Meta-analyses and large randomized, controlled trials.
Explore the source record for details and available documents.
Bias↗
Biomedical subjects
J C Cappelleri
Publications and source records attributed to J C Cappelleri.
Correlation of biochemical response to interferon alfa with histological improvement in hepatitis C: a meta-analysis of diagnostic test characteristics.
The current goal of interferon treatment for chronic hepatitis C is to normalize alanine aminotransferase (ALT) and to eradicate detectable viral RNA. Many patients do not achieve this objective during treatment, and most do not sustain these outcomes after interferon is discontinued. However, biochemical or virological responses to interferon may not reflect accurately the histological consequences of therapy. The aim of this study was to determine the extent to which the biochemical measures reflect the histological outcomes in the treatment of hepatitis C with interferon alfa using a meta-analysis of diagnostic test characteristics. The data sources were English and non-English language studies retrieved from Medline (from 1966 to December 1995). The study selection included studies in which interferon alfa was used for treatment of chronic hepatitis C with liver biopsies performed before and after therapy. Data on histological and biochemical outcomes were extracted independently by two reviewers. Two separate criteria were used for defining histological response. When strict definitions of histological improvement were considered, histology improved in 28% (95% confidence interval [95% CI], 17%-43%) of patients after interferon treatment. The sensitivity and specificity of the ALT for determining histological change were 70% (95% CI, 56%-81%) and 66% (95% CI, 56%-75%), respectively. As many as 17% (95% CI, 9%-30%) of subjects with an abnormal ALT at the end of therapy may have improved histologically after interferon therapy. When less stringent definitions of histological improvement were considered, 62% (95% CI, 51%-72%) improved after therapy. The sensitivity and specificity of the ALT for determining histological change were 55% (95% CI, 44%-65%) and 75% (95% CI, 67%-81%), respectively. As many as 51% (95% CI, 38%-64%) may have improved, despite failure to normalize ALT. A substantial number of patients may improve histologically after interferon therapy. The significance of histological changes observed after interferon therapy must be weighed against the limitations of liver biopsy and the uncertain natural history of hepatitis C. Nevertheless, the ALT does not always reflect liver histology accurately after interferon alfa treatment and may underestimate histological improvement.
Alanine Transaminase↗
The relationship between study design, results, and reporting of randomized clinical trials of HIV infection.
We examined whether the study design of randomized clinical trials for medications against human immunodeficiency virus (HIV) may affect the results and whether the outcomes of these trials affect reporting and publication. We used a database of 71 published randomized HIV-related drug efficacy trials and considered the following study design factors: endpoint definition and method of analysis, masked design, sample size, and duration of follow-up. Large variation was noted in the methods of analysis for surrogate endpoints. Often statistical significance for a surrogate endpoint was not associated with statistical significance for the clinical endpoint or for survival in the same trial, although disagreements in the direction of the treatment effect for surrogate endpoints and survival within individual trials were uncommon. Open-label design seemed to affect the magnitude of the treatment effect for two treatments. The magnitude of the treatment effect in trials of zidovudine monotherapy was inversely related to their sample size, but this probably reflected the confounding effect of longer duration of follow-up in large trials (with a resulting loss of efficacy) rather than publication bias. There was, however, evidence for potential bias in reporting and publication of HIV-related trials. Meta-analyses of published trials for specific treatments demonstrated a sizable treatment benefit for all the examined medications regardless of whether these medications were officially approved, controversial, or abandoned, raising concerns about either publication bias or unjustifiable rejection of potentially useful medications. Compared with trials published in specialized journals, trials published in journals of wide readership were larger (p = 0.001) and 4.4 times more likely to report "positive" results (p = 0.01). We identified several examples of trials with "negative" results that have remained unpublished for a long time. In conclusion, study design factors may have an impact on the magnitude and significance of the treatment effect in HIV-related trials. Bias in reporting can further affect the information that these studies provide.
Anti-HIV Agents↗
Impact of epidemic and individual heterogeneity on the population distribution of disease progression rates. An example from patient populations in trials of human immunodeficiency virus infection.
