Statistical approaches in the development of clinical practice guidelines from expert panels: the case of laminectomy in sciatica patients

Determining Indications for Care Common to Competing Guidelines by Using Classification Tree Analysis: Application to the Prevention of Venous Thromboembolism in Medical Inpatients

Background. Substantial variations have been reported in the advice given by competing guidelines addressing the same clinical problem. Objective. This study aimed to assess the usefulness of classification tree analysis in comparing competing guidelines. Method. The authors implemented a classification tree–growing algorithm on cross-sectional data from 818 patients to determine indications for prophylactic heparin treatment common to 4 competing guidelines disseminated between 1998 and 2000 and addressing the prophylaxis of venous thromboembolism in medical inpatients. Results. The resulting classification tree involved 10 terminal nodes. Its mean accuracy estimated by performing 10-fold cross-validation was 82% (s = 3). The guidelines consistently supported prophylactic heparin treatment for 5 indications: a previous episode of deep vein thrombosis or pulmonary embolism, recent paralysis of lower limb(s), congestive heart failure with one or more risk factors, recent myocardial infarction, and malignancy with one or more risk factors. These indications involved 257 patients (31.4%) and were supported by robust scientific evidence. Deep vein thrombosis was detected in 27 of these patients (10.5%). Two consistent negative indications involved 347 patients (42.4%). Deep vein thrombosis was detected in 9 of these patients (2.6%). Three indications involving 214 patients (26.2%) were discordant over the 4 guidelines. Conclusion. Classification tree analysis of real patient data is a useful strategy to identify indications common to competing guidelines. These indications should be considered for inclusion when updating guidelines. The findings of recently completed randomized trials have partly resolved the disagreement among the 4 guidelines. This approach may be helpful when developing new guidelines or for identifying topics warranting further complementary clinical trials.

Influence of Clinical and Economical Factors on the Expert Rating of Appropriateness of Preoperative Use of Recombinant Erythropoietin in Elective Orthopedic Surgery Patients

To evaluate the relative impact of clinical factors and health care environment resources on the expert ratings of appropriateness of preoperative erythropoietin in elective orthopedic surgery, the authors analyzed 6905 individual votes on 496 clinical scenarios from 14 experts, applying a multivariate logistic model. Sixty-six percent of the indications were appropriate when resource constraints (RC) were not considered and 53% when they were, resulting in a drop in the median vote of 2 points on a 9-point scale (P < 0.05). Initial hemoglobin level, expected perioperative blood loss, and RC were by far the most significant contributors to the model (P < 0.01), but other factors (i.e., clinical specialty of the expert, prior history of transfusion reactions, patient age, cardiovascular disease, anemia of chronic disease) also contributed significantly (P< 0.01). For assessing appropriateness of care, this study confirms the need for detailed clinical scenarios and a multidisciplinary panel carefully selected to reflect those involved in the interventions under consideration.

[Analysis of the most appropriate surgical treatment for acute cholecystitis by applying the RAND/UCLA method]

INTRODUCTION

Acute cholecystitis (AC) is a common indication for cholecystectomy. Local circumstances and certain patient characteristics lead to high failure rates and complications in laparoscopic cholecystectomy (LC), and despite the experience gained, we still do not have a detailed list of indications which could minimise them.

MATERIAL AND METHOD

We used the RAND/UCLA Appropriateness Method (RAM) to evaluate 2 options, LC and open cholecystectomy (OC). An expert panel analysed its suitability after a literature review, a consensus meeting, and 2 rounds of scores on different clinical situations. The score of each scenario was analysed to establish the appropriateness level of each option.

RESULTS

At the end of the meeting there were 64 defined scenarios, with an agreement being reached on the indications in 67.18% of them. In 86.04% of the scenarios, the agreement was due to the appropriateness of the indications. When cholecystectomy was indicated, it was always by laparoscopy, while it was only occasionally by laparotomy. In patients with less than 72 h of onset, LC was always considered appropriate when there was sepsis, or even without this if the ultrasound data showed complicated AC.

CONCLUSIONS

There is still uncertainty as regards the management of AC, especially as regards the timing of the operation and the surgical approach, particularly in frail patients and with a clinical onset greater than 72 h. The RAND method can help to make decisions on the appropriateness of different therapeutic options.

Appropriateness criteria for surgery improve clinical outcomes in patients with low back pain and/or sciatica

STUDY DESIGN

Prospective, controlled, observational outcome study using clinical, radiographic, and patient/physician-based questionnaire data, with patient outcomes at 12 months follow-up.

OBJECTIVE

To validate appropriateness criteria for low back surgery.

