Predictors of survival following in-hospital cardiopulmonary resuscitation. A moving target.

BACKGROUND: Counseling patients about the risks and benefits of in-hospital cardiopulmonary resuscitation (CPR) can potentially reduce patient suffering and hospital costs. However, there is currently much disagreement regarding the overall rate of in-hospital CPR survival and characteristics that identify patients more or less likely to survive CPR. METHODS: The charts of all adults who were pulseless and received basic CPR at a 720-bed university hospital during 1990 and 1991 were reviewed. Patients were excluded if cardiac arrest occurred outside the hospital or in the emergency department, operating room, recovery room, or cardiac catheterization laboratory. Each patient's chart was reviewed to determine the presence of explicitly defined clinical characteristics. RESULTS: Overall, 50 (16.0%) of 313 patients survived to discharge. Before arrest, only impaired functional capacity and sepsis identified patients unlikely to survive CPR. Of adults suffering cardiac arrest during the study period, only 22% underwent CPR, including 13.0% of those with cancer and 18.1% of those 70 years or older. CONCLUSIONS: The use of do-not-resuscitate orders to exclude patients who were inappropriate candidates for CPR may explain why the survival rate reported here is higher than similar reports and why more clinical characteristics were not found to predict CPR survival. Investigators of in-hospital CPR should use explicit criteria to describe the conditions studied and report survival for patients who receive basic CPR. The impact of do-not-resuscitate orders on survival rates must be considered. Functional capacity deserves further investigation as a predictor of CPR survival.

The impact of informed consent on patient interest in prostate-specific antigen screening.

BACKGROUND: Because of the many uncertainties surrounding screening for prostate cancer, authorities recommend that patients be involved in the screening decision. OBJECTIVE: To determine the impact of informed consent on patient interest in undergoing prostate-specific antigen (PSA) screening. METHODS: Men 50 years or older with no prior PSA testing and no history of prostate cancer presenting to 1 of 4 university-affiliated primary care practices were eligible for enrollment. Patients were randomized to receive either a scripted informational intervention simulating an informed consent presentation (intervention group, n = 103) or a single sentence about the PSA (control group, n = 102). The main outcome measure was patient interest in undergoing PSA screening measured on a 5-point Likert scale. RESULTS: Patients who received the informational intervention were significantly less interested in undergoing PSA screening than controls (mean difference in interest, 0.8 on 5-point scale, P < .001). Informed patients were much less likely to indicate high interest in screening (odds ratio, 0.34; 95% confidence interval, 0.19-0.60; P < .001). In a multivariate model, family history of prostate cancer was associated with increased interest and advancing age with decreased interest in PSA screening, but the informational intervention remained the strongest predictor of interest. CONCLUSIONS: Among primary care patients of predominantly lower socioeconomic status, those who received informed consent were significantly less interested in PSA screening than those who did not. For physicians who offer the PSA as a screening test, this finding highlights the importance of apprising patients of the associated benefits, burdens, and uncertainties and allowing them to participate in the screening decision.

Is "sugar" the same as diabetes? A community-based study among rural African-Americans.

OBJECTIVE: To determine if differing beliefs about high blood glucose exist and are associated with blood glucose control among rural African-Americans. RESEARCH DESIGN AND METHODS: A community-based sample of rural African-Americans completed a survey, and a subsample underwent a subsequent screening that included glucose and GHb measurement. Participants were asked if they thought they had diabetes or sugar-diabetes on the survey; "sugar" was added to the screening along with specific questions about this condition. RESULTS: A total of 1,031 people completed the survey, and 403 the screening exam. The total prevalence of diabetes was 13.6% for men and 15.5% for women. Among those who reported having one of the three conditions, 64% said they had diabetes, 7% sugar-diabetes, and 29% sugar. There was a discrepancy between the survey and screening in that 31% of subjects who answered "yes" to whether they had sugar at the screening had answered "no" to the survey question about diabetes. Subjects who believed they had sugar felt their condition was less serious and had higher glucose levels than those who said they had diabetes. CONCLUSIONS: Diabetes was very common in this population. Over one-fourth of those with diabetes believed they had the condition "sugar." Efforts are needed to improve control of diabetes in this population and should consider these disparate health beliefs.

