Statistics for Healthcare Professionals

An Introduction

Ian Scott, Debbie Mazhindu

• Publisher: SAGE Publications Ltd
• Year: 2014
• ISBN:
  9781446298039
• Edition: 2
• Topic: Medicine
• Subtopic: Medical Theory, Practice & Reference

This book introduces you to the basic rules and principles of statistics and shows you how straightforward statistical tests can be performed and analysed in relation to healthcare.

Key features of the text are: - Examples which show how statistics are used in the real world
- Exercises to test your understanding and develop your proficiency
- A glossary to remind you of key terms.
- Additional online material including guides to using Microsoft Excel and SPSS and sample datasets.

1

Statistics for health care research

Areas of learning covered in this chapter

• How can you use this book and use statistics to enhance practice?

• What is evidence-based practice (EBP)?

• How are statistics used to enhance evidence-based health care?

• How can statistics help you to evaluate professional knowledge and professional values as the evidence base for practice?

Scope of the book

This book aims to expand health care students’ and professionals’ knowledge and understanding of statistics within health care practice. We hope that through reading and using this book you will be encouraged to evaluate statistical analysis as a technique and its relationship to evidence-based practice, and develop a practical understanding of basic statistics. There are many different approaches to investigating questions in health care practice. We deal with quantitative approaches to investigating problems in health care, which require an understanding of some of the rules and principles of statistics. Understanding research in health care requires appreciating both qualitative and quantitative approaches to analysing data. Students of health care professions, both medical and non-medical, can expect to encounter a range of research approaches, which use particular statistical techniques to derive answers to particular patient problems. To this end, health care practice involves understanding of a range of research techniques within health-related subjects, including the use of statistics.

Glossary When you see a term in bold you may want to look it up in the glossary of terms at the end of the book. This will explain terms used and help you to remember key words.

For many students and practitioners of health care, statistics is a subject that can often appear incomprehensible, daunting and, worse still, far removed from the real issues and problems encountered when caring for people. Within the book we try to demystify some of the more commonly used statistical tests and help readers understand the language of research that uses statistics, which both novice and advanced researchers find very off-putting to read. Undertaking the exercises within and at the end of each chapter will encourage you to think critically and reflectively about the research that you encounter, both in terms of the statistical process and the dynamic health care context in which studies are executed. The book can be used as a basis for taught courses on statistics at both pre-registration and post-registration level and as a reference guide for those using statistics in health care settings. The book is an introduction and aims to take you from novice to advanced beginner. Statistics is a vast subject which many people find difficult. We hope to make the learning process easier, but we can’t promise an effort-free process.

The book is divided into twenty chapters. Each contains explanations of its terms and use for reference. A glossary has been provided and every time you encounter a word in bold print you can look up a short definition in the glossary. At the end of each chapter you will find questions and exercises. Using these exercises will allow you to practise, become more proficient and develop your understanding.

The book takes a ‘how to’ approach, which explains how specific statistical tests can be performed. We have included details of how to calculate many of the statistics by hand, because doing at least some of the statistics by hand will help you to get a feel for the processes. As the statistical techniques become more advanced, readers are directed towards suitable computer packages and other literature. Understanding statistics can sometimes be difficult for students of health care, as statistics often appear far removed from immediacy of patient problems. Examples of statistical techniques commonly used will be explored and the results of a hypothetical questionnaire on sexual health and a clinical trial are used to encourage you to practise and explore statistics. We encourage you to try calculating statistics by hand, at least at first as this will help you to develop a feel for what is going on. Computers with statistical packages are now more readily available in some health care settings, particularly in the more industrialized parts of the world. As tests become more advanced and you feel more confident, do try to use the statistical computer packages that are available. (A short guide to using SPSS and Excel can be found at www.sagepub.co.uk/scott2e.)

The use of clinical and community-based studies will form a central thread to allow data analysis to be explored from the perspective of differing subject areas. As well as analysing data from the studies provided we encourage you to analyse your own data, collected in response to some sample questions.

We encourage you to think critically about data analysis and research design and how appropriate research design impacts upon evidence-based practice, because an understanding of statistics is essential if the numerous reports and documents issued within the health industries are to be scrutinized and considered in a critical manner.

Box 1.1 Reflective exercise

Take a few moments to think about the decisions made in health care today.

• How do we make decisions about what is the best care for patients and clients?

• How do you make decisions about what is best for you?

• How do you feel about making decisions that may affect another person’s life?

• How do you feel about people’s health?

• How do you feel about circumstances which may affect a person’s ability to: give birth, father or mother children, recover from cancer, cope with devastating trauma, cope with loss or grief, help someone integrate back into society after mental illness?

How do we as health care providers analyse such problems and then come to reach decisions that have an impact on many people’s lives? The answers to these questions lie in the basis we use for our professional knowledge, the power we have to implement changes in practice and what constitutes our evidence base for our practice.

