Intended for healthcare professionals


Income, health, and the National Lottery

BMJ 2001; 323 doi: (Published 22 December 2001) Cite this as: BMJ 2001;323:1438

This article has a correction. Please see:

The lottery is one of the world's largest randomised trials

  1. Anthony Rodgers (a.rodgers{at}, co-director
  1. Clinical Trials Research Unit, Department of Medicine, University of Auckland, Auckland, New Zealand

    Most people in the United Kingdom have taken part in one of the world's largest trials of one of the most important determinants of health. Unfortunately, neither the participants nor the organisers know about the trial and no one has collected follow up data.

    Each month, more than £150m is randomly redistributed among 60% of the adult population in the National Lottery.1 Over £16bn has been redistributed since the lottery began in 1994. Changing the redistribution of a small fraction of this money could create a randomised trial that reliably assessed the speed and extent to which increases in income improve health. The basic study design would be simple. Instead of lump sums, winners would receive regular, income-like payments (such as £40, £80, or £160 a month for a decade). Follow up of these winners, and a large random selection of non-winners, would assess effects on outcomes such as diet, smoking, admission to hospital and broader indicators like employment, social participation, and entrepreneurship.2 Most people buy lottery tickets and winning is purely chance, so the study would be, in effect, a randomised trial of income supplementation in a group drawn at random from the majority of the UK population.

    If just 5% of one year's worth of prize money (2.5% of sales) was redistributed in this way the resulting fund of more than £100m, and the interest it would generate, could pay for all study costs and prizes. The prizes could, for example, allow 10 000 people to receive £40 a month for 10 years, 5000 to receive £80, and 2500 to receive £160. Standard tickets that didn't win the first time could be re-entered into the extra “pay packet winner” draw (which would allow enrolment targeting if reuse was restricted, for example, to those receiving a social security benefit). Alternatively, a new type of ticket could be sold in the usual outlets.

    The project would be entirely self funding and be “win-win” for all parties:

    • Some participants might appreciate the new type of prize, especially if there is a higher chance of winning and they can reuse a non-winning normal ticket. Certainly, regular payment prizes are accepted overseas.

    • People concerned about possible adverse effects3 might welcome the chance to assess the benefits of more modest winnings.

    • The lottery operator may welcome making more than 10 000 people better off for 10 years, as well as making millionaires. The novelty could revitalise sales. A television documentary following some participants could generate renewed interest.

    • The government's stated aim is to direct lottery funds more actively to health, education, and the environment “with particular focus upon the needs of those who are most disadvantaged in society.” This proposal fits that aim. It is also in keeping with the move towards evidence based social policy.4 If a government alters fiscal policy and increases expendable income among, for example, those receiving benefits, how much is needed to see health and other improvements? And how soon are the effects seen?

    The appropriate role of controlled evidence in social policy is controversial.5-8 Here, it would be important to recognise that income is only one component of socioeconomic status and that while evidence is gained at an individual level, policy implications would be at a societal level.9 The evidence base guiding such policy decisions clearly has to be more diverse than, for example, that guiding clinical practice. Nevertheless, there may be particular value in new sources of evidence relating more distant determinants, such as raising household income, to health, especially from a study that is local, straightforward, and robust. Substantial changes in income are regularly made by governments via taxation, benefits, minimum wage policies, etc, but in such a way that their effects on health cannot be evaluated reliably. This initiative would increase the consideration of potential changes in health status during deliberations about such policies.

    A recent non-randomised analysis suggested that windfalls (lottery wins or inheritances) of about £50 000 were associated with moderate improvements in mental wellbeing in following years.11 However, much more modest increases in income, if sustained, could lead to important health benefits. 12 13 Reliably detecting how much and how quickly those benefits are realised ideally requires large scale, long term randomised evidence. The UK lottery provides a unique opportunity to generate such evidence, given its size, coordination, and high participation rate. If successful, the project could also be adopted elsewhere: most countries have lotteries and well over $100bn is spent on lottery sales worldwide each year. However, the UK lottery provides the best opportunity, given its size, coordination, and high participation rate: it could provide valuable evidence for future social policies, particularly those aiming to lift people out of poverty.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.
    View Abstract