i Research Methods for Medical Graduates ii iii Research Methods for Medical Graduates Abhaya Indrayan iv CRC Press Taylor & Francis Group 6000 Broken Sound Parkway NW, Suite 300 Boca Raton, FL 33487-2742 © 2020 by Taylor & Francis Group, LLC CRC Press is an imprint of Taylor & Francis Group, an Informa business No claim to original U.S Government works International Standard Book Number-13: 978-1-138-35181-3 (Hardback) This book contains information obtained from authentic and highly regarded sources While all reasonable efforts have been made to publish reliable data and information, neither the author nor the publisher can accept any legal responsibility or liability for any errors or omissions that may be made The publishers wish to make clear that any views or opinions expressed in this book by individual editors, authors or contributors are personal to them and not necessarily reflect the views/opinions of the publishers The information or guidance contained in this book is intended for use by medical, scientific, or health-care professionals and is provided strictly as a supplement to the medical or other professional’s own judgment, their knowledge of the patient’s medical history, relevant manufacturer’s instructions, and the appropriate best practice guidelines Because of the rapid advances in medical science, any information or advice on dosages, procedures, or diagnoses should be independently verified The reader is strongly urged to consult the relevant national drug formulary and the drug companies’ and device or material manufacturers’ printed instructions, and their websites, before administering or utilizing any of the drugs, devices, or materials mentioned in this book This book does not indicate whether a particular treatment is appropriate or suitable for a particular individual Ultimately it is the sole responsibility of the medical professional to make his or her own professional judgments, so as to advise and treat patients appropriately The authors and publishers have also attempted to trace the copyright holders of all material reproduced in this publication and apologize to copyright holders if permission to publish in this form has not been obtained If any copyright material has not been acknowledged please write and let us know so we may rectify in any future reprint Except as permitted under U.S Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers For permission to photocopy or use material electronically from this work, please access www.copyright.com or contact the Copyright Clearance Center, Inc (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400 CCC is a not-for-profit organization that provides licenses and registration for a variety of users For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe Library of Congress Cataloging‑in‑Publication Data Names: Indrayan, Abhaya, 1945– author Title: Research methods for medical graduates / by Dr Abhaya Indrayan Description: Boca Raton : CRC Press, [2020] | Includes bibliographical references and index | Summary: “This book discusses the why and how of each step of data-based medical research that can provide basic information to emerging researchers and medical graduate students who write theses or publish articles The chapters are arranged in the sequence of steps for data-based research – Provided by publisher Identifiers: LCCN 2019034459 (print) | LCCN 2019034460 (ebook) | ISBN 9781138351813 (hardback ; alk paper) | ISBN 9780429435034 (ebook) Subjects: MESH: Research Design | Education, Medical, Graduate | Research–education Classification: LCC R834 (print) | LCC R834 (ebook) | NLM W 20.5 | DDC 610.71/1–dc23 LC record available at https://lccn.loc.gov/2019034459 LC ebook record available at https://lccn.loc.gov/2019034460 Visit the Taylor & Francis Web site at www.taylorandfrancis.com and the CRC Press Web site at www.crcpress.com v Contents Preface xi Author xiii Basics of Medical Research 1.1 What Is Medical Research? 1.1.1 Medical Research and Empiricism 1.1.2 Types of Medical Research and the Scope of This Book 1.1.3 Levels of Medical Research 1.2 Uncertainties in Medical Research 1.2.1 Epistemic Uncertainties 1.2.2 Aleatory Uncertainties 11 1.2.3 Managing Uncertainties in Empirical Medical Research 14 1.3 Broad Steps in Medical Research 14 1.3.1 Pre-Investigation Steps 14 1.3.2 Investigation Steps 17 1.3.3 Post-Investigation Steps 18 1.4 Quality of Medical Research 21 1.4.1 What Qualifies Good Research? 21 1.4.2 Quality of a Good Researcher 21 1.4.3 Pleasures and Frustrations of Medical Research 22 The Topic of Medical Research 2.1 Selection of the Topic of Research 2.1.1 What Is a Problem? 2.1.2 Review of Literature and Databases, and Their Critique 2.2 Feasibility and Resources 2.2.1 Ethical Considerations 2.2.2 Resources 2.3 Objectives and Hypotheses 2.3.1 Broad and Specific Objectives 2.3.2 Hypotheses 27 27 27 29 34 34 35 36 36 37 Study Designs: An Overview 3.1 What Is a Design of an Empirical Study? 3.1.1 Elements of a Design 3.1.2 Types of Designs 3.2 Descriptive Studies 3.2.1 Sample Surveys and Their Designs 3.2.2 Case Studies and Case Series 3.2.3 Census 3.3 Analytical Studies 3.3.1 Choice of Strategy for Analytical Studies 3.3.2 Some Useful Terms and Concepts for Analytical Studies 39 39 39 40 42 43 44 45 46 46 47 v vi vi Contents 3.4 Essentials of Intervention Studies 3.4.1 Medical Experiments 3.4.2 Clinical Trials 3.5 Essentials of Observational Studies 3.5.1 Prospective Studies 3.5.2 Retrospective Studies 3.5.3 Cross-Sectional Studies 3.6 Reliability and Validity of Designs, and Biases 3.6.1 Reliability of a Design 3.6.2 Validity of a Design 3.6.3 Biases in Medical Studies and Their Control 3.7 Where to Use Which Design? 