SUMMARY OF BACKGROUND DATA

Most surgical treatment failures are attributed to poor patient selection, but no widely accepted consensus exists on detailed indications for appropriate surgery.

METHODS

Appropriateness criteria for low back surgery have been developed by a multispecialty panel using the RAND appropriateness method. Based on panel criteria, a prospective study compared outcomes of patients appropriately and inappropriately treated at a single institution with 12 months follow-up assessment. Included were patients with low back pain and/or sciatica referred to the neurosurgical department. Information about symptoms, neurologic signs, the health-related quality of life (SF-36), disability status (Roland-Morris), and pain intensity (VAS) was assessed at baseline, at 6 months, and at 12 months follow-up. The appropriateness criteria were administered prospectively to each clinical situation and outside of the clinical setting, with the surgeon and patients blinded to the results of the panel decision. The patients were further stratified into 2 groups: appropriate treatment group (ATG) and inappropriate treatment group (ITG).

RESULTS

Overall, 398 patients completed all forms at 12 months. Treatment was considered appropriate for 365 participants and inappropriate for 33 participants. The mean improvement in the SF-36 physical component score at 12 months was significantly higher in the ATG (mean: 12.3 points) than in the ITG (mean: 6.8 points) (P = 0.01), as well as the mean improvement in the SF-36 mental component score (ATG mean: 5.0 points; ITG mean: -0.5 points) (P = 0.02). Improvement was also significantly higher in the ATG for the mean VAS back pain (ATG mean: 2.3 points; ITG mean: 0.8 points; P = 0.02) and Roland-Morris disability score (ATG mean: 7.7 points; ITG mean: 4.2 points; P = 0.004). The ATG also had a higher improvement in mean VAS for sciatica (4.0 points) than the ITG (2.8 points), but the difference was not significant (P = 0.08). The SF-36 General Health score declined in both groups after 12 months, however, the decline was worse in the ITG (mean decline: 8.2 points) than in the ATG (mean decline: 1.2 points) (P = 0.04). Overall, in comparison to ITG patients, ATG patients had significantly higher improvement at 12 months, both statistically and clinically.

CONCLUSION

In comparison to previously reported literature, our study is the first to assess the utility of appropriateness criteria for low back surgery at 1-year follow-up with multiple outcome dimensions. Our results confirm the hypothesis that application of appropriateness criteria can significantly improve patient outcomes.

A comparison of regression trees, logistic regression, generalized additive models, and multivariate adaptive regression splines for predicting AMI mortality

Clinicians and health service researchers are frequently interested in predicting patient-specific probabilities of adverse events (e.g. death, disease recurrence, post-operative complications, hospital readmission). There is an increasing interest in the use of classification and regression trees (CART) for predicting outcomes in clinical studies. We compared the predictive accuracy of logistic regression with that of regression trees for predicting mortality after hospitalization with an acute myocardial infarction (AMI). We also examined the predictive ability of two other types of data-driven models: generalized additive models (GAMs) and multivariate adaptive regression splines (MARS). We used data on 9484 patients admitted to hospital with an AMI in Ontario. We used repeated split-sample validation: the data were randomly divided into derivation and validation samples. Predictive models were estimated using the derivation sample and the predictive accuracy of the resultant model was assessed using the area under the receiver operating characteristic (ROC) curve in the validation sample. This process was repeated 1000 times-the initial data set was randomly divided into derivation and validation samples 1000 times, and the predictive accuracy of each method was assessed each time. The mean ROC curve area for the regression tree models in the 1000 derivation samples was 0.762, while the mean ROC curve area of a simple logistic regression model was 0.845. The mean ROC curve areas for the other methods ranged from a low of 0.831 to a high of 0.851. Our study shows that regression trees do not perform as well as logistic regression for predicting mortality following AMI. However, the logistic regression model had performance comparable to that of more flexible, data-driven models such as GAMs and MARS.

Report of the National Consensus Conference on Family Presence During Pediatric Cardiopulmonary Resuscitation and Procedures

Representatives from 18 national organizations were convened for a conference to develop recommendations regarding family presence (FP) during pediatric procedures and cardiopulmonary resuscitation. Before the conference, invitees were given a questionnaire and provided with current literature regarding FP. A modified Delphi process was used to develop consensus, including use of multiple questionnaires and breakouts for discussion of specific issues. Participants were encouraged to develop consensus recommendations based on the literature and discussions. Changes in attitude were tracked with repeat questionnaires. Results of the conference were circulated to participants for review and revision. Consensus recommendations include (1) consider FP as an option for families during pediatric procedures and cardiopulmonary resuscitation, (2) offer FP as an option after assessing factors that could adversely affect the interaction, (3) if family is not offered the option for FP, document the reasons why, (4) always consider the safety of the health care team, (5) develop in-hospital transport and transfer policies and procedures for FP, such as family member definition, preparation of the family, handling disagreements, and providing support for the staff, (6) obtain legal review of policies, (7) include education in FP in all core curricula and orientation for health care providers, (8) promote research into best methods for education; effects of FP on patients, family, and staff; best practices for FP; and legal issues regarding FP, among others. These recommendations were approved in concept by the American Academy of Pediatrics and the Ambulatory Pediatrics Association.