Patients at the same stage of chronic disease may have had different rates of disease progression. The authors developed a mathematical modeling approach that allows reconstructing and comparing populations in terms of the disease progression rates of their participants when the disease onset and progression rates are unknown for individual patients. Human immunodeficiency virus 1 infection was used as an example. Both published and hypothetical models were used to describe the human immunodeficiency virus 1 epidemic (epidemic heterogeneity) and incubation and survival functions for different disease stages (individual heterogeneity). Reconstructions of populations with late disease (e.g., acquired immunodeficiency syndrome patients) show a marked predominance of rapid progressors, unless the incidence of new infections has been decreasing for a long time. Rapid progressors would also predominate in populations of acute seroconverters, unless diagnosis is based on repeated serologic screening rather than symptoms. Populations of patients who have not progressed beyond an early stage of the disease (e.g., patients with CD4 cell counts > 500/microliter) tend to overrepresent slow progressors, especially if the epidemic has been decreasing for a long time. With this approach, one can assess whether the target population of a clinical trial is comparable with other patient populations at different places and times. Epidemic and individual diversity may even affect trial results if patients with different progression rates experience different benefits from a treatment. By modeling the targeted populations in trials of early versus deferred antiretroviral treatment, the authors observed larger treatment benefits in trials in which rapid progressors probably predominated, compared with trials of slow progressors.
Acquired Immunodeficiency Syndrome↗
Viral load and response to treatment of HIV.
Explore the source record for details and available documents.
Antiviral Agents↗
Single or multiple daily doses of aminoglycosides: a meta-analysis.
OBJECTIVE: To assess relative efficacy and toxicity of aminoglycosides given by single daily dose compared with multiple daily doses. DESIGN: Meta-analysis of 21 randomised trials identified through MEDLARS (1966 to January 1995). Data were overviewed with fixed effects and random effects models and with meta-regression analysis. SUBJECTS: Total of 3091 patients with bacterial infection, most without pre-existing renal disease. INTERVENTIONS: Patients were randomized to receive aminoglycosides once daily or multiple times daily with similar total daily dose. MAIN OUTCOME MEASURES: Clinical failure of treatment, nephrotoxicity, ototoxicity, and mortality. RESULTS: Single daily dose regimen produced a non-significant decrease in risk of antibiotic failures (random effects risk ratio 0.83 (95% confidence interval 0.57 to 1.21)). Benefit of once daily dosing was greater when the percentage of pseudomonas isolates in a trial was larger. Once daily administration reduced risk of nephrotoxicity (fixed effects risk ratio 0.74 (0.54 to 1.00)). Similar trends were noted for patients with febrile neutropenia and for children. There was no significant difference in ototoxicity between the two dosing regimens, but the power of the pooled trials to detect a meaningful difference was low. There was no significant difference in mortality. CONCLUSIONS: Once daily administration of aminoglycosides in patients without pre-existing renal impairment is as effective as multiple daily dosing, has a lower risk of nephrotoxicity, and no greater risk of ototoxicity. Given the additional convenience and reduced cost, once daily dosing should be the preferred mode of administration.
Aminoglycosides↗
A meta-analysis of the relative efficacy and toxicity of Pneumocystis carinii prophylactic regimens.
BACKGROUND: Finding the optimal strategy for Pneumocystis carinii prophylaxis in patients with human immunodeficiency virus infection can be problematic. Several prophylactic regimens are available, but their relative efficacy and tolerance are not well understood. METHODS: A meta-analysis overviewed 35 randomized trials comparing different regimens for P carinii prophylaxis directly or with placebo. Analyses were based on intention-to-treat. On-treatment data were also analyzed when available. RESULTS: Regardless of dose, sulfamethoxazole-trimethoprim was almost universally effective for patients who tolerated it. The risk of discontinuing sulfamethoxazole-trimethoprim because of side effects decreased by 43% (95% confidence interval, 30% to 54%) if one double-strength tablet was given three times a week instead of daily. For dapsone, among 100 patients given 100 mg daily instead of twice a week for 1 year (primary prophylaxis), seven fewer patients would develop P carinii pneumonia, but 17 more would have significant toxic reactions. Aerosolized pentamidine was well tolerated regardless of the dose used. Prophylaxis failures might be halved if the dose of aerosolized pentamidine were doubled. Compared with aerosolized pentamidine, oral regimens prevented 73% (95% confidence interval, 57% to 82%) of toxoplasmosis events by on-treatment analysis, but only 33% (95% confidence interval, 12% to 50%) by intention-to-treat. No significant difference in mortality was demonstrated between different regimens. CONCLUSIONS: Sulfamethoxazole-trimethoprim is the superior regimen, and low doses could improve tolerance without losing effectiveness for primary prophylaxis. Low doses of dapsone reduce toxic effects, but at the expense of some loss of efficacy. There are few data on the use of low-dose regimens for secondary prophylaxis. High doses of aerosolized pentamidine may improve the efficacy of this regimen. Aerosolized pentamidine is inadequate for prevention of toxoplasmosis, and strategies that improve the tolerance of oral regimens may increase effectiveness in preventing toxoplasmosis.
