Guest Editorial: Ethics in Research: What Is It and Who Cares?

   Traditionally, we think of ethics in research in terms of protecting human subjects. Maintaining the safety of human subjects is important; ensuring research integrity is equally imperative.

Research integrity includes: making an unbiased sample selection and logical data decisions, reporting the results correctly (and completely), and including correct (and complete) citations from previously published work. Unfortunately, the more one reads and uses research, the more one realizes that exercising research integrity does not appear to come "naturally" to everyone.

   Human subject protection means the study participant will be protected from potential harm, that the benefits of research participation outweigh the risks, and that subjects are fully informed about the purpose of the project. To that end, Institutional Review Board (IRB) Review is a must. Researchers who receive federal funding are required to complete a human subject's protection course. It is hoped that these measures will prevent the recurrence of some of the more serious violations in the treatment of study participants that have occurred in the past.

   Selecting an appropriate study sample also falls in the category of "research integrity." After all, anyone with expertise in a particular clinical area can select a study sample that will make a product or service look good. Ideally, a sample includes everyone with the problem - eg, all people with pressure ulcers in the world. Because this option is not very realistic, most researchers will study those with pressure ulcers in the nursing home next door. Using this so-called convenience sample is convenient, but the people selected do not represent everyone everywhere with the problem. A stronger form of intervention sampling is random selection. This allows available subjects to have an equal chance of receiving the treatment or intervention. While random selection eliminates selection bias on the part of the researchers, it does not guarantee that the groups are similar. What happened to people who didn't complete the study? Were they different or similar? Would their continuation in the study have changed the results? As consumers of research, clinicians need to know what happened to everyone selected to participate in the study.

   Another ethical issue facing researchers relates to missing or outlier data decisions. Is the data missing because subjects chose not to answer the questions? Is it missing because they were too ill to undergo all tests? The researcher must decide whether to include such data in the final analysis or how to statistically control for possible variables. The same is true when dealing with results outside of the expected range. For example, if I were doing research involving a small sample size, one subject could have a serum albumin level of 6.5 g/dL (normal is 3.5 to 5.0 g/dL). If I reported the average albumin on data that includes this very high value, my study sample would appear to have adequate nutrition when, in fact, it did not. The decision to include or exclude patient data in the final analysis can dramatically change the study's results. These decisions should be made before initiating the research project and be based on sound objective reasoning.

   Research integrity extends beyond conducting a study and analyzing the results; it also applies to presenting and publishing your work. Publishing only expected or positive study results seriously compromises scientific integrity, progress, and patient care. For instance, three publications may provide information about the effectiveness of a particular wound dressing, but few people may know that five unpublished studies indicate the product is not effective (or worse, is not safe). In the real world, the results of most studies are mixed. Some findings are expected, others are not. Some unexpected or negative findings may be related to problems with the intervention itself and others may be the result of poor study design, a problem with the sample, or an unanticipated event at the study site. Not sharing all information is a serious ethical concern and violates research integrity.

   Finally, all healthcare professionals, whether publishing the results of a study or writing a review article, should be concerned about the ethics of citations. This includes plagiarism, slanting the words of others to fit one's own needs, or selectively citing previously published data. For example, finding misrepresentations of previously published research results or seeing statements such as, "The majority of ulcers are preventable" translated into "All ulcers are avoidable" is not uncommon practice.

   The first time someone told me they read my work and used my results scared me to death. I was overwhelmed with the enormity of the realization that my work had the potential to change practice. Articles were more than just words on paper. My results were important to people's lives. I went home and rechecked my work to make sure I had done everything right.

   Researchers are encouraged to publish, publish, publish. But quality of the research is more important than quantity. Novice and experienced researchers need to maintain high standards of ethical integrity. No one involved in medical care would purposefully harm patients by using inappropriate treatments, medications, or procedures. Although researchers do not provide direct hands-on patient care, their work touches the lives of individuals indirectly. Unethical research work or research reporting can harm people unintentionally. Consumers and practitioners need to critically evaluate what they read and how that information translates into their practice.