattooing in commercial tattoo parlors is known to transmit blood-borne viral infections, including hepatitis C virus (HCV), in other countries, but its contribution to the high population prevalence of HCV infection in the United States has been incompletely evaluated. Risk factors for blood-borne infection were assessed by physician's interview of 626 consecutive patients undergoing medical evaluation for spinal problems in 1991 and 1992 while unaware of their HCV status. Later all were screened for HCV infection with enzyme-linked immunosorbent assay (EIA-1 and EIA-2), and positives were confirmed with second-generation recombinant immunoblot assay (RIBA). Forty-three patients were seropositive for HCV (sample prevalence 6.9%, population-standardized prevalence 2.8%). Logistic regression analysis identified 4 independent risk factors for HCV infection: injection-drug use (adjusted prevalence odds ratio [OR] = 23.0; 95% confidence intervals [CI] = 7.5-70.6), ancillary hospital jobs held by men (OR = 9.6; 95% CI = 3.8-24.3), tattoos from commercial tattoo parlors (OR = 6.5; 95% CI = 2.9-14.8), and drinking > or = 3 6-packs of beer per month (OR = 4.0; 95% CI = 1.8-8.7). If causal, these 4 risk factors account for 91% of HCV infections, with tattooing explaining 41%, heavy beer drinking 23%, injection-drug use 17%, and ancillary health care jobs for men 8%. Transfusions, promiscuous sexual activity, bone grafts, acupuncture, perinatal or intimate transmission in families, and other modes were not independently associated with serologic evidence of HCV infection. Unlikely to be explained by confounding or incomplete disclosure of other risk factors, tattooing in commercial tattoo parlors may have been responsible for more HCV infections than injection-drug use.

Use of a secure Internet Web site for collaborative medical research

Researchers who collaborate on clinical research studies from diffuse locations need a convenient, inexpensive, secure way to record and manage data. The Internet, with its World Wide Web, provides a vast network that enables researchers with diverse types of computers and operating systems anywhere in the world to log data through a common interface. Development of a Web site for scientific data collection can be organized into 10 steps, including planning the scientific database, choosing a database management software system, setting up database tables for each collaborator's variables, developing the Web site's screen layout, choosing a middleware software system to tie the database software to the Web site interface, embedding data editing and calculation routines, setting up the database on the central server computer, obtaining a unique Internet address and name for the Web site, applying security measures to the site, and training staff who enter data. Ensuring the security of an Internet database requires limiting the number of people who have access to the server, setting up the server on a stand-alone computer, requiring user-name and password authentication for server and Web site access, installing a firewall computer to prevent break-ins and block bogus information from reaching the server, verifying the identity of the server and client computers with certification from a certificate authority, encrypting information sent between server and client computers to avoid eavesdropping, establishing audit trails to record all accesses into the Web site, and educating Web site users about security techniques. When these measures are carefully undertaken, in our experience, information for scientific studies can be collected and maintained on Internet databases more efficiently and securely than through conventional systems of paper records protected by filing cabinets and locked doors. JAMA. 2000;284:1843-1849.

Alternative case definitions of ventilator-associated pneumonia identify different patients in a surgical intensive care unit

Diagnostic criteria that define ventilator-associated pneumonia (VAP) remain controversial. The purpose of this study was to evaluate common definitions of VAP and determine their relationship to each other and clinical treatment. This study prospectively evaluated several diagnostic criteria that define VAP in a cohort of 255 consecutive SICU patients ventilated for < 48 h. Definitions evaluated include the CDC definitions, the Johanson definitions which do not rely on culture data, the Physician's Probable diagnosis which relies on positive quantitative cultures, and the antibiotic treatment group. Forty-four patients (17%) received antibiotic treatment for VAP. Depending on the definition evaluated, criteria were met for a diagnosis of VAP from as low as 4% of patients by the Johanson definition to as high as 48% of patients by the CDC definition. There was poor agreement among the definitions in their ability to select the same patient as having VAP. Besides duration of mechanical ventilation and tube feeding, which were risk factors that predicted meeting the criteria for all groups, risk factors predicting VAP varied among the definitions. This study demonstrates that in a surgical ICU, the candidate definitions of pneumonia evaluated show little agreement. The particular case definition chosen to diagnose VAP will determine the incidence rate of pneumonia, the time to onset of pneumonia, and the risk factors of the type of patient treated.