Identifying problem drinkers: lack of sensitivity of the two-question drinking test.

BACKGROUND AND OBJECTIVE: The two-question drinking test ("Have you ever had a drinking problem?" and "When was your last drink?") is frequently cited as an accurate method of screening for alcoholism, yet its performance has never been validated. This study evaluated its utility among patients in two primary care settings. PATIENTS AND METHODS: The two-question drinking test was compared to the Michigan Alcohol Screening Test (MAST) as the standard criterion for the diagnosis of alcoholism. The subjects were 334 nonpregnant English-speaking patients over the age of 18 seen in two university outpatient teaching clinics. RESULTS: The overall prevalence of alcoholism based on a MAST score > or = 5 was 18%. The sensitivity of the two-question test was 53% and its specificity 93%. The question "When was your last drink?" added little to the question "Have you ever had a drinking problem?" The latter question alone had a sensitivity of 50% and a specificity of 97%. CONCLUSIONS: The two-question test is not sensitive enough to be used to screen for alcohol abuse among outpatients. However, because of the low false positive rate, a positive response to the question "Have you ever had a drinking problem?" may help identify alcoholic patients.

Predictors of interest in prostate-specific antigen screening and the impact of informed consent: what should we tell our patients?

PURPOSE: Screening for prostate cancer with the prostate-specific antigen (PSA) remains highly controversial. We sought to discern which patient factors predict interest in the PSA and how informed consent impacts these predictors. PATIENTS AND METHODS: In a randomized trial that found that informed consent decreases patient interest in PSA screening, potential predictors of interest were analyzed separately in the uninformed (n = 102) and informed (n = 103) cohorts to examine the effects of the informational intervention. RESULTS: Univariate predictors of PSA screening interest (P < 0.05) among uninformed patients included perceived efficacy of screening, perceived seriousness of an abnormal PSA, and willingness to accept treatment risks. Among patients who had been informed about PSA screening, univariate predictors included family history of prostate cancer, perceived susceptibility to prostate cancer, age (inverse association), and perceived efficacy, although informed patients rated PSA efficacy significantly lower than uninformed patients (P < 0.001). In multivariate logistic regression modeling for the uninformed cohort, perceived screening efficacy (P < 0.001), perceived seriousness (P < 0.05), and willingness to accept treatment risks (P < 0.05) together were significant predictors of PSA screening interest. Among informed patients, perceived efficacy (P < 0.001), perceived susceptibility (P = 0.01), and younger age (P = 0.01) together predicted interest in screening. CONCLUSIONS: In contrast to uninformed patients, patients given information about PSA screening and prostate cancer are more likely to be interested in screening if they have a family history of prostate cancer, are younger, or otherwise consider themselves susceptible to developing prostate cancer. Uninformed patients are more likely to base their screening interest on the perceived seriousness of prostate cancer and on their willingness to accept treatment risks.

Evaluation and diagnosis of acute infectious diarrhea.

The appropriate approach to the diagnosis and management of acute infectious diarrhea is determined by the frequency and setting of the illness, the recognizable causes or syndromes, the cost and yield of available diagnostic tests, and the treatability of the disease. Acute diarrhea affects everyone throughout the world from one to more than six times each year, depending on age, location, and living conditions. The range of identifiable viral, bacterial, and parasitic etiologies is great, and the cost of indiscriminate use of etiologic studies for diagnosis is prohibitive. Because of its insensitivity for many organisms and poor selection of cases for testing, routine stool culture has been one of the most costly and ineffective microbiologic tests; the cost per positive result has traditionally exceeded $900 to $1,000. The appropriate treatment for the vast majority of cases (independent of their cause) is simple and effective: oral glucose- and electrolyte-containing rehydration solution. On the basis of an appropriate history and understanding of pathogenesis, fecal specimens can be selectively obtained and promptly examined for leukocytes and parasites, and the common noninflammatory diarrheas can be separated from the inflammatory infections in order to focus further studies on the latter group. The bacteria for which specific antimicrobial therapy should be considered usually cause inflammatory diarrhea in the United States. Therefore, only when the history or fecal leukocyte findings indicates an inflammatory process is it appropriate to culture for the routine invasive bacterial pathogens. In sporadic inflammatory diarrhea, culture methods should include those for Campylobacter jejuni as well as Salmonella and Shigella. Several special circumstances may prompt a consideration of parasites (including Giardia, Entamoeba, Strongyloides, Cryptosporidium), Vibrio, Yersinia, Clostridium difficile, enterotoxigenic Escherichia coli, food-borne agents, or sexually transmitted pathogens. The practical value of specific identification of rotaviruses (by enzyme-linked immunosorbent assay, Rotazyme, or electron microscopy) is primarily epidemiologic, particularly in hospitalized infants or young children. Using such a selective approach to fecal culture will greatly increase its yield and can reduce the cost per positive result from $1,000 to less than $150.

