Biomedical subjects
Publications and source records attributed to M Schemper.
Whether asymptomatic patients with severe aortic stenosis benefit from surgery remains unclear. We report our data recently published in the New England Journal of Medicine on the natural history of this disease and predictors of outcome.
The phenomenon of monotone likelihood is observed in the fitting process of a Cox model if the likelihood converges to a finite value while at least one parameter estimate diverges to +/- infinity. Monotone likelihood primarily occurs in small samples with substantial censoring of survival times and several highly predictive covariates. Previous options to deal with monotone likelihood have been unsatisfactory. The solution we suggest is an adaptation of a procedure by Firth (1993, Biometrika 80, 27-38) originally developed to reduce the bias of maximum likelihood estimates. This procedure produces finite parameter estimates by means of penalized maximum likelihood estimation. Corresponding Wald-type tests and confidence intervals are available, but it is shown that penalized likelihood ratio tests and profile penalized likelihood confidence intervals are often preferable. An empirical study of the suggested procedures confirms satisfactory performance of both estimation and inference. The advantage of the procedure over previous options of analysis is finally exemplified in the analysis of a breast cancer study.
BACKGROUND: Whether to perform valve replacement in patients with asymptomatic but severe aortic stenosis is controversial. Therefore, we studied the natural history of this condition to identify predictors of outcome. METHODS: During 1994, we identified 128 consecutive patients with asymptomatic, severe aortic stenosis (59 women and 69 men; mean [+/-SD] age, 60+/-18 years; aortic-jet velocity, 5.0+/-0.6 m per second). The patients were prospectively followed until 1998. RESULTS: Follow-up information was available for 126 patients (98 percent) for a mean of 22+/-18 months. Event-free survival, with the end point defined as death (8 patients) or valve replacement necessitated by the development of symptoms (59 patients), was 67+/-5 percent at one year, 56+/-5 percent at two years, and 33+/-5 percent at four years. Five of the six deaths from cardiac disease were preceded by symptoms. According to multivariate analysis, only the extent of aortic-valve calcification was an independent predictor of outcome, whereas age, sex, and the presence or absence of coronary artery disease, hypertension, diabetes, and hypercholesterolemia were not. Event-free survival for patients with no or mild valvular calcification was 92+/-5 percent at one year, 84+/-8 percent at two years, and 75+/-9 percent at four years, as compared with 60+/-6 percent, 47+/-6 percent, and 20+/-5 percent, respectively, for those with moderate or severe calcification. The rate of progression of stenosis, as reflected by the aortic-jet velocity, was significantly higher in patients who had cardiac events (0.45+/-0.38 m per second per year) than those who did not have cardiac events (0.14+/-0.18 m per second per year, P<0.001), and the rate of progression of stenosis provided useful prognostic information. Of the patients with moderately or severely calcified aortic valves whose aortic-jet velocity increased by 0.3 m per second or more within one year, 79 percent underwent surgery or died within two years of the observed increase. CONCLUSIONS: In asymptomatic patients with aortic stenosis, it appears to be relatively safe to delay surgery until symptoms develop. However, outcomes vary widely. The presence of moderate or severe valvular calcification, together with a rapid increase in aortic-jet velocity, identifies patients with a very poor prognosis. These patients should be considered for early valve replacement rather than have surgery delayed until symptoms develop.
We suggest a new measure of the proportion of the variation of possibly censored survival times explained by a given proportional hazards model. The proposed measure, termed V, shares several favorable properties with an earlier V1 but also improves the handling of censoring. The statistic contrasts distance measures between individual 1/0 survival processes and fitted survival curves with and without covariate information. These distance measures, Dx and D, respectively, are themselves informative as summaries of absolute rather than relative predictive accuracy. We recommend graphical comparisons of survival curves for prognostic index groups to improve the understanding of obtained values for V, Dx, and D. Their use and interpretation is exemplified for a Yorkshire lung cancer study on survival. From this and an overview for several well-known clinical data sets, we show that the likely amount of relative or absolute predictive accuracy is often low even if there are highly significant and relatively strong prognostic factors.