OncoSurge: a strategy for improving resectability with curative intent in metastatic colorectal cancer

PURPOSE

Most patients with colorectal liver metastases present to general surgeons and oncologists without a specialist interest in their management. Since treatment strategy is frequently dependent on the response to earlier treatments, our aim was to create a therapeutic decision model identifying appropriate procedure sequences.

METHODS

We used the RAND Corporation/University of California, Los Angeles Appropriateness Method (RAM) assessing strategies of resection, local ablation and chemotherapy. After a comprehensive literature review, an expert panel rated appropriateness of each treatment option for a total of 1,872 ratings decisions in 252 cases. A decision model was constructed, consensus measured and results validated using 48 virtual cases, and 34 real cases with known outcomes.

RESULTS

Consensus was achieved with overall agreement rates of 93.4 to 99.1%. Absolute resection contraindications included unresectable extrahepatic disease, more than 70% liver involvement, liver failure, and being surgically unfit. Factors not influencing treatment strategy were age, primary tumor stage, timing of metastases detection, past blood transfusion, liver resection type, pre-resection carcinoembryonic antigen (CEA), and previous hepatectomy. Immediate resection was appropriate with adequate radiologically-defined resection margins and no portal adenopathy; other factors included presence of < or = 4 or > 4 metastases and unilobar or bilobar involvement. Resection was appropriate postchemotherapy, independent of tumor response in the case of < or = 4 metastases and unilobar liver involvement. Resection was appropriate only for > 4 metastases or bilobar liver involvement, after tumor shrinkage with chemotherapy. When possible, resection was preferred to local ablation.

CONCLUSION

The results were incorporated into a decision matrix, creating a computer program (OncoSurge). This model identifies individual patient resectability, recommending optimal treatment strategies. It may also be used for medical education.

Report of the National Consensus Conference on Family Presence During Pediatric Cardiopulmonary Resuscitation and Procedures

Representatives from 18 national organizations were convened for a conference to develop recommendations regarding family presence (FP) during pediatric procedures and cardiopulmonary resuscitation. Before the conference, invitees were given a questionnaire and provided with current literature regarding FP. A modified Delphi process was used to develop consensus, including use of multiple questionnaires and breakouts for discussion of specific issues. Participants were encouraged to develop consensus recommendations based on the literature and discussions. Changes in attitude were tracked with repeat questionnaires. Results of the conference were circulated to participants for review and revision. Consensus recommendations include (1) consider FP as an option for families during pediatric procedures and cardiopulmonary resuscitation, (2) offer FP as an option after assessing factors that could adversely affect the interaction, (3) if family is not offered the option for FP, document the reasons why, (4) always consider the safety of the health care team, (5) develop in-hospital transport and transfer policies and procedures for FP, such as family member definition, preparation of the family, handling disagreements, and providing support for the staff, (6) obtain legal review of policies, (7) include education in FP in all core curricula and orientation for health care providers, (8) promote research into best methods for education; effects of FP on patients, family, and staff; best practices for FP; and legal issues regarding FP, among others. These recommendations were approved in concept by the American Academy of Pediatrics and the Ambulatory Pediatrics Association.

Prognosis or "curabo effect?": physician prediction and patient outcome of surgery for low back pain and sciatica

STUDY DESIGN

Prospective study with patient and physician questionnaires, clinical records, and imaging.

OBJECTIVE

To compare physician expectations of surgery for sciatica and patient outcome.

SUMMARY OF BACKGROUND DATA

Physician accuracy in identifying individual patient prognosis is important for therapeutic decisions.

METHODS

A total of 197 consecutive patients with low back pain and/or sciatica who underwent low back surgery in the University Hospital of Lausanne, Switzerland.