Effect of basal ganglia injury on central dopamine activity in Gulf War syndrome: correlation of proton magnetic resonance spectroscopy and plasma homovanillic acid levels

BACKGROUND

Many complaints of Gulf War veterans are compatible with a neurologic illness involving the basal ganglia.

METHODS

In 12 veterans with Haley Gulf War syndrome 2 and in 15 healthy control veterans of similar age, sex, and educational level, we assessed functioning neuronal mass in both basal ganglia by measuring the ratio of N-acetyl-aspartate to creatine with proton magnetic resonance spectroscopy. Central dopamine activity was assessed by measuring the ratio of plasma homovanillic acid (HVA) and 3-methoxy-4-hydroxyphenlyglycol (MHPG).

RESULTS

The logarithm of the age-standardized HVA/MHPG ratio was inversely associated with functioning neuronal mass in the left basal ganglia (R(2) = 0.56; F(1,27) = 33.82; P<.001) but not with that in the right (R(2) = 0. 04; F(1,26) = 1.09; P =.30). Controlling for age, renal clearances of creatinine and weak organic anions, handedness, and smoking did not substantially alter the associations.

CONCLUSIONS

The reduction in functioning neuronal mass in the left basal ganglia of these veterans with Gulf War syndrome seems to have altered central dopamine production in a lateralized pattern. This finding supports the theory that Gulf War syndrome is a neurologic illness, in part related to injury to dopaminergic neurons in the basal ganglia.

Brain abnormalities in Gulf War syndrome: evaluation with 1H MR spectroscopy

PURPOSE

To test for neuronal brain damage in the basal ganglia and brainstem in Gulf War veterans by using magnetic resonance (MR) spectroscopy.

MATERIALS AND METHODS

Twenty-two Gulf War veterans with one of three factor analysis-derived syndromes (case patients); 18 well veterans matched for age, sex, and education level (control subjects); and six Gulf War veterans with syndrome 2 from a different population (replication sample) underwent long echo time (272 msec) proton (hydrogen 1) MR spectroscopy on a 4 x 2 x 2-cm voxel in the basal ganglia bilaterally and a 2 x 2 x 2-cm voxel in the pons. Syndromes 1-3 are described as "impaired cognition," "confusion-ataxia," and "central pain," respectively.

RESULTS

The N-acetylaspartate-to-creatine (NAA/Cr) ratio, which reflects functional neuronal mass, was significantly lower in the basal ganglia and brainstem of Gulf War veterans with the three syndromes than in those structures of the control subjects (P =.007). The finding was corroborated in the replication sample (P =.002). Veterans with syndrome 2 (the most severe clinically) had evidence of decreased NAA/Cr in both the basal ganglia and the brainstem; those with syndrome 1, in the basal ganglia only; and those with syndrome 3, in the brainstem only.

CONCLUSION

Veterans with different Gulf War syndromes have biochemical evidence of neuronal damage in different distributions in the basal ganglia and brainstem.

Stressful manipulations that elevate corticosterone reduce blood-brain barrier permeability to pyridostigmine in the Rat

Pyridostigmine bromide (PB), a reversible inhibitor of acetylcholinesterase (AChE), is used for the treatment of myasthenia gravis. PB has also been provided to military personnel for preexposure protection against potential soman release. The entry of PB into the brain is typically minimal, but recently published data in mice suggest that a brief forced swim stress increases the permeability of the blood-brain barrier to PB. From these results, PB administered under stressful conditions was proposed to induce long-lasting central cholinergic deficits, potentially explaining the neurological and neuropsychological symptoms presented by some Gulf War veterans. In undertaking to replicate these results in the Long-Evans rat, no evidence of a stress-potentiated central effect of PB, administered at doses up 5.0 mg/kg ip, was found. Three stress protocols were used: restraint, forced swim, or a combined restraint/forced swim. Wistar rats were also tested in some of the protocols to ensure that the results were generalizable across rat strains, and plasma corticosterone levels were measured to test the effectiveness of the stressors employed. In contrast to the previously reported findings in the mouse, stress significantly reduced the entry of PB into rat brain, as measured by reduced inhibition of AChE activity: a 12.5% reduction in whole brain AChE activity after treatment with 5.0 mg/kg PB under control conditions declined to 9% after stress exposure. It is apparent, therefore, that the interaction between stress and PB requires further study, and previous data should be reassessed before they are used as a basis for interpreting symptoms presented by veterans.