OBJECTIVE: To compare the therapeutic efficacy of narrowband (TL-01) UV-B phototherapy vs photochemotherapy (psoralen-UV-A [PUVA]) in patients with chronic plaque-type psoriasis. DESIGN: Open, nonrandomized, intraindividually controlled paired comparison study. SETTING: Phototherapeutic unit in a university hospital. PATIENTS: Twenty-five patients with chronic plaque-type psoriasis. INTERVENTIONS: Paired irradiations with threshold erythemogenic doses of narrowband UV-B and PUVA were given to the patients' dorsal aspect including the arms and legs. Treatment was performed 3 times weekly until complete or almost complete clearing with one or both regimens or over a maximum period of 18 exposures. MAIN OUTCOME MEASURES: Assessment of the Psoriasis Area and Severity Index (PASI) in each half of the patient's dorsal aspect before and after treatment with the 2 regimens. RESULTS: The median pretreatment PASI score of 16 (range, 6.2-23.4) was reduced by 84% to 2.5 (range, 0-12.6) by the narrowband UV-B treatment and by 89% to 1.8 (range, 0-8.2) by the PUVA treatment. Statistical analysis of these data showed a tendency for PUVA being superior to narrowband UV-B although the difference remained below the level of significance (P = .17). However, a clear effect of the pretreatment PASI score on the therapeutic outcome was found. Patients with higher baseline PASI scores responded significantly better to PUVA than to narrowband UV-B (P = .03). CONCLUSIONS: Our data demonstrate that in many patients with plaque-type psoriasis, narrowband UV-B is comparably as effective as PUVA and, given the lack of photosensitizer-related adverse reactions and the possibly lower long-term cancer risk, can be considered as first-line treatment. Treatment with PUVA, on the other hand, remains the mainstay for patients with high PASI scores who do not respond or whose psoriasis cannot be controlled adequately by narrowband UV-B.
The proportion of explained variation (R2) is frequently used in the general linear model but in logistic regression no standard definition of R2 exists. We present a SAS macro which calculates two R2-measures based on Pearson and on deviance residuals for logistic regression. Also, adjusted versions for both measures are given, which should prevent the inflation of R2 in small samples.
The identification of individuals who 'died far too early' or 'lived far too long' as compared to their survival probabilities from a Cox regression can lead to the detection of new prognostic factors. Methods to identify outliers are generally based on residuals. For Cox regression, only deviance residuals have been considered for this purpose, but we show that these residuals are not very suitable. Instead, we develop and propose two new types of residuals: the suggested log-odds and normal deviate residuals are simple and intuitively appealing and their theoretical properties and empirical performance make them very suitable for outlier identification. Finally, various practical aspects of screening for individuals with outlying survival times are discussed by means of a cancer study example.
Psychosocial stress has been shown to contribute to neurodegenerative changes and has been discussed as a pathogenic element in Alzheimer's disease (AD). However, studies investigating this aspect are rare. We performed a case-control study on 50 clinically diagnosed probable AD patients and 90 controls consisting of surgical patients. Interviews were performed by trained personnel, using a questionnaire, a semi-structured interview, and a psychosocial risk list protocol. Findings are presented as marginal and partial odds ratio from linear logistic regressions. Adapting to an active but unproductive working style as well as living with a dominant spouse is associated with significant risk for AD. Protective factors are high self-esteem and working in one's desired job. Our results indicate psychosocial factors as a possible agent in the latent development of AD and may shift the focus from more traditional risk factors to hitherto almost neglected psychosocial factors in a disease of still unknown etiology.