3.7.1 Recommended Designs for Different Types of Research Questions 3.7.2 Levels of Evidence for Cause–Effect Relationships 51 51 52 53 53 54 54 54 55 55 56 63 63 63 Clinical Trials 4.1 Types of Clinical Trials 4.1.1 Therapeutic Trials –Efficacy and Safety 4.1.2 Clinical Trials for Diagnostic and Prophylactic Modalities 4.1.3 Field Trials for Screening, Prophylaxis, and Vaccines 4.1.4 Superiority, Equivalence, and Noninferiority Trials 4.1.5 Other Types of Clinical Trials 4.2 Basics of Clinical Trials 4.2.1 Arms of a Trial 4.2.2 Phases of a Clinical Trial 4.2.3 Randomization and Matching 4.2.4 Control Group in a Clinical Trial 4.3 Validity of a Clinical Trial 4.3.1 Selection of Participants 4.3.2 Blinding, Concealment of Allocation, and Masking 4.3.3 Compliance 4.3.4 Uncertainties in Clinical Trials 4.4 Choosing a Design for an Efficacy Trial 67 67 68 69 70 72 73 76 76 76 78 82 83 83 84 87 87 88 Observational Studies 91 5.1 Prospective Studies 91 5.1.1 Subjects in a Prospective Study 93 5.1.2 Potential Biases in Prospective Studies and Their Merits and Demerits 95 5.1.3 Cohort Studies 97 5.1.4 Longitudinal Studies 98 5.1.5 Repeated Measures Studies 99 5.2 Retrospective Studies 100 5.2.1 Case–Control Design 101 5.2.2 Selection of Cases and Controls 103 5.2.3 Merits and Demerits of Retrospective Studies 103 5.3 Cross-Sectional Studies 104 5.3.1 Merits and Demerits of Cross-Sectional Studies 105 vii Contents 5.4 Comparative Performance of Prospective, Retrospective, and Cross-Sectional Studies 5.4.1 Performance of a Prospective Study 5.4.2 Performance of a Retrospective Study 5.4.3 Performance of a Cross-Sectional Study vii 106 108 108 108 Assessment of Medical Factors 6.1 Intricacies of Assessment 6.1.1 Univariate and Multifactorial Assessments 6.1.2 Assessment in the Implementation Phase and the Results Phase 6.2 Types of Medical Factors 6.2.1 Distal and Proximal Factors 6.2.2 Physiological and Pathophysiological Factors 6.2.3 Pathological Factors and Disease 6.3 Assessment of Mortality, Duration, and Quality of Life 6.3.1 Assessment of Mortality 6.3.2 Quality of Life and Duration 111 111 112 Methodology of Data Collection 7.1 Types of Measurements 7.1.1 Nominal, Metric, and Ordinal Measurements 7.1.2 Other Types of Scales for Measurement 7.1.3 Continuous and Discrete Variables 7.2 Tools of Data Collection 7.2.1 Questionnaires, Schedules, and Proforma 7.2.2 Interview, Examination, and Investigation 7.3 Quality of Data 7.3.1 Errors in Medical Data 7.3.2 Reliability, Validity, and Accuracy of Data 7.3.3 Other Aspects of Data Quality 7.4 Validity of the Tools 7.4.1 Pilot Study and Pretesting 7.4.2 Sensitivity and Specificity of Medical Tests 7.4.3 ROC Curves and Youden Index 7.4.4 Predictivities and Prevalence 121 122 122 125 126 127 127 130 131 131 133 135 136 136 138 141 142 Sampling and Sample Size 8.1 Sampling Methods and Sampling for Descriptive Studies 8.1.1 Purposive Sampling (Nonrandom Methods) 8.1.2 Random Sampling 8.2 Sampling for Analytical Studies 8.2.1 Sampling Methods in Observational Studies 8.2.2 Sampling Methods in Clinical Trials 8.3 Sampling and Nonsampling Errors 8.3.1 Sampling Errors 8.3.2 Nonsampling Errors 147 148 149 150 154 154 155 156 156 157 112 113 114 116 116 118 118 119 viii viii Contents 8.4 Sample Size 8.4.1 Sample Size for Descriptive Studies 8.4.2 Sample Size for Analytical Studies and Clinical Trials Appendix 1: Some Sample Size Formulas 158 160 160 164 Research Protocol 9.1 Structure of the Protocol 9.1.1 Title, Researchers, Supervisors, and Collaborators 9.1.2 Executive Summary 9.1.3 Main Body of the Protocol 9.1.4 Logistics and Appendices 9.2 Main Body of the Protocol 9.2.1 Specifics of the Content of the Main Body of the Protocol 9.2.2 Further Details of the Contents of the Main Body of the Protocol 169 170 171 172 172 172 174 174 178 10 Processing of Data 10.1 Collation of Data and Scrutiny 10.1.1 Uniformity of the Process of Data Collection 10.1.2 Data Validation 10.1.3 Master Chart and Data Entries 10.1.4 Indexes and Scores for Individual Subjects 10.2 Epidemiological Indices 10.2.1 Rates and Ratios 10.2.2 Prevalence and Incidence 10.2.3 Risk, Hazard, and Odds 10.3 Representative Summary Measures 10.3.1 Summary Measures for Quantitative Data 10.3.2 Summary Measures for Qualitative Data 10.4 Tabulation and Graphics 10.4.1 Categorization of Data and the Choice of Class Intervals 10.4.2 Types of Data Tables 10.4.3 Graphs and Diagrams 10.4.4 Statistical Distribution of Medical Measurements 10.4.5 Normal versus Abnormal Levels 181 182 182 182 183 184 186 186 187 188 192 192 195 195 195 196 199 204 206 11 Statistical Analysis 11.1 Confidence Intervals, P-Values, and Power 11.1.1 CI for Proportion and Mean 11.1.2 CI for Relative Risk and Odds Ratio 11.1.3 Statistical Significance, P-Value, and Power 11.2 Some Basic Statistical Tests 11.2.1 Tests for Qualitative Data 11.2.2 Tests for Quantitative Data 11.3 Relationships and Regressions 11.3.1 Dependent and Independent Variables 11.3.2 Basics of Logistic Regression 11.3.3 Ordinary Least Square Regression 11.3.4 Correlation and Agreement 209 210 211 212 212 217 217 220 223 224 225 227 229 ix Contents ix 11.4 Cause–Effect Relationships and Validation of Results 11.4.1 Evidence of Cause–Effect 11.4.2 Validation of the Findings 11.5 Statistical Fallacies 11.5.1 Cherry-Picking the Statistical Indices 11.5.2 Fallacious Interpretation 11.5.3 Statistical Errors Can Cause Many Deaths 231 231 234 236 236 238 240 12 Writing a Thesis or a Paper, and Oral Presentation 12.1 Effective Scientific Writing 12.1.1 Text Style 12.1.2 Tables 12.1.3 Illustrations 12.1.4 Format of a Manuscript (IMRaD) 12.2 Preliminaries of a Manuscript 12.2.1 Title 12.2.2 Authorship Credits 12.2.3 Keywords 12.2.4 Abstract and Summary 12.3 Main Body of the Report 12.3.1 Writing a Suitable Introduction 12.3.2 Explaining Materials and Methods 12.3.3 Describing the Results 12.3.4 Discussion of Findings and Conclusion 12.4 End Features of a Report 12.4.1 Acknowledgment Ethics 12.4.2 Key Messages 12.4.3 References 12.4.4 Contribution of Authors and Conflict of Interest 12.4.5 Appendix 12.5 Oral Presentation 12.5.1 Essentials of Effective Presentation 12.5.2 Poster Presentation 243 243 244 245 246 247 248 248 251 252 252 254 254 255 257 260 262 262 263 263 265 265 265 266 269 13 Reporting Guidelines 13.1 Guidelines for Reporting of Clinical Trials (CONSORT Statement) 13.2 Reporting of Observational Studies (STROBE Statement) 13.3 Reporting of Diagnostic Accuracy Studies (STARD Statement) 13.4 Guidelines for Reporting of Statistical Methods (Revised SAMPL Statement) 273 273 274 277 14 Reporting Ethics and Peer Reviews Covering Letter 14.1 Duplication 14.1.1 Duplicate Publication 14.1.2 Plagiarism 14.1.3 Copyright and Permissions 283 284 284 284 285 286 277 290 290 Research Methods for Medical Graduates because that can delay acceptance Some suggested changes may not be to your liking Explain the changes you have made in the manuscript in response to each comment, and also explain if you find that changes as per any comment will not be appropriate Do not worry if your response is long For