Intended for healthcare professionals

Feature Briefing

Citizens’ juries for health policy

BMJ 2017; 357 doi: https://doi.org/10.1136/bmj.j2650 (Published 02 June 2017) Cite this as: BMJ 2017;357:j2650
  1. Jacqui Wise, journalist
  1. London
  1. jacquiyoung1{at}gmail.com

A representative citizens’ group recently recommended that the Irish government permit abortion in early pregnancy. Jacqui Wise considers how such panels can influence health policy

What is a citizens’ jury?

Citizens’ juries (or community juries or citizens’ assemblies) aim to give ordinary people a role in democratic decision making. They usually consist of 12-20 randomly selected and demographically representative people, but some have had as many as 100. They explore difficult policy questions for government, charities, or think tanks. They differ from focus groups in that participants should be given reliable information and time to deliberate.

Malcolm Oswald, director of the Citizens’ Juries community interest company, a social enterprise supported by the University of Manchester, says, “Citizens deliberate among themselves, and, as they become better informed over several days, their views often change.”

When were they first used?

The process was devised in the US in 1974 by the Jefferson Center, a Minneapolis based non-profit organisation.1 And the UK prime minister Gordon Brown set up several juries in 2007, on crime, immigration, education, health, and transport.2

How do they work?

The Jefferson Center recommends that jurors take 4-7 days to consider independently verified facts and figures, as well as expert testimony. After discussion the jurors reach a conclusion or give recommendations for policy makers.

How widely are they used in health?

Many countries, including Australia, Brazil, Canada, Ireland, Italy, New Zealand, the UK, and the US, have used citizens’ juries to consider health policies on, for example, genetic testing, priority setting, provision of mental health services, and environmental health.

They are particularly helpful for considering controversial and value laden questions. For example, the Irish parliament formed a 99 member jury to advise elected representatives on ethical and political dilemmas including abortion, climate change, and provision for an ageing population. In April 2017, after 10 days’ debate, the jury voted by 64% to 34% to have no restrictions on terminations in early pregnancy.3 Parliament will now consider whether to amend the law.

What are their strengths?

Citizens’ juries involve the public in decision making, providing diverse experiences and perspectives, and the process can be thorough. For example, in 2015 in Victoria, Australia, 78 randomly selected people took part in a citizens’ jury on obesity set up by VicHealth, the state’s health promotion foundation.4 Jurors were given more than 60 background papers from various interest groups. After six weeks of online deliberation they met for a weekend of discussion and debate, and they voted on experts they wanted to hear from. Their final report contained 20 “asks,” including a 20% tax on sugary drinks and removing the unhealthiest foods from points of sale. VicHealth said that it would use the report to inform its healthy eating action plan.

A study published in BMJ Open concluded that a community jury gave insights into the priorities and concerns of men weighing up the benefits and harms of prostate specific antigen screening.5 The jury of 11 men from Queensland, Australia, unanimously concluded that governments should not invest in a screening programme despite their own divergent views on whether they would seek screening themselves.

Citizens’ juries can change policy. For example, juries have considered Melbourne’s city budget. External evaluation found that the juries made a tangible difference to the 10 year financial plan and achieved broad community engagement.6

What are their weaknesses?

A systematic review in 2014 that looked at 66 citizens’ juries found that many were of shorter duration than the Jefferson Center recommends, limiting the time for discussion. And recruitment practices vary.7 For example, some have used stratified random sampling through the electoral roll or random telephone digit dialling; others used newspaper advertising or direct recruitment of people in the street.

The University of Manchester’s Mary Tully is director of public engagement for Connected Health Cities, a project that uses health data and technology to improve health services for patients in northern England. It established a citizens’ jury to consider the extent to which patients should control access to their medical records.

“There is a real risk of bias if the juries are not run in a way that ensures inclusivity, deliberation, and active citizenship,” says Tully. “It is very easy to select only the people that you want; to give them only the subset of the evidence you want them to consider; or to ignore the results if they don’t fit with what you wanted to do in the first place. Using the Jefferson Center model means that the risk of bias is minimised.”

Oswald concurs that it is important to monitor, minimise, and report potential bias—for example, by appointing an independent oversight panel to review the choice of expert witnesses and the jury’s materials and other documentation. In addition, all jury documentation should be published after the event so that it is open to scrutiny, he says.

One citizens’ jury, tasked with helping to rewrite a leaflet about breast cancer screening in the UK in 2012, was heavily criticised in letters to The BMJ.8910 The jury, set up by King’s Health Partners, an academic health science centre in London, disagreed about whether the screening leaflet should put reassurance before accuracy.

One criticism was that the jurors were not representative because the 25 women were recruited on the streets of London and outside community centres. Another was that they had been given biased information: for example, the jury was told from the start that breast cancer screening saves 1300 lives a year, as stated in the Marmot review of breast cancer screening,11 which did not consider all cause mortality.

References

View Abstract