BACKGROUND: Recently, the serotonin antagonist ondansetron has been reported to have a positive effect on cholestasis-associated pruritus. OBJECTIVES: To study the effect of orally administered ondansetron on pruritus in chronic liver disease in a randomized, placebo-controlled, double-blind, cross-over study. METHODS: Subjective severity of pruritus was assessed using a visual analogue scale (VAS) recorded four times daily by the patients. After a one week pretreatment baseline period the patients were randomized to receive ondansetron tablets 8 mg tds or placebo tablets tds for one week. Following a one week wash-out period patients were switched to the other treatment for one week. The study was ended by an additional follow-up week without medication. For each day peak VAS values were determined and the mean value of the last five days of each week was calculated and referred to as the composite peak VAS score. RESULTS: We observed a significant but moderate reduction of the composite peak VAS score of 1.34 points (CI(95%): 0.12-2.56; P=0.033) during treatment with ondansetron as compared to placebo (treatment effect). In addition, a period effect was observed: a reduction of composite peak VAS score by 1.26 points (C1(95%): 0.04-2.48; P=0.044) was seen in the second treatment period as compared to the first period, irrespective of the kind of treatment. Although under treatment with ondansetron a significant improvement of itching as assessed by the VAS score was demonstrated, this treatment was not preferred over placebo by the patients. CONCLUSIONS: The 5-hydroxytryptamine receptor type 3 antagonist ondansetron has a small, but significant positive effect on pruritus in chronic liver disease as compared to placebo.
OBJECTIVE: To determine more precisely the effect of prophylactic low dose aspirin on intrauterine growth retardation and perinatal mortality. DESIGN: Meta-analysis of 13 published randomised clinical trials. METHODS: We searched 18 medical databases, including MEDLINE since 1964 and EMBASE since 1974, review articles and the references from each retrieved report to identify all studies evaluating the effect of aspirin in pregnancy and including both intrauterine growth retardation and perinatal mortality as outcome measures. RESULTS: Among 13,234 women from 13 studies between 1985 and 1994, aspirin showed a significant reduction in intrauterine growth retardation (IUGR) (OR 0.82; 95% CI 0.72-0.93; P = 0.003) and a nonsignificant reduction in perinatal mortality (OR 0.84; 95% CI 0.66-1.08; P = 0.18). Subgroup analyses revealed that aspirin was effective at lower doses between 50 and 80 mg/day (IUGR: OR 0.87; 95% CI 0.76-0.99; mortality: OR 0.90, 95% CI 0.70-1.16), but that the preventive effect was greater at higher doses between 100 and 150 mg/day (IUGR: OR 0.36, 95% CI 0.22-0.59; mortality: OR 0.40, 95% CI 0.16-0.97) and among women entered before the 17th week of gestation (IUGR: OR 0.35, 95% CI 0.21-0.58; mortality: OR 0.43, 95% CI 0.17-1.06). We did not identify any specific subgroup of women most likely to benefit from aspirin treatment. CONCLUSION: The results of this meta-analysis showed that early aspirin treatment reduced the risk of intrauterine growth retardation. Low dose aspirin should not be used routinely in pregnant women until those most likely to benefit from aspirin treatment have been clearly identified.
Different measures of the proportion of variation in a dependent variable explained by covariates are reported by different standard programs for logistic regression. We review twelve measures that have been suggested or might be useful to measure explained variation in logistic regression models. The definitions and properties of these measures are discussed and their performance is compared in an empirical study. Two of the measures (squared Pearson correlation between the binary outcome and the predictor, and the proportional reduction of squared Pearson residuals by the use of covariates) give almost identical results, agree very well with the multiple R2 of the general linear model, have an intuitively clear interpretation and perform satisfactorily in our study. For all measures the explained variation for the given sample and also the one expected in future samples can be obtained easily. For small samples an adjustment analogous to Radj2 in the general linear model is suggested. We discuss some aspects of application and recommend the routine use of a suitable measure of explained variation for logistic models.
