- Jeffrey K Aronson
Centre for Evidence Based Medicine, Nuffield Department of Primary Care Health Sciences, University of Oxford, Oxford, UK
- Twitter @JKAronson
Better than cure
The proverb that prevention is better than cure, in its modern form, or something like it, first appeared in English in the 17th century, but evidence of the idea can be found in many ancient Chinese texts, dating as far back as Huang Ti, the Yellow Emperor (2697–2597 BC). “The skilful doctor treats those who are well, while the less skilful treats those who are ill” is a saying attributed to Ch’in Yueh-jen (ca. 225 BC). Henry de Bracton in De Legibus et Consuetudinibus Angliae (ca. 1240) wrote “cum melius et utilius sit in tempore occurrere quam post causam vulneratam quaerrere remedium” (“It is better and more advantageous to deal with something in good time, than to pursue a remedy after the harm has been done.”) And a similar sentiment appeared in an early 13th century book of Jewish aphorisms, the Sefer Hasidim (the Book of the Pious): “Who is a skilled physician? He who can prevent sickness.”
Preventing future deaths
Searching for examples of the phrase “prevent future deaths” in biomedical publications, I have found none before 1900, amid many instances of preventing a range of other future concerns, such as alienation, breaches of the law, collisions, conception, crimes, danger, disappointment, disaster, discord, discussions, disputes, errors, famines, fears, ill results, inconveniences, inroads, interments, litigation, mistakes, offences, outbreaks, overflows, probate after death, punishment, seditions, sin, transgressions, and waste.
The nearest medical example comes from the immediate forerunner of The New England Journal of Medicine, The Boston Medical and Surgical Journal, in 18781: “Acting upon the advice of members of the American Public Health Association, the Surgeon-General of the Marine Hospital Service has organized a commission to gather and record all important facts relating to the commencement and spread of the present epidemic of yellow fever, with the view of establishing truths upon which the theory and practice of prevention of future epidemics may rest.” Not the exact phrase, but preventing future epidemics implies preventing future deaths, as we have recently become more aware.
I did find an example of “prevent further deaths” in The American Lancet of 1886, albeit not a medical one2: “A disease which would destroy in a few weeks seventy-five people, would awaken a wide interest, and stimulate vigorous efforts to remove its cause. But within this time seventy-six persons have died at the hand of the gaming tables of Monte Carlo. Still no effort is made to prevent further deaths at the instance of these deadly tables.”
But in 1900, following a report of the death of a young married woman in Highbridge after chloroform anaesthesia, included in a report of inquests of deaths from anaesthetics, James Edmunds wrote to The Lancet: “Not infrequently such reports as this are all that emerges when someone has been killed by an anaesthetic. This is not for the public good. It is not helpful to science. It does not prevent future deaths.”3
Later examples do not always refer to coroners’ reports. Here is an example from 19374: “To prevent future deaths in coal mine cave-ins, Dr Helmut Landsberg, assistant professor of geophysics at Pennsylvania State College, has contrived a simplified seismograph to predict otherwise imperceptible movements of the ground long before the crash.”
But there are examples in the context of postmortem findings5: “Modern medicine is unthinkable without the contributions which previous autopsies have made possible. They may free the mind of a research scientist to new concepts of diagnosis and treatment and help prevent future deaths from this disease.”
However, it was not until 1984 in the UK that the idea emerged more fully that the circumstances that caused a death might stimulate actions to be taken in order to prevent future ones, prompted by coroners’ concerns,6 forming the basis of Coroners’ Rule 43, “Prevention of similar fatalities,”7 which was later replaced by the Coroners and Justice Act 20098 and the Coroners (Investigations) Regulations, 2013.9 Under the regulations, coroners are required to report any action that, in the coroner’s opinion, should be taken to eliminate or reduce the risk of death created by the kinds of circumstances that emerged from an inquest, and to send the report to a person or persons that the coroner believes may be able to take such an action.
Similar legislation has been introduced in Australia10 and New Zealand.11
Efficacy
Unfortunately, current evidence shows that coroners’ reports, here and elsewhere, do not have as much in the way of beneficial effects as would be hoped.
For example, in our own recent survey of 3785 reports from coroners, the coroners received responses to only 49% of the medicines-related reports,12 despite the fact that by law responses must be received within 56 days.
