The Role of the Media in Public Health Crises: Perspectives from the UK and Europe
The Role of the Media in Public Health Crises: Perspectives from the UK and Europe
Abstract and Keywords
This chapter presents a series of case studies of the role of the media in public health crises. The case studies illustrate factors that drive media interest in public health stories. They show how they are told in newspapers and on radio and television. They also demonstrate that microbes attract the media because they have particular properties.
What is a public health crisis? Academics who study crisis management have formally defined a crisis as ‘a serious threat to the basic structures or the fundamental values and norms of a system which under time pressures and highly uncertain circumstances necessitates making critical decisions’ (Rosenthal et al., 2001). Rightly, this is a very broad definition. A simpler way to emphasize the many different forms that a crisis can take and the difficulty in formulating a precise and concise operational definition is to draw an analogy with US Supreme Court Justice Potter Stewart’s opinion of what is obscene: ‘I shall not today attempt further to define the kinds of material I understand to be embraced … but I know it when I see it …’ (Jacobellis v. Ohio, 378 US184, 197 (1964)).
Nevertheless, whatever kind of definition is adopted, it is beyond dispute that for public health crises and the events that follow their initiation, the explanatory power of the Thomas theorem is very great: ‘if men define situations as real, they are real in their consequences’ (Thomas and Thomas, 1928); and it is also universally agreed that the media play an exceptionally important role as one of the forces that drives the theorem. In other words, ‘if CNN defines a situation as a crisis, it will indeed be a crisis’. (Rosenthal et al., 2001).
This account presents a series of case studies of the role of the media in public health crises. It is a personal one. As a commentator on microbiological issues (and from time to time a direct participant in crises), my involvement with the UK media has been extensive. Chairing a group that prepared a report in 1997 for the Scottish Office into a large E. coli O157 outbreak (Pennington Group, 1997) led to a high public profile. A readiness to respond to media inquiries (helped by residence close to two TV and two radio broadcasting stations), the regular and continuing occurrence of food poisoning outbreaks and other (p.82) newsworthy events, and the operation of the Matthew effect (‘For unto every one that hath shall be given, and he shall have abundance …’ or, to put it another way, the continued accrual of greater increments of recognition to those who already have recognition (Merton, 1968)), have further increased this public visibility. Four other factors have been important in assisting interaction with the media: being an academic gave me an almost untrammelled freedom to comment as an individual—one that is denied, for example, to a British civil servant (the UK Civil Service Code says that civil servants may not disclose official information without authority); being a Professor at an ancient Scottish university was an affiliation that endorsed a position as an expert; my career development (15 years in virology followed by 24 years in bacteriology) has allowed me to present myself as a medical microbiologist in the broad rather than as a narrow specialist; and a shortage of ‘public’ scientists has pushed the media in my direction. The use of the term ‘public’ to define someone who carries their professional work to non-academic audiences through media work and by publishing in outlets that reach non-specialists was coined by US sociologists two decades ago. It has been vigorously debated by them (Burawoy, 2005). Vaughan’s vigorous defence of this kind of work is compelling (Vaughan, 2006).
The case studies that follow illustrate the interactions between the media and those involved in public health crises. The factors that determine the final media product, topic selection, and the treatment and the telling of the story (Hartley, 1982), are illustrated. The first case study, of the British necrotizing fasciitis ‘outbreak’ in May 1994, provides a proof for the Thomas theorem; the media was causative.
Necrotizing fasciitis is a rapidly spreading and aggressive infection that leads to massive tissue destruction and gangrene. It was first described in 1924 by an American surgeon working in Beijing (Meleney, 1924). The causative organism in his cases was Streptococcus pyogenes, long known to be the cause of scarlet fever, erysipelas, and puerperal sepsis. A perception in the USA that the virulence of this organism was increasing, and the occurrence of a cluster of cases of necrotizing fasciitis and other severe infections caused by S. pyogenes in northern Scotland in the 1980s (Upton et al., 1995) were the stimuli for my research group in Aberdeen to obtain research funding in 1992 to develop new fingerprinting methods to test the hypothesis that ‘hotter’ strains were in circulation in Scotland. My name was on a list held by the British Medical Association of experts willing to talk to the media, with streptococci being named as a particular interest.