those comments that look appropriate to you, express gratitude to the reviewer for pointing them out Some journals require that changes in the manuscript be made in “track changes” mode Experience suggests that it is expedient to comply with the comments of the reviewers and modify your manuscript accordingly instead of raising a counter-argument If you feel strongly about a comment, give a polite explanation why it is not appropriate, or cannot be complied with, and cite supportive evidence See if your writing lacks clarity, which led to this kind of comment Merely giving an explanation in reply to the comments is not enough: the text of the manuscript should also be changed accordingly so that it reflects the reply Each comment must be individually attended to Get help from the other authors if you have others for the paper Do not be disappointed if the paper is not accepted even after complying with all the suggestions This can happen, although it is rare Some journals may not provide comments and only ask you to reduce the size of the paper because they have space limitations Think critically regarding which part of the text, which table, and which figure can be safely deleted without altering the essential content Do so with discretion, because continuity and flow have to be maintained and no vital information should be lost Sometimes there is a conflict in the comments themselves One reviewer may ask to delete some portion and the other may say that the portion is good One reviewer may also provide conflicting advice –for example, to cut short the paper and to add some explanation Make the best of such conflicts without raising controversy Although the decision to accept or reject is supposed to be based entirely on the scientific merit of the paper, the editors are not above the board either They are perceived to be in a tremendously advantageous position Perhaps there is no author who has no grievances against one editor or another British and American journals are sometimes accused of being biased, although they claim to bend over backward to accept quality papers from developing countries Some journals (e.g., Lancet) have appointed an ombudsman for redressal of grievances You can file an appeal if you feel strongly about unfair rejection or any such occurrence Journals are supposed to clearly state the process of appeals and have a system for redressal 14.3 Confidentiality and Misreporting We have discussed ethics in conducting medical research earlier in Chapter and the present chapter is on ethics to be followed in reporting of research findings While confidentiality was discussed there, this needs to be reemphasized in the context of reporting Misreporting is in any case is a serious issue globally 14.3.1 Confidentiality In the context of publishing research, the question of confidentiality arises at least at two levels First is the respect for privacy of patients or subjects No information that can identify any individual should be part of your manuscript The names certainly cannot 291 Reporting Ethics and Peer Reviews 291 be revealed, but things like photos and pedigrees should be sufficiently masked Only in cases where necessary for scientific reasons can this identity be revealed with the full and expressed consent of the person concerned This consent should be free of any duress or pressure The second is at the level of the authors and reviewers All reviews must respect the confidentiality of the authors and the names of the reviewers also are not disclosed Authors have the right not to be discussed for their work unless it is published Reviewers cannot take advantage of their privilege of knowing the contents before publication Nonetheless, the confidentiality of the authors can be breached in a rare case of fraud, when established, with respect to subjects, methods, statistical misappropriation, and other such lapses This brings us to misreporting 14.3.2 Misreporting Scientific misconduct in terms of misreporting can go unnoticed in subtle ways This is manipulating the results and includes bogus research This not only impedes the progress of science but also sometimes misleads All such instances must be reported and investigated when noticed Researchers are not supposed to keep silent or sweep it under the carpet Whistle-blowers are generally protected in all countries Result Manipulation This can occur in one or more of the following ways Changing the endpoint or outcome of interest from, say, death to complication or vice versa when results for the planned outcome fail to meet the expectations of the investigator This can also be in terms of looking at 1-year outcome in place of a 2-year outcome as originally planned, or any such deviation Presenting the results groupwise instead of combined, or vice versa, when such transposition helps to provide findings in support of the investigator’s hypothesis Presenting results for a subgroup of patients, pretending that other groups were not there Presenting univariate (unadjusted) results when the objective was to present results adjusted for confounders Presenting results for proportions when averages fail to serve the hypothesis of the investigator, or vice versa; or using proportion where odds ratio should be used Arbitrary merging of categories of data to provide specific results Altering or omitting part of the data to manufacture the results These are examples of the avenues that can be misused to cook up the desired results Some will select part of the data to get “evidence” in support of their hypothesis There is a famous saying that if someone searches hard enough, part of the data can be located that would support almost any hypothesis The difficulty is that such misreporting is hard to detect in a finished report, and can go unnoticed for a long time It is up to you to be true to yourself and science, and report the full facts Correctly reported results replicate well and endure while incorrectly reported results are soon forgotten They may help to improve your career résumé, but at a substantial cost to science A true doctor would never do this Misreporting includes both intentional and unintentional lapses The unintentional could be due to carelessness or honest errors The latter could be due to miscoding, 292 292 Research Methods for Medical Graduates miscalculation, or inadequate method of statistical analysis Honest deviation from the design or