Paracetamol is the drug most commonly taken in overdose in the United Kingdom,1 causing a substantial number of deaths.2 We have investigated the impact of the fictional portrayal of a potentially fatal paracetamol overdose in the television drama Casualty (seen by 12.8 million viewers) on short and long term knowledge related to paracetamol poisoning.
Subjects, methods, and results
The episode (described in detail in accompanying article3) depicted a man suffering potentially fatal liver damage after an untreated paracetamol overdose. Although the particular dose (50 tablets) and delay before presentation (2 days) were mentioned, the episode did not specify minimal toxic doses or maximum safe delays.
At one week and 32 weeks after this episode was broadcast, we sent questionnaires to members of the BBC Television Opinion Panel. Panel members are recruited by structured sampling to be representative of the adult UK population and are sent weekly questionnaires related to their viewing. At one week after the broadcast, we asked them whether they had viewed the relevant Casualty episode and questions to test their knowledge of the delayed hepatotoxic effects of paracetamol in overdose, maximum safe delays before seeking help, and fatal doses of several drugs commonly used for self poisoning (see table). The test questions were repeated at 32 weeks after thebroadcast, together with further questions about members’ interest in medical matters and viewing of other medical dramas and documentaries. General medical knowledge was tested by two multiple choice questions: “What is a crash team?” and “What is a laparotomy?” Responses from the two surveys were linked. At one week, 2792 out of 3115 panel members participated, 1030 (37%) of whom reported watching the episode. At 32 weeks, 1282 of these participants remained in the panel, of whom 475 (37%) had seen the episode.
Non-viewers were more likely not to respond or respond “don’t know” to all questions in both surveys. At one week, significantly more viewers (85%) than non-viewers (45%) correctly identified paracetamol as having hepatotoxic effects. By 32 weeks, this knowledge had declined by 12% in viewers (P<0.0005) and increased by 5% in non-viewers (P=0.004). The effects were little changed by using a logistic regression model to adjust for demographic factors and measures of medical interest, knowledge, and viewing habits.
Viewers indicated longer minimal safe delays before seeking help than non-viewers, the differences being small but significant. There was no difference between viewers and non-viewers in lethal doses of paracetamol or the ranking of paracetamol toxicity compared with that of other drugs.
Comment
Rates of deliberate self harm continue to increase: overdose is the most common method, and paracetamol the most commonly used substance.1 Baseline knowledge in this study was high: 45% of those who did not see the episode knew of the delayed hepatotoxicity of paracetamol, possibly reflecting recent extensive media attention.
Television is an important potential source of medical information,4 with programmes such as Casualty attracting audiences of over 10 million. Our study showed that viewers of a Casualty episode registered and retained information about paracetamol toxicity presented in the programme among other distracting story lines. Interestingly, it also revealed that the knowledge obtained was strictly restricted to the presented facts and that incomplete messages might have been misinterpreted. There was also an increase in overdose presentations to general hospitals after the broadcast.3 Medical messages broadcast within television programmes are likely to have an impact on the knowledge of the general public: editors should be aware of this and ensure that they are accurate and complete.
Table.
Responses to the survey questions testing knowledge of paracetamol toxicity 1 week and 32 weeks after broadcast of an episode of Casualty depicting a paracetamol overdose: comparison of the knowledge of those who did and did not see the episode. Values are numbers (percentages) of those who responded to question unless stated otherwise
| Question and responses | Responses 1 week after broadcast
|
Responses 32 weeks after broadcast
|
|||||||
|---|---|---|---|---|---|---|---|---|---|
| Viewers (n=1030) | Non-viewers (n=1762) | Odds ratio (95% CI) for viewers v non-viewers; P value
|
Viewers (n=475) | Non-viewers (n=807) | Odds ratio (95% CI) for viewers v non-viewers; P value
|
||||
| Crude (n=2792) | Adjusted* (n=1168) | Crude (n=1282) | Adjusted* (n=1168) | ||||||
| Delayed effects of paracetamol | |||||||||
| Question: An overdose of one of the following drugs will make a person unwell to start with, then they may feel better, but a few days later they may die from liver damage. Can you say which drug has this effect? Answers: antibiotics, aspirin, Brufen, paracetamol, sleeping tablets, tranquillisers, don’t know | |||||||||
