Clinical-Pathologic Conference: Use and choice of statistical methods for the clinical study, “Superficial adenocarcinoma of the esophagus” Eugene H.

Slides:

Advertisements

Similar presentations

Copyright © 2013, 2009, and 2007, Pearson Education, Inc. 1 PROBABILITIES FOR CONTINUOUS RANDOM VARIABLES THE NORMAL DISTRIBUTION CHAPTER 8_B.

Advertisements

Orlando Petrucci, MD, PhD, Philip R. Khoury, MS, Peter B

Interpreting Statistics in the Urological Literature

Society of Thoracic Surgeons 2008 cardiac risk models predict in-hospital mortality of heart valve surgery in a Chinese population: A multicenter study

Characterization and Importance of Air Leak After Lobectomy

Generating New Knowledge in Cardiac Interventions

Rounding numbers The Journal of Thoracic and Cardiovascular Surgery

Advising complex patients who require complex heart operations

Should lung transplantation be performed for patients on mechanical respiratory support? The US experience David P. Mason, MD, Lucy Thuita, MS, Edward.

Siva Raja, MD, PhD, Thomas W. Rice, MD, John Ehrlinger, PhD, John R

Orlando Petrucci, MD, PhD, Philip R. Khoury, MS, Peter B

Srdjan V. Kostic, MDa, Thomas W. Rice, MDa, Mark E

Kathy J. Jenkins, MD, MPH, Kimberlee Gauvreau, ScD, Jane W

Can pulmonary conduit dysfunction and failure be reduced in infants and children less than age 2 years at initial implantation? Tara Karamlou, MD, Eugene.

Effect of changes in postoperative spirometry on survival after lung transplantation David P. Mason, MD, Jeevanantham Rajeswaran, MSc, Liang Li, PhD,

Cumulative sum techniques for assessing surgical results

Endovascular versus open elephant trunk completion for extensive aortic disease Eric E. Roselli, MD, Sreekumar Subramanian, MD, Zhiyuan Sun, BS, Jahanzaib.

Factors influencing early and late outcome of the arterial switch operation for transposition of the great arteries Gil Wernovsky, MD* (by invitation),

Emergency Physician Risk Estimates and Admission Decisions for Chest Pain: A Web- Based Scenario Study David L. Schriger, MD, MPH, Michael Menchine, MD,

Duration of inotropic support after left ventricular assist device implantation: Risk factors and impact on outcome Soren Schenk, MD, Patrick M. McCarthy,

Reoperation After Mitral Valve Repair for Degenerative Disease

Bruce W. Lytle, MD, Eugene H. Blackstone, MD, Joseph F

Sina Ercan, MD, Thomas W. Rice, MD, Sudish C. Murthy, MD, PhD, Lisa A

Clostridium difficile infection after cardiac surgery: Prevalence, morbidity, mortality, and resource utilization Suresh Keshavamurthy, MBBS, MS, Colleen.

Prosthesis size and long-term survival after aortic valve replacement

Too High for Transplantation

Does a similar procedure result in similar survival for women and men undergoing isolated coronary artery bypass grafting? Tamer Attia, MD, MSc, Colleen.

Aortic valve replacement: Is valve size important?

Prediction of rupture risk in abdominal aortic aneurysm during observation: Wall stress versus diameter Mark F. Fillinger, MDa, Steven P. Marra, PhDa,b,

William G. Williams, MDa, Jan M. Quaegebeur, MDb, John W

Gain and subsequent loss of lung function after lung volume reduction surgery in cases of severe emphysema with different morphologic patterns Konrad.

The Relationship Between Hospital Surgical Case Volumes and Mortality Rates in Pediatric Cardiac Surgery: A National Sample, 1988–2005 Karl F. Welke,

Atrial fibrillation complicating lung cancer resection

Gary L. Grunkemeier, PhD, Ruyun Jin, MD, Albert Starr, MD

Stage III non–small cell lung cancer and metachronous brain metastases

Reduced survival in women after valve surgery for aortic regurgitationEffect of aortic enlargement and late aortic rupture Monica L. McDonald, MDa, Nicholas.

Superficial adenocarcinoma of the esophagus

Eugene H. Blackstone, MDa,b, Bruce W. Lytle, MDb

What Are the Odds? The Annals of Thoracic Surgery

Distal aortic interventions after repair of ascending dissection: The argument for a more aggressive approach Eric E. Roselli, MD, Gabriel Loor, MD,

Value of accelerated multimodality therapy in stage IIIA and IIIB non–small cell lung cancer Malcolm M DeCamp, MD, Thomas W Rice, MD, David J Adelstein,

Support Your Specialty

Two internal thoracic artery grafts are better than one

Are allografts the biologic valve of choice for aortic valve replacement in nonelderly patients? Comparison of explantation for structural valve deterioration.

Long-term effectiveness of operations for ascending aortic dissections

Decision support in surgical management of ischemic cardiomyopathy

Bahaaldin Alsoufi, MD, Cedric Manlhiot, BSc, Brian W

Intermediate survival in neonates with aortic atresia: A multi-institutional study Marshall L. Jacobs, MD, Eugene H. Blackstone, MD, Leonard L. Bailey,

Ruyun Jin, MD, Anthony P. Furnary, MD, Stephanie C. Fine, MA, Eugene H

Lars G. Svensson, MD, PhD, Fernando A. Atik, MD, Delos M

Jeevanantham Rajeswaran, PhD, Eugene H. Blackstone, MD

Mark K. Ferguson, MD, Amy E. Durkin, MS, PA-C

Isolated bypass grafting of the left internal thoracic artery to the left anterior descending coronary artery Rosalyn Scott, MDa, Eugene H. Blackstone,

Mohammed A Quader, MD, Patrick M McCarthy, MD, A

Is repair preferable to replacement for ischemic mitral regurgitation?