subjective deviations in those aspects which were not thought of earlier at the time of protocol also are not misconduct According to the publication guidelines, a correction can be published if these errors not change the essential results Otherwise, the paper may have to be retracted or republished Whereas substantial deviation can occur from established medical practice due to new research with the wrong results that can adversely affect many patients, Indrayan [9] has pleaded that benign aspects such as wrong statistical methods can produce results that can jeopardize health and life of many persons This further underscores the need to be careful about reporting of research results Intentional errors, in all cases, are inexcusable Fanelli [10] reported that nearly 2% of scientists admitted to have fabricated, falsified, or modified data or results at least once in their career, and up to 38% admitted other questionable research practices Having said all that, there might be valid reasons in rare cases to report part of the findings These could be that you merged categories or groups as these had small numbers, discovered an interesting finding for a specific subgroup, found that the data for a particular group are erroneous –not properly collected, or any such reason In such cases, explain what happened and why only part of the results is being reported The keyword is the intention of the researcher to commit the misconduct If it is established as being unintentional, it can probably be excused Bogus Research At the extreme is research based on fake data No or very little data are collected and the major part is imputed at will to get the desired results This is fraud, to say the least, and puts the reputation at stake not only of the researcher and institution but of the whole country Fraud in the context of research is falsification, fabrication, or deception This can prove to be a setback to the research endeavors or detrimental to science Many patients may be affected and some may even be victims of death when wrong results are used for treatment Persons reporting such research can be severely punished by their employer If the results have wider implications, even the police can take action Known frauds may be just the tip of the iceberg as most of them remain unknown 14.4 The Last Word Medical researchers have a tall order of producing a worthwhile result when uncertainties are numerous and many are insurmountable The core of success lies in being able to identify the sources of uncertainties and using a methodology that can take care of most of them Time spent in developing the proposal is well spent Do not shy away from working hard at that stage Constantly remind yourself about the identification and management of medical uncertainties Credibility of research mostly depends on sound methodology rather than on results The objective is to reach reliable and valid conclusions that are verifiable by repeating the investigation Check that all known epistemic and aleatory uncertainties have been tackled, and the design has taken care of various sources of known bias Those that creep 293 Reporting Ethics and Peer Reviews 293 in despite the best design should be dealt with at the analysis stage If this is achieved to the satisfaction of all concerned, consider that the research has attained a high quality Do not succumb to the pressure of producing quantity at the expense of quality A lengthy list of publications may give some mileage in the beginning but ultimately a reputation cannot be built on shoddy research Haste at the expense of care cannot pay Most employments, promotions, and recognitions are based on, say, the five best publications rather than the full list In some instances, an extremely long list of low-quality publications is disliked rather than appreciated Thus, resist the temptation to publish or present research that you know is substandard Quality work gives immense satisfaction that sloppy work cannot provide In the end, we must emphasize that there is no alternative to common sense Tools such as protocol, design, and statistical analysis are only aids to carry out research in a systematic manner They must only complement common sense and not replace it References Altman DG, Bland JM Measurement in medicine: The analysis of method comparison studies Statistician 1983; 32:307–317 http://people.stat.sfu.ca/~raltman/stat300/AltmanBland.pdf Bland JM, Altman DG Statistical methods for assessing agreement between two methods of clinical measurement Lancet 1986;i:307–310 www-users.york.ac.uk/~mb55/meas/ba.pdf Gold EB Epidemiology of and risk factors for pancreatic cancer Surg Clin North Am 1995;75:819–843 www.ncbi.nlm.nih.gov/pubmed/7660248 Gold EB, Goldin SB Epidemiology of and risk factors for pancreatic cancer Surg Oncol Clin N Am 1998;7:67–91 www.sciencedirect.com/science/article/pii/S0039610916467307 Patel SV, Hodge DO, Bourne WM Corneal endothelium and postoperative outcomes 15 years after penetrating keratoplasty Trans Am Ophthalmol Soc 2004;102:57– 65 www.ncbi.nlm.nih.gov/pmc/articles/PMC1280087/ Patel SV, Hodge DO, Bourne WM Corneal endothelium and postoperative outcomes 15 years after penetrating keratoplasty Am J Ophthalmol 2005;139:311–319 www.ncbi.nlm.nih.gov/ pubmed/15733993 Smith R The Trouble with Medical Journals Royal Society of Medicine Press, 2006 Cummings P, Rivara FP Responding to reviewers’ comments on submitted articles Arch Pediatr Adolesc Med 2002;156:105–107 https://jamanetwork.com/journals/jamapediatrics/ article-abstract/191489 Indrayan A Statistical fallacies and errors can jeopardize life and health of many Indian J Med Res 2018;148:677–679 www.ijmr.org.in/text.asp?2018/148/6/677/252165 10 Fanelli D How many scientists fabricate and falsify research: A systematic review and meta- analysis of survey data PLoS One 2009;4(5):e5738/https://journals.plos.org/plosone/ article?id=10.1371/journal.pone.0005738 294 295 Index Acknowledgment ethics, 262–263 Adaptive trials, 75–76 Additive effects, 49 Adjusted odds ratio, 226 Agreement, assessment of, 229–231 Aleatory uncertainties, 11–13, 14 Alpha error, 214–215, 237 Alpha level, 214–215 Alternative hypothesis, 213–214 one-sided, 214 two-sided, 214 Analysis of variance (ANOVA), 99 Analytical studies, 10, 39–40, 46–51 cause–effect relationship, 46 choice of strategy for, 46–47 ecological studies, 46 sample size for, 158–164, 164–165 