AIDS-Related Opportunistic Infections↗
A personal view of some controversies in allocating treatment to patients in clinical trials.
Explore the source record for details and available documents.
Confounding Factors, Epidemiologic↗
Predictive value of viral load measurements in asymptomatic untreated HIV-1 infection: a mathematical model.
OBJECTIVE: To model the predictive value of viral load measurements in asymptomatic patients with HIV-1 infection, who have CD4 cell counts > 500 x 10(6)/l and no prior antiretroviral therapy, when the time of seroconversion and the prior levels of viremia are unknown. DESIGN: A mathematical model was constructed for the changes in HIV RNA load over time based on data from cohorts of HIV-infected patients followed since the time of seroconversion. METHODS: For different values of viral load, the time to progression to AIDS or an equivalent state [progression to AIDS equivalent (PAE)] was calculated using a wide range of estimates for the time since seroconversion and the rate of change of the viral load over time. RESULTS: In the absence of antiretroviral treatment, patients with a viral load of 10(5) copies/ml serum are at risk for PAE in less than 3 years (0-3 years) and patients with a viral load half a log higher are at risk in less than 1 year. In contrast, patients with a viral load of 10(4.5) have at least 1.9 years and may have up to 8 years before risk of PAE. Patients with a viral load of 10(4) RNA copies/ml have at least 2.8 years and may have up to 19 years before risk of PAE. The rate of change of the viral load was an important predictor of outcome; the time since seroconversion had only a minor effect. CONCLUSIONS: The viral load in the plasma or serum has predictive value even if the time of seroconversion is unknown. The rate of change of viral load over time may also be an important predictive factor. Serial measurements of viral load over time may provide therapeutic guidance.
CD4 Lymphocyte Count↗
Early or deferred zidovudine therapy in HIV-infected patients without an AIDS-defining illness.
PURPOSE: To do a meta-analysis on the efficacy of early or deferred zidovudine monotherapy in patients with human immunodeficiency virus (HIV) infection but not the acquired immunodeficiency syndrome (AIDS). DATA SOURCES: Articles on zidovudine monotherapy published through May 1994. STUDY SELECTION: Double-blind, randomized, placebo-controlled trials addressing the efficacy of zidovudine monotherapy in HIV-infected persons without an AIDS-defining illness. DATA EXTRACTION: Progression to any primary trial end point; any clinical end point; and AIDS or death. Data were stratified according to disease stage at study entry and duration of follow-up (short-term, < 14 months; long-term, > 21 months). DATA SYNTHESIS: Early initiation of zidovudine therapy was of short-term benefit for all the end points evaluated (for example, the risk ratio for progression to any primary end point was 0.51; 95% CI, 0.41 to 0.64). Long-term trials showed a marginally significant trend of decreased progression to any primary end point (risk ratio, 0.73; CI, 0.52 to 1.03). The trend was not significant for other end points. With further stratification according to disease stage, progression to AIDS or death in the short term was significantly decreased for both symptomatic and asymptomatic patients with CD4 cell counts of less than 500 x 10(6)/L (risk ratios, 0.26 [CI, 0.13 to 0.56] and 0.43 [CI, 0.30 to 0.64], respectively). A regression analysis indicated a larger relative benefit in short-term trials and symptomatic patients than in long-term trials and asymptomatic patients. CONCLUSIONS: Early initiation of zidovudine therapy offers a benefit that decreases over time. Symptomatic patients experience a larger benefit than asymptomatic patients. The implications beyond 3 years of follow-up remain unknown.
Acquired Immunodeficiency Syndrome↗
Efficacy and safety of different aspirin dosages on vascular diseases in high-risk patients. A metaregression analysis.