(p.83) Two patients from the Stroud area of Gloucestershire underwent elective surgery in the same operating theatre, one on the 4th February 1994 and one on the 7th. Both developed necrotizing fasciitis (Cartwright et al., 1995). Subsequent cases of necrotizing fasciitis in west Gloucestershire presented on the 18th of February, the 7th and 15th of April, and the 11th of May. One of these patients lived in Stroud and the others lived 1, 15, and 40 km away. Two died. The cluster of cases first received media attention outside the local area at the beginning of May, when the BBC South of England health correspondent broadcast a radio item about them. The story was run by the Press Association on the 11th of May and attracted the attention of a Daily Sport journalist. This paper carried a report about it on the 13th of May, replacing its usual front page sex story with the headline ‘BUG THAT EATS YOU ALIVE’, and, in lower case, ‘Killer virus scoffs three’. Two weeks later the story had spread worldwide, with items being carried on Canadian television and Australian radio. Its end was marked by leaders in the Lancet, the British Medical Journal, Nature, and Science in early June. The acute microbiology and epidemiology investigations were handled by the Public Health Laboratory Service. Its London HQ handled about 1000 enquiries during the incident and its Gloucester laboratory received 200 requests for information.
The number of media outlets (newspapers, radio programmes, and television stations) telephoning me for microbiological information between the 12th and the 28th of May is shown in Figure 6.1. Enquiries from newspapers happened in three phases. The first enquiry (The Daily Sport) was followed by
Television and radio enquiries lagged behind those from the press, peaking on Wednesday the 25th of May. A graph showing the cumulative number of calls is shown in Figure 6.2. It shows an S-shaped, or logistic, curve.
It is clear that the necrotizing fasciitis media story appeared and disappeared with a dynamic not unlike the rise and fall of cases in an epidemic caused by a virus or a bacterium. It even had an incubation period. Rather than an outbreak of infection, it was an outbreak of media interest. It scored highly in the fright factors and media triggers listed in Chapter 1. For fright factors its acquisition seemed involuntary because it attacked victims at random, and its distribution was inequitable—why so many victims at Stroud, which had also had the misfortune between 1981 and 1986 to be the centre of a meningitis epidemic with more than 60 cases and two deaths? It was inescapable because risk factors appeared to be poorly defined, and it was novel because its clustering seemed to defy explanation. It caused irreversible damage through massive tissue loss, and caused death in a dreadful way—‘Bug that eats you alive’ (The Daily Sport). Its victims were identifiable. For the media triggers, it scored highly in the human interest and visual impact factors, with powerful images being conveyed by text, such as ‘Thank God I’M FAT’ (Take a Break), and ‘BUG EATS HUMAN FLESH’ (Sunday Mail).
(p.85) It is probable that the coming together of all these features accounted for its selection by the media, its strength as a story, and, in part, its persistence. This was also aided by the discovery and description of previously unreported cases by journalists (the media were able to find a number of victims or relatives of victims willing to speak about their experiences—photographs of at least seven were published and television programmes about two were made, a significant tally considering the rarity of the disease), by the raising of uncertainties about its incidence, and by the rehearsal of arguments for and against notification. The lack of data about the incidence of the condition made it impossible to make evidence-based comments about whether it was changing, so it was not possible to curb speculation.
The dynamics of media interest in necrotizing fasciitis and the shift of interest from newspapers to broadcasting can be analysed both in terms of the bureaucratic setting which produced it and its mathematical epidemiology. Thus journalists make extensive use of media organizations other than their own in determining what is news. The influence of the early morning BBC Today radio programme in setting the news agenda for the rest of the day is well known. TV and radio newsrooms are always littered with newspapers. In a study of crime reporting, Fishman (1981) showed that this mutual dependence between different branches of the media is an important factor in encouraging the spread of a news theme through the community of a news organization, in his case converting a crime theme into a crime wave. There is a clear similarity between such events and the spread of necrotizing fasciitis as a news theme and its conversion thereby into an ‘outbreak’. In terms of its mathematics, it is highly probable that the basic principles governing the spread of a news theme and the spread of an infective agent are the same. The course of the latter is governed by the mass action principle. This indicates that the rate of spread is proportional to the product of the density of susceptibles multiplied by the density of sources of infection. The pattern of spread following the introduction of a small nucleus of infection (the initial story) into a population of uninfected humans (journalists) and vectors (media outlets) would, if plotted, give an S-shaped curve identical in general form to that shown in Figure 6.2.
