 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
a Division of General Internal Medicine, Brigham and Women's Hospital, Boston, Massachusetts
b Department of Public Health, Cornell Medical School, New York, New York
c Division of General Internal Medicine, Harvard Medical School, Boston, Massachusetts
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS. To address these issues, we analyzed newspaper coverage of pediatric medication errors and adverse drug events. We searched Lexis Nexis for newspaper articles on pediatric medication safety from the United States, Canada, United Kingdom, Australia, and Ireland, during a 10-year period (1994–2004), by using specific keywords. Main outcome measures were the number of articles (adjusted for population), the type of events covered, and article slant. We also examined qualitatively the overall themes and the extent to which these articles portrayed a culture of safety to the public.
RESULTS. Throughout the world, there was a steady increase in articles on pediatric medication safety, peaking in 2003, with the highest per-capita rate in Canada. Approximately 65% of articles were about patient incidents, 20% mentioned policy, and 25% discussed research. Of the reported events judged to be negative for patient safety, 75% were covered in a neutral manner and 19% were covered in an unduly negative manner.
CONCLUSIONS. Media coverage of pediatric medication safety has increased in the past 10 years. Reporting of patient safety failures was generally fair, and reports were generally framed in light of a culture of safety.
Key Words: safety • media • public opinion
The media play a pivotal role in forming public opinion by presenting news and information to the public that shape their views. In addition, the degree of coverage of news stories reflects the public's interests.1–4 Newspapers represent a key part of the media, and it is relatively straightforward to assess what information has been presented during a defined period, compared with other media such as television and radio.
The public is justifiably concerned about medical safety. In a Kaiser Family Foundation survey in 2000, 47% of respondents reported that they were "very concerned" about an error resulting in injury happening to them or their family members, when receiving health care in general.5 When a child dies or is injured unnecessarily, it is especially heart-wrenching. Adverse drug events occur in 2.3% to 6% of all pediatric inpatient admissions.6,7 In ambulatory pediatrics, 16% of patients experienced an adverse drug event.8
Providers and even more the public often blame individuals when an accident occurs. In a study comparing the views of the public and physicians, Blendon et al9 found that members of the public were more likely to think that the party responsible for the error (ie, the care provider) should be sued for malpractice than were the physicians surveyed. Four percent of physicians thought that surgeons should be sued, whereas 30% of the public supported this (P < .001). The public also endorsed suggestions that care providers should be fined by a government agency and have their licenses suspended and involved institutions (such as hospitals) should risk the loss of accreditation.9 This conflicts with increasing evidence suggesting that improved safety is most likely to be achieved in nonpunitive cultures in which mistakes are seen as opportunities for improvement.10 The importance of a safety culture has been demonstrated in the aviation industry.11,12 News editors commission stories and reporters develop stories that are topical (in the public eye) and will interest readers.2,13 Therefore, it is not surprising that children's deaths or injuries receive a great deal of coverage.14 These articles may be influential, affecting not only the public but also policymakers when the topic ignites sufficient public outcry.2 These articles either can motivate hospitals to improve or can influence hospitals or providers to cover up future events.4
In this study, we attempted to understand how the public is presented with information about pediatric medication safety. Our aims were (1) to quantify the amount of newspaper reporting of pediatric medication safety issues and to compare international rates, (2) to identify how the issue is framed to the public, and (3) to elucidate the key themes within the articles.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Coding Variables
We noted the country of origin, date of publication, newspaper, article type (news article, editorial, or letters). One of the authors (C.S.) categorized the articles according to event type and article slant. Event type classified the actual stories reported into 4 categories, namely, negative, positive, mixed, or neutral; for example, a patient death would be classified as a negative event. The article slant is the skew of the report written about the event; for example, if the article overstated the event, then this would represent a negative slant.17 To account for varying country size, the total number of articles for each country was divided by the country population. The population of each country was derived from national census data. We also coded the articles with a more qualitative approach. First, articles were coded according to the 4 main themes of patient incident, research, policy, or other (or combinations of themes). To assess overall classification reliability, a second independent researcher reassessed a random sample of 30 of the articles. The 
score for event slant was 0.55; values were 0.33 for article slant and 0.54 for theme.