PURPOSE: To compare the efficacy and safety of different aspirin dosages in trials with patients at increased risk of vascular disease. DATA SOURCES: Pertinent studies were selected using MEDLINE (1966 through 1992), weekly reviews of Current Contents (1970 through 1992), and references from review articles and editorials. STUDY SELECTION: Thirty-six randomized control trials of aspirin compared with another dosage of aspirin or with placebo. METHODS: The Mantel-Haenszel method of pooling odds ratios and metaregression involving log odds ratio (and the risk difference) on aspirin dosage, adjusting for the control rate and the mean length of follow up of the studies. RESULTS: For all patients and for subgroups of patients with previous vascular conditions, there was no relationship between dose and vascular events. For all patients, a dose-response relation was not found with gastrointestinal hemorrhages and hemorrhagic stroke, but was found with gastrointestinal symptoms and withdrawals from side effects. For every 25 mg/day increase in aspirin dosage, the odds ratio of gastrointestinal symptoms and withdrawals increased, respectively, by 0.87% (99% Cl, 0.18 to 1.57%) and 0.94% (99% Cl, 0.06 to 1.82%). The corresponding absolute risk increases were 0.58 and 0.78 per 1,000 patients. CONCLUSIONS: Direct and indirect comparisons of high-risk patients suggest no statistical differences in efficacy, gastrointestinal bleeds, and hemorrhagic strokes across aspirin dosages. These comparisons, however, suggest decreased risk of gastrointestinal symptoms and of withdrawals with lower doses of aspirin. More direct comparison studies are warranted that should contrast the benefits and risks to determine the net benefit.
Aspirin↗
Efficacy and safety of combined anticoagulant and antiplatelet therapy versus anticoagulant monotherapy after mechanical heart-valve replacement: a metaanalysis.
We performed a metaanalysis of five randomized controlled trials to compare the efficacy and safety of combined oral anticoagulant and antiplatelet therapy versus oral anticoagulants alone after prosthetic heart-valve replacement. The combined regimen reduced embolism and overall mortality by approximately 67% (pooled odds ratio [OR] 0.33; 95% confidence interval [CI] 0.16 to 0.69; p = 0.0032) and 40% (OR 0.60; 95% CI 0.32 to 1.12; p = 0.11), respectively, but increased the risk of hemorrhage by approximately 65% (OR 1.65; 95% CI 1.15 to 2.39; p = 0.0069) and of major gastrointestinal hemorrhage by approximately 250% (OR 3.47; 95% CI 1.43 to 8.40; p = 0.0058). It is estimated that for every 1.6 patients who had their stroke prevented by combination therapy, there was an excess of one major gastrointestinal bleed. This metaanalysis suggests that the benefits derived from the enhanced antithrombotic potential of combined therapy outweigh the toxic effects resulting from the enhanced anticoagulant potential of this regimen.
Administration, Oral↗
An illustrative statistical analysis of cutoff-based randomized clinical trials.
Cutoff-based randomized clinical trials (RCTs) are designed to balance ethical and scientific concerns. Patients scoring below a cutoff score on a baseline measure (i.e. the least severely ill) are assigned to the control-treated group, those scoring above a second cutoff score (i.e. the most severely ill) are assigned to the test-treated group, and those scoring within the interval (i.e. the moderately ill) are randomly assigned. This paper provides a formal illustration on the statistical analysis of cutoff-based RCTs using data from the Xanax Cross-National Collaborative Study. To overcome problems specific to cutoff-based designs, we generally recommend a backward elimination approach that tests interactions before main effects.
Humans↗
The epidemiology of child abuse: findings from the Second National Incidence and Prevalence Study of Child Abuse and Neglect.
The epidemiology of child abuse was investigated with data from the Second National Incidence and Prevalence Study of Child Abuse and Neglect. A statistical comparison of incidence rates suggested that age, family income, and ethnicity were risk factors for both sexual abuse and physical abuse, but county metrostatus was not. Gender was a risk factor for sexual abuse but not for physical abuse. A logistic regression analysis showed that ethnicity, county metrostatus, and a gender-by-income interaction distinguished sexual abuse from physical abuse.
Adolescent↗
Cutoff assignment strategies for enhancing randomized clinical trials.
