SEARCH · PubMed Health

Results for “Internal validation”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,099 records · Page 61Linked to original sources

PubMed2005

Hyperbaric oxygen therapy for stroke: a systematic review of the evidence.

OBJECTIVE: To identify the benefits and harms of using hyperbaric oxygen therapy to treat acute or subacute stroke or the chronic effects of a stroke. We aimed to identify any gaps in the evidence to provide guidance for future research. DESIGN: A systematic review of the evidence. SEARCH STRATEGY: We searched MEDLINE, EMBASE, the Cochrane Library, HealthSTAR, CINAHL, MANTIS, bibliographic databases from professional societies and hyperbaric oxygen therapy practitioners, and reference lists. Databases were searched from inception to December 2003. INCLUSION CRITERIA STUDIES: Controlled clinical trials and observational studies published in English. PARTICIPANTS: Patients with ischaemic stroke in any inpatient or outpatient setting. OUTCOMES: Mortality, functional health outcomes and adverse events. DATA COLLECTION AND ANALYSIS: Using predetermined criteria, two reviewers assessed each study for inclusion, and abstracted data about study design, population, interventions, and outcomes. We assigned an overall quality rating (good, fair, or poor) based on internal validity. RESULTS: We identified only four randomized controlled trials and one controlled clinical trial. The best evidence shows no benefit to hyperbaric oxygen therapy in patients with stroke, but because the stage of patients enrolled (acute, subacute, or chronic), the documentation of type and severity of stroke, and the dosage of hyperbaric oxygen therapy given varied considerably, the generalizability of these results is limited. We identified 17 observational studies; all were poor quality. CONCLUSIONS: The overall evidence is insufficient to determine the effectiveness of hyperbaric oxygen therapy in any subgroup of stroke patients. To determine if hyperbaric oxygen therapy for stroke provides any benefit and that these outweigh potential harms, good quality studies are needed.

Brain Ischemia↗

PubMed2005

Predicting survival in terminal cancer patients: clinical observation or quality-of-life evaluation?

INTRODUCTION: This study compares the relative prognostic power of clinical variables and quality-of-life (QoL) measures in a population of terminal cancer patients. METHODS: A prospective cohort study in 58 Italian Palliative Care Units. Of the 601 randomly selected terminal cancer patients, 574 were followed until death in order to compare clinical and QoL variables (using the Therapy Impact Questionnaire (TIQ) as predictors of survival, and assess whether their combined implementation makes prediction more accurate. RESULTS: The clinical variables most strongly associated with survival were dyspnoea, cachexia, Katz's ADL, oliguria, dysphagia, dehydration, liver and acute kidney failure and delirium (hazard ratios (HR) ranging from 2.10 to 3.01). Only the first four kept their strength once introduced in the Cox model (HRs ranging from 1.95 to 2.22). In the TIQ primary scale the strongest predictors were physical wellbeing, fatigue, functional status and cognitive status (HRs ranging from 1.42 to 1.71), but only fatigue showed an independent prognostic relevance (90% of selection). In the TIQ global scales, the Physical Symptom Index showed a stronger association with survival (HR 1.71) than the Therapy Impact Index (HR 1.47). The former marginally improved the prognostic power of the model when added to clinical variables. Internal validation confirmed that the results were not spurious. CONCLUSIONS: In terminal cancer patients, clinical variables are better predictors of survival than QoL. The large residual variability not accounted for by the model (approximately 70%) suggests that survival is also influenced by factors unlikely to be identified in a survey.

Aged↗

PubMed2006

Advances and applications of multiplexed diagnostic technologies in autoimmune diseases.

There is a rapid proliferation of new technologies to identify an ever increasing spectrum of autoantibodies in diverse medical conditions that range from organ-specific autoimmune diseases to systemic rheumatic diseases. Although many laboratories have adopted diagnostic platforms, such as enzyme linked immunoassays (ELISAs), to improve turn around times and meet budget constraints, the prevailing evidence is that the rapid adoption of new technologies is not attended by an appropriate balance of assay sensitivity and specificity. Emerging diagnostic technologies include addressable laser bead immunoassays, microarrays in microfluidics platforms and nanobarcode particles. Although these technologies provide advantages of high-throughput, multiplexed autoantibody assays that can be coupled to other disease specific biomarkers (ie, cytokines, single nucleotide polymorphisms) there is a clear need for standardization and internal validation before they are adopted into the clinical diagnostic laboratory.