sampling methods for, 154–156 terms and concepts for, 47–49 ANOVA-based F-test, 222–223 Antagonism, 48 Antecedents, definition, 49, 50–51 Applied research, 3–5 Arithmetic mean, 192 Arms of a trial, 76 control arm, 76 multi-arms, 76 test arm, 76 Attributable risk (AR), 93, 191 Authorship, ghost, 251 Average (main) effect, 49 Bar diagrams, 200, 202 Basic (pure) research, 3, 4 Bayes’ rule, 143–145 Beta error, 213–214, 240 Bias in medical studies, 13, 43, 157–158 chance of bias in different study designs, 64–65 compliance issues in clinical trials, 87 list of, 56 possible sources of bias, 56–61 randomization and, 78 steps to control bias, 61–62 Bibliography, 180 Bimodal distribution, 192 Binary (variables), 123 Binomial distribution, 126, 195 BIOSIS Previews, 30, 31 Biostatisticians, 239–240 Biostatistics, 8 Blinding (clinical trials), 52, 55, 84–85, 86 Black-box approach, 239–240 Block randomization, 80 Bogus research, 291, 292 Bonferroni adjustment, 161, 162 Box-and-whiskers plot, 203 British Medical Journal, 29 Broad steps in medical research, 14–20 investigation steps, 17–18 post-investigation steps, 18–20 pre-investigation steps, 14–17 Case–control design, 101–103 nested, 102–103, 152–153 Case-fatality rates, 118 Case-referent study, 102 Case report forms, 174 Case reports, 44, 64–65 Case series, 41, 44, 64–65, 148 Case studies, 43, 44 Categorical data, dichotomous, 122 polytomous, 122, 195 Causal inference, 73, 74 Cause–effect relationship, assessment of, 233 criteria of, 232–233 evidence of, 231–234 levels of evidence for, 63, 64 Causes, definition, 47 Censored values, 120 Census, 41, 45–46, 148 Central value, 192 Chance variability, influence on research outcomes, 10 Chi-square test, 217–221 Circular sampling, 152 Class interval, 195–196 Clinical equipoise, 87–88 Clinical trials, 40–41, 46, 52, 67–89 adaptive trials, 75–64 allocation of subjects to treatment groups, 78–82 arms of a trial, 76 basics of, 76–83 blinding, 84–85, 86 choosing an appropriate design, 88–89 compliance issues, 87 concealment of allocation, 85–86 295 296 296 Clinical trials (cont.) control group, 50, 74, 82–83, 92–93 diagnostic trials, 69 equivalence trials, 72–73 ethical issues, 50, 67 explanatory trials, 73–74 field trials, 70–71 masking, 89, 87 matching of subjects, 81–82 multicentric trials, 74–75 noninferiority trials, 72–73 phases of a trial, 76–78 placebos, 50, 80–81 pragmatic trials, 73–74 prophylactic trials, 69, 71 randomization, 78–82 reporting guidelines (CONSORT statement) for, 273–274 safety of the regimen, 69 sample size for, 158–164, 165–166 sampling methods for, 152–153 screening trials, 70–71 selection of participants, 83–42 size of the trial, 83–84 specification of outcome for measuring efficacy, 68 superiority trials, 72–73 therapeutic trials, 68–69 treatment group, 55 two-stage design, 76 types of, 67–76 uncertainties in, 87–88 vaccine trials, 71 validity of, 83–88 Cluster random sampling, 147, 152 Cluster randomization, 70, 80 Cluster sampling, 43, 105 Cochrane Reviews, 30 Coefficient of variation (CV), 194–195 Cohort, birth, 97 concurrent, 97 inception, 97 retrospective, 97, 98 Cohort studies, 97–98 Comparison group, 54, 93, 94–95 Concealment of allocation (clinical trials), 52 Concomitant variables, 224 Confidence intervals (CI), 43, 209–212 aleatory uncertainties and, 13 for difference, 211–212 for mean, 211–212 for odds ratio, 212 Index for proportion, 211 for relative risk, 212 Confidentiality in research, 290–291 Conflict of interest, 283–284 Confounding, 49–51 cross-sectional studies, 104 Confounding variables, 224 CONsolidated Standards Of Reporting of Trials (CONSORT) statement, 33–34, 273–274 Contingency tables for qualitative data, 197–198 Continuous variables, 126–127 Control group in a clinical trial, 52, 82–83, 94–95 historical, 83 parallel, 76, 82 Control strategy, 74 Convenience sample, 149, 150 Copyright, 283 Correlation, 227–228 Correlation coefficient, 228 Covariates, statistical tests for, 218 Covering letter with submission for publication, 284–285 Cox regression, 227 CREAM indicators, 116 Cross-sectional studies, 40–42, 53, 54, 63, 64–65, 104–106 basics, 92 comparative features, 106–109 merits and demerits, 10–104 performance comparison, 106, 107–109 sampling for, 105, 153 C-statistic, 235 Current Contents –Clinical Medicine, 30, 31 Data, definition, 121 Data, types of, primary, 121, 130 secondary, 15, 33–34, 121, 203 Data analysis, see Statistical analysis Databases, electronic search of medical literature, 30–31 Data collation, 181–186 Data collection, assessment of medical factors, 111–122 methodology, 121–145 uniformity of the process, 182 Data dredging, 239–240 Data Monitoring and Safety Board, 75, 173 Data processing, 181–206 categorization of data, 195–196 choice of class intervals, 195–196 collation and scrutiny of data, 182–186 epidemiological indices, 186–192 297 Index grouped data, 196 master chart and data entries, 181–184 missing values, 183 outliers, 183 record linkage, 182 representative summary measures, 192–195 scores, 185–186 scrutiny 182–186 statistical distribution of medical measurements, 204–205 uniformity of the data collection process, 182 validation of the data, 182–183 Deciles, 194 Decimal places, guidance on use of, 237–238 Dependent variable, 224 Descriptive studies, 10, 40–46 case series, 41, 45 case studies, 43, 44 census, 41, 45–46 sample size for, 160, 164 sample surveys and their designs, 31, 43–44 sampling methods for, 148–154 Design effect, 43, 160 Determinants, definition, 46 Deviation from the mean, 194 Diagnostic accuracy studies, reporting guidelines (STARD statement), 273, 277 Diagnostic trials, 67–68 Diagrams, 199–206 Dichotomous categories, 123 Distal factors, 113, 114–115 Discrete variables, 126–127 Dissertation defence, 265 Distribution of medical measurements, 204–207 binomial, 126, 195 Gaussian, 127, 204–205 multinomial, 126, 195 Doctoral dissertations, 6, 7–8 defense of, 265 Double-blind randomized controlled trials (RCTs), 52, 62, 63–64, 84, 85 Dunnett procedure, 222 Duplicate publication, 283–284 Ecological studies, 46, 63–64 Effectiveness, definition, 73 use effectiveness, 73 Effect modification, 48 Effects, definition, 47 Efficacy definition, 71 sample size for estimation of, 160 297 Embase, 30, 31 Empiricism in medical research, 2, 3 Epidemiological indices, 186–191 Epidemiological studies, 53, 91 Epistemic gaps in research results, 10–11 Epistemic uncertainties, 