Summerfield suggests that the World Health Organization ^{Reference Demyttenaere, Bruffaerts, Posada-Villa, Gasquet, Kovess and Lepine1} and other survey data that we use seem ‘preposterous’ to him as a doctor and a citizen. His quarrel, then, is not with us but with the psychometric testing of the diagnostic interviews used by the WHO and other epidemiological surveys of mental illness. But even in terms of our own personal experience, we are not at all surprised at the 23% annual period prevalence of any mental illness in the UK. Many of us have felt incapacitated by depression or anxiety, and among our acquaintances we can count episodes of self-harm, eating disorders, addictions, behaviour problems and autism-spectrum disorders. As we mentioned in our paper, episodes of severe mental illness are also strongly correlated with income inequality. ^{Reference Pickett and Wilkinson2} Both sets of data suggest that inequality is related to mental health, however we choose to label the symptoms.

It is wrong to suggest that the correlations reflect only the high measured prevalence of mental illness in the English-speaking countries. Although these countries do indeed have higher prevalence of mental illness, and higher levels of income inequality, they are not outliers and do not appear to represent a distinct group: they are simply the countries at one end of the distribution. Indeed, if we look only at the subsample of the English-speaking countries, income inequality remains significantly correlated, and is an important explanatory factor for mental illness just among them (r = 0.95, P = 0.01).

It would be odd if the relationships we showed with mental illness existed in a vacuum but of course they do not. Our research focuses on problems with social gradients, and we find that more unequal societies also have lower levels of trust and social capital, poorer physical health, higher rates of obesity and teenage pregnancies and births, low child well-being, educational achievement and social mobility, and higher levels of violence and imprisonment. ^{Reference Wilkinson and Pickett3} Against that background it would be surprising if mental health were not also affected by wider income differences.

Until the rise of neoliberal economic policy in the 1980s, the UK was a much more equal society and it could be so again. We are optimistic that societies can change. There are numerous mechanisms through which governments and institutions can promote greater equality, and a wider recognition of the harm caused by inequality is an essential prerequisite.

The reality is that inequality causes real suffering – regardless of labels. Those of us concerned with the mental health of the public need to address its structural, as well as its individual, context.