What are the consequences when doctors strike?
BMJ 2015; 351 doi: https://doi.org/10.1136/bmj.h6231 (Published 25 November 2015) Cite this as: BMJ 2015;351:h6231- David Metcalfe, research fellow,
- Ritam Chowdhury, research associate,
- Ali Salim, professor of surgery
- Correspondence to: D Metcalfe dmetcalfe{at}bwh.harvard.edu
- Accepted 17 November 2015
The right to strike is recognised as a fundamental human right by the United Nations, the Council of Europe, and the European Union.1 Most European countries enshrine the right to strike in their national constitutions.2 It has been argued that collective bargaining amounts to nothing more than “collective begging” for employees without the option to strike.3 This is because employees have no choice but to accept the final terms imposed by their employer. It has even been claimed that an individual’s right to withdraw his or her labour is a feature that distinguishes employment from slavery.4
Industrial action by doctors is, however, complicated by their professional values and ethical framework. Most strikes are effective because they harm a neutral third party, who is then motivated to pressure the employer to accede to strike demands. Unfortunately, when doctors strike, this third party is often their patients.5 For many people, such behaviour is inconsistent with the over-riding duty of doctors to advocate for their patients.6 Others have claimed that doctor strikes inevitably expose patients to risk of serious harm.7 The situation may be further complicated by doctors striking to oppose policies that are perceived to threaten the standard of care they are able to deliver.
Strikes by doctors highlight the conflict between doctors’ rights as employees and their duty to patients. They are, however, a global phenomenon, with strikes already reported this year in Australia, India, Ghana, Nigeria, the United States, and Venezuela. In a recent British Medical Association ballot, junior doctors in England voted overwhelmingly in favour of strike action over the government’s threat to impose a new contract next year.8 We examine data from previous strikes for evidence to support claims that industrial action harms patients.
Professional and legal consequences of strike action
Medical regulators often oppose strikes by doctors.6 For example, the Delhi Medical Council has stated that “under no circumstances [should doctors] resort to strike.”9 In the UK, it is an offence under trade union legislation to “wilfully and maliciously [break] a contract of service … knowing or having reasonable cause to believe that the probable consequences … will be … to endanger human life or cause serious bodily injury.”10 Similarly, the General Medical Council (GMC) requires doctors to ensure that patients are not harmed or put at risk by industrial action. It has stated that doctors should “satisfy themselves that arrangements are in place to care for their patients” during industrial action and “must not harm patients or put them at risk.”11 After a UK doctors’ strike in 2012, the GMC received complaints about three doctors.12 We could find no details of these complaints under the decisions of interim order and fitness to practise panels posted on the GMC and Medical Practitioner Tribunal Service websites. It was also not possible to identify these cases when we contacted the GMC directly. As details of complaints become public only when they lead to a warning or a hearing, it is unlikely that these complaints resulted in disciplinary action. Similarly, threats to sanction striking general practitioners for breach of contract were dropped following the 2012 action.13
Although legal action is sometimes brought against doctors’ unions (for example, to break strikes), there are few documented cases of action against individual doctors. An 18 day strike by 500 interns in Chicago in 1975 led to brief jail terms for seven of the strike leaders. However, this followed their decision to ignore a court order to end the action.14 We could not find any cases brought against individual doctors following the 2012 UK strike, despite a comprehensive search using the Westlaw UK legal database.
Harm to patients
A recent systematic review reported mortality data from five doctor strikes, all of which saw patient mortality remain the same or fall during industrial action (table⇓).15 Two further studies have been published since that review.16 17
In 1976, between 25% and 50% of physicians in Los Angeles County, California withheld care for all but emergency cases over five weeks. Three studies used a range of approaches to examine the consequences of this strike, and all found that mortality fell during the strike period.18 19 20
In 1983, 73% of doctors in Jerusalem refused to treat patients inside hospitals over a salary dispute. During this four month action, emergency departments were staffed as on weekends and many doctors provided care for ambulatory patients in tents outside hospitals for a fee. A subsequent analysis of death certificates found no excess mortality during the strike.21 A second action in Jerusalem, in 2000, led to the cancellation of all elective hospital admissions. There were fewer funerals held in Jerusalem during these three months than during the same period in the preceding year.23
Junior doctors in Spain went on strike for nine non-consecutive days in 1999. A study from one emergency department (in which all resident doctors ceased treating patients) reported no mortality difference between strike and non-strike periods.22
National mortality data have been studied for only two countrywide doctor strikes.17 24 In 2003, most doctors in Croatia went on strike for four weeks, during which they provided only emergency care and at the level usually available at weekends. A study that analysed both total and cause specific mortality found no significant association between the industrial action and patient deaths.24
In 2012, the BMA organised a single “day of action” as a response to government pension reforms. The aim was to boycott non-urgent care but many doctors continued working as normal; the government estimated that only 8% of the medical workforce participated.25 There were fewer in-hospital deaths on this day, both among elective and emergency populations, although neither difference was significant.17
The only report of increased mortality associated with strike action comes from South Africa. In 2010, all the doctors in one province ceased to provide any treatment to their patients for 20 consecutive days. Only one hospital continued to provide services during this period to an estimated population of 5.5 million people. Although their data are poorly reported, authors from this hospital found that the number of emergency admissions fell during the strike period but that the odds of death for these patients increased by 67%.16 This may be because patients delayed seeking treatment and so were more likely to present in extremis during the strike.
