A RISK COMMUNICATIONS BIBLIOGRAPHY Explaining Risks to the Public INTRODUCTION Why a Risk Communication Bibliography Is Necessary Science is continually adding to our store of knowledge about health and safety problems. Researchers learn about the prevalence and consequences of various hazards, the risk factors that determine individual vulnerability, and the choices available for prevention and treatment. The need to share this information with the public, health professionals, the media, and policy makers is obvious, but it is also becoming clear that effective transmission of these research findings is a serious challenge. The probabilities and odds that form the language of scientific research are seldom understandable to nonscientists, and even when understandable, nonscientists may not find such statistics very helpful in making actual decisions. Much communication about risk is patently unsuccessful. Because considerations of risk are relevant to so many human endeavors, attempts to improve risk communication are presently occurring in a variety of different professions and disciplines. As a consequence, the literature on this topic is unusually dispersed. Discovering what is known about any particular issue in risk communication is extremely difficult. SCOPE The Scope of the Bibliography This bibliography provides an introduction to the risk communication literature through the spring of 2000. The bibliography does not pretend to be a complete collection of all relevant articles, and many valuable sources have undoubtedly been missed. Nevertheless, its nearly 400 listings should provide a good starting point for searching the field. Emphasizing recent publications, it includes reports of experiments, theoretical discussions, case histories, how-to manuals, dissertations, and reviews. The bibliography is not limited to cancer risks because much of use to cancer risk communication can be gleaned from lessons learned in other domains. To set realistic boundaries, we focused this bibliography on publications that relate to the specific task of explaining the nature and magnitude of hazards to the public. This focus unavoidably neglects important social, legal, political, ethical, and institutional issues in risk communication. (A non-annotated list of references covering the broad field of risk communication may be found at the National Library of Medicine: www.nlm.nih.gov/pubs/cbm/health_risk_communication.html.) The bibliography does, however, include many articles that deal with risk perception, even ones lacking any mention of communication, because learning how people think about risk should help us discover how well they understand our risk messages and help us find ways to increase their risk understanding. ENTRIES How Entries Were Located The two primary sources of the items in this bibliography, PsycINFO(c), the database in psychology and Medline(c), the database in medicine, were searched using appropriate keywords. Another valuable resource was the compendium of articles on risk communication, edited by Barbara Rimer and J. Paul Van Nevel, that was published as the Journal of the National Cancer Institute, Monograph 25 (1999). This bibliography includes the articles that appeared in that monograph. We also used the reference lists at the ends of those articles to find additional relevant publications, and received help from the authors of the articles, several of whom shared with us their personal collections of risk communication articles. The Information Provided about Each Entry A simple list of references would have been relatively easy to compile, but its usefulness would have been limited by the fact that a title is not always a good guide to the content of a document. This bibliography provides citation information, but it also contains a one-sentence summary of each entry. Sentences were written to emphasize the issues that are addressed by the author(s) rather than the conclusions that are reached. In addition, articles were coded on a number of dimensions ("coding categories") so that those articles dealing with particular issues will be easier to locate. This coding system was especially designed to classify empirical research articles, describing the setting, communicator, audience, hazard, outcome variable, and so forth. Many of these categories do not pertain to other types of articles or documents, but they were used whenever they were applicable. Because dissertations are difficult to obtain, those appearing here have not received the detailed coding given to journal articles, books, chapters, and reports. The coding categories and an example of a code used in each category follow in Table 1. If a category proved to be relevant to a publication, the appropriate code (or codes) was assigned. If the category was only discussed in general terms or if many different aspects of the category topic were examined, the name of the category, rather than specific codes, was entered to indicate this situation. If a category did not prove to be relevant to a particular publication, no code was assigned for that category. A complete list of all the categories (Bibliography Coding System) and all the codes used within categories appears at the end of this document. Table 1 Coding category--------------------Example Publication type-------------------discussion Focus------------------------------format Communicator-----------------------celebrity Audience role----------------------health/safety professional Audience ethnicity-----------------African-American Audience gender--------------------female Setting for risk communication-----health care setting Channels of risk communication-----radio Content of risk message------------definition of hazard Form of risk information-----------graphical display Outcomes---------------------------recall of message Hazard-----------------------------cancer Methodological issues--------------evaluation criteria Bibliography Users For Whom the Bibliography Is Intended This bibliography is intended for anyone with an interest in risk communication, including health educators, media professionals, psychologists, physicians, nurses, social workers, environmental