We then evaluated the content in more detail. These codes were developed from initial assessment of the news articles with ATLAS.ti software (version 4.2; ATLAS.ti, Eden Prairie, MN). These codes allowed us to assess the extent to which the media framed articles within the context of a culture of safety. We looked at whether the media presented the public with the 3 key tenets of a culture of safety. First, to what extent did the news articles portray adverse events as systems failures? Second, did the cases described in the news articles illustrate best practices for providers for dealing with adverse events? It is acknowledged that, after an adverse event occurs, there should be an apology to the family or patient, a thorough investigation of why the event occurred, and institution of polices and procedures to prevent repeat occurrences.18 This is linked directly with the final concept we examined. Third, to what extent was the concept of shared learning prominent within articles? Shared learning is the term used to explain that adverse events and near misses must be seen as opportunities for learning. The care providers involved directly in the adverse event should learn from the event, but there is also a duty to disseminate this knowledge. Learning cannot occur unless adverse events and near misses are reported, and we coded articles for incorporation of this concept.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
The slant in the majority of articles (71%) was neutral. Of the events judged to be negative for patient safety, 75% were covered in a neutral way and only 19% were reported in an unduly negative manner. Of the articles covering a negative event, 4% had a positive orientation (Tables 1 and 2).
|
|
In a qualitative analysis, we sought to determine whether we could find examples suggesting that news coverage was representing safety issues in medicine in the ways in which safety experts view most accidents, in particular, that multiple defects can be found when most accidents are evaluated and that improving safety requires changing systems.18 A number of anecdotes illustrated that this is indeed the case. For example, Claire Lewis was an 11-year-old girl undergoing surgery for treatment of a benign brain tumor who died at MacMaster University Medical Center, in part as a result of inappropriate use of desmopressin. Chief of Staff Andrew McCallum was quoted as follows: "We didn't find a single person or a single event that led to this. We found a series of events and occurrences, slips, lapses, errors in judgment that led to this tragic outcome."19 In another newspaper, he explained, "It's important to understand that saying we are moving from a culture of blame to one of understanding and learning doesn't mean everybody gets off."20 This coverage highlights the understandable difficulty parents have in accepting accidents. Claire Lewis' father noted, "I guess what the health-care industry needs to ask itself is, how does the family of, say, an 11-year-old child killed by physicians feel about the death being regarded as a learning experience for physicians, nurses, and the hospital?"20
Four clear steps have been identified and accepted as best practice after an adverse event, namely, timely apologies, thorough honest investigation, institution of policy and practice changes, and shared learning.18 These are goals that parents, care providers, and the wider public seemed to share in newspaper articles on pediatric medication safety. An article describing the death in 1997, in British Columbia’s Children’s Hospital, of a young girl as a result of intrathecal administration of vincristine, instead of intravenous administration, demonstrated that the care providers seemed to accept this approach. The hospital president explained, "Our only options are to understand it and to do everything we can to reduce risks in the future."21 Lisa Shore died at the Toronto Sick Kids Hospital, after being administered morphine with insufficient monitoring. The coroner's jury in that case noted that hospitals must respond "quickly, accurately, and openly."22 This example highlights how members of the public view the responsibility of care providers and, importantly, how the public expects institutions to respond.22
The first step in dealing with an adverse event is to recognize the event and to apologize to the patient and family. In doing so, health care providers need to recognize that there are 2 victims. The patient clearly is the primary victim, but the provider often suffers substantially and has been dubbed the "second victim."23 The media have recognized the second victim, in particular by focusing on the difficulty practitioners face in apologizing. Although apologizing is extremely important, it is very difficult, both because of the potential legal corollaries and because it is hard to find a way to apologize. The British Columbia hospital president summed this up, "This is the most difficult thing I have ever had to tell a family: we failed, and as a result, your child died.... There is no way to adequately apologize for this failure."21 However, news articles also identify examples of poor or absent apologies. For example, Sharon Shore, the mother of Lisa Shore, explained, "... nobody wanted to talk to us. Nobody would acknowledge that anything was done in error."24