8–11, 12–13, 14, 49, 52 Equipoise, 87–88 Equivalence trials, 72–73 Errors in medical data, 131–133 Errors in research, 23; see also Bias in medical studies non-sampling errors, 156–157 sampling error, 155–156 sources of, 62 Ethical issues clinical trials, 52 considerations in medical research, 33, 34 reporting ethics, 283–293 research protocol, 169–170 Evidence, levels of, 63–64 Evidence base for medical research, 2, 3 ExcerptaMedica (EM), 30 Experiments, see Medical experiments Expert opinion, 64–65 Explanatory trials, 73–74 External validity, 235–236 F-test, 220–222 Factors, definition, 108; see also Medical factors Field trials, 70–71 Figures (in reports), 193–194, 199–206 Fisher exact test, 218, 219 Flip-flop in medical research results, 23 Follow-up studies, see Prospective studies Force of mortality, 189 Frequency curves, 200, 201 Frequency matching, 82, 103 Frequency tables for quantitative data, 196–197 Gaussian (normal) distribution, 127, 204–205 Generalized estimating equations, 93 Geometric mean, 192 g-index, 32 Gold standard, 59, 84, 88, 138, 139, 143, 231, 253 Goodness-of-fit test, one-way tables, 218–214 Google Scholar 31 Graduate theses, 5–6 choosing a thesis topic, 28 confirmatory research, 1–2 review of, 283–284 writing a thesis or paper, 243–265 Graphs, 195–196, 199–206 Grouped data, 196 298 298 Group matching, 82, 103 Guttman scale, 125–126 Haphazard sampling, 149, 150 Harmonic mean, 192 Hawthorne effect, 74, 84, 97 Hazard, 186, 227 ratio, 186, 197, 281 Helsinki declaration, 34 Hirsch index (h-index), 32 Histograms, 200, 201 Historical prospective studies, 92 Honest errors, 23 Hypotheses, 37 generating, 45 sample size for testing, 161 types of, 37 Impact Factor, 31 IMRaD format, 247–248, 248, 249 Incidence, 187–188 Incidence density, 188 Inclusion and exclusion criteria, 175 Independent variables, 221 Indexes, 184–185 Indicators, definition, 111 Informed consent, 34, 35, 83 Inherent validity, 138 Institutional studies, 7, 8 Intercept (in regression), 228 Intentional errors, 23 Intention-to-treat analysis, 210 Interactions, statistical, 47–49 Interdisciplinary research, 5 Internal validity, 235 Interval scale, 123–124 Intervention studies, 39–41, 46, 51–52 Investigation steps, 17–18 collect the data, 18 data cleaning, 18 handle the ethical issues, 18 handle the nonresponse issue, 18 pretest and the pilot study, 17–18 scrutinize the data, 18 Journal Citation Reports, 31 Kaplan–Meier curves, 120 Key messages, 31, 249, 262 Keywords, 245, 249, 251, 252 Knowledge gaps, 8, 9 Lancet, 29 Index Levels of evidence for cause–effect relationship, 63, 64 Levels of medical research, 5–8 Likert scale, 59, 125 Line diagrams, 200, 201 Literature review, 29–33 Logistic regression, 102, 105, 225–227 conditional, 226 multivariable, 226 Log-scale graphs, 200 Longitudinal studies, 98–99 McNemar’s test, 219–220 Main (average) effect, 49 Malpractice, see Misconduct Masking (clinical trials), 52, 85, 86 Master chart, 19, 127, 129, 181, 183–184 Master’s theses, 5–6, 7 Matching (clinical trials), 52, 81–82 case–control design, 103 frequency matching, 82, 103 group matching, 82 matched pairs, 81 MD Consult, 31 Mean, 188–189 misuse of, 231, 232 Measurements, types of, 119–123 Median, 192–193 Mediators, 50 Medical experiments, 40–41, 46, 51–52; see also Clinical trials Medical factors assessment in results phase, 112–113 assessment in the implementation phase, 112–113 classification, 114 distal factors, 113–114 intricacies of assessment, 111–113 multifactorial assessment, 112 pathological factors and disease, 113, 116–117 pathophysiological factors, 113, 116 physiological factors, 113, 116 proximal factors, 113, 114–115 types of, 113–118 univariate assessment, 111 Medical research basic features, 1–8 broad steps, 14–20 empiricism and, 2, 3 evidence base, 2, 3 flip-flop in medical research results, 23 fruits of, 24 levels of, 5–8 299 299 Index pleasures and frustrations of, 22–24 producing high quality research, 287–288 purpose of, 1–2 quality of, 21–24 responsibility of researchers, 24 types of, 2–5 uncertainties in, 8–14 Medical significance, statistical power and, 215–217 Medical tests, see Methodology of data collection MedLine, 30, 31 Medscape, 31 Meta-analysis, 30 Methodological articles, 30 Methodology, importance of, 23 Methodology of data collection, 121–144 accuracy of measurements and recording, 134–135 completeness of data, 135 definition of data, 121 discrete variables, 16–127 errors in medical data, 131–133 measurement types, 122–127 physical examination, 130 pilot study, 136–137 positive predictivity, 138–145 predictivities of medical tests, 138–145 pretesting of tools, 137–138 primary data, 121 proformas, 127, 129 quality of data, 131–136 questionnaires, 127, 128–129 reliability of data, 133–134 schedules, 127, 129 scoring systems for soft data, 123 secondary data, 121 tools of data collection, 127–131 training and supervision, 136 validity of data, 133–134 Metric measurement, 121 Metric scales, 122–123 Misconduct intentional errors in research, 23 misreporting of research, 284–285 use of “copy-paste” technology, 7 Missing values, 181 Misuse of percentages, 237–238 of P-values, 238 of statistical tools, 238–239 Mixed diagrams, 200, 201 Mode, 192–193 Mortality, 113, 114 assessment of, 118–119 Multicentric studies, 7, 8, 74–75 Multifactorial causation, 233–234 Multifactorial characteristics, 124 Multinomial distribution, 126, 195 Multiple comparison procedure, 222 Multiple regression, 105, 224–225 Multistage random sampling, 43, 105, 148, 142 Negative trials, 217 Negligent errors, 23 Nested case–control design, 100–101, 151–152 New England Journal of Medicine, 29 Nominal measurement, 121 Nominal scale, 121–122 Noninferiority trials, 72–73 Nonparametric methods, 19, 221 Nonresponse in medical studies, 18, 55, 156 Normal distribution, see Gaussian distribution Normal values, 204, 205 Null hypothesis, 211, 212 Objectives broad and specific, 36–37 formulating research, 36–37 Observational studies, 39–41, 46, 52–53, 91–109 comparative performance of prospective, retrospective, and cross-sectional studies, 106–109 cross-sectional studies, 91, 104–106 prospective studies, 91–100 reporting guidelines (STROBE statement), 274–276 retrospective studies, 91, 100–104 sampling methods, 154–155 Odds ratio (OR), 101, 105, 188 adjusted, 226 confidence intervals, 209 from logistic regression, 223–224 sample size for estimation of, 