In an earlier study in Australia, of 185 coroners’ reports, including 485 recommendations, the proportions of recommendations implemented varied from 27% in Victoria to 41% in Tasmania, 48% in New South Wales, 50% in Western Australia, 52% in South Australia, 65% in the Northern Territory, and 70% in the Australian Capital Territory.13
A few years later a prospective Australian study showed that only 90 of 153 recipient organisations surveyed (59%) responded to 164 recommendations from 74 cases.14 In all, 60/164 (37%) of the recommendations were accepted and implemented, 27% (45/164) were rejected, and in 59/164 (36%) the recommended action had reportedly already been taken.
Problems
In a study of 99 coroners’ reports in which medicines or parts of the medication process or both were mentioned, most of the reports went either to NHS hospital trusts or to local trusts and not more widely; the responses of addressees were rarely published and the reports rarely identified new hazards.15
In a report from Queensland, Australia, the ombudsman concluded that, among other things, a significant reason for the failure of public sector agencies to implement coroners’ recommendations is that the recommendations are not regarded as being sound or practicable. In the Australian study mentioned above,14 in nearly half of the rejected recommendations (18/45), the recipient organisations reported that implementation was not practicable. While 67 of the 90 organisations (75%) regarded the introduction of a mandatory response as a good idea, fewer regarded the recommendations they received as appropriate (52/90) or likely to be effective in preventing deaths and injuries (45/90). The same authors also later reported that the objectives of the state of Victoria’s mandatory response system were being compromised by the opacity of many of the response letters.16
In a review of 159 coroners’ reports there were differences in the amount of detail in the reports and in the frequencies of reporting between different regions.17 Almost 40% of the reports did not include the age of the deceased. There were errors in the reports, including incorrect dates, and disparities in the ways in which different coroners framed their reports. No responses were listed in about 23% of cases.
In another study there were regional variations in the frequencies of reports, a poor rate of responses, especially from small organisations, and little evidence of remedial actions.18
Elsewhere it has been suggested that generic recommendations should not be framed based on a single unusual case. While this is a reasonable conclusion, in several systematic reviews of coroners’ reports, members of our research group have found that there are instances of many similar cases in different jurisdictions, which together can provide useful information; these include deaths from cardiovascular disease involving anticoagulants,19 deaths attributed to the use of medications purchased online,20 and deaths from hand sanitizers during the recent pandemic.21 And in some cases of suspected adverse drug reactions single reports can be afforced by reports to regulatory authorities, as in the cases of tramadol22 and diclofenac.23 Furthermore, in a study of 710 reports, 36 expressed concerns about having to issue repeat reports to the same organisation for the same or similar concerns.24
A final thought
Whatever the reasons for the poor implementation of coroners’ recommendations, it is not unreasonable to hope that better outcomes might be achieved by improving the ways in which coroners’ reports are produced and made available. This can be done by ensuring that responses are forthcoming from all those to whom reports are sent, by distributing reports more widely, and by preparing systematic reviews of similar reports to reinforce the concerns expressed.
Footnotes
-
Acknowledgments: Thanks to Robin Ferner for ferreting out the Lancet citation.3
-
Competing interests: JKA is a member of a group studying Prevent Future Deaths reports (PFDs) and has published papers on the subject.
-
Provenance and peer review: not commissioned; not externally peer reviewed.
References
- ↵
- ↵
Anonymous. The American Lancet 1886 Nov; NS10(11): 427.
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
Anis A, Heneghan C, Aronson JK, DeVito NJ, Richards GC. Deaths from cardiovascular disease involving anticoagulants: a systematic synthesis of coroners’ case reports. BJGP Open 2022; 6(1): BJGPO.2021.0150. doi:10.3399/BJGPO.2021.0150
- ↵
Aronson JK, Ferner RE, Richards GC. Deaths attributed to the use of medications purchased online. BMJ Evid Based Med 2022; 27(1): 60-4. doi:10.1136/bmjebm-2021-111759
- ↵
- ↵
- ↵
Thomas ET, Richards GC. Diclofenac in adolescents: diagnosing and treating gastrointestinal adverse drug reactions can prevent future deaths. BMJ Evid Based Med 2021: bmjebm-2020-111640. doi:10.1136/bmjebm-2020-111640.
- ↵