The second step that best practice demands is a thorough investigation of events. Often this occurs only because of prolonged parental campaigns, frequently involving the media.24 The third and fourth steps are learning from the mistake and disseminating this knowledge. Time after time, media reports stress that parents whose children have suffered adverse events want health care in general and the organization in particular to learn from their mistakes. This idea of shared learning is exemplified by the parents of the child who died in Vancouver, "We desire our daughter's memory should be honored by the knowledge that some good will come from this tragedy."21 News articles articulate that institutions are aware of the need for this approach. As Phillip Herbert, a family physician and bioethicist at the Sunnybrook and Women's College Health Sciences Center (Toronto, Canada), explained, "It sure would be nice to learn what a hospital in Saskatchewan is doing, and what a hospital in Halifax is doing, so that people aren't required to reinvent the wheel at every institution."20
Stressing shared aims helps to indicate to the public that providers and parents are able to share goals and to work together to reduce adverse events. Previously there would have been more emphasis on identifying the providers at fault and then punishing them. This is not to say that there are not parents calling for accountability, sometimes very strongly, and sometimes still pointing fingers at individuals; understandably, this is reported, because it is often very newsworthy. This is particularly true if the parents are forced into a position of crusading for information and allegations of cover-ups start to circulate, such as in the case of Lisa Shore. The Vancouver Province reported that Lisa's mother "immediately demanded a police investigation, the dismissal of hospital staff involved in her daughter's care, and a public inquiry."25
Reporting adverse event rates is a prerequisite of shared learning. Examples exist of misinterpretation of research by the media. For example, when an Australian hospital reported its data, the Herald Sun summarized, "The hospital's figures reveal dozens of the state's sickest children have fallen victim to potentially disastrous drug mix-ups."26 In a similar vein, subtle criticism is aimed at hospitals for requiring such research. The Daily Mail reported, "DOZENS of seriously-ill infants and children are being put at risk every week because hospital doctors give them the wrong drug dosages, it is revealed today. Disturbing evidence uncovered by the Daily Mail has exposed a widespread problem, which could be solved by simple checking procedures. Ministers are so concerned they have commissioned a study which is expected to reveal the extent of the scandal."27
Although overall the media may be fair and frame the news within the context of a culture of safety, misinterpretation of research findings is not the only example of skew. This occurs against a background in which it may be difficult for the public to develop an overall sense of individual cases, because of the unavoidably piecemeal nature of reporting. In the case of Lisa Shore's parents' court battle, readers would need to follow a story daily (or be lucky enough to read the report on the final day of court) to develop a clear understanding of events. It should be noted that cases that lead to lengthy court battles provide not only more newsworthy opportunities for newspapers but also very powerful stories, and perhaps unbalanced presentation can thus be explained. The opposite is also true. The media might even be excluded actively from cases in which information, apologies, and policy/practice changes occur speedily, which would prejudice their coverage heavily. In the British Columbia case, the hospital apologized, investigated and disclosed fully to the parents the causes of the incident, and enacted a series of policy changes, such as warning stickers and a training videotape (which was dedicated to the child's memory). The family's response to this was to request privacy. The family stated, "While we understand that the hospital's error is newsworthy, it is our choice not to participate in your coverage."21
Although the media cannot control some of the imbalance, some of the language used is dramatic. Ross Woolard died as a result of complications of a pheochromocytoma, which was diagnosed in the postmortem examination. The article in the Daily Record reported, "Less than 24 hours later, he was dead after appalling blunders by hospital staff who failed to diagnose a rare tumor. Not realizing it was totally the wrong thing to do, they pumped so much fluid into Ross that, ultimately, he drowned."28 The headline alone may be sufficiently skewed to affect the public, for example, "Doctor Zombie: with no training in the files, Andrew Holton misdiagnosed 618 children as epileptic then numbed their minds with drugs that made their lives a misery."29 Often, however, the media do go beyond the expected in a positive way and attempt to educate members of the public regarding how to be advocates for their children and to decrease adverse events, such as those attributable to inappropriate prescription of antibiotics in the presence of allergies.