scientists, public officials, and others. In other words, anyone who has to communicate risk information and anyone who would like to study issues in risk communication. Still, users should recognize that because there is so much we do not yet know, someone who looks to this bibliography for detailed guidance in the design of effective messages is likely to be disappointed. In fact, a great many of the entries are reports of original research investigations, each focusing on a very specific issue. There are many valuable insights and helpful principles in this research literature, but no cook-book answers. We know more today about what does not work than about what does. CODING SYSTEM Categories and Examples of Coded Entries These coding categories were developed after examining many bibliography documents. Nevertheless, during the actual coding, issues suggesting new categories or new codes within categories sometimes appeared. Unfortunately, the system could not be changed without recoding everything, which was not feasible. Thus, there is room for further development and improvement of this coding system. Also, despite attempts to create clear definitions that would produce coding reliability, there is unavoidably some subjectivity in how each entry was finally coded. Thus, when looking for publications on a given topic it is best to search in multiple ways. For completeness, one can use not only the existing coding categories and specific codes but also search the entire record (including the title and sentence summary) for words or phrases that might also describe the same idea. BIBLIOGRAPHY SEARCH Alternative Versions of the Bibliography You can search the HTML version of the Risk Communication Bibliography online or you can download a copy of the bibliography to the hard drive of your personal computer in text (riskcombib.txt), Microsoft Word for Windows Version 2000 (riskcombib.doc), or Wordperfect for Windows Version 9 (riskcombib.wpd) file types. There is also an instruction file to help you search the downloaded bibliographies. File Name------------Version--------------Size Riskcombib.wpd-------Word Perfect 9-------1,049K Riskcombib.doc-------MS Word 2000---------473K Riskcombib.txt-------Text-----------------329K Instructions.txt-----Text-----------------19K Searching the Downloaded Bibliography Files Searching by First Author's Name If you want to locate a document and know the name of the first author, you can use the letters at the top of each section to quickly move to the area of the bibliography containing documents whose authors' names start with this letter. However, you may ignore these letters if you are performing any of the searches described below. If you would like to search by the last name of any author other than the first author, please see the next paragraph. Searching for Specific Words It is easy to search for any word or sequence of words. Simply type the word or phrase you wish to locate into the "Find" window of your word processor and proceed to search for the first occurrence of the search text you entered. (The "Find" or "Search" tool is accessed in most programs with the Ctrl-F key combination or under the "Edit" menu.) Use the "Find Next" or "Find Again" option to locate the next instance of this text, and continue to find all instances of this text. For example, if you want to find all documents on which a particular person is listed as an author, you can type his or her last name in the "Find" or "Search" window. All the articles of which that person was author, not just first author, will be located. Searching for Particular Codes Often, however, you will want to search for a particular code within a particular category. For example, the word "cancer" appears many times in the bibliography in many different fields, but you may wish to locate only those documents in which cancer was the topic of the risk communication; in other words, those articles in which "cancer" was the code assigned to the category "hazard." Because each code within a category is preceded by a number (e.g., "1. Alcohol") and such number-plus-text combinations are unique, you can search for a particular code in a specific category by looking for the entire text string (number plus text) with the "Find" command. In this case, the sequence of number and letters, including the blank spaces between the number and letters, must be exactly the same in your Find or Search command window as in the bibliography for the search to be successful. Thus, it is strongly recommended that you copy the string from the detailed list of codes in the Bibliography Coding System, that appears at the beginning of the bibliography and also at the end of this introduction. Once you have copied the string of text, paste it to the search field of your browser or word processor. To do this easily, go to the complete list of codes and highlight the specific code you want to find. Then press Ctrl-C to copy this string; press Ctrl-F to activate the Find procedure; and press Ctrl-V to copy the string into the Find window. Then click the appropriate button to begin the search, and use the "Find Next" or "Find Again" options to locate additional documents in which this code appears. Working with a Printed Copy of the Bibliography If you have only a printed copy of the bibliography, it is necessary to page through it to find the types of documents you are looking for. Limits on Searching You cannot search the downloaded bibliography for several codes at once (e.g., Hazard coded as "colon cancer" and audience coded as "teen"), nor can you easily create a list or file of all the documents that match your search criterion. (It is possible to achieve such a file, however, by copying each document you find to a separate file.) These search capabilities are available on the online version. ACKNOWLEDGMENTS The Risk Communications Bibliography was created by Professor Neil D. Weinstein, Department of Human Ecology, and Cara L. Cuite, a doctoral student in the Department of Psychology, Rutgers, The State University of New Jersey. Barbara Rimer and Robert Croyle, director and associate director respectively in the Division of Cancer Control and Population Sciences at the National Cancer Institute were the original catalyst behind producing the Bibliography. Invaluable assistance and contributions to the Bibliography were received from Jessica Heppen, Alison Smith, and Diane Vreeland. The National Cancer Institute, National Institutes of Health, U.S. Department of Health and Human Services provided funding for the Bibliography under the contract mechanism 263-MQ-915306-1. Coding System Note that category names are given first, with additional clarifying information sometimes appearing in brackets. The category codes include the text in parentheses (if any is given), but not clarifying text occasionally present in brackets. Publication Type [If multiple codes apply, the code assigned is the highest that describes the publication.] 