Animals↗

PubMed2004

The effects of an open design on trial participant recruitment, compliance and retention--a randomized controlled trial comparison with a blinded, placebo-controlled design.

BACKGROUND: In randomized trials there may be no overriding reason whether or not to have a placebo control. PURPOSE: We assessed the effects of an open trial design (no placebo and people know what tablets they are given) compared with a blinded, placebo-controlled design on recruitment, compliance and retention within a randomized trial of secondary osteoporotic fracture prevention. METHODS: We undertook a randomized controlled comparison nested within a placebo-controlled trial of nutritional supplementation amongst people aged 70 years or over who had previously sustained a fracture, recruited in a UK teaching hospital. Randomization was 2:1 in favour of the blinded, placebo-controlled trial design. RESULTS: From 180 eligible participants randomized to receive information based on the open trial design, 134 (74.4%) consented to take part, compared with 233 (65.1%) of 358 people randomized to the blinded, placebo-controlled design (difference 9.4%, 95% confidence interval 1.3-17.4%). Reluctance to take a placebo and the desire to know tablet allocation were reasons given for not taking part in the blinded, placebo-controlled design. There was no significant difference in tablet compliance. Open trial participants were more likely to remain in the trial for one year (difference 13.9%, 95% confidence interval 3.1-24.6%), mainly reflecting the high retention of the open trial no tablet group compared to the open trial tablet group (difference 23.6%, 95% confidence interval 11.9-35.2%). The odds ratio for reporting an adverse event in the open trial compared to the blinded, placebo-controlled design was 0.64 (95% confidence interval 0.28-1.49), and for reporting a fracture was 0.81 (0.36-1.85). CONCLUSIONS: We conclude that using an open trial design may enhance participant recruitment and retention and thus improve generalizability and statistical power, but withdrawal rates may differ between the study allocations and may threaten the internal validity of the trial.

Aged↗

PubMed2005

Informed patient consent to participation in cluster randomized trials: an empirical exploration of trials in primary care.

Cluster randomized trials are increasingly common. Obtaining informed patient consent to participation in these trials raises practical challenges and ethical issues. The aims of this paper were to 1) develop a typology of interventions employed in cluster randomized trials in primary care; 2) assess whether the likelihood of seeking individual consent to participation varies by intervension type; 3) assess whether this likelihood has increased over time; 4) assess evidence for under reporting of consent procedures; 5) articulate reasons for not obtaining consent; and 6) make recommendations for future trial investigators. We collected data on trial interventions and consent procedures from reports of 152 recently published trials, and 47 unpublished trials. We develop a typology of interventions based on reasons for adopting a clustered design. We examine proportions seeking individual consent to participation among trials involving different types of intervention, in different periods, and among published and unpublished trials. Two-thirds of the trials had multifaceted interventions. Trials involving different types of intervention had different propensities to seek consent, largely because of practical obstacles to obtaining consent. Obtaining consent can compromise internal validity. More recent trials are no more likely to obtain consent than past trials. There was no evidence of under-reporting of consent procedures in publications. In conclusion, future trial investigators should consider both practical reasons and scientific arguments for not obtaining individual patient consent for all interventions in their trials. Where feasible, they should allow patients to opt out of the trial. Lay individuals should represent trial participants as part of the process of cluster consent to participation, and lay individuals could also be involved in considering ethical issues during trial planning. A more public debate may clarify the general acceptability of not obtaining consent in certain situations.

Biomedical Research↗

PubMed1989

Measuring behaviour disorder in mental handicap.

By use of latent-trait item analysis (the Rasch model), the internal validity of the behaviour section of the MRC-HBS schedule was analysed. The set of behavioural symptoms in this rating scale was found to measure a one-dimensional structure. Thus addition of item scores provides a meaningful index. Such a numerical measure for the degree of behaviour disorder facilitates comparison between different patients or groups, and furthermore, forms the basis for quantitative description of the outcome of therapeutic intervention. A diagnosis of behaviour disorder was established in 17.2% of 302 mentally-handicapped adults. However, some kind of deviant behaviour was found in 41%, and was correlated with origin and degree of retardation, epilepsy, and place of living. The distribution of the symptoms strongly indicates that organic brain damage is a major aetiological cause.

Adult↗

PubMed2000

Psychotherapy for patients with complex disorders and chronic symptoms. The need for a new research paradigm.