Potential role of adherence traits of Escherichia coli in persistent diarrhea in an urban Brazilian slum.

We examined stools from 40 children with persistent diarrhea (duration, 14 days or more), from 50 children with acute diarrhea and from 38 control children to determine infectious etiologies for persistent diarrhea in Goncalves Dias, an urban favela (slum) in Fortaleza, Ceara, Brazil. Children with persistent diarrhea and children with acute diarrhea had similar rates of isolation of routine viral, bacterial and parasitic enteric pathogens. Routine pathogens were identified in at least 20% of cultures done more than 14 days into the diarrheal illness. We examined Escherichia coli isolated from these stools for adherence potential. Enteroaggregative E. coli were isolated significantly more often from children with persistent diarrhea than from control children or children with acute diarrhea (P less than 0.05). E. coli with hemagglutination patterns suggestive of adherence pili were also isolated more often from children with persistent diarrhea than from children with acute diarrhea (38% vs. 18%; P less than 0.05). Enterotoxigenic E. coli were isolated in combination with rotavirus more often from children with persistent diarrhea than from children with acute diarrhea. E. coli which were hydrophobic or exhibited hemagglutination were also seen more often in association with Giardia in children with persistent diarrhea. These findings suggest that the etiology of persistent diarrhea in children is complex and that the aggregative E. coli are associated with prolonged diarrheal illness. Although routine diarrheal pathogens may be present for more than 14 days, combinations of pathogens, including E. coli with adherence potential, may also contribute to prolonged diarrheal disease.

Patterns of antibiotic use among children in an urban Brazilian slum.

Antibiotics are utilized excessively in many areas of the world. To better define how often and why these drugs are used, we prospectively studied antibiotic use among a cohort of 105 children less than five years of age in a poor area of the northeastern Brazilian city of Fortaleza. During a 16-week period, 65 children took 137 courses of antibiotics. Physicians recommended 54% of these, mothers or their associates 39%, and pharmacy workers 7%. Mothers and pharmacy workers recommended drugs for shorter courses than physicians, and were more likely to recommend drugs inappropriate for children. Duration of illness was not a risk factor for antibiotic use, but both poor nutritional status and poor socioeconomic status were. Health care seeking behaviour was further studied in 58 diarrhoea episodes. The type of care sought was related to the duration of illness. Overall, antibiotic use was very common and often inappropriate among children in this poor urban area. More controlled use might decrease the use of potentially dangerous drugs and the use of these drugs when no benefit is likely.

Malnutrition is associated with increased diarrhoea incidence and duration among children in an urban Brazilian slum.

This study was undertaken to further define the relationship between malnutrition and subsequent diarrhoeal illness among children. A cohort of 61 children under five years of age was followed for two years in an urban Brazilian slum. Nutritional status was determined at two-month intervals and was used to predict the subsequent occurrence of diarrhoea. A significant, graded association between worsened nutritional status, as measured by length- or weight-for-age, and diarrhoea incidence was found. This relationship was present for both two-month and one-year periods following nutritional assessment. The average duration of diarrhoea was also significantly longer during the two-month periods which were preceded by the worst nutritional status. Overall, the most malnourished children had nearly twice the total number of days of diarrhoea that better nourished children had. These results provide additional evidence that a significant association between malnutrition and both increased diarrhoea incidence and duration exists.