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Leape18 set out the key elements for reducing adverse events and suggested, in particular, that adopting a culture of safety is crucial for error reduction. However, work by the Kaiser Family Foundation5 and Blendon et al9 demonstrated that the public in general has not yet endorsed this concept. Ryan2 described 3 models in which the media may affect public opinion. First, according to the "hypodermic needle theory," the media inject ideas directly into the public psyche. The "minimal effects model" suggests that the public plays a limited role in modifying media ideas. Finally, the "constructionist model," which was developed originally by William Gamson and Andre Modigliani, suggests that the public decides actively what to accept from the media. Whichever theory is correct, it seems clear that the media play a major role in affecting public opinion.2
This study attempted to answer 3 questions. First, we assessed the extent of newspaper coverage of pediatric medication safety and found that pediatric medication safety is a topic of increasing interest to the media. The dip in 2004 may be attributable to the lag time in loading data in Lexis Nexis. In the past 5 years, the number of articles covering research and policy has increased considerably. This time period coincides with increased research and policy interest in patient safety since publication of the Institute of Medicine report on the topic.16 As an example, in 2005 the Agency for Healthcare Research and Quality spent more than $84 million on safety research, an increase of $4 million in comparison with the previous year.31 When raw figures were adjusted for country population, Canada demonstrated the greatest coverage, followed by the United Kingdom and the United States. Relatively few studies estimating the prevalence of adverse drug events have been conducted throughout the world, particularly looking at pediatrics.6 Without national data on errors or adverse events, it is difficult to explain this as being solely attributable to variations in adverse events. There are likely to be other factors, such as the type of newspapers that exist in each country and national efforts to produce reductions in adverse outcomes. Although attempts were made to produce per-capita data, this was a crude estimate. More-complex techniques are available, but these do not work smoothly when the rarity of articles forces international collection of data. Furthermore, even the most complex techniques have yet to be adapted for a world in which newspapers can be read online, which makes readership numbers even more difficult to assess. In addition, this assessment of the effect of coverage is potentially an underestimate, because public opinion regarding pediatric medication safety is molded by exposure to more-general articles on patient safety and by exposure to more than just newspapers.
Second, we examined how the topic is framed to the public. We found that, even when an event that was negative for patient safety was reported, 75% of the articles had a neutral slant. It could be argued that, because most of the articles dealt with negative events such as child death, additional negative skew was not necessary to interest the public.
Third, we addressed key themes. The content analysis showed that overall the media are attempting to present cases in light of a culture of patient safety and not blame and that occasionally papers go further and provide detailed tips on how parents can be advocates for safety. This is almost beyond normal practice expected of the media, which have a widely acknowledged duty to present the news, with some arguing that it is not an important role to educate the public.1,32
These data suggest that the media may be helping to close the gap between the expert approach to reducing adverse events, through the culture of safety, and public opinion. This is an important message for care providers, and these data also suggest that the efforts to reduce adverse event rates should be publicized to the public through the media.
The study has a number of limitations. One group is related to the Lexis Nexis database and search engine. Lexis Nexis is the largest newspaper database, but there are complex inclusion biases. Publishers are in control of the number of articles given to Lexis Nexis, and this varies widely; for copyright reasons, no articles written by freelance journalists are included. Little assessment has been made of the accuracy of alternative strategies, such as use of news clipping services. Our current study is also unable to express the number of articles published on pediatric medication safety as a proportion of the total number published. Although we attempted to overcome the subjective nature of assessments of theme and content by asking a second independent reviewer to analyze a subset of articles, this is also a potential limitation of the study.
This study provides evidence that the topic of pediatric medication safety is of increasing interest to the public and the media. Overall, the information is provided in a fair manner by the media, in ways that should make it possible to build a culture of safety in health care. Health providers have a duty to maximize the potential benefits of this by contributing to research, striving to incorporate the culture of safety into everyday practice, teaching this key message to junior staff members, and educating the public on how to interpret media commentaries, in a manner similar to the instructions given for medical website information. On the basis of this research, the media seem to be moving in the same direction as researchers, policymakers, and health care providers, in a direction considered necessary to improve patient safety.
|
ACKNOWLEDGMENTS |
|---|
Thanks go to R.E., the independent observer, for her help.
|
FOOTNOTES |
|---|
Address correspondence to David Bates, MD, MSc, Division of General Internal Medicine, BWH, 1620 Tremont St, Boston, MA 02120. E-mail: dbates{at}partners.org
The authors have indicated they have no financial relationships relevant to this article to disclose.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. J. Smith, S. S. Ellenberg, L. M. Bell, and D. M. Rubin Media Coverage of the Measles-Mumps-Rubella Vaccine and Autism Controversy and Its Relationship to MMR Immunization Rates in the United States Pediatrics, April 1, 2008; 121(4): e836 - e843. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Suresh If it's in the paper, it must be true: newspaper reporting of pediatric medication errors. Pediatrics, June 1, 2006; 117(6): 2281 - 2282. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2006 American Academy of Pediatrics. All rights reserved. |
||
|
||