The end of the media ‘outbreak’ was quite sudden. It was finally killed on 28 May by a much stronger story, an attack by Prime Minister John Major on beggars. It is likely that the absence of such rival stories during the previous week had made a significant contribution to its longevity.
It is possible to read a message about risk into the Banx cartoon (Figure 6.3) published at the end of the episode—that being attacked by the flesh-eating bug and winning the lottery were equally improbable. It also indicates that by (p.86)
E. coli O157 outbreaks in Scotland and Wales
No humour, black or otherwise, is associated with E. coli O157. Although much rarer than the usual causes of bacterial gastroenteritis, Campylobacter and Salmonella, the infections caused by it are often severe. Complications (which cannot be prevented by any specific treatment once an infection has been established) are more frequent in young children and the elderly—renal failure (which may be permanent) and brain damage are the commonest. Cardiac involvement is not uncommonly a cause of death. The incidence of human infections is higher in the UK than anywhere else in the world.
On the afternoon of Friday 22 November 1996, the Public Health Department of Lanarkshire Health Board became aware of several cases of infection with E. coli O157 in residents of Wishaw in the central belt of Scotland. By the evening, histories from confirmed or suspected cases indicated that 14 of the 15 who were ill had consumed food obtained directly or indirectly from J. Barr and Son, Butchers of Wishaw. Although outwardly a small local butcher with a bakery shop adjoining, the business was involved at the time of the outbreak in a substantial wholesale and retail trade involving the production and distribution of raw and cooked meats and bakery products from the Wishaw premises. The epidemic curve of the outbreak (Figure 6.4) shows that the number of suspected or confirmed cases increased dramatically from its onset.
Epidemiological and subsequent microbiological evidence shows that the outbreak was made up of several separate but related incidents. The largest of these were a lunch attended by more than 70 frail elderly people held in Wishaw Parish Church Hall on the 17th of November, a birthday party that took place in the Cascade Public House on the 23rd of November, and retail sales in Lanarkshire and the Forth Valley. The outbreak was declared over on Monday the 20th of January 1997, although further deaths occurred following prolonged illness. The final tally of cases was 501 (the largest ever outbreak of infection with the organism in the UK). Of these, 127 were admitted to hospital, of whom 13 required dialysis. Of those infected, 21 died; the deaths of 17 of them were caused directly by E. coli O157.
(p.88) On 28 November 1996 the Secretary of State for Scotland announced the establishment of an Expert Group with me as chairman. Our remit was ‘to examine the circumstances which led to the outbreak in the central belt of Scotland and to advise on the implications for food safety and the general lessons to be learned’. In early 1997 another E. coli O157outbreak occurred in a nursing home in Tayside and in its deliberations the group was asked to take account of this as well as other outbreaks that had occurred in the Borders and in Lothian. The Group convened from the beginning of December 1996 until the end of March 1997. An interim report with recommendations was submitted to the Secretary of State on Hogmanay 1996. He responded to it in the House of Commons on 15 January 1997. Our final report was submitted at the end of March and was published, with the government’s response, on 8 April 1997.
In 1992 I had established the Scottish Reference Laboratory for E. coli O157 in Aberdeen, and so was known as a source of information about the organism. Before the Wishaw outbreak, the frequency of media enquiries in 1996 ranged from one to six per month. Eighteen of the 31 enquiries were from Scottish organizations. At the beginning of November the BBC current affairs programme Newsnight came to my department to film the work of a visiting Public Health Laboratory Service staff member (and E. coli O157 expert) as part of an item about the fiftieth anniversary of the Service, and BBC Scotland was planning with us a programme about the transmission of E. coli O157 to shepherds at lambing. Filming took place in the department on the 28th of November, the day when the establishment of the Expert Group was announced.