Early childhood diarrhoea and helminthiases associate with long-term linear growth faltering.

BACKGROUND: Although the acute mortality from diarrhoeal diseases is well recognized, the potentially prolonged impact of early childhood diarrhoea on background growth and development is often overlooked. To examine the magnitude and duration of the association of early childhood enteric infections with growth faltering in later childhood, we investigated associations of early childhood diarrhoea (0-2 years) and intestinal helminthiases with nutritional status from age 2 to 7 years. METHODS: Twice-weekly diarrhoea surveillance and quarterly anthropometrics were followed from 1989 to 1998 in 119 children born into a Northeast Brazilian shantytown. RESULTS: Diarrhoea burdens at 0-2 years old were significantly associated with growth faltering at ages 2-7 years, even after controlling for nutritional status in infancy, helminthiases at 0-2 years old, family income, and maternal education by Pearson correlation, multivariate linear regression, and repeat measures analysis. The average 9.1 diarrhoeal episodes before age 2 years was associated with a 3.6 cm (95% CI : 0.6-6.6 cm) growth shortfall at age 7 years. Early childhood helminthiasis was also associated with linear growth faltering and a further 4.6 cm shortfall (95% CI : 0.8-7.9 cm) at age 7 years. CONCLUSIONS: Early childhood diarrhoea and helminthiases independently associate with substantial linear growth shortfalls that continue beyond age 6 years. Targeted interventions for their control may have profound and lasting growth benefits for children in similar settings.

Preferences of elderly men for prostate-specific antigen screening and the impact of informed consent.

BACKGROUND: Use of the prostate-specific antigen (PSA) as a screening test remains highly controversial, particularly in older men. This study was undertaken to assess the impact of information on the preferences of older men for such screening. METHODS: The elderly cohort (age > or = 65 years) of a larger randomized trial was studied to determine the effect of a 3-minute scripted informational intervention on primary care patients' interest in PSA screening and on potential predictors of screening interest. RESULTS: Informed patients were significantly less interested in screening than were uninformed patients (p = .006). Informed patients considered PSA screening to be significantly less efficacious than did uninformed patients (p = .004), but among both uninformed and informed patients, perceived efficacy correlated with interest in screening (multivariate OR 2.3, 95% CI 1.5-3.8 for uninformed patients; OR 2.2, 95% CI 1.3-3.9 for informed patients). Perceived seriousness of prostate cancer predicted interest in screening among uninformed patients (OR 1.8, 95% CI 1.3-2.6), but not among informed patients. Informed patients who were married were less interested in screening than those who were single, divorced, or widowed (OR 0.3, 95% CI .08-0.9). Marital status did not predict screening interest among uninformed patients. CONCLUSIONS: Involving elderly patients in the decision whether to screen with the PSA by providing them with information leads to a significant reduction in interest in such screening. Factors that appear to influence the screening preferences of informed elderly patients include perceived efficacy of screening and marital status, whereas uninformed patients are more likely to weigh the perceived seriousness of prostate cancer in their screening decision.

Diarrhea as a cause and an effect of malnutrition: diarrhea prevents catch-up growth and malnutrition increases diarrhea frequency and duration.

Diarrhea and malnutrition, alone or together, constitute major causes of morbidity and mortality among children throughout the tropical world. Data from northeast Brazil, taken with numerous other studies, clearly show that diarrhea is both a cause and an effect of malnutrition. Diarrheal illnesses impair weight as well as height gains, with the greatest effects being seen with recurrent illnesses, which reduce the critical catch-up growth that otherwise occurs after diarrheal illnesses or severe malnutrition. Malnutrition (whether assessed by impaired weight or height for age) leads to increased frequencies and durations of diarrheal illnesses, with a 37% increase in frequency and a 73% increase in duration accounting for a doubling of the diarrhea burden (days of diarrhea) in malnourished children. A multi-pronged approach focusing on those with prolonged diarrhea and severe malnutrition is suggested.

Association of early childhood diarrhea and cryptosporidiosis with impaired physical fitness and cognitive function four-seven years later in a poor urban community in northeast Brazil.