Vestibular dysfunction in Gulf War syndrome

METHODS

Vestibular complaints of Gulf War veterans were characterized by a nested case-control study of 23 veterans with 3 different Gulf War syndromes and 20 matched control subjects. All subjects completed a standardized symptom questionnaire and underwent standard audiovestibular tests administered by audiologists blinded to group identities.

RESULTS

The prevalence of reported dizzy spells was higher in veterans with Gulf War syndromes 1 (100%), 2 (85%), and 3 (100%) than in controls (25%, P < 0.0001). Dizzy spells were more frequent, lasted longer, and involved a wider variety of accompanying symptoms in veterans with syndrome 2 than in those with syndromes 1 and 3. Audiovestibular testing showed greater interocular asymmetry of nystagmic velocity on sinusoidal harmonic acceleration in syndromes 1 (P = 0.015) and 2 (P = 0.002), greater asymmetry of saccadic velocity in syndrome 2 (P = 0.4), diminished nystagmic velocity after caloric stimulation bilaterally in syndrome 3 (P = 0.02 to 0.04), more subjects with pathologic nystagmus (P = 0. 09), and greater interside asymmetry of wave I to III interpeak latency on auditory brain stem response in syndromes 1 (P = 0.005) and 2 (P = 0.07). Asymmetry of gain on sinusoidal harmonic acceleration and pathologic nystagmus were most strongly associated with symptoms of paroxysmal vertigo (P = 0.002 and 0.07, respectively); asymmetry of saccadic velocity, with the severity of vertigo (P = 0.004); and abnormal caloric response, with chronic dysequilibrium (P = 0.006).

CONCLUSIONS

The findings are compatible with a subtle neurologic injury from organophosphate-induced delayed neurotoxicity.

Association of low PON1 type Q (type A) arylesterase activity with neurologic symptom complexes in Gulf War veterans

Previously Haley et al. described six possible syndromes identified by factor analysis of symptoms in Gulf War veterans and demonstrated that veterans with these symptom complexes were more neurologically impaired than age-sex-education-matched well controls. They also uncovered strong associations (relative risks 4-8) suggesting that these symptom complexes were related to wartime exposure to combinations of organophosphate pesticides, chemical nerve agents, high concentration DEET insect repellant, and symptoms of advanced acute toxicity after taking pyridostigmine. Here we have shown that compared to controls, ill veterans with the neurologic symptom complexes were more likely to have the R allele (heterozygous QR or homozygous R) than to be homozygous Q for the paraoxonase/arylesterase 1 (PON1) gene. Moreover, low activity of the PON1 type Q (Gln192, formerly designated type A) arylesterase allozyme distinguished ill veterans from controls better than just the PON1 genotype or the activity levels of the type R (Arg192, formerly designated type B) arylesterase allozyme, total arylesterase, total paraoxonase, or butyrylcholinesterase. A history of advanced acute toxicity after taking pyridostigmine was also correlated with low PON1 type Q arylesterase activity. Type Q is the allozyme of paraoxonase/arylesterase that most efficiently hydrolyzes several organophosphates including sarin, soman, and diazinon. These findings further support the proposal that neurologic symptoms in some Gulf War veterans were caused by environmental chemical exposures.

Is Gulf War syndrome due to stress? The evidence reexamined

Medical policy-makers have concluded that stress from wartime trauma and deployment constitutes an important cause of the chronic physical symptoms observed in US veterans who served in the Persian Gulf War. The author reviewed scientific articles from peer-reviewed journals referenced in the final report of the Presidential Advisory Committee on Gulf War Veterans' illnesses and conducted a MEDLINE literature search. All reported prevalence rates of post-traumatic stress disorder (PTSD) in Gulf War veterans were defined by critical cutpoints on psychometric scales constructed by summing veterans' responses on standardized symptom questionnaires rather than by clinical psychiatric interviews. Observed PTSD rates varied from 0% to 36% (mean, 9%). Correcting for measurement errors with previously determined values of the sensitivity (range 0.77 to 0.96) and specificity (range 0.62 to 0.89) of the psychometric tests yielded estimated true PTSD rates of 0% for 18 of the 20 reported rates. Mean scores on the Mississippi PTSD scale in all subgroups of Gulf War veterans were within the range of values for well-adjusted Vietnam veterans (50-89) and far below that of Vietnam veterans with psychiatrically confirmed PTSD (120-140). Most PTSD and "stress-related symptoms" reported in studies of Gulf War veterans appear to represent false-positive errors of measurement reflecting nonspecific symptoms of other conditions.