The number of telephone enquiries from the media about E. coli O157 during and following the outbreak is shown in Figure 6.5. Calls are plotted day by day, starting on the 25th of November 1996 (day 1) and finishing on the 16th of July 1997 (day 230). On nine occasions, enquiries peaked at 10 or more a day. The first peak, labelled (a), reflected media activity on the day of the announcement of the Expert Group. The next peak (b) coincided with the meeting at the Scottish Office at which the composition of the group was decided, and five peaks (c–f and h) occurred on the day before the meetings of the group. Media pressure was particularly sustained just before and on the day of the meeting at which the report was finalized (peak h). The largest peak of all—36 enquiries on the 6th of March 1997 (peak g)—was linked to the leaking to the press of the first draft of the Swann Report, a review by the Meat Hygiene Service of abattoir hygiene practice conducted in 1995. Highly critical of aspects of abattoir practice, the report underwent significant drafting changes and was eventually put into the public domain in a low key way in (p.89)
Fright factors for E. coli O157include its involuntary acquisition (by consuming seemingly safe but microscopically contaminated ready-to-eat cold meats, for example), its inequitable distribution (four times commoner in Scotland than in England), its ability to cause irreversible damage to kidneys and brain, particularly in small children, and its ability to cause death in a painful way (Jeremy Bray, MP: ‘As the families of too many of my constituents have discovered, dying from E. coli is a horrible way to die’; Hansard, 15 January 1997). As factors determining media activity, these are almost certainly sufficient to explain the persistence of a long-term interest. But with the exception of the twentieth death, none of them seemed to play a role in determining the major fluctuations in the number of enquiries that I received after the beginning of the outbreak. These related either to the deliberations and conclusions of the Expert Group or to political events. The first five of the media triggers listed in Chapter 1 map well on to these circumstances; thus the alleged concealment of the Swann Report by the Meat Hygiene Service and the Ministry of Agriculture and the Scottish Secretary provide questions of: (1) blame, (2) cover-ups, (3) human interest, (4) links with high-profile personalities, and (5) evidence of conflict.
The Expert Group met in private. On arrival in Edinburgh from Aberdeen to chair its first meeting I was met at the railway station ticket barrier by two civil servants to shield me from reporters. At this meeting it was agreed that our discussions would be confidential. For my part, this remained the case. But there were leaks to the media. The source was never established. My interpretation of them was that they were manifestations of the battle being fought over the recommendations that the Group was developing; the representatives on the Group of London-based government departments (Agriculture and (p.91) Health) were unhappy about the radical form they were taking. The leaks were not helpful to their case. At the end of the day, although by convention not admitted, the arguments were settled by a Cabinet committee in favour of the more radical Scottish position (Pennington, 2000).
The 2005 South Wales outbreak
The first cases of gastroenteritis presented on Wednesday 14 September 2005. By Friday the 16th it was clear that an outbreak affecting children was in progress in the Merthyr Tydfil and Rhondda Cynon Taf areas, and that microbiological evidence was pointing to E. coli O157 as its cause. Within a week, 150 cases had been identified in the South Wales valleys with 42 schools affected. Following the death of 5-year-old Mason Jones on the 4th of October 2005, a police investigation was launched. By the 14th of November, 168 cases compatible with an E. coli O157 infection had been identified. On the 5th of October the National Assembly for Wales established a committee to consider the establishment of a public inquiry under the 2005 Inquiries Act. They recommended that one should be held and that I should chair it, with a remit ‘To inquire into the circumstances that led to the outbreak of E. coli O157 infection in South Wales in September 2005 and into the handling of the outbreak, and to consider the implications for the future and make recommendations accordingly’ (Pennington, 2009).
A public inquiry is inquisitorial, not adversarial. It is very different from the internal government inquiry that I chaired in 1996–97. The 2005 Act confers powers to compel the production of documents and the attendance of witnesses to give evidence. Its hearings are held in public and the evidence it gathers is published.