RESULTS

Physicians predicted "a great improvement" of quality of life after surgery for 79% and "moderate improvement" for 20% (1% others); 39% of patients had no "minimal clinically important difference" in back pain after surgery, despite physician prediction of "great improvement." Correlations between physician expectation and various dimensions of patient outcome were not significant, and agreement with patient global judgment of 1-year outcome was poor (kappa = 0.03). However, in a subgroup where the indication for treatment was not considered appropriate, physician prediction of "great improvement" was followed by greater improvement outcome on SF-36 mental component score (P = 0.05), mental health (0.02), and general health (0.03) compared with patients where the physician did not predict "great improvement."

CONCLUSION

Despite clear average improvement, surgeons tended to give overly optimistic predictions that were not correlated with patient outcome. For patients receiving a treatment not meeting explicit criteria of appropriateness, more optimistic physician expectation was associated with better improvement of psychological dimensions. Besides prognostic ability, the influence of physician expectation on patient outcome is discussed and the concept of "curabo effect" (differentiated from "placebo effect") proposed.

Agreement of four competing guidelines on prevention of venous thromboembolism and comparison with observed physician practices: a cross-sectional study of 1,032 medical inpatients

OBJECTIVE

To assess the degree of agreement between four competing guidelines regarding the recommendation for prophylactic heparin therapy and to report to what extent actual practice agreed with or differed from the recommendations made before these guidelines were disseminated.

DESIGN

Four French guidelines were applied to data from a cross-sectional study conducted before their dissemination.

SETTING

Twenty-six medical units of a teaching and a nonteaching hospital.

PATIENTS

One thousand thirty-two medical inpatients.

MAIN MEASUREMENTS

Interguideline agreement rated by the kappa coefficient and percentage of patients receiving prophylactic heparin treatment.

RESULTS

The percentage of patients requiring prophylactic treatment ranged from 35.4% to 54.6% (overall kappa coefficient, 0.65 [0.63 to 0.68]), depending on the guideline. The four guidelines agreed in recommending prophylactic heparin treatment in 330 patients (32.0%). The corresponding rate of prophylactic treatment use was 57.0% (188/330). None of the guidelines recommended prophylactic heparin treatment in 385 patients (37.3%). The physicians did not order prophylactic treatment in 80.3% of these patients (309/385). The guidelines disagreed in recommending prophylactic treatment in 317 patients (30.7%). The corresponding rate of prophylactic treatment use was 32.8% (104/317).

CONCLUSION

The four guidelines agreed in 69.3% of patients but physician practices were already quite appropriate in these patients before the guidelines were disseminated. Active dissemination of the guidelines can be expected to improve physician practices in the treatment of these patients, but likely with limited impact. In contrast, the four guidelines disagreed in 30.7% of patients. Further clinical trials are needed in this subgroup of patients.

Classification and regression tree analysis in public health: methodological review and comparison with logistic regression

BACKGROUND

Audience segmentation strategies are of increasing interest to public health professionals who wish to identify easily defined, mutually exclusive population subgroups whose members share similar characteristics that help determine participation in a health-related behavior as a basis for targeted interventions. Classification and regression tree (C&RT) analysis is a nonparametric decision tree methodology that has the ability to efficiently segment populations into meaningful subgroups. However, it is not commonly used in public health.

PURPOSE

This study provides a methodological overview of C&RT analysis for persons unfamiliar with the procedure.

METHODS AND RESULTS

An example of a C&RT analysis is provided and interpretation of results is discussed. Results are validated with those obtained from a logistic regression model that was created to replicate the C&RT findings. Results obtained from the example C&RT analysis are also compared to those obtained from a common approach to logistic regression, the stepwise selection procedure. Issues to consider when deciding whether to use C&RT are discussed, and situations in which C&RT may and may not be beneficial are described.

CONCLUSIONS

C&RT is a promising research tool for the identification of at-risk populations in public health research and outreach.

Hérnia discal: procedimentos de tratamento

A hérnia de disco é um processo em que ocorre a ruptura do anel fibroso, com subsequente deslocamento da massa central do disco nos espaços intervertebrais. É considerada uma patologia extremamente comum, que causa séria inabilidade em seus portadores. Estima-se que 2 a 3 % da população sejam acometidos desse processo, cuja prevalência é de 4,8% em homens e 2,5% em mulheres, acima de 35 anos. São fatores de risco, causas ambientais, posturais, desequilíbrios musculares e possivelmente, a influência genética. A terapia conservadora tem sido a preferida como a primeira escolha de tratamento, cujos objetivos são o alívio da dor, o aumento da capacidade funcional e o retardamento da progressão da doença. Nesta revisão, são abordadas as principais metodologias, de acordo com a literatura, dando ênfase ao uso de fármacos analgésicos e anti-inflamatórios, o uso de órteses, a acupuntura, o repouso e a adoção de um programa de exercícios adequados.