Evaluation of neurologic function in Gulf War veterans. A blinded case-control study

OBJECTIVE

To determine whether Gulf War-related illnesses are associated with central or peripheral nervous system dysfunction.

DESIGN

Nested case-control study.

PARTICIPANTS

Twenty-three veterans with factor analysis-derived syndromes (the cases), 10 well veterans deployed to the Gulf War (the deployed controls), and 10 well veterans not deployed to the Gulf War (the nondeployed controls).

METHOD

With investigators blinded to group identities, participants underwent objective neurophysiological, audiovestibular, neuroradiological, neuropsychological, and blood tests.

MAIN OUTCOME MEASURES

Evidence of neurologic dysfunction.

RESULTS

Compared with the 20 controls, the 23 cases had significantly more neuropsychological evidence of brain dysfunction on the Halstead Impairment Index (P=.01), greater interside asymmetry of the wave I to wave III interpeak latency of brain stem auditory evoked potentials (P=.02), greater interocular asymmetry of nystagmic velocity on rotational testing, increased asymmetry of saccadic velocity (P=.04), more prolonged interpeak latency of the lumbar-to-cerebral peaks on posterior tibial somatosensory evoked potentials (on right side, P=.03, and on the left side, P=.005), and diminished nystagmic velocity after caloric stimulation bilaterally (P values range from .02 to .04). Cases (n=5) with syndrome 1 ("impaired cognition") were the most impaired on brain stem auditory evoked potentials (P=.005); those (n=13) with syndrome 2 ("confusion-ataxia") were the most impaired on the Halstead Impairment Index (P=.006), rotational testing (P=.01), asymmetry of saccadic velocity (P=.03), and somatosensory evoked potentials (P< or =.01); and those (n=5) with syndrome 3 ("arthro-myo-neuropathy") were the most impaired on caloric stimulation (P< or =.01).

CONCLUSIONS

The 3 factor-derived syndromes identified among Gulf War veterans appear to represent variants of a generalized injury to the nervous system.

Neuropsychological correlates of Gulf War syndrome

As part of a comprehensive multispecialty project, the present study reports on the neurocognitive and psychological function of veterans who report Persian Gulf War-related symptoms. The neuropsychological and psychological performances of 26 ill Gulf War veterans were compared to 20 well veterans from the same military unit. Neurocognitive functions assessed included intelligence, abstraction and problem-solving, attention and concentration, memory and learning, language and visual-spatial function, and sensorimotor abilities. Psychological function was measured by self-report questionnaires. Results indicated global and consistently poorer intellectual and neurocognitive function among the ill veterans compared to the control veterans. A generalized pattern of neuropsychological deficit was evident for the ill veterans. Psychological profiles of the ill veterans were similar to those in general medical patients. Based on these findings and results from the multispecialty investigation, we conclude that some of the ill veterans have experienced neurotoxic injury resulting in chronic neuropsychological impairment that is related to their service in the Persian Gulf War.

The scientific basis for using surveillance and risk factor data to reduce nosocomial infection rates

Research over the past 20 years has demonstrated that an active programme of surveillance with feedback of surgical wound infection rates to surgeons can reduce subsequent rates by 30-40%. For surveillance data and feedback to be meaningful and influential, however, certain rigorous methodological principles must be observed. First, surveillance data must be collected in an accurate, efficient and confidential manner. This requires written definitions of infection, regular clinical case-finding, post-discharge follow up for short-staying patients, and computer storage, analysis and reporting of the data in coded form that does not publicly identify individuals. Second, the variation in intrinsic risk of the patients of the various surgeons must be controlled for by stratifying the final infection rates on a multivariate risk index, which combines the traditional classes of wound contamination with measures of intrinsic patient susceptibility. This can be accomplished with a relatively small commitment of time by the Infection Control Nurse with the aid of sophisticated computer software that is now available.