160, 161 unadjusted, 226 Oral presentation, 265–269 Ordinal measurement, 122 Ordinal scale, 124–125 Ordinary least square regression, 227–229 Outcomes, definition, 47 Outliers, 181 Ovid (search platform), 31 Parallel control, 82 Patient, incomplete information on, 9–10 Patient equipoise, 87–88 300 300 Pearson correlation coefficient, 230 Peer review, 288–290 Percentages, misuse of, 237 Percentiles, 193 Personal equipoise, 87–88 Person-years, 188 Pharmacokinetic studies, 99 Phases of trials, 76–77 Pie diagrams, 202–203, 204 Placebo, 52, 82–83 Plagiarism, 283–274 Pilot study, 136–137 Point estimate, 156–157 Poisson distribution, 127 Polygon graphs, 200, 201 Polytomous data, 122, 195 Population, statistical definition, 148 Poster presentation, 269–270 Post-investigation steps, 18–20 analyze the data, 19 interpret the results, 19–20 monitor the reactions, 20 write and disseminate the report, 20 Post-marketing surveillance of drugs, 78 Power (statistical), 212–214 Pragmatic trials, 52, 73–74 Predictive validity, 138–139 Predictivities of medical tests, 59, 136–145 negative, 143–14 positive, 142–144 Predisposing factors, 47 Pre-investigation steps, 14–17 choose the sampling plan, 17 collect and evaluate existing information, 15 decide on the sample size, 17 develop the tools, 16–17 formulate research objectives and hypotheses, 15–16 identify the problem, 15 identify the study subjects, 16 think of a study design, 16 write the protocol, 17 Pretesting, 136–137 Prevalence, 142–143, 186 Prevalence rate ratio (PRR), 105 Prevalence studies, 42 Primary research, 4–6 Probability proportional to size sampling, 148, 153 Problem for research, defining, 27–28 Proforma, 127, 129 Proof of concept phase, 77 Prophylactic trials, 69, 70 Proportion (statistics), 192 Index misuse of, 236, 237 Proportionate sample, 151 Prospective studies, 40–42, 52–53, 91–100 basics, 90 cohort studies, 97–98 comparative features, 106–109 comparison group, 94–95 longitudinal studies, 98–99 merits and demerits, 96–97 performance comparison, 107, 108 potential biases, 95–96 repeated measure studies, 99–100 salient features, 93 sampling methods for, 154–155 selection of subjects, 93–94 Prospective surveys, 63, 64–65 Protocol main body (elements of), 169, 175–180 structure, 170–171 Proximal factors, 113, 114, 115 PubMed database, 30, 45 Pure research, 3, 4, 5 Purposive (nonrandom) sampling, 147, 148–149 P-value, 212–215 misuse of, 238 Qualitative data contingency tables, 197–198 statistical tests for, 217–220 summary measures for, 192 Qualitative nominal measurement, 122 Qualitative ordinal measurement, 122 Quality of a good researcher, 21, 22 Quality of data, 131–136 errors in medical data, 131–133 other aspects, 135–136 reliability, validity and accuracy of data, 133–132 Quality of life, 117, 118–119 Quality of medical research, 21–24 assessment of, 21, 22 characteristics of good research, 21, 22 errors in research, 23 fruits of medical research, 24 nature of truth, 21 pleasures and frustrations of medical research, 22–24 responsibility of researchers, 24 Quantitative data frequency tables, 196–197 statistical tests for, 215, 218–222 summary measures for, 192–195 Quantitative measurement, 121 Quartiles, 194 301 Index Quasi-experiments, 81 Quasi-randomization, 80 Questionnaires, 125, 126–127 structured, 128–129 Quota sampling, 149, 150 Random allocation, 32, 52, 57, 60, 62, 78–80, 85, 155 Randomization, 52, 70, 74, 78–82 block, 80 cluster, 80 simple, 80 statified, 81 Randomized controlled trials (RCTs), 52, 62, 63–64, 78 Random sampling, 43, 149, 150–153; see also Sampling methods Rate ratio, 190–191 Rates, 186 Rationalism, 2, 3 Ratios, 186–187 Ratio scales, 123–124 Recall lapse/bias, 101, 102, 103, 104, 130 Record linkage, 181, 182 References, 175 citation systems, 180 format of, 264–265 selection of, 263–264 Reference values, 205 Regression, 223–229 logistic, 102, 103, 225–227 multiple, 105, 228–229 ordinary least square, 227–229 simple, 227 Regression coefficient, 223 Regression models, 11–13 Relationships, 229–231 statistical tools to assess, 230 Relative risk (risk ratio) (RR), 93, 187, 189 confidence intervals, 209 sample size for estimation of, 161, 162 Reliability of clinical tools, 10 of data, 131–132 definition, 10 measures of, 10 of a study design, 54–55, 56 Repeated measure studies, 98–99 Repeated measures analysis of variance (ANOVA), 99, 220 Reporting, 243–271 oral presentation, 265–269 poster presentation, 269–271 writing a thesis or paper, 243–265 301 Reporting ethics, 283–293 confidentiality, 290–291 conflict of interest, 286–287 copyright, 286 covering letter with submission, 284–285 duplication, 284–285 misreporting of research, 291–292 permissions, 286 plagiarism, 285–286 Reporting guidelines, 273–281 clinical trials (CONSORT statement), 273–274 diagnostic accuracy studies (STARD statement), 277, 278–279 observational studies (STROBE statement), 274–276 reporting of statistical methods (Revised SAMPL statement), 2778, 280–281 Research see also Clinical trials; Medical research applied, 3, 4 basic, 3, 4 definition, 1 inter-disciplinary, objective of, 21 secondary, 3–4 translational, 5 Research methodology, training in, 5 Research protocol, 169–180 compliance with regulations, 1173 definition and purpose, 169 development of data collection form, 173–174 ethics, 173–174 executive summary, 170, 162 funding and training, 171–173 informed consent form, 173–174 logistics, 170, 172–174 main body, 172, 174–180 references and citation systems, 180 researcher(s), 171, 172 structure, 170–174 supervisors and collaborators, 170, 171–172 title, 170, 171 Resources for medical research, 34–36 Results, definition, 45 Retrospective follow-up studies, 90 Retrospective studies, 38–40, 51, 52, 98–102 basics, 90 case–control design, 99–101 comparative features, 104–107 merits and demerits, 101–102 performance comparison of, 105, 106 salient features, 99 sampling for, 98–99 sampling methods, 151–152 302 302 Retrospective studies (cont.) selection of cases and controls for, 101 Retrospective surveys, 63, 64–65 Review articles, 29–30 Risk measurement, 186–187 Risk ratio, see Relative risk ROC curves, 141–142 SAMPL statement (Revised), 273, 274–276 Sample, definition, 147 Sample size, 158–166 additional considerations for determination, 162–163 advantages of a large sample, 159–160 analytical studies, 160–164, 165–166 clinical trials, 160–164, 165–166 criteria for adequacy, 