Several measures of explained variation have been suggested for the Cox proportional hazards regression model. We have categorized these measures into three classes which correspond to three different definitions of multiple R2 of the general linear model. In an empirical study we compared the performance of these measures and classified them by their adherence to a set of criteria which we think should be met by a measure of explained variation for survival data. We suggest that currently there is no uniformly superior measure, particularly as the concepts of either uncensored or censored populations may lead to different choices. For uncensored populations, a measure by Kent and O'Quigley and the squared rank correlation between survival time and the predictor from a Cox regression model appear recommendable choices. For the latter, censored survival times are terminated using a very recent data augmentation algorithm for multiple imputation under proportional hazards. With censored populations, Schemper's measure, V2, could be considered. We give an introductory example, discuss aspects of application and stress the desirability of routinely evaluating explained variation in studies of survival.
OBJECTIVE: To estimate the effect of prophylactic antibiotics on neonatal mortality and morbidity in patients with preterm labor, based on a meta-analysis of seven published randomized clinical trials. DATA SOURCES: We searched 18 medical data bases, including MEDLINE from 1964 and EMBASE from 1974, to identify all literature included under preterm or premature labor and antibiotics. We scanned all abstracts from the computer printouts, the retrieved full-text reports, the references from each retrieved report, and review articles to determine whether studies met our inclusion criteria. METHODS OF STUDY SELECTION: The following criteria were used to select studies for inclusion: article-original published report written in English; study design-randomized controlled trial; population-patients with preterm labor, defined as labor before 37 weeks' gestation; intervention-antibiotic treatment; and one or more of the following outcomes-neonatal mortality, sepsis, pneumonia, respiratory distress syndrome, intraventricular hemorrhage, and necrotizing enterocolitis. TABULATION, INTEGRATION AND RESULTS: We analyzed study patients and methods, and abstracted quantitative outcome data. For each outcome, both odds ratio (OR) and 95% confidence interval (CI) were calculated. Seven trials, published between 1989 and 1995 included a total of 795 patients. Adjunctive antibiotic therapy appeared to reduce the risk of pneumonia (OR 0.45, 95% CI 0.12-1.72) and necrotizing enterocolitis (OR 0.38, 95% CI 0.14-1.08) and to increase the risk of neonatal mortality (OR 3.25, 95% CI 0.93-11.38), but it had no effect on neonatal sepsis (OR 0.98, 95% CI 0.34-2.83), respiratory distress syndrome (OR 0.93, 95% CI 0.54-1.87), and intraventricular hemorrhage (OR 1.01, 95% CI 0.20-5.10). None of the effects observed reached a significance level of P < .05. CONCLUSION: The results of this meta-analysis do not support the routine use of adjunctive antibiotic treatment in patients with preterm labor diagnosed on the basis of subjective uterine contractions and the resulting cervical changes.
Explore the source record for details and available documents.
OBJECTIVE: We performed a metaanalysis of seven published randomized clinical trials to estimate more precisely the effect of prophylactic antibiotics on neonatal mortality, clinical sepsis of the neonate, respiratory distress syndrome, intraventricular hemorrhage, and necrotizing enterocolitis. STUDY DESIGN: To evaluate the effect of antibiotic treatment unaffected by other forms of treatment such as tocolytics or corticosteroids, investigations in which these additional measures were used were not included. We analyzed study patients and methods and abstracted quantitative outcome data. For each outcome both odds ratios and 95% confidence intervals were calculated. RESULTS: Among the 657 patients from seven trials published between 1989 and 1994, antibiotic therapy significantly reduced the risk of neonatal sepsis by 68% (odds ratio 0.32, 95% confidence interval 0.16 to 0.65, p=0.001) and that of intraventricular hemorrhage by 50% (odds ratio 0.50, 95% confidence interval 0.28 to 0.89, p=0.019). In contrast, no significant effect of antibiotics on overall neonatal mortality (odds ratio 0.92, 95% confidence interval 0.46 to 1.81), respiratory distress syndrome (odds ratio 0.84, 95% confidence interval 0.58 to 1.22), or necrotizing enterocolitis (odds ratio 1.27, 95% confidence interval 0.61 to 2.62) was found. CONCLUSION: This metaanalysis supports an improvement of neonatal morbidity in mothers with preterm premature rupture of membranes treated prenatally with different antibiotic regimens.