After reading this book, we hope that your skills of critical analysis will have become more refined. We anticipate that you will have a better understanding of the process of research and the use of statistics and all that is involved in getting answers to problems. We hope that you will be able to understand the importance of carefully reading and reviewing research reports in order to come to conclusions about their relevance to your area of practice. It is also hoped that you will begin to feel comfortable talking the language of quantitative research, which can assist you when presenting a case for changing or improving an area of practice.

The use of statistics in health care research

There has been an increase in the development of research-based and evidence-based practice (EBP) in health care over the last two decades. For students undertaking professional health care education and those practising as professionals in health care, understanding the statistical terms used in research is of paramount importance when evaluating research studies.

One of the key aims of this text is to enable the development of a greater understanding of the process and practice of using statistics in order to find answers to complex health problems. After the initial preparation to practise health care, qualified practitioners are charged with a duty to care for patients and clients in the best manner possible. This means: taking account of patients’ and clients’ family life and their involvement with others, the knowledge and skills attributed to practice, medical orders and the role of the allied health professions (AHPs). Understanding statistical terms and analysis helps practitioners to make sense of the many research studies that underpin EBP, takes account of professional values as service providers to the public and enables practitioners to undertake studies themselves.

What are the goals of health care research?

The primary goal of health care research is to aid patient and client care by developing a scientifically based body of knowledge, which can be used to help us with decision making, to develop new practice and promote the professional role. The degree to which a body of people such as health care professionals can be judged as professional resides to a large extent on the body of specialist knowledge that they can draw on for practice. Developing a specialist knowledge base by a process of scientific enquiry can aid the professional standing of a professional health care provider. The ultimate aim of health care providers is the delivery of safe, effective care for patients and clients. This is an essential prerequisite in all health care professionals’ codes of conduct.

Understanding statistics is just one way of ensuring that professional practice is based on the best available evidence to date by which to treat and help the wider community. Health care research shares many of the qualities of practice interventions, as they are both practical and intellectual activities, which have a defined language to learn. Good practice and good research do not spring from ad hoc or sloppy practice. Health care practice and health care research both use processes which when applied rigorously can improve patient care. Health care practice that is ritualized, not research-based and not assessed, not planned and (more importantly) not evaluated to see if it makes a difference to patient care can kill people, as it leads to unsafe practice. Evidence of this is to be found in the various catalogues of disciplinary hearings concerned with the safe conduct of health care providers of all disciplines.

Box 1.2 Professional values

Take a look at this set of values, which are common to most health care professions; how do you think a knowledge of statistical analysis could help you to maintain these values?

• To practise within the various codes of conduct, which regulate your chosen health care discipline.

• Be aware of cause and effect of health and ill health.

• Track progression and regression of health states and disease.

• Serve society by implementing the best practice based on available evidence.

• Teach new generations of health care professionals by using up-to-date research findings.

• To be self-governing and self-regulating in order to protect and serve the public.

Randomized control trials (RCTs)

Currently in health care practice, the ‘gold standard’ type of research is the randomized control trial (RCT) but even this kind of research has its critics and is not immune from accusations concerning its truthfulness and the way RCTs are conducted. For example, there are disagreements about how many RCTs should be undertaken before the outcomes can be used as evidence, how big the samples are to be legitimized as ‘representative’ and how ‘international’ a range of samples are before data is acceptable as ‘significant’. An RCT is a highly specialized research undertaking as usually the patients are carefully selected prior to randomization; usually they are younger and healthier than the people who will actually be taking the drug, and the people in the sample selected will only usually have one thing wrong with them and only be taking that one drug. Conversely, people in the ‘real world’ will be older and very likely have two or three conditions for which they are also taking five or more drugs making comparisons difficult if not impossible. RCTs do, however, remain the ‘gold standard’ form of research in health care.

The fact remains that research, from whichever perspective, that is not rigorous, systematic, ethical and well designed can have devastating effects on people’s lives. The thalidomide drug research is a case in point. Maynard (2003) suggested that medication errors occurring in health care in the United States kill twice the number of people per year than th...

Table of contents

• Cover Page
• Halftitle
• Advertisement
• Title
• Copyright
• Contents
• About the authors
• Acknowledgements
• Additional online material
•   1   Statistics for health care research
•   2   The statistical approach: when should it be applied?
•   3   Measuring, sampling and error
•   4   Questionnaires
•   5   The studies
•   6   Descriptive statistics
•   7   Displaying data
•   8   Lies, damned lies and statistics
•   9   An introduction to hypothesis testing
• 10   Distributions and probabilities
• 11   Making predictions
• 12   Testing for differences between means
• 13   Errors and ANOVAs
• 14   Not normal
• 15   Non-parametric tests
• 16   Tests for association: chi-square
• 17   Tests for association: correlation and regression
• 18   How big a sample: power analysis
• 19   Analysing data from systematic reviews
• 20   Choosing test statistics
• Appendix 1  The common symbols and abbreviations used in statistics
• Appendix 2  A guide to critically analysing statistics
• Appendix 3  Statistical tables
• Appendix 4  Answers to exercises
• Glossary
• References
• Index