147, 148 descriptive studies, 160, 162 formulas, 164–165 reliability and, 54 rules of thumb, 163–164 Sample surveys, 41, 43–44, 63, 64–65, 148–150 Sampling, 55, 147–158 for analytical studies, 154–155 cross-sectional studies, 104 for descriptive studies, 148–153 for observational studies, 154–156 for retrospective studies, 100–101 role in medical research, 147 where to use which random sampling method, 152 Sampling error, 62, 156–157 Sampling fluctuation, 156 Sampling fraction, 151 Sampling frame, 149, 150 Sampling methods cluster random sampling, 149, 152 consecutive sampling, 78–79, 83, 145 haphazard sampling, 149, 151 multistage random sampling, 43, 105, 149, 152 probability proportional to size sampling, 149, 153 proportionate sampling, 151 purposive sampling, 149, 150–151 quota sampling, 149, 150 random sampling, 43, 149, 150–153 simple random sampling, 43, 149, 150–151 snowball sampling, 149, 150 stratified random sampling, 149, 151 systematic random sampling, 149, 151–152 Sampling strategies, 43 Sampling unit, 149, 150 Scales of measurement, 122–126 metric, 122, 123–124 Index nominal, 122–123 ordinal, 122, 124–125 semiordered, 123 Scatter plots, 200, 201 Schedule, 127, 129 Scientific writing, text style, 243–244 SciSearch, 30–31 Scores, 184 Scoring systems, 234 Screening trials, 70–71 Secondary data, cautions in using, 33–34 Secondary research, 3–4 Semiordered measurements, 124 Sensitivity analysis, 229 Sensitivity and specificity, 138–145 Serial surveys, 149 Significant digits, 237–238 Simple random sampling, 43, 149, 150–151 Single blinding, 84 Skewed distribution, 120 left skewed, 205 right skewed, 205 Slope (in regression), 228 Smoke-pipe distribution, 205 Snowball sampling, 149, 150 Spearman correlation coefficient, 230 Standard deviation, 194 Standard error, 43 of an estimate, 157 Standardization of values, 185 Standardized deviates, 185 Standardized variates, 185 STatement for Reporting studies of Diagnostic Accuracy (STARD), 273, 277–278 Statistical analysis, 209–240 errors can cause many deaths, 240 validation of methods, results, and conclusions, 234–236 Statistical distribution of medical measurements, 204–206 Statistical fallacies, 236–240 Statistical interaction, 47–48 Statistical methods avoiding bias, 61 guidelines for reporting (Revised SAMPL statement), 277, 280–281 Statistical models, 11 Statistical significance, 88, 209–217 alpha level of significance, 214–215 Statistical tests basic tests, 217–223 philosophical basis of, 213 Statistical uncertainties, 88 Stratified randomization, 80 303 303 Index Stratified random sampling, 151, 154 STrengthening of Reporting of OBservational studies in Epidemiology (STROBE) statement, 274–276 Student’s t-test, 218–219 Study designs analytical studies, 40–42, 46–54 clinical trials, 40–41, 42, 55, 67–89 cross-sectional studies, 40–42, 53, 54 definition of the design of an empirical study, 38–4 descriptive studies, 40–46 double-blind randomized controlled trial (RCT), 52 elements of a design, 39–40 epidemiological studies, 53 follow-up studies (prospective studies), 40–42, 53–54 intervention studies, 40–42, 47, 51–53 medical experiments, 41–42, 47, 51–52 observational studies, 40–42, 45, 53–54, 91–109 overview, 39–63 prospective studies, 40–42, 53–54 randomized controlled trials (RCT), 52 reliability of a design, 54–55, 56 retrospective studies, 40–42, 53, 54 selection of an appropriate design, 63–65 types of designs, 40–42 validity of a design, 53, 54–56 Superiority trials, 72–73 Survival duration, 114, 117, 118, 119 Survival analysis, 93, 119 Synergism, 48 Systematic random sampling, 148, 151–152 Systematic sampling, 105 Tables composite, 199 contingency, 197–198 features of, 198–199 frequency, 19, 196–197 Target population, 148 Tertiles, 194 Therapeutic equivalence, 72–73 Therapeutic trials, 68–70 Thresholds of normal levels of quantitative measurement, 206 Tools for the clinician imperfect nature of, 10 validity and reliability of, 10 Topic of medical research, 27–37 availability of facilities, 35–36 availability of resources, 34–36 choosing a problem for research, 27–28 choosing a thesis topic, 28 defining a problem, 27–28 graduate theses, 36 statement of the problem, 28 time considerations, 35 Training in research methodology, 5 Translational research, 5 Treatment group, 52 Trials, see Clinical trials Triple-blind trials, 84, 85 T-scores, 185 Tukey, Bonferroni, and Scheffe procedure, 222 Two-stage design, 76 Two-way tables, 219–220 Type I (alpha) errors, 213–214, 240 Type II (beta) errors, 213–214, 215–216, 240 Unbiased estimate, 43 Uncertainties in clinical trials, 87–88 Uncertainties in medical research, 8–14 aleatory uncertainties, 11–13, 14 chance variability, 10 epistemic gaps in research results, 10–11 epistemic uncertainties, 8–11, 12–13, 14 imperfect nature of clinical tools, 10 inadequate knowledge, 8, 9 incomplete information on the patient, 9–10 management of, 14 Uncertainty analysis, 229 Uncertainty principle, 87 Unintentional errors, 23 Use effectiveness, 73 U-shaped distribution, 205 Vaccine trials, 71 Valid test, definition, 138 Validity of clinical tools, 10 of clinical trials, 83–89 of data, 133–134 of data analysis methods, results, and conclusions, 32, 234–236 of data collection tools, 136–145 definition, 10 external validity, 235–236 internal validity, 235 of study design, 54, 55–56 Vancouver format for listing references, 180 Variables, 224 concurrent, 57, 94, 232 confounding, 224 continuous, 126–127 dependent, 224 304 304 Variables (cont.) discrete, 126–127 independent, 224 Variance, 194–195 Venn diagrams, 200, 202 Index end features, 248, 262–265 main text, 248, 254–262 preliminaries of a manuscript, 248–253 Wrong assessment, 158 Youden index, 141–142 Whistle-blowers, for misreporting of research, 291 Writing a thesis or paper, 243–265 effective scientific writing, 243–248 z-score, 185–186 Z-test, 219 ...i Research Methods for Medical? ?Graduates ii iii Research Methods for Medical? ?Graduates Abhaya Indrayan iv CRC? ??Press Taylor & Francis Group 6000 Broken... and for other such small-scale endeavors Advanced methods would be different for, say, cancer research than for tuberculosis research and for a drug trial than for behavioral research For such... 1.3) 6 Research Methods for Medical Graduates BOX 1.3 LEVELS OF PRIMARY MEDICAL RESEARCH First Level of Research – Master’s Thesis • Generally a small-scale investigation that puts forward