Eradication of endemic methicillin-resistant Staphylococcus aureus infections from a neonatal intensive care unit

To control infections with endemic methicillin-resistant Staphylococcus aureus (MRSA) in a neonatal intensive care unit (NICU), triple dye was applied to the umbilical cords of infants in the intermediate-care but not the intensive-care area. The rate of MRSA infection, adjusted for time and intensity of care, decreased in the intermediate-care area (rate ratio, 0.35; 95% confidence interval [CI], 0.14-0.87; P < .01) but not in the intensive-care area (rate ratio, 0.92; 95% CI, 0.41-2.24; P = .48). After 22 months, the rate increased in both areas (Mantel-Haenszel rate ratio, 1.7; 95% CI, 1.0-2.8; P < .05) after overcrowding and understaffing increased. After temporary reduction of overcrowding and understaffing, extension of triple dye use to the intensive-care area and dedication of an infection control nurse to the NICU, MRSA colonization and infection rates decreased to near zero in both areas (infection rate ratios, 0.09 and 0.11, respectively; P < .005). The endemic MRSA strain, identified by pulsed-field gel electrophoresis, was eradicated.

Initial STD visits by adolescent females, Dallas County, Texas

Sexually active adolescents are at high risk for unintended pregnancy and sexually transmitted diseases. We identified a cohort of 6089 women aged 13 to 21 years who in 1985 either delivered an infant at Parkland Hospital (3154) or used the outpatient clinics (2935). We reviewed those records and others in the Dallas County sexually transmitted disease (STD) clinic for the period 1983 through 1988 to determine the prevalence of visits to the Dallas County STD clinic by pregnant and nonpregnant adolescents, STD diagnoses at initial visit, whether STD visits and diagnoses varied by pregnancy status or ethnic background, and how the women were referred to the STD clinic. Similar percentages of nonpregnant (13%) and pregnant (11%) women had an STD visit, and most adolescents who visited the STD clinic had an STD diagnosis made. Gonorrhea and syphilis were the STDs diagnosed most frequently in both pregnant and nonpregnant adolescents. Regardless of pregnancy status, black adolescents were 7.3 times more likely to have an STD visit. Referral by a sexual partner was the second most common reason that adolescents sought STD care. Our study provides only minimal estimates of STDs among adolescents but indicates STDs, regardless of pregnancy status, are a more common medical problem than has been reported previously. Referral by sexual partners may be an important way to reach adolescents at risk for STDs.

The necessity and efficiency of wound surveillance after discharge

A surgical wound surveillance program followed up 16,453 consecutive patients from 1983 through 1988. Patients were followed up for 30 days after operation, and 516 (35%) of the surgical wound infections first became manifest after discharge. In-hospital surveillance alone would have estimated the surgical wound infection rate to be 5.8% when the true rate was 8.9%. Infections that occurred after discharge were more likely in clean operations, in shorter operations, in obese patients, and in nonalcoholic patients. The probability that infections would begin after discharge was inversely associated with the duration of postoperative stay in the hospital. Postdischarge follow-up of patients who previously have undergone surgery is necessary to avoid underestimated of the infection rates and biases related to known risk factors. The most efficient time to survey patients appears to be at 21 days after the operation, at which time 90% of surgical wound infections have occurred.

Nosocomial infections in surgical patients: developing valid measures of intrinsic patient risk

For surgeons or hospitals to compare their rates of wound infection meaningfully, the analysis must first control for the mix of intrinsic infection risk of their patients. Research over the past century has led to the development of several intrinsic risk indexes that can be used to stratify the wound infection rates so that valid comparisons can be made within risk strata. For an intrinsic risk index to be useful for comparing rates, it must control for all of the important intrinsic risk constructs; merely being statistically associated with infection rates does not ensure that a risk index will be useful. Understanding how a risk index can be both parsimonious and comprehensive requires consideration of the competing principles of multicollinearity and orthogonality. Various techniques of multivariate analysis are used to develop multivariate risk indexes, but the success of the process depends on having all of the important orthogonal risk constructs represented in the pool of predictor variables available for the analysis, either directly by variables in the pool or by demonstrated multicollinearity. Despite recent advances in risk measurement, many important questions remain.