The hearings were held in Cardiff from the 12th of February to the 19th of March 2008. Journalists from the print and electronic media had full access. The proceedings were not televised. Transcripts went on the Inquiry website within hours. The daily newspapers that gave the most detailed coverage of the hearings were the Western Mail and the South Wales Echo. Both are published in Cardiff by the Trinity Mirror Group. The Echo is described as ‘tabloid’. It has a circulation of about 50 000. The Western Mail is produced in ‘compact’ format and has a circulation of about 40 000. The hearings were covered mainly by two reporters. They wrote for both papers.
Three individuals had their photographs reproduced many times in both papers alongside the reports of the hearings. In the Echo, Mason Jones (the 5-year-old boy who died) was shown on 6 days and his mother, Sharon Mills, on 7. But William Tudor, the butcher who had supplied contaminated meat to the schools in the valleys, had his photograph in the paper on 8 days. He had (p.92) pleaded guilty to six charges of supplying contaminated meat in July 2007, and in August to an additional charge of supplying contaminated meat. In September he was sentenced to 12 months in prison.
Headlines tell the story.
12 February—Western Mail. ‘Mum wants answers from E. coli inquiry.’ ‘People need to be reminded of the human cost of E. coli.’ ‘Inquiry team has collected 36,000 pages of evidence.’
13 February—Western Mail. ‘Councils chose “lowest cost” Tudor’s meat despite scores of complaints.’
14 February—Western Mail, front page. ‘E. coli butcher’s laid bare by inquiry.’ (Double page spread inside) ‘Tudor “lied to officers” about vac-packer that spread contamination to meat.’ (Echo, front page) ‘SECRETS AND LIES.’ ‘E. coli butcher hid factory filth and falsified hygiene records.’ (Double page spread inside) ‘If it didn’t smell too nice just bag it and send it out.’
15 February—Echo, front page. ‘A TRAVESTY OF JUSTICE—Families’ anger as E. coli butcher William Tudor is released after just three months of a year-long sentence.’ (Inside) ‘Mason’s life was snatched away—yet Tudor’s got his life back.’
16 February—Western Mail. ‘Council officials admit falling short on checking of E. coli butcher’s cleanliness standards.’ (Echo) ‘E. coli butcher: How the system failed.’
19 February—Western Mail. ‘Inspector defends her view that butchers premises were safe when she visited.’
21 February—Western Mail. ‘Using vac-pack for dual meat role “like Russian roulette”.’
28 February—Western Mail. ‘School contracts awarded on price despite complaints.’
29 February—Western Mail. ‘Food Standards Agency blames WAG under-funding for delaying report.’
4 March—Western Mail. ‘One in five schools had no hot water for children to wash their hands.’ (Echo) ‘Vital hot water supplies shock.’
6 March—Echo, front page. ‘MASON DENIED HOME VISIT “THREE TIMES”.’ (Inside) ‘Desperately-ill Mason offered Calpol by doc.’
7 March—Western Mail. ‘Mamma I’m dying.’
11 March—Echo. ‘Groundhog Day for chair of E. coli inquiry. Hygiene failures similar to Scottish case.’
18 March—Western Mail. ‘Tudor abattoir scored the ‘lowest ever’ hygiene rating.’
The way these headlines and the photographs accompanying them highlighted the events and issues being revealed at the public hearings followed the (p.93) patterns found by Petts and her colleagues in their study of risk reporting in the UK (Petts et al., 2001). Risk event causation was personalized to individuals. There was a powerful use of visual images. Lay experiences, initiatives, and voices figured large.
One of the reasons for setting up a public inquiry is to ‘give an opportunity to all who reasonably have an interest in making representations to do so. It thus has a cathartic effect for victims, relatives and, via the media, the public in regard to distress, recriminations, speculations and rumours’ (Blom-Cooper, 1992). In this inquiry the media did what was hoped—and expected—that it would do towards fulfilling this aim.