1. Discussion 2. Review 3. Case study 4. Empirical 5. Intervention Focus [The code assigned indicates the coder's judgment as to the issue or issues most emphasized. Sometimes several of these topics receive detailed attention in an article. A maximum of three focus codes were assigned to each article. If more than three topics received emphasis the code "11. Multiple" was used.] 1. Communicator 2. Audience 3. Setting 4. Channel 5. Content 6. Format 7. Outcome 8. Measurement/evaluation 9. Risk perception/comprehension 10. Decision aid 11. Multiple 12. General Communicator 0. Communicator (general) 1. Authority figure (non-expert) 2. Business/corporation/commercial group 3. Celebrity 4. Expert/researcher/scientist 5. Governmental organization (e.g., NCI) 6. Investigator/experimenter 7. Health professional/worker in health care setting 8. Non-governmental organization (e.g., Am. Lung Assoc.) 9. Parent 10. Patient/peer/community member (affected) 11. Peer/community member (not affected) 12. Unspecified 13. Other 14. Media Audience role 0. Audience role (general) 1. Health/safety professional 2. Child/Pre-teen 3. Teen 4. College student/Graduate student 5. Adult 6. Parent or prospective parent 7. Senior citizen 8. At risk/exposed-child or teen 9. At risk/exposed-adult 10. At risk/exposed-parent of child 11. At risk/exposed-senior citizen 12. Ill-child 13. Ill-adult 14. Ill-parent of ill child 15. Ill-senior citizen 16. General public 17. Other 18. Patient (general) 19. Other professionals and experts Audience gender 1. Female 2. Male Audience ethnicity 1. African-American 2. Asian 3. Caucasian 4. Hispanic 5. Native American 6. Mixed/varied/multiple 7. Other Setting 0. Setting (general) 1. Armed Forces 2. Community meeting/community group 3. Focus group/small group 4. General community 5. Health care setting 6. Laboratory 7. Prison 8. Educational setting: school or university 9. Workplace 10. Other 11. Home Channels of risk communication 0. Channels (general) 1. Computer program (non-internet) 2. Internet (listserves, web pages) 3. Lecture 4. One-on-one (face-to-face) 5. Pamphlets, newsletters, fact sheets, handouts, or questionnaires 6. Print media (newspaper or magazine) 7. Radio 8. Small groups/focus groups 9. Telephone 10. TV 11. Video 12. Other 21. Letter 22. Community meetings/community groups 23. Audiotape 24. Product label/packaging 25. Media-general Content of risk message 0. Content (general) 1. Cause/source/communicability of risk/mode of transmission 2. Communication style 3. Definition of hazard 4. Expense of hazard 5. Likelihood-absolute 6. Likelihood-comparative (across people or hazards) 7. Possible health outcomes 8. Possible psychosocial outcomes 9. Preventability of hazard/personal control 10. Risk factors (exposure levels, blood pressure, genetic risk) 11. Severity of outcome 12. Time course of hazard 13. Treatability of hazard consequences 14. Other 15. Costs/barriers/losses 16. Benefits/gains 17. Skill building/instructions 18. Reminder 19. General information about hazard 20. General information about recommended action 21. Test results 22. Personalized information Form of risk information 0. Format (general) 1. Decision aid 2. Graphical display 3. Interactive 4. Text-numerical 5. Text-verbal 6. Other 7. Oral/spoken Outcomes 0. Outcomes (general) 1. Attitude 2. Comprehension of message 3. Knowledge 4. Communicator-Audience relationship (e.g., trust) 5. Psychological distress/affect/concern/preoccupation 6. Recall of message 7. Risk perception-likelihood 8. Risk perception-severity 9. Risk perception-other attribute of risk 10. Satisfaction/communication appraisal 11. Actual information seeking behavior 12. Actual preventive health behavior/risky behavior 13. Actual testing/screening 14. Actual treatment behavior 15. Actual other behavior (e.g., activism) 16. Intended information seeking behavior 17. Intended preventive health behavior/risky behavior 18. Intended testing/screening 19. Intended treatment behavior 20. Intended other behavior (e.g., activism) 21. Other 22. Perceptions of recommended action/treatment 23. Perceptions of efficacy/control 24. Community or public response 25. Policy/regulation changes 26. Corporate response 27. Information processing/decision-making process 28. Information/communication preferences Hazard 0. Hazard (general) 1. Alcohol 2. Arthritis 3. Asthma/Emphysema/Chronic lung disease 4. Automobile safety 5. Birth defects 6.0 Cancer [Note: If a specific type of cancer was coded, the general cancer code 6. 0 was also assigned to the publication.] 6.1 Breast cancer 6.2 Cervical cancer 6.3 Ovarian cancer 6.4 Prostate cancer 6.5 Skin cancer 6.9 Lung cancer 6.10 Pancreatic cancer 6.11 Bladder cancer 6.12 Colon cancer 6.13 Other cancer 7. Diabetes 8. Drug abuse 9. Pollutants/chemical exposure 10. Other environmental (flooding, climate change/not pollutants) 11. HIV/AIDS/STDs 12. Other infectious diseases (e.g., Lyme disease) 13. Hypertension/heart disease/stroke 14. Hypothetical risk 15. Injury/safety (non-auto) 16. Radiation (radon, sun exposure, x-ray) 17. Smoking 18. Multiple (greater than 5) 19. General, unspecified 20. Other 21. Dental issues 22. Treatment related risks 23. Food related risks 24. Menopause Methodological issues 1. Evaluation criteria (how to determine effectiveness of risk communication) 2. Measurement (studies of different response formats) 3. Other