Measuring the costs of nosocomial infections: methods for estimating economic burden on the hospital

To compete more effectively for resources, it is increasingly important for infection control practitioners to estimate the costs of nosocomial infections and the amount of money their infection control programs save the hospital. Studies on costs should estimate both extra length of stay and extra costs attributable to infectious complications. Cost estimates should either adjust charges by a cost to charge ratio, which is relatively easy, or estimate hospital costs directly by detailed cost-accounting, which is comparatively more difficult. If there is insufficient time to measure costs concurrently in every infected patient, comparative studies can be done by comparing infected and uninfected patients matched on characteristics that control for the preexisting differences between them. Diagnosis-related groups and the number of diagnoses appear to be useful matching variables because they are strongly associated with both nosocomial infection and length of stay. The final results should be expressed as either potential or actual savings to the hospital, depending on whether significant reductions in nosocomial infection rates have been achieved.

Tuberculosis epidemic among hospital personnel

Six employees of the emergency department at Parkland Memorial Hospital developed active tuberculosis in 1983-1984. Five of the cases occurred four to 12 months after exposure to the index case, a patient with severe cavitary tuberculosis seen in the emergency department in April 1983. One resident physician developed cavitary disease after exposure to this patient. An additional employee case may have resulted from transmission from one of the initial employee cases. One immunocompromised patient may have acquired tuberculosis as a result of exposure to the index case. In addition, the tuberculin skin tests of at least 47 employees exposed to the index case converted from negative to positive. Of 112 previously tuberculin-negative emergency department employees who were tested in October 1983, 16 developed positive skin tests, including the 5 employees with active disease. Fifteen of these new positives had worked on April 7, 1983, while the index case was in the emergency department (X2 = 20.6, P less than 0.001). Factors related to the genesis of the epidemic included the disease characteristics in the index case and the recirculation of air in the emergency department. This investigation indicates that city-county hospital emergency department employees should be screened at least twice a year for evidence of tuberculosis and that the employee health services of such hospitals should regard the surveillance of tuberculosis infection among personnel at a high-priority level.

Surgical wound infections

At the last meeting of the SURGIKOS Operating Room/Infection Control Combined Advisory Panel, guest speakers reviewed current thinking and practices that have an impact on surgical wound infections. Presentation topics included the role of the CDC, nursing and classification systems in controlling wound infections; cost effective epidemiologic methods of reducing infections; and outcomes of surgical wound infection prevention practices deemed critical by the Joint Commission on Accreditation of Healthcare Organizations. The following is a brief overview of the presentations that guided panel discussions on these topics.

Factors which influence the risk of wound infection in trauma patients

Surgical wound infections following traumatic injury remain a source of morbidity and mortality. A simple system for estimating the risk of infectious complications was evaluated in 949 trauma patients requiring operative therapy. The majority of cases were caused by penetrating trauma (784). Truncal, neck, and extremity procedures were included. The overall wound infection rate was 7%. Infection rates were related to amount of bacterial contamination and mechanism of injury. Age, type of antibiotics, and delay time from injury to operation were not risk factors for any injury type. Wound classification, shock, blood loss, number of organs injured, and operative time were significant risk factors, but had different effects on infection rate related to injury type. Multivariate analysis revealed no significant infectious risk factors for stabwounds. Significant factors were wound class (p = 0.02) and shock (p = 0.001) for gunshot wounds, wound class (p = 0.03) and number of organs injured (p = 0.01) for blunt trauma, and blood loss (p = 0.01) for shotgun wounds. This classification system can be used to review outcome and compare trauma patient populations for infectious morbidity in a more uniform fashion.

The financial incentive for hospitals to prevent nosocomial infections under the prospective payment system. An empirical determination from a nationally representative sample

To clarify the financial incentives for hospitals to prevent nosocomial infections, we analyzed 9423 nosocomial infections identified in 169 526 admissions selected randomly from the adult admissions to a random sample of US hospitals. By classifying each admission into a baseline diagnosis related group (DRG) (after first excluding all diagnoses of nosocomial infection) and a final DRG (after including these diagnoses), we found that only 5% to 18% of nosocomial infections would have caused the admission to be reclassified to a higher-paying DRG, depending on the extent to which physicians recorded nosocomial infection diagnoses in patients' medical records. The extra payment from the reclassification, averaged over all nosocomial infections, would have been no more than $93 per infection (in 1985 reimbursement rates), constituting only 5% of the hospitals' costs for treating these infections. Thus, at least 95% of the cost savings obtained from preventing nosocomial infections represents financial gains to the hospital.