The Inquiry used a public relations specialist to advise it on media matters. So did the Outbreak Control Team. The Head of Communications at the National Public Health Service for Wales joined the team on Saturday 17 September 2005, the day after the declaration of the outbreak. It was decided that the Rhondda Cynon Taf local authority would lead on communications in general, and that a public health doctor would give broadcast interviews. In the event, because of illness, the lead for newspaper inquiries passed to the NPHS communications head. He kept a log of media inquiries and interview bids. The number of inquiries peaked at 72 on Monday the 19th of September. It remained in double figures daily until the end of the month, totalling 459. Bids (16) also peaked on the 19th; by the end of the month there had been 102. The number of enquiries rose from 6 on the 3rd of October to 58 on the 4th, when Mason Jones died. The Inquiry report was published on the 19th of March 2009. There were many media bids. My interviews were marshalled and timetabled by the Inquiry media adviser; those for the electronic media (BBC Wales, ITV Wales, ITN, Channel 4) were shorter than those given by Sharon Mills and Lisa Bray (who had given oral evidence to the Inquiry about the illnesses of herself, her daughter, and her son). On the next day, the Western Mail front page headline was ‘We must never forget’, with a picture of Mason Jones. Sharon Mills’ face dominated the front page of the South Wales Echo with Mason’s photograph in the lower corner and the headline ‘Failed’.
Foot and mouth disease in England, 2007
Memories of the draconian measures used to control the 2001 UK foot and mouth disease outbreak (which cost more than £3 billion) were still fresh when the disease was confirmed in cattle in Surrey on Friday 3 August 2007. It soon became clear that virus had escaped from nearby Pirbright, the site of the only laboratories in the UK allowed to handle it (Pennington, 2007a). The outbreak was very small, but its media impact was very great. It was the first item on the BBC 6 O’clock news on four consecutive nights. BBC News 24 (p.94) (television) contacted me on the evening of the 3rd and interviewed me. I became their expert, and was interviewed by them on the 4th, 5th, 6th, 7th, 8th and 14th of August. They came back on September 12th and 13th when a second cluster of cases occurred. As well as giving interviews, the BBC reporter standing at the gate outside the Pirbright laboratory site (he was not allowed inside) telephoned me on several occasions to run past me what he was about to say on the news to check it for scientific accuracy.
I have never worked on the foot and mouth virus, although I have worked for and with ex-Pirbright virologists and was familiar with veterinary virology issues because my PhD was on Newcastle disease, an infection of birds. I guess that I was the port of call for the media (appearing on BBC News 24 and Newsnight on the 3rd of August was followed by interview bids from BBC Breakfast, BBC 5 Live, Sky and STV on the 4th) because all the real experts on foot and mouth virus in the UK were inside Pirbright and for obvious reasons were unlikely to appear at its gate to give free-ranging interviews any time soon.
The European perspective
My personal experience of the media outside the UK is limited. Interviews at their request by Norwegian television on E. coli O157 and by Radio Telefis Eireann on foot and mouth disease, waterborne Cryptosporidium infections, and on the significance of low levels of dioxins in Irish pork gave no surprises about topic choice by journalists.
A survey in September and October 2005 of 25 642 people in the 25 Member States of the European Union commissioned by the Directorate General Health and Consumer Protection and the European Food Safety Authority (Special Eurobarometer 238, ‘Risk Issues’) showed that 42% of the population considered that food would damage their health (ranking fourth as a risk after environmental pollution, car accidents, and serious illness), with food poisoning coming first to mind most often (before chemicals and obesity). Top worries included new viruses such as avian influenza and the contamination of food by bacteria. Paradoxically, while 16% permanently changed their eating habits and 37% avoided food temporarily as a consequence of media coverage, only 17% regarded the media as their most trusted source of information, behind consumer groups (32%), doctors (32%), scientists (30%), and public authorities (22%).