Why don’t patient deaths increase during doctor strikes?
Many explanations have been proposed for why doctor strikes in high income countries have not been found to increase patient mortality. Importantly, all such strikes guaranteed provision of emergency care, at least at the level usually available at weekends. In addition, many were incomplete, with physicians declaring a strike but continuing to provide routine services to patients. This was most apparent during the 2012 UK strike when it was sometimes difficult to determine which doctors were actually taking action.17 Similarly, during the 1983 Jerusalem strike, the provision of care to ambulatory patients may have permitted hospitals to focus on treating the most urgent cases. It was also suggested that many Israeli doctors continued responding to emergencies within hospitals even though they were officially striking.26
Emergency care may even improve during industrial action. For example, during the 1999 strike in Spain, junior doctors in the emergency department were replaced by more senior physicians.22 The cancellation of elective admissions may also increase the number of doctors available to treat emergency patients. This phenomenon was noted after the 1983 strike in Jerusalem, even though medical staffing levels had fallen to 30% of the usual workforce.21
It is likely that temporary reductions in mortality are related to the cancellation of elective surgery.5 6 15 Death rates increased immediately after the LA County strike, which was attributed to hospitals resuming elective operations.19 20 Other possibilities are that doctors are better rested during strike periods and that the number of staff required to avert patient deaths is comparatively low.21
How disruptive are doctor strikes?
Although doctor strikes do not seem to increase patient mortality, they can disrupt delivery of healthcare. The extent of such disruption depends on the healthcare setting, strike duration, and the extent of doctor participation. Most strikes have led to widespread cancellation of elective operations and non-urgent hospital consultations.17 18 19 20 22 Service disruption can be substantial, even when comparatively few doctors participate, as in the 2012 UK strike over pensions. Although elective admissions fell by only 12.8%, outpatient cancellations increased by 45.5% on the day of the strike.17 This disruption was exacerbated by managers needing to overestimate strike participation and to ensure adequate staffing of essential services.
Importantly, most previous strikes have not impaired the quality of care. Despite disruption to UK elective services in 2012, patients were not discouraged from attending NHS emergency departments.17 Care quality indicators were unchanged, or even improved, during junior doctor strikes in Spain and New Zealand.22 27 28 In New Zealand, both emergency department waiting times and hospital length of stay fell significantly, which was attributed to the replacement of junior doctors by senior physicians.27 28 One study reported that each senior physician carried the workload of two junior doctors during this period.28 The deployment of senior physicians is, however, likely to be at the expense of disruption to elective care.
Conclusion
Doctors must carefully balance their duties to patients with their rights as individuals.6 Previous strikes have shown that it is possible to disrupt elective services while ensuring that emergencies are treated promptly and effectively. The strikes in New Zealand showed that senior physicians can provide emergency care in the absence of junior doctors.
It would, nevertheless, be naive to imagine that industrial action can be undertaken without causing any harm to patients. No study to date has explored the effect of industrial action on patients’ quality of life or confidence in the medical profession. The effectiveness of strikes depends on causing disruption to a third party.5 In the case of doctor strikes, this disruption is likely to manifest itself as financial loss for hospitals,17 inconvenience for patients, and political pressure on employers.6 This pressure may be increased by the popular perception that doctor strikes risk patient safety. Nevertheless, doctors should ensure that industrial action is organised in such a way that patient safety is not compromised. The consequences of doctors withdrawing all services and overwhelming the only available emergency facility are clearly shown by the 2010 strike in South Africa.16
Some doctors will always feel that industrial action is fundamentally inconsistent with their professional obligations because of its inevitable impact on patients. However, in balancing their competing priorities, doctors in high income countries can be reassured by the consistent evidence that patients do not come to serious harm during industrial action provided that provisions are made for emergency care.
What you need to know
Industrial action by doctors can substantially affect elective services, even when many do not participate
Within developed healthcare systems, doctor strikes have not been found to affect mortality
Emergency care has been maintained in most previous strikes
Full strikes by all doctors in resource poor healthcare settings endanger patient safety
Notes
Cite this as: BMJ 2015;351:h6231
Footnotes
Contributors and sources: DM is trained in medicine and law. He is a Royal College of Surgeons of England Fulbright Scholar working in health policy at Harvard Medical School. RC is a physician, epidemiologist, and biostatistician. AS is a trauma surgeon and critical care intensivist, whose research interests include the development of effective health systems. DM performed the literature search, drafted the manuscript, and is the guarantor. RC and AS made important revisions to the manuscript.
Competing interests: We have read and understood BMJ policy on declaration of interests and declare the following interests: DM is a specialty registrar in trauma and orthopaedic surgery and a member of the BMA. He has not held a formal position or had a political advocacy role within the BMA.
Provenance and peer review: Not commissioned; externally peer reviewed.