BACKGROUND: A clear distinction has been made between efficacy and effectiveness in relation to the methods of evaluation of new psychological treatments in psychiatry. Efficacy trials target patients with relatively pure conditions, who may not be representative of the patients who are usually referred for psychological treatment in a clinical setting. Few studies have explored the benefits of psychotherapy in patients with complex disorders and enduring symptoms. AIMS: To explore the rationale for the distinction between efficacy and effectiveness, particularly in relation to outcome studies of patients with complex and enduring disorders. METHOD: A narrative review with examples drawn from the literature, and an illustration of a recent naturalistic outcome study which combines features of both efficacy and effectiveness. RESULTS: Studies of patients with complex and mixed disorders can be designed so that they retain internal validity, but also have external validity and are relevant to clinical practice. CONCLUSION: Studies which evaluate psychological interventions should be carried out in populations of patients clinically representative of those who are likely to receive the intervention, should it be shown to be of benefit.

Adult↗

PubMed1999

Quality-of-life measurement in advanced cancer: assessing the individual.

PURPOSE: Despite the increasing importance of assessing quality of life (QoL) in patients with advanced cancer, relatively little is known about individual patient's perceptions of the issues contributing to their QoL. The Schedule for the Evaluation of Individual Quality of Life (SEIQoL) and the shorter SEIQoL-Direct Weighting (SEIQoL-DW) assess individualized QoL using a semistructured interview technique. Here we report findings from the first administration of the SEIQoL and SEIQoL-DW to patients with advanced incurable cancer. PATIENTS AND METHODS: QoL was assessed on a single occasion using the SEIQoL and SEIQoL-DW in 80 patients with advanced incurable cancer. RESULTS: All patients were able to complete the SEIQoL-DW, and 78% completed the SEIQoL. Of a possible score of 100, the median QoL global score was as follows: SEIQoL, 61 (range, 24 to 94); SEIQoL-DW, 60.5 (range, 6 to 95). Psychometric data for SEIQoL indicated very high levels of internal consistency (median r =.90) and internal validity (median R(2) = 0.88). Patients' judgments of their QoL were unique to the individual. Family concerns were almost universally rated as more important than health, the difference being significant when measured using the SEIQoL-DW (P =.002). CONCLUSION: Patients with advanced incurable cancer were very good judges of their QoL, and many patients rated their QoL as good. Judgments were highly individual, with very high levels of consistency and validity. The primacy given to health in many QoL questionnaires may be questioned in this population. The implications of these findings are discussed with regard to clinical assessment and advance directives.

Adult↗

PubMed1999

Survival and prognostic stratification of 670 patients with advanced renal cell carcinoma.

PURPOSE: To identify prognostic factors and a model predictive for survival in patients with metastatic renal-cell carcinoma (RCC). PATIENTS AND METHODS: The relationship between pretreatment clinical features and survival was studied in 670 patients with advanced RCC treated in 24 Memorial Sloan-Kettering Cancer Center clinical trials between 1975 and 1996. Clinical features were first examined univariately. A stepwise modeling approach based on Cox proportional hazards regression was then used to form a multivariate model. The predictive performance of the model was internally validated through a two-step nonparametric bootstrapping process. RESULTS: The median survival time was 10 months (95% confidence interval [CI], 9 to 11 months). Fifty-seven of 670 patients remain alive, and the median follow-up time for survivors was 33 months. Pretreatment features associated with a shorter survival in the multivariate analysis were low Karnofsky performance status (<80%), high serum lactate dehydrogenase (> 1.5 times upper limit of normal), low hemoglobin (< lower limit of normal), high "corrected" serum calcium (> 10 mg/dL), and absence of prior nephrectomy. These were used as risk factors to categorize patients into three different groups. The median time to death in the 25% of patients with zero risk factors (favorable-risk) was 20 months. Fifty-three percent of the patients had one or two risk factors (intermediate-risk), and the median survival time in this group was 10 months. Patients with three or more risk factors (poor-risk), who comprised 22% of the patients, had a median survival time of 4 months. CONCLUSIONS: Five prognostic factors for predicting survival were identified and used to categorize patients with metastatic RCC into three risk groups, for which the median survival times were separated by 6 months or more. These risk categories can be used in clinical trial design and interpretation and in patient management. The low long-term survival rate emphasizes the priority of clinical investigation to identify more effective therapy.

Adult↗

PubMed2000

The Multinational Association for Supportive Care in Cancer risk index: A multinational scoring system for identifying low-risk febrile neutropenic cancer patients.