To determine potential, long-term deficits associated with early childhood diarrhea and parasitic infections, we studied the physical fitness (by the Harvard Step Test) and cognitive function (by standardized tests noted below) of 26 children who had complete surveillance for diarrhea in their first 2 years of life and who had continued surveillance until 6-9 years of age in a poor urban community (favela) in Fortaleza in northeast Brazil. Early childhood diarrhea at 0-2 years of age correlated with reduced fitness by the Harvard Step Test at 6-9 years of age (P = 0.03) even after controlling for anthropometric and muscle area effects, anemia, intestinal helminths, Giardia infections, respiratory illnesses, and socioeconomic variables. Early childhood cryptosporidial infections (6 with diarrhea and 3 without diarrhea) were also associated with reduced fitness at 6-9 year of age, even when controlling for current nutritional status. Early diarrhea did not correlate with activity scores (P = 0.697), and early diarrhea remained significantly correlated with fitness scores (P = 0.035) after controlling for activity scores. Early diarrhea burdens also correlated in pilot studies with impaired cognitive function using a McCarthy Draw-A-Design (P = 0.01; P = 0.017 when controlling for early helminth infections), Wechsler Intelligence Scale for Children coding tasks (P = 0.031), and backward digit span tests (P = 0.045). These findings document for the first time a potentially substantial impact of early childhood diarrhea and cryptosporidial infections on subsequent functional status. If confirmed, these findings have major implications for calculations of global disability adjusted life years and for the importance and potential cost effectiveness of targeted interventions for early childhood diarrhea.

The stages of change of rural African-American smokers.

Cigarette smoking remains the most important preventable cause of death in the United States, and the burden is especially great for African Americans. A promising approach to better understanding smoking behavior involves the application of a stages-of-change model. The purpose of this study was to analyze the smoking behavior of a population-based sample of African Americans in two rural southern counties using this model. Data were collected using a door-to-door household survey. The study sites were two rural counties in central Virginia. A total of 556 African-American smokers and recent quitters completed a questionnaire concerning their smoking behavior and beliefs about quitting. Overall, 51% of the respondents were in the pre-contemplation stage, 28% in the contemplation stage, 17% in the preparation stage, and 4% in the action stage. In a multivariate model, eight factors were significantly associated with the stage of change among current smokers: age, number of prior cessation attempts; confidence in becoming a nonsmoker; scales measuring both the pros and cons of smoking; perceived desires of others; believing quitting would improve health; and number of doctor visits in the past year. This study provides support for applying a stages-of-change model to African Americans who smoke. Many of the predictors of the stage of change are the same as those found in other populations. In addition, finding an association between stages and both physician visits and believing stopping smoking can improve health has important implications for providing smoking cessation services in health care settings.

Screening for alcohol and drug abuse.

The purpose of this article is to review screening for substance use disorders in health care settings. The epidemiology of alcohol and other drug abuse is briefly reviewed, followed by a discussion of the principles underlying whether or not screening is warranted. Different screening instruments and strategies are then described. Finally, current recommendations for screening for alcohol and other drug abuse are discussed.

Screening for alcohol problems in primary care.

Alcohol problems are common among patients seen in primary care settings, yet they are often missed by physicians. This article offers a model for alcohol screening designed to facilitate early identification of alcohol problems. This approach emphasizes the heterogeneity of alcohol problems and looks at alcohol use along a spectrum, with different risks depending on where in the spectrum a patient falls. The authors provide practical suggestions on integrating alcohol screening and early intervention into routine care.

Changes over time in the epidemiology of diarrhea and malnutrition among children in an urban Brazilian shantytown, 1989 to 1996.

