The dangers of good intentionsBMJ 1994; 309 doi: https://doi.org/10.1136/bmj.309.6958.883 (Published 01 October 1994) Cite this as: BMJ 1994;309:883
- I Robertson
It is 1939 in urban Massachusetts. A group of social workers and social psychologists devise a plan which will take young delinquents off the streets and give them a stake in society. The scientists among them, however, insist that, though the scheme is self evidently worth while and of benefit to the boys, there must be a control group against which to compare the benefits of this programme of counselling and practical help.
Seven hundred boys are therefore randomly assigned to a control group and to an experimental group respectively. The fortunate 50% receive a twice monthly visit from a counsellor and get extra academic tutoring and family assistance whenever necessary. Most are introduced to the YMCA, boy scouts, summer camp, and other community programmes. The poor control group receive nothing except requests to supply information about themselves at regular intervals.
Thirty years on, Joan McCord and colleagues decided to find out what had happened to the 700 young rogues and embarked on a painstaking search of hospital, police, and other records to find out how beneficial the programme had been in the long term. She found an astounding difference between the two groups. One had a much higher incidence of diseases such as arthritis, emphysema, depression, ulcers, high blood pressure, and other circulatory disorders. The same group had significantly higher rates of alcoholism and psychiatric disturbance, and fewer of them were in white collar jobs. This group also committed more offences.
A resounding endorsement of compassionate commitment to enlightened treatment of social problems? Not quite, for it was the treatment group who were sicker, drunker, poorer, and more criminal 30 years after this exemplary exercise in blameless good will. One possible explanation for this finding is that the programme created dependency and a self perception by its beneficiaries that they were people in need of help - and hence not fully in control of their own lives. Such attitudes are known to be very corrosive not only of psychological wellbeing, but also of physical health.
What this study does not show, however, is that interventionism is necessarily harmful and laissez faire approaches wonderful: one wouldn't place bets on the future mental and physical health of the young beggars on London streets in 30 years' time, for instance. What it does show is that nothing can be taken for granted when it comes to devising ways of influencing people's lives. If such interventions can do good, then they must also be capable of causing harm. Research in psychotherapy has shown this for decades - some people are harmed by psychotherapy and would have been better off without it, while others get better. No one yet knows who or why. The same is probably true for health education, social work, probation services, and many medical services. Too often it is assumed that such social endeavours are self evidently desirable and in need of no further justification than the assenting smiles of like minded colleagues. Just as well there were a few scientists around in Boston in 1939 who mistrusted their own gut feelings and good intentions.