Fever in hospitalized patients. With special reference to the medical service

Fever (oral temperature of 38 degrees C or more on two or more consecutive days) during the hospital stay of 4,065 patients admitted to Grady Memorial Hospital during an 11-week period was studied. At least one episode of fever occurred in 1,194 patients (29 percent). Rates of fever were highest on medical and surgical services. Review of 341 episodes of fever in 302 patients on the medical service identified a single potential cause in 56 percent. Multiple factors were present in 26 percent, and no potential causes were found in 18 percent. Of 390 factors identified, 44 percent were community-acquired infections, 9 percent were nosocomial infections, 20 percent possibly involved infection, and 26 percent were noninfectious processes. Fever is a frequent finding in hospitalized patients. Both infectious and noninfectious processes play important roles. Determining the cause of fever is complicated by the multiplicity of possible causes.

Update from the SENIC project. Hospital infection control: recent progress and opportunities under prospective payment

From a survey of all U.S. hospitals in 1976 and of a random sample in 1983, we found that the intensity of infection surveillance and control activities greatly increased, and the percentage of hospitals with an infection control nurse per 250 beds increased from 22% to 57%. The percentage with a physician trained in infection control remained low (15%), and there was a drop in the percentages of hospitals doing surgical wound infection surveillance (from 90% down to 79%) and reporting surgeon-specific rates to surgeons (from 19% down to 13%). There was an increase in the percentage of hospitals with programs shown to be effective in preventing urinary tract infections, bacteremias, and pneumonias, but not surgical wound infections. The percentage of nosocomial infections being prevented nationwide appears to have increased from 6% to only 9%, whereas 32% could be prevented if all hospitals adopted the most effective programs.

A new approach to the isolation of hospitalized patients with infectious diseases: alternative systems

A new guideline developed by the Centers for Disease Control suggests that hospitals adopt one of two alternative isolation systems: the category system or the disease-specific system. The older category system has been modified to reflect current knowledge; for example, the category of protective isolation has been deleted, new categories for contact precautions and tuberculosis precautions have been added, the specific precautions indicated in the other categories have been substantially modified, and many infections have been assigned to new categories. The disease-specific system, a newly developed approach, lists the specific isolation precautions indicated for each infectious disease. Whereas the revised category system offers greater simplicity in practice, the disease-specific system minimizes unnecessary precautions. Both systems allow patient-care personnel more decision-making authority in determining which precautions to apply.

How frequent are outbreaks of nosocomial infection in community hospitals?

A statistical algorithm was used to identify potentially important clusters among nosocomial infections reported each month by 7 community hospitals. Epidemiologic review and on-site investigations distinguished outbreaks of clinical disease from factitious clusters. In 1 year, 8 outbreaks were confirmed. They involved 82 patients--approximately 2% of patients with nosocomial infections and 0.09% of all discharges. One true outbreak occurred for every 12,000 discharges--at least 1 outbreak per year for the average community hospital. Five (63%) outbreaks were recognized independently by the hospitals' infection control personnel. Four (50%) resolved spontaneously; the hospitals' own control measures were necessary in 2; and 2 resolved only after an outside investigation. Organized surveillance appears necessary to detect some outbreaks, and control measures are needed to stop many. Since, however, outbreaks account for such a small proportion of nosocomial infections, infection control programs should be sufficiently staffed and managed so that most of the effort is directed toward the surveillance and control of endemic infection problems, but with adequate resources remaining to respond to outbreaks when they occur.

The nationwide nosocomial infection rate. A new need for vital statistics

From a random sample of patients and hospitals and extrapolation ratios derived from the best available sources of data, the authors estimate that the nationwide nosocomial infection rate among the 6,449 acute-care US hospitals in 1975-1976 was 5.7 nosocomial infections per 100 admissions and that over 2 million nosocomial infections occurred in a 12-month period in these hospitals. Nosocomial urinary tract infections constituted 42% of the infections, surgical wound infections 24%, nosocomial pneumonia 10%, nosocomial bacteremia 5%, and nosocomial infections at all other sites 19%. If adjustments are made for the accuracy of the diagnostic method, the increasing nationwide secular trend, and the number of nosocomial infections in nursing homes, however, as many as 4 million nosocomial infections per year may now be occurring. This greatly exceeds previous estimates and calls for timely and accurate vital statistics on the problem.