The case studies described above illustrate factors that drive media interest in public health stories. They show how they are told in newspapers and on radio (p.95) and television. They also demonstrate that microbes attract the media because they have particular properties. They cause horrible diseases, like necrotizing fasciitis. E. coli O157 ranks high because it can kill, and because it targets children and the elderly. ‘Killer viruses’ initiate stories more easily if they have a high previous profile; the confirmation that the week-old corpse of a whooper swan washed up at Cellardyke on the East coast of Scotland had tested positive for H5N1 bird flu in early April 2006 caused a brief—but massive—invasion of the East Neuk of Fife by London-based correspondents. This story illustrates one of the most difficult things for a ‘public’ microbiologist to get right, a prediction. Is an influenza pandemic imminent? How many will it kill? Unfortunately, the randomness of evolution makes such questions trans-scientific (Weinberg, 1972); they can be stated in the language of science but cannot be answered by it. Evolution is at the core of many other public health stories as well. E. coli O157 only evolved to be a human pathogen 30 years ago. The 2008 pandemic norovirus strain only emerged in 2006 (Siebenga et al., 2008). Clostridium difficile ribotype 027 only started to be an important pathogen in 2002 (Kuijper et al., 2007), going on to be pandemic and dramatically lethal (Pennington, 2007b). So in the year that we celebrate the bicentenary of Darwin’s birth and the 150th fiftieth anniversary of the publication of The Origin of Species, it is right and appropriate to enter evolution into the list of major causes of current media-worthy events. A safe prediction is that it will be thus for all time coming.
Blom-Cooper, L. (1992). Report into the Committee of Inquiry into Complaints about Ashworth Hospital. (1992). Volume 1. London, HMSO.
Burawoy, M. (2005). For public sociology. American Sociological Review, 70: 4–28.
Cartwright, K., Logan, M., McNulty, C.M. et al. (1995). A cluster of cases of streptococcal necrotising fasciitis in Gloucester. Epidemiology and Infection, 115: 387–97.
Fishman, M. (1981.) Crime waves as ideology. In: S. Cohen and J. Young (ed.) The manufacture of news: deviance, social problems and the mass media. London, Constable, pp. 98–117.
Hartley, J. (1982). Understanding news. London, Methuen.
Kuijper, E. J., Coignard, B., Brazier, J. et al. (2007). Update of Clostridium difficile-associated disease due to PCR ribotype 027 in Europe. Eurosurveillance, 12: 163–6.
Meleney, F. L. (1924). Hemolytic Streptococcus gangrene. Archives of Surgery, 9: 317–64.
Merton, R. K. (1968). The Matthew effect in science. Science, 159: 56–63.
Pennington, T. H. (2000). Recent experiences in food poisoning: science and policy, science and the media. In: D.F. Smith & J. Phillips (eds). Food, Science, Policy and Regulation in the Twentieth Century. London, New York, Routledge, pp. 223–38.
Pennington, T. H. (2007a). Biosecurity101: Pirbright’s lessons in laboratory security. BioSocieties, 2: 449–53.
Pennington, T. H. (2009). The Public Inquiry into the September 2005 Outbreak of E. coli O157 in South Wales. (HMSO, also http://www.ecoliinquirywales.org).
Pennington Group Report (1997). On the circumstances leading to the 1996 outbreak of infection with E. coli O157 in Central Scotland, the implications for food safety and the lessons to be learned. Edinburgh, The Stationery Office.
Petts, J., Horlick-Jones, T. & Murdock, G. (2001). Social amplification of risk: The media and the public. Contract Research Report 329/2001. Sudbury, HSE Books.
Rosenthal, U., Boin, R. A. & Comfort, L. K. (2001). Managing crises. Threats, dilemmas, opportunities. Springfield, C. C. Thomas.
Siebenga, J., Kroneman, A., Vennema, H., Duizer, E. & Koopmans, M. (2008). Food-borne viruses in Europe network report: the norovirus GII.42006B (for US named Minerva-like, for Japan Kobe 034-like, for UK V6) variant now dominant in early seasonal surveillance. Eurosurveillance, 13(2): pii=8009.
Thomas, W. I. & Thomas D. S. (1928). The child in America. Behavior problems and programs. New York, A. A. Knopf.
Upton, M., Carter, P. E., Morgan, M., Edwards, G. P. & Pennington, T. H. (1995). Clonal structure of invasive Streptococcus pyogenes in Northern Scotland. Epidemiology and Infection, 115: 231–41.
Vaughan, D. (2006). NASA revisited: Theory, analogy, and public sociology. American Journal of Sociology, 112: 353–93.
Weinberg, A. M. (1972). Science and trans-science. Minerva, 10: 209–22.