PURPOSE: To evaluate the efficacy of topical anesthesia as an alternative to peribulbar anesthesia in clear corneal cataract surgery. SETTING: Department of Ophthalmology, University of Vienna, Austria. METHODS: In this prospective, double-blind clinical trial, 36 patients had bilateral cataract surgeries performed from 1 to 3 months apart. Half of the patients had topical anesthesia for the first surgery and peribulbar anesthesia for the second surgery. The other half had peribulbar first and then topical. All surgery was done using a temporal clear corneal approach and bimanual phacoemulsification followed by in-the-bag intraocular lens implantation. Subjective pain was assessed using a visual analog scale of no pain (0%) to worst pain imaginable (100%) and intraoperative motility using a rank scale of adverse motility (-5) to ideal patient cooperation (+5). RESULTS: Subjective pain was comparable whether topical or peribulbar anesthesia was used (mean 10.75 versus 10.97%; P > .6). Patient cooperation (motility) was significantly better when topical anesthesia was used (+2.16 versus +1.11; P = .03). There were no significant differences in complications. A peribulbar block was given in addition to the topical anesthesia in two cases. CONCLUSIONS: Topical anesthesia is a safe, effective alternative to peribulbar anesthesia in clear corneal cataract surgery.
BACKGROUND: The significance of oral psoralen photochemotherapy (PUVA) for the development of nonmelanoma skin cancers (NMSC) is still controversial. OBJECTIVE: We evaluated 496 psoriatics, who received PUVA treatment according to the European PUVA protocol in order to reassess the influence of the cumulative UVA dose on the development of NMSC and to answer the question if there is a UVA threshold dose above which the carcinogenic risk is increased. METHODS: The study was conducted as a retrospective investigation. All patients were seen personally. Age, sex, skin type, cumulative UVA dose and carcinogenic risk factors (arsenic, X-rays, tar, UVB, methotrexate) were recorded and investigated by marginal (MA) and partial effects analyses (PA) according to the Cox regression model. RESULTS: In 14 patients (2.8%), one or multiple histologically confirmed NMSC were diagnosed. Nine patients (1.8%) had squamous cell carcinoma (SCC), 5 patients (1.0%) had basal cell carcinoma (BCC). No patient had both types of NMSC. None of the SCC had metastasized. By taking the appearance of BCC and SCC as the endpoint, arsenic [MA: relative risk (RR) = 7.62; PA:RR = 5.36], tar (MA:RR = 4.51; PA:RR = 3.83) and methotrexate (MTX; MA:RR = 4.97; PA:RR = 4.07) appear to produce strong and significant effects (p < 0.05), both in MA and PA. Using the endpoint SCC only, the effect of the natural logarithm of UVA (ln UVA; RR = 2.47), arsenic (RR = 11.2), tar (RR = 9.92) and MTX (RR = 7.1) is significant (p < 0.05) in MA. In PA, only the effect of arsenic (RR = 5.19) is strong and significant (p < 0.05) while the effects of tar (RR = 7.85), MTX (RR = 3.22) and ln UVA (RR = 2.77) are strong but of borderline significance (p = 0.05-0.11). Nonlinear effects of ln UVA on the risk of SCC were far from significant (p > 0.2). CONCLUSION: PUVA with the European treatment protocol appears to be only a weak carcinogen by itself for SCC with a linear increase in tumor risk but not for BCC development.
From 1972 to 1991 primary dilatation therapy was investigated in a prospective study in 77 patients suffering from achalasia of the gastroesophageal sphincter. The patients were followed up continuously and the median follow-up period was ten years. 70% of patients were symptom-free after dilatation therapy at the time of the last follow-up examination, whereby 50% were cured of symptoms after a single dilatation. 27% required surgery after unsuccessful dilatation therapy. The probability of avoiding operation after dilatation therapy within 15 years of diagnosis of achalasia was 73%. The clinical symptomatology at diagnosis graded according to a scheme from I to IV, was significantly worse in patients requiring an operation (Wilcoxon test, p < 0.0005). The data of esophageal manometry were of no prognostic relevance.