PURPOSE: Febrile neutropenia remains a potentially life-threatening complication of anticancer chemotherapy, but some patients are at low risk for serious medical complications. The purpose of this study was to develop an internationally validated scoring system to identify these patients. MATERIALS AND METHODS: Febrile neutropenic cancer patients were observed in a prospective multinational study. Independent factors assessable at fever onset, predicting low risk of complications, on a randomly selected derivation set, were assigned integer weights to develop a risk-index score, which was subsequently tested on a validation set. RESULTS: On the derivation set (756 patients), predictive factors were a burden of illness indicating absence of symptoms or mild symptoms (weight, 5; odds ratio [OR], 8.21; 95% confidence interval [CI], 4.15 to 16.38) or moderate symptoms (weight, 3; OR, 3.70; 95% CI, 2.18 to 6.29); absence of hypotension (weight, 5; OR, 7.62; 95% CI, 2.91 to 19.89); absence of chronic obstructive pulmonary disease (weight, 4; OR, 5. 35; 95% CI, 1.86 to 15.46); presence of solid tumor or absence of previous fungal infection in patients with hematologic malignancies (weight, 4; OR, 5.07; 95% CI, 1.97 to 12.95); outpatient status (weight, 3; OR, 3.51; 95% CI, 2.02 to 6.04); absence of dehydration (weight, 3; OR, 3.81; 95% CI, 1.89 to 7.73); and age less than 60 years (weight, 2; OR, 2.45; 95% CI, 1.51 to 4.01). On the validation set, a Multinational Association for Supportive Care in Cancer risk-index score >/= 21 identified low-risk patients with a positive predictive value of 91%, specificity of 68%, and sensitivity of 71%. CONCLUSION: The risk index accurately identifies patients at low risk for complications and may be used to select patients for testing therapeutic strategies that may be more convenient or cost-effective.

Anti-Bacterial Agents↗

PubMed2001

Improved prognostication of renal cell carcinoma using an integrated staging system.

PURPOSE: To integrate stage, grade, and Eastern Cooperative Oncology Group (ECOG) performance status (PS) into a clinically useful tool capable of stratifying the survival of renal cell carcinoma (RCC) patients. PATIENTS AND METHODS: The medical records of 661 patients undergoing nephrectomy at University of California Los Angeles between 1989 and 1999 were evaluated. Median age was 61 years, male-to-female ratio was 2.2:1, and median follow-up was 37 months. Survival time was the primary end point assessed. Sixty-four possible combinations of stage, grade, and ECOG PS were analyzed and collapsed into distinct groups. The internal validity of the categorized was challenged by a univariate analysis and a multivariate analysis testing for the accountability of each UCLA Integrated Staging System (UISS) category against independent variables shown to have impact on survival. RESULTS: Combining and stratifying 1997 tumor-node-metastasis stage, Fuhrman's grade and ECOG PS resulted in five survival stratification groups designated UISS, and numbered I to V. The projected 2- and 5-year survival for the UISS groups are as follows for the groups: I, 96% and 94%; II, 89% and 67%; III, 66% and 39%; IV, 42% and 23%; and V, 9% and 0%, respectively. UISS accounted for the significant variables in the variate analysis. CONCLUSION: A novel system for staging and predicting survival for RCC integrating clinical variables is offered. UISS is simple to use and is superior to stage alone in differentiating patients' survival. Our data suggests that UISS is an important prognostic tool for counseling patients with various stages of kidney cancer. Further prospective large-scale validation with external data is awaited.

Adult↗

PubMed2003

Postoperative nomogram for disease-specific survival after an R0 resection for gastric carcinoma.

PURPOSE: Few published studies have addressed individual patient risk after R0 resection for gastric cancer. We developed and internally validated a nomogram that combines these factors to predict the probability of 5-year gastric cancer-specific survival on the basis of 1,039 patients treated at a single institution. METHODS: Nomogram predictor variables included age, sex, primary site (distal one-third, middle one-third, gastroesophageal junction, and proximal one-third), Lauren histotype (diffuse, intestinal, mixed), number of positive lymph nodes resected, number of negative lymph nodes resected, and depth of invasion. Death as a result of gastric cancer was the predicted end point. The concordance index was used as an accuracy measure, with bootstrapping to correct for optimistic bias. Calibration plots were constructed. RESULTS: Gastric cancer-specific survival at 5 years was 50%. A nomogram was constructed on the basis of a Cox regression model. The bootstrap-corrected concordance index was 0.80. When compared with the predictive ability of American Joint Committee on Cancer stage, the nomogram discrimination was superior (P <.001). Nomogram calibration appeared to be excellent. CONCLUSION: A nomogram was developed to predict 5-year disease-specific survival after R0 resection for gastric cancer. This tool should be useful for patient counseling, follow-up scheduling, and clinical trial eligibility determination.