| Correct (paracetamol) | 877 (85) | 786 (45) | OR for giving correct answer† | 361 (76) | 392 (49) | OR for giving correct answer† | |||
| Incorrect (other drug) | 61 (6) | 275 (16) | 5.03 (3.75 to 6.75); P<0.0005 | 4.79 (2.63 to 8.71); P<0.0005 | 40 (8) | 113 (14) | 2.60 (1.77 to 3.83); P<0.0005 | 2.32 (1.42 to 3.81); P=0.001 | |
| Don’t know or no answer‡ | 92 (9) | 701 (40) | 74 (16) | 302 (37) | |||||
| Delay in presentation§ | |||||||||
| Question: What is the latest response time after an overdose of this drug that a person can seek help to avoid this fatal effect? Answers: 6, 12, 24, 48 or 72 hours, or don’t know | |||||||||
| 6 hours | 305 (35) | 292 (37) | 138 (38) | 133 (34) | |||||
| 12 hours | 172 (20) | 159 (20) | OR for reporting shorter times¶ | 89 (25) | 91 (23) | OR for reporting shorter times¶ | |||
| 24 hours | 121 (14) | 59 (8) | 0.67 (0.54 to 0.83); P<0.0005 | 0.62 (0.40 to 0.94); P=0.02 | 43 (12) | 41 (10) | 0.84 (0.62 to 1.14); P=0.3 | 0.69 (0.45 to 1.05); P=0.08 | |
| 48 hours | 55 (6) | 26 (3) | 20 (6) | 5 (1) | |||||
| 72 hours | 13 (1) | 10 (1) | 5 (1) | 3 (1) | |||||
| Don’t know or no answer‡ | 211 (24) | 240 (31) | 66 (18) | 119 (30) | |||||
| Fatal dose of paracetamol | |||||||||
| Question: What number of tablets of paracetamol do you think could kill a person? Answers: 15, 30, 50, or 100 tablets, or don’t know | |||||||||
| 15 | 258 (25) | 369 (21) | 140 (29) | 214 (27) | |||||
| 30 | 233 (23) | 365 (21) | OR for reporting lower doses¶ | 120 (25) | 171 (21) | OR for reporting lower doses¶ | |||
| 50 | 276 (27) | 247 (14) | 0.91 (0.77 to 1.06); P=0.2 | 0.80 (0.57 to 1.11); P=0.18 | 72 (15) | 100 (12) | 0.94 (0.74 to 1.21); P=0.6 | 0.75 (0.53 to 1.04); P=0.09 | |
| 100 | 67 (7) | 144 (8) | 29 (6) | 43 (5) | |||||
| Don’t know or no answer‡ | 196 (19) | 637 (36) | 114 (24) | 279 (35) | |||||
| Comparative toxicity of paracetamol | |||||||||
| Question: What number of tablets of each of the following drugs do you think could kill a person? Antibiotics, aspirin, Brufen, paracetamol, sleeping tablets, tranquillisers. Answers: For each drug dose stated as 15, 30, 50 or 100 tablets, or don’t know. Drugs then ranked according to responses | |||||||||
| Paracetamol singly most toxic | 145 (14) | 177 (10) | OR for ranking paracetamol as more toxic¶ | 88 (19) | 82 (10) | OR for ranking paracetamol as more toxic¶ | |||
| Paracetamol jointly most toxic | 570 (55) | 796 (45) | 1.04 (0.86 to 1.26); P=0.7 | 0.92 (0.63 to 1.35); P=0.7 | 236 (50) | 399 (49) | 1.40 (1.04 to 1.87); P=0.03 | 1.16 (0.79 to 1.71); P=0.4 | |
| Other drug more toxic | 119 (12) | 152 (9) | 37 (8) | 47 (6) | |||||
| Don’t know or no answer‡ | 196 (19) | 637 (36) | 114 (24) | 279 (35) | |||||
Adjusted for age, sex, social class, region, medical viewing habits, self reported medical interest, and general medical knowledge assessed by responses to two test questions.
Odds ratios (OR) estimated from logistic regression analyses for dichotomous outcomes.
For all questions, viewers were significantly less likely (P<0.001) not to respond or to respond “don’t know” than non-viewers. These responses were excluded from further analyses.
Only participants who correctly identified paracetamol as the drug causing liver damage are included in this analysis.
Odds ratios (OR) estimated from proportional odds models for ordinal outcomes, comparing the odds of viewers and non-viewers responding with more appropriate responses of any degree.
Acknowledgments
We thank the BBC for its help with this project, in particular, Barbara Machin (script writer Casualty), Christine Hamar Brown (script editor Casualty), the Casualty production team, the BBC television opinion panel, the BBC Information Department, and Dr G Hughes (consultant to Casualty).
Footnotes
Funding: The study was funded by the Nuffield Trust. KH and SS were also supported by Anglia and Oxford NHS Executive Research and Development Committee.
Competing interests: None declared.
References
- 1.Hawton K, Fagg J, Simkin S, Bale E, Bond A. Trends in deliberate self harm in Oxford 1985-95. Br J Psychiatry. 1997;171:556–560. doi: 10.1192/bjp.171.6.556. [DOI] [PubMed] [Google Scholar]
- 2.Bray G. Liver failure induced by paracetamol. BMJ. 1993;306:157–158. doi: 10.1136/bmj.306.6871.157. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Hawton K, Simkin S, Deeks JJ, O’Connor S, Keen A, Altman DG, et al. Effects of a drug overdose in a television drama on presentations to hospital for self poisoning: time series and questionnaire study. BMJ. 1999;318:972. doi: 10.1136/bmj.318.7189.972. ;7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Gunnell D. Reporting suicide. The effects of media coverage on patterns of self harm. BMJ. 1994;308:1446–1447. [Google Scholar]