BACKGROUND: Endemic diarrhea and its associated malnutrition remain leading causes of childhood morbidity and mortality in developing countries. This study was undertaken to describe changes in the incidence of diarrhea and prevalence of malnutrition among children in an urban Brazilian shantytown from 1989 to 1996. A secondary purpose was to examine associations between malnutrition and increased incidence and duration of diarrhea. METHODS: From August 1989 through December 1996 a dynamic birth cohort of 315 children was followed for surveillance of diarrhea and nutrition. Study homes were visited twice or thrice weekly to assess the occurrence of diarrhea. Length and weight of the subjects were measured quarterly. Poisson regression was used to test for associations between prior nutritional status and subsequent diarrhea during a quarter. Multiple regression was used to test for an association between nutritional status and episode duration. RESULTS: Declines in both age-adjusted attack rates (6.0 episodes/child-year in study year 3 [1991] to 2.5 episodes per child-year in study year 8 [1996] and days of diarrhea per child-year (30.8 days/child-year in year 3 to 8.5 days/child-year in year 8) were correlated with yearly improvements in mean nutritional status (R2= 0.84, P < 0.05, for mean length-for-age with mean number of episodes/child-year [corrected]. Both length- and weight-for-age were significant predictors of diarrhea incidence, including persistent episodes (> or =14 d), but not duration. CONCLUSIONS: These results demonstrate marked changes over time in the diarrhea burden and nutritional status of children in this population and provide further evidence of a significant association between malnutrition and increased incidence of diarrhea.

Prenatal alcohol education for low-income women with interactive multimedia.

Prenatal alcohol use is a major cause of birth defects and other adverse outcomes of pregnancy. Educating expectant mothers concerning the need to limit alcohol consumption has been advocated as a method to decrease alcohol-related birth problems, however providing this service for low-income women is increasingly difficult due to ongoing changes in obstetrical practice patterns. The purpose of this work was to develop and test a computer-based multimedia prenatal alcohol education program which would be appropriate for use in a poor rural patient population. We first review the effect of alcohol consumption during pregnancy and previously employed intervention strategies. Then the design, development, and evaluation of the program is described. The content of the program reflects user beliefs and attitudes and takes advantage of a computer-based multimedia approach. This format provides numerous opportunities for user interaction, requires little reading ability and offers the potential to increase levels of health education efforts with no increase in staff. The program was found to be well accepted, to provide new information and to demonstrate potential for effecting change in behavioral intention. Possible uses of this technology for educating similar populations and the use of computers for health screening are discussed.

Methacholine provocation of Prinzmetal's variant angina pectoris: a revised perspective.

We report 13 patient with unequivocal Prinzmetal's variant angina pectoris as the entire experience with this syndrome during a 7-year period in a single institution. The clinical diagnosis of this relatively uncommon disorder is emphasized. Five patients were given 10 mg of methacholine subcutaneously. Three demonstrated subsequent delayed appearance of chest pain, ECG change, and coronary vasospasm following early appearance of muscarinic effects. Two Prinzmetal patients had no provocation of variant angina following methacholine, though they did experience significantly less blood pressure fall in response to muscarinic provocation. Another 23 subjects with incompletely explained chest pain given methacholine had neither ECG change nor spasm. Methacholine provocation of variant angina need not necessarily implicated a parasympathomimetic mechanism for otherwise spontaneous episodes. Rather, provocation would appear to occur via the customary reflex adrenergic response to drug-induced hypotension. Methacholine is probably safe though unreliable as an agent to be used for spasm provocation.

The prevention of prenatal alcohol use: a critical analysis of intervention studies.

Alcohol is a major cause of birth defects and mental retardation in the U.S. The yearly cost to care for those affected has been estimated to be over $300 million. Prenatal education and counseling have been advocated as techniques to reduce alcohol use in pregnancy. The purpose of this study was to critically review investigations that used these methods in order to determine their effectiveness. Reports of relevant studies were found by searching Medline, the ETOH database and bibliographies of primary sources. Only five studies were identified that met predetermined selection criteria. As judged by usual methodologic standards for intervention programs, only one was of acceptable quality. None employed a randomized design, and only two compared the treatment group to a control group. Both these studies found no difference in alcohol use between control and intervention groups. An ethical argument was used by several authors to justify performing inadequately controlled studies. Since no intervention has proven to be superior to the usual care delivered to pregnant women, this problem can be overcome through the use of controls who receive this level of care. Despite the public health importance of prenatal alcohol use, interventions to alter this behavior have not been rigorously evaluated, and the benefits of any specific approach are unclear. Randomized trials are ethically justified and guidelines for undertaking such studies are proposed.