Adult↗

PubMed2004

Interleukin-6, interleukin-10, and vascular endothelial growth factor in metastatic renal cell carcinoma: prognostic value of interleukin-6--from the Groupe Francais d'Immunotherapie.

PURPOSE: Few clinical prognostic factors have been identified for patients with metastatic renal cell carcinoma (MRCC), and no biomarker is known in this disease. Several endogenous cytokines have demonstrated interesting and significant correlations with survival in these patients. Our objective was to analyze the prognostic value of circulating vascular endothelial growth factor (VEGF), interleukin-10 (IL-10), and interleukin-6 (IL-6). PATIENTS AND METHODS: Serum levels of IL-6, IL-10, and VEGF were measured in patients with MRCC. Their prognostic value for response to treatment and progression-free and overall survival was evaluated. Pretreatment samples were obtained from 138 patients of a large randomized multicentric trial. Endogenous cytokine levels were determined using immunoassays. Univariate and multivariate analyses were performed to evaluate the prognostic value of each factor further controlled by an internal validation test. Threshold values for serum IL-6 and VEGF were determined using the quartile method. RESULTS: Serum IL-6 was detectable in 70% of the patients. IL-10 and VEGF were elevated in 8% and 71% of the patients, respectively. None of these circulating factors was correlated with response to treatment. IL-10 was not significantly correlated with progression-free or overall survival. Despite significant correlation with survival, VEGF was not an independent prognostic factor in the multivariate analysis. Finally, IL-6 was significantly correlated with progression-free survival and overall survival, and has prognostic value for overall survival. CONCLUSION: Circulating IL-6 level appears to be an important independent prognostic factor in patients with MRCC; if confirmed in further studies, it could be considered for treatment decisions in these patients.

Adult↗

PubMed2004

RE-AIM: evidence-based standards and a Web resource to improve translation of research into practice.

BACKGROUND: Health services data indicate that under present conditions evidence-based medical and preventive practices are not consistently implemented in clinical practice and affect the quality of care provided to patients. Operating with similar conditions and resources, it is unlikely that evidence-based behavioral medicine (EBBM) practices will be more successfully implemented. PURPOSE: In this article we propose ways to help improve the implementation of EBBM practice. METHODS: This article describes the RE-AIM (Reach, Efficacy/Effectiveness, Adoption, Implementation, and Maintenance) framework that is available on a free-use Web site (http://www.re-aim.org), which offers practical research translation tools, resources, and support for program planners, community leaders, and researchers. The material located at www.re-aim.org can be used to help anticipate and overcome likely barriers to dissemination and to estimate eventual public health impact. RESULTS: Data on Web site utilization and lessons learned thus far in its implementation are presented. CONCLUSIONS: Scientists and public health leaders should devote greater attention to reporting practice-oriented issues such as generalizability, breadth of application, and pragmatic and setting or contextual issues in addition to the current focus on internal validity issues. We hope that this and similar efforts will assist EBBM interventions to have broader applications, be consistently implemented, and be sustained.

Behavioral Medicine↗

PubMed2004

Networks of attention in children with the 22q11 deletion syndrome.

The 22q11 chromosomal deletion syndrome (22q11 DS) is associated with learning disabilities and a complex neuropsychological profile. Previous findings have suggested that executive attention deficits might underlie other neurocognitive anomalies. We administered the child Attention Network Test (ANT) to 52 children ages 5.0 to 11.5, 32 22q11 DS children (19 girls) and 20 controls (13 girls) and assessed the efficiency of segregated executive, orienting, and alerting networks. We hypothesized that 22q11 DS children have impaired executive network efficiency as compared to control siblings. The internal validity of the child ANT was confirmed for this population. Analysis of variance results showed significant main effects for flanker and cue types and no interaction effect in either 22q11 DS children or control siblings. Compared to control siblings, 22q11 DS children had significantly larger (less efficient) executive network scores, significantly increased errors on only incongruent trials, and a significant correlation between executive network scores and accuracy. The implications of these findings for future neurocognitive studies of 22q11 DS children are considered.

Analysis of Variance↗

PubMed1998

Empirically supported psychosocial interventions for children: an overview.