N1 esophageal carcinoma: The importance of staging and downstaging

Douglas R. Johnston, MD, Edward G

Passing the torch The Journal of Thoracic and Cardiovascular Surgery

Turki B. Albacker, MD, MSc, Eugene H. Blackstone, MD, Sarah J

Mitral valve repair with aortic valve replacement is superior to double valve replacement A.Marc Gillinov, MD, Eugene H Blackstone, MD, Delos M Cosgrove,

Siva Raja, MD, PhD, Jay J. Idrees, MD, Eugene H

Arvind K. Agnihotri, MD (by invitation), David C

The Journal of Thoracic and Cardiovascular Surgery

Discussion The Journal of Thoracic and Cardiovascular Surgery

Pulmonary hypertension is associated with worse early and late outcomes after aortic valve replacement: Implications for transcatheter aortic valve replacement

The Journal of Thoracic and Cardiovascular Surgery

Richard P Anderson, MD, Ruyun Jin, MD, Gary L Grunkemeier, PhD

External model validation of binary clinical risk prediction models in cardiovascular and thoracic surgery Graeme L. Hickey, PhD, Eugene H. Blackstone,

International Normalized Ratio Self-Management Lowers the Risk of Thromboembolic Events After Prosthetic Heart Valve Replacement Thomas Eitz, MD, Soren.

Fernando A. Atik, MD, Lars G. Svensson, MD, PhD, FACC, Eugene H

Presentation transcript:

Clinical-Pathologic Conference: Use and choice of statistical methods for the clinical study, “Superficial adenocarcinoma of the esophagus” Eugene H. Blackstone, MDa,b, Thomas W. Rice, MDa The Journal of Thoracic and Cardiovascular Surgery Volume 122, Issue 6, Pages 1063-1076 (December 2001) DOI: 10.1067/mtc.2001.119858 Copyright © 2001 American Association for Thoracic Surgery Terms and Conditions

Drs Rice and Blackstone The Journal of Thoracic and Cardiovascular Surgery 2001 122, 1063-1076DOI: (10.1067/mtc.2001.119858) Copyright © 2001 American Association for Thoracic Surgery Terms and Conditions

Fig. 1 Relation of an unlimited scale of risk, here expressed in logit units, to the probability of occurrence of an event. The logistic equation mapping logit units to probability is shown. ln, Natural logarithm. The Journal of Thoracic and Cardiovascular Surgery 2001 122, 1063-1076DOI: (10.1067/mtc.2001.119858) Copyright © 2001 American Association for Thoracic Surgery Terms and Conditions

Fig. 2 Cumulative distribution of age at operation for superficial adenocarcinoma of the esophagus. Each patient's age is represented on the curve from youngest to oldest. Each unique age value increments to the curve by 1/n, where n is the total number of patients. Notice that 50% of the patients were younger than 64.5 years and 50% were older. This is the median age, expressed by horizontal and vertical solid straight lines. The coarse dashed lines enclose the 25th and 75th percentiles (or quartiles ), meaning that 25% of patients were younger than the 25th percentile and 75% were younger than the 75th percentile. For consistency with the standard deviation, which encloses about 70% of the ages, the 15th and 85th percentiles (15% of ages above and 15% below) are shown by fine dashed lines. The Journal of Thoracic and Cardiovascular Surgery 2001 122, 1063-1076DOI: (10.1067/mtc.2001.119858) Copyright © 2001 American Association for Thoracic Surgery Terms and Conditions

Fig. 3 Histogram of age and superimposed Gaussian distribution curve. Ages of patients have been categorized and counted in 5-year intervals. The smooth curve is a parametric summary of the distribution of age, expressed as the mean (the vertical line at the peak of the curve) and standard deviation (enclosed by the finely dashed vertical lines ). It is questionable whether the mean and standard deviation represent the distribution of these ages well; yet the Gaussian distribution is often used in this setting because of tradition or ease of calculation. On the other hand, notice that the mean on this figure and median on Figure 2 are similar; the ages enclosed within ±1 standard deviation of this figure are similar to those within the 15th and 85th percentiles of Figure 2. The Journal of Thoracic and Cardiovascular Surgery 2001 122, 1063-1076DOI: (10.1067/mtc.2001.119858) Copyright © 2001 American Association for Thoracic Surgery Terms and Conditions

Fig. 4 Calibration of 1-second forced expiratory volume (FEV1) to risk. A scale of risk is given on the vertical axis (akin to the logit units of Figure 1), and 8 groups of equal numbers of patients according to the value for FEV1 along the horizontal axis. Their mortality, converted to the risk scale, is shown at each closed circle. (The 8th closed circle cannot be shown because there were no deaths in the 8th group with the highest FEV1s.) A, Linear scale of FEV1. Clearly, there is a decreasing (more negative) value of risk at higher FEV1 (simple regression line shown, with explained scatter for these points of 80%). B, Inverse scale of FEV1. Because of the inverse transformation, the lower FEV1s are to the right of the scale and the higher FEV1s to the left. Risk falls from left to right , unlike in Figure 4, A. There is now tighter correspondence of risk to this rescaling of FEV1 (85% of scatter explained) than the conventional scale of Figure 4, A. The Journal of Thoracic and Cardiovascular Surgery 2001 122, 1063-1076DOI: (10.1067/mtc.2001.119858) Copyright © 2001 American Association for Thoracic Surgery Terms and Conditions