The randomized clinical trial (RCT) is the preferred method for assessing the efficacy of treatments. Recent ethical and logistical criticisms suggest that new variations of the traditional RCT are needed. Some of these criticisms may be addressed with new hybrid designs that combine random assignment with assignment by one or more cutoff values on a baseline variable (e.g., severity of illness). In a simple version of such a "cutoff-based" RTC, persons scoring below a cutoff score on a baseline measure (i.e., the least severely ill) are automatically assigned to the control-treated group, those scoring above a second, higher cutoff (i.e., the most ill) are automatically assigned to the test-treated group, and those scoring in the interval between the cutoff scores (i.e., the moderately ill) are randomly assigned to either group. Depending on the baseline score, the patient is assigned to treatment either randomly or by the need-based, clinically related baseline score. Six cutoff-based design variations are studied via simulations and compared with the traditional RCT and the single-cutoff (i.e., regression-discontinuity) design. All variations yield unbiased estimates of the treatment effect but estimates differ in efficiency, with the RCT being most efficient and the single-cutoff design being least efficient. Secondary analyses of data from the Cross-National Collaborative Study of the Effects of Alprazolam (Xanax) on panic are conducted for each variation by selectivity discarding cases from the original dataset to stimulate cutoff-based assignment. The results confirm the simulations and illustrate how cutoff-based designs might look with real data.
Alprazolam↗
Validation of the behavioural activity rating scale (BARS): a novel measure of activity in agitated patients.
We report psychometric results of the Behavioural Activity Rating Scale (BARS) using data from three Phase III clinical trials of intramuscular ziprasidone in acutely agitated patients with psychosis (Studies 1 and 2) or in stable psychotic patients (Study 3). Convergent validity and divergent validity were assessed with baseline data from Studies 1 and 2 in subjects with acute agitation. To investigate convergent validity, we sought Pearson and Spearman correlation of BARS scores with scores on the Clinical Global Impression of Severity (CGI-S) Scale and a predefined cluster of agitation-related items from the Positive and Negative Syndrome Scale (PANSS). For divergent validity, we sought Pearson and Spearman correlation between BARS scores and a predefined cluster of PANSS items measuring negative symptoms. Discriminant validity was investigated with the help of subjects with moderate psychopathology (Study 3). Wilcoxon rank-sum and two-sample t tests determined whether mean (or median) BARS scores differed between subjects with acute agitation (Studies 1 and 2) and moderate psychopathology (Study 3). Responsiveness to treatment effect and rater reliability were also evaluated. In Study 2, Pearson correlation coefficients of BARS scores with PANSS agitation items and CGI-S were moderate (convergent validity) and statistically significant (P<0.005). The correlation between BARS scores and PANSS negative component scores was low (divergent validity). Treatment effect size was larger for BARS than for PANSS agitation items and CGI-S (responsive to treatment differences). Virtually perfect inter- and intra-rater reliability was achieved. Study 1 produced similar results. BARS showed psychometrically valid properties for measurement of behavioral activity in acutely agitated patients with psychosis.
Cluster Analysis↗
Information processing speed deficits may be better correlated with the extent of white matter sclerotic lesions in multiple sclerosis than previously suspected.
Despite frequent use of the Paced Auditory Serial Addition Test (PASAT; Gronwall, 1977) for examining information processing speed (IPS) deficits in multiple sclerosis (MS), prior literature on the relationship between PASAT performance and severity of brain disease has been contradictory. In the present study, we found that PASAT performance is moderately well correlated with the total area of sclerotic brain lesions in MS patients only if a modified scoring method (mean "dyad" score; Snyder et al., 1993), but not the standard scoring method, is applied. We conclude that the PASAT remains a useful measure of IPS in MS, if a flaw in the typical scoring technique for this test is corrected. In our sample of 41 MS patients, for every one point decrease in mean dyad score, total lesion area increased by 364.08 mm2 on average, after controlling for age, sex, and education. IPS deficits are more severe for patients with greater brain disease, due to increased disruption of the white matter pathways that likely support the parallel distributed processing of complex information by geographically distant brain regions.
Adult↗
The Sexual Health Inventory for Men (SHIM): a 5-year review of research and clinical experience.
The Sexual Health Inventory for Men (SHIM) is a widely used scale for screening and diagnosis of erectile dysfunction (ED) and severity of ED in clinical practice and research. In reviewing the SHIM-related literature, we sought to provide a compendium of studies in which the SHIM was used, to provide a systematic framework for organizing and evaluating the studies, and to provide a status report on the SHIM and its impact on the management of male sexual dysfunction. Using a Medline search, we found that the SHIM was an integral measure in at least 21 studies on the prevalence of ED, 23 studies on the efficacy of ED interventions, and eight other (mainly correlational) studies. The quantity of research and quality of scholarship on the SHIM provide testimony to its positive impact on understanding and improving male sexual function. These scientific contributions are likely to remain influential in coming years.
Erectile Dysfunction↗