Discusses issues related to the identification of psychosocial interventions for children that have demonstrated efficacy. Recent debate concerning differences between clinical trials research and clinical practice is summarized, including the tradeoff between interpretability (internal validity) and generalizability (external validity) of outcome studies. This article serves as an introduction to the special issue containing articles that have as their focus the identification of empirically supported psychosocial interventions for children as part of a task force. The article provides an overview of the history, agenda, and methodology used by the task force to define and identify specific empirically supported interventions for children with specific disorders. Whereas a number of well-established or probably efficacious interventions are identified within the series, more work directed at closing the gap between research and practice is needed.

Adolescent↗

PubMed2004

Efficacy of pergolide in treatment of restless legs syndrome: the PEARLS Study.

OBJECTIVE: To evaluate the short- and long-term safety and efficacy of pergolide therapy for restless legs syndrome (RLS) in a double-blind, placebo-controlled, randomized trial (Pergolide European Australian RLS [PEARLS] study). METHODS: We randomized 100 patients with idiopathic RLS were randomized to pergolide, 0.25 to 0.75 mg, in the evening or placebo for 6 weeks (phase 1); thereafter, patients with response on the Patient Global Impression (PGI) scale continued on double-blind pergolide or placebo, and nonresponders received open-label pergolide up to 1.5 mg/d for 12 months of treatment (phase 2). Sleep efficiency (SE) and periodic limb movements during sleep (PLMS) arousal index were monitored by centrally evaluated polysomnography (PSG). The severity of RLS was assessed using the validated International RLS Scale (IRLS). RESULTS: In phase 1 (change from baseline to week 6), pergolide reduced PLMS arousal index vs placebo (mean +/- SD, -12.6 +/- 10.0 vs -3.6 +/- 15.9; p = 0.004), and SE did not improve (mean +/- SD, +11.3 +/- 11.9% vs +6.1 +/- 18.6%; p = 0.196). Pergolide improved RLS severity score (-12.2 +/- 9.9 vs -1.8 +/- 7.5 placebo; p < 0.001) and was associated with a higher PGI response (68.1% vs 15.1%; p < 0.001) and improvements in periodic limb movements (PLM) index, PGI improvement scale, Clinical Global Impression improvement, and IRLS (all p < 0.001), patient-reported SE (p = 0.019), and quality of sleep (p < 0.001). After 12 months (phase 2), double-blind pergolide maintained improvements in PLMS arousal index and PLM index. Placebo patients switched to open-label pergolide in phase 2 exhibited marked improvements in these measures that were maintained at 12 months. Pooled results from the blinded and open-label pergolide groups demonstrated improvements at 12 months in the PLMS arousal index (p = 0.028) and PLM index (p < 0.0001) compared with placebo. Nausea and headache were more frequent with pergolide than with placebo treatment. CONCLUSIONS: Pergolide substantially improves periodic limb movement measures and subjective sleep disturbance associated with restless legs syndrome. Low-dose pergolide was well tolerated and maintained its efficacy in the long term.

Abdominal Pain↗

PubMed2005

Prediction of outcome in traumatic brain injury with computed tomographic characteristics: a comparison between the computed tomographic classification and combinations of computed tomographic predictors.

BACKGROUND AND OBJECTIVE: The Marshall computed tomographic (CT) classification identifies six groups of patients with traumatic brain injury (TBI), based on morphological abnormalities on the CT scan. This classification is increasingly used as a predictor of outcome. We aimed to examine the predictive value of the Marshall CT classification in comparison with alternative CT models. METHODS: The predictive value was investigated in the Tirilazad trials (n = 2269). Alternative models were developed with logistic regression analysis and recursive partitioning. Six month mortality was used as outcome measure. Internal validity was assessed with bootstrapping techniques and expressed as the area under the receiver operating curve (AUC). RESULTS: The Marshall CT classification indicated reasonable discrimination (AUC = 0.67), which could be improved by rearranging the underlying individual CT characteristics (AUC = 0.71). Performance could be further increased by adding intraventricular and traumatic subarachnoid hemorrhage and by a more detailed differentiation of mass lesions and basal cisterns (AUC = 0.77). Models developed with logistic regression analysis and recursive partitioning showed similar performance. For clinical application we propose a simple CT score, which permits a more clear differentiation of prognostic risk, particularly in patients with mass lesions. CONCLUSION: It is preferable to use combinations of individual CT predictors rather than the Marshall CT classification for prognostic purposes in TBI. Such models should include at least the following parameters: status of basal cisterns, shift, traumatic subarachnoid or intraventricular hemorrhage, and presence of different types of mass lesions.

Adolescent↗