Is deterioration in adult mental health linked to worsening childhood conditions?

A paper I have recently come across (and tweeted a link to) has showed us that mental health in their early 40s is quite a bit worse among members of the 1970 British Birth Cohort than it was at the same age in the 1958 birth cohort. This was a very welcome analysis. Many years ago, in response to a request from the old Health Education Authority and the Department of Employment (now DWP) our ESRC Resilience Network team set about a similar task in a rather rushed manner and I always wondered about the results.

The DWP was very worried about the large & increasing numbers of young-ish working age people who were on long term sickness benefits because of mental health issues. At a meeting we had agreed this might happen in (at least) 2 ways: 1. young people were arriving at school leaving and labour market entry with a heavier burden of mental health problems in 1986 (when 1970 cohort would be 16) than they did in the mid 1970s; 2. something was different in the lives of early labour market entrants .

I personally was quite sure that changes for the worse in childhood, such as higher levels of divorce and single parenthood, would have produced a less mentally fit cohort of labour market entrants. How wrong I was! To our surprise the data gave absolutely no sign of this. So we went on, in an even greater rush, to have a quick look (using different data from the BHPS youth boost survey) to look for a mismatch between job aspirations before age 16 and destinations at the time when adult mental health was measured. I was sorry to have too little time to go further into this and arguably we should have when we started work as the ESRC International Centre for Life Course Studies (ICLS). But at the time we wrote the research application for ICLS, 2006-7, funders' priorities had shifted away from unemployment & non-employment (don't laugh).

In thsi paper "Psychological distress in mid-life: evidence from the 1958 and 1970 British birth cohorts" Psychological Medicine Volume 47, Issue 2January 2017 , pp. 291-30, Ploubidis et al. take great care to make sure their measures of adult distress are truly comparable between the 2 cohorts. They also did a far more sophisticated analysis, by considering a lot more mediating and confounding factors than we did. But their analysis did not indicate that childhood factors such as parental divorce, maaternal employment or childhood behavioural maladjustment accounted to any large extent for the higher distress scores in the later cohort. They comment:

The 1958 cohort are part of the ‘Lucky Generation’ of post-war baby boomers, who experienced high absolute levels of social mobility, and lower levels of social inequality, whereas the 1970 cohort are part of ‘Generation X’, who have experienced greater uncertainty and insecurity over the whole of their adult lives and a more individualistic ideological climate (Sullivan et al., 2015). If these generational changes lie behind the increase in psychological distress, then we would predict that future generations will be worse off still if such trends were to continue

This is a sobering conclusion. Even more so when one considers that the study participants who suffered the worst from mental distress are likely to be  found among people "sanctioned" by today's workfare policies.

Reference:

SULLIVAN, A., BROWN, M. & BANN, D. 2015. Guest Editorial: Generation X enters

middle age. Longitudinal and Life Course Studies, 6, 120-130.

Access to the paper:

Printed version, paywalled at

https://www.cambridge.org/core/journals/psychological-medicine/article/psychological-distress-in-midlife-evidence-from-the-1958-and-1970-british-birth-cohorts/43F04DCFE2E3771DED949BB02068F02D

and for free preprint:

https://discovery.ucl.ac.uk/id/eprint/1522682/1/PloubidisPsychMed_Accepted.pdf

Looking back at The Black Report

The publication of the latest in Michael Marmot's reports on health inequality has aroused the curiosity of people who either don't remember the original Black report on heath inequality published in 1980, or only vaguely remember it.

At the time I was working as a clerical assistant on the British Regional Heart Study (BRHS) (my 1st experience of epidemiology), but also hanging out with various left wing people who were into the politics of health. I had studied for the Bedford College MSc in Medical Sociology between 1971-2 and so had an initiation into the fact that health and life expectancy varied according to social class. It was pretty stunning to think that social inequality could actually influence your chances of a long life. The BRHS also studied some aspects of what would now be called "social determinants of health" although only as a side issue. The main hypothesis of the study was that heart disease had something to do with how hard the water was and the policy implications would be whether measures should be taken to harden water in the areas where water was softest. So social conditions and even behaviours were not the main focus, altough the study did concentrate mainly on health behaviour eventually. But it meant that both friends and colleagues were informed observers of the publication of the Black Report.

The Black Report had been commissioned by a Labour Secretary of State at the Department of Health and Social Security (DHSS) , David Ennals, after he had read an article in New Society by Richard Wilkinson. It happened at the time that the DHSS had become disillusioned with the fact that the Medical Research Council (MRC) spent too much money on obscure diseases and genetics (yes, already) and so removed 30% of the MRC budget to be spent on "cinderella specialties" such as the social factors in health and illness. The rumour was that Sir Douglas Black, a very senior medical figure, had been given the funding for the Black Report to make up for the loss of power of the medical royal colleges over the health budget.

In those days, scientists interestd in health inequality had to contend with 2 major strands of thought whose implications were that it did not exist at all (you might say it was fake news). One of these held that health inequality was a normal part of a kind of Darwinian process of selection: fitter people found themselves in the more advantaged social classes because they were fitter & more intelligent, and it was this fitness that also made them healthier (nowadays we might draw a DAG for this). The 2nd was the "artefact explanation", which held that the appearance of health inequality in the official statistics was due to statistical artefacts which I wont go into unless anyone asks me.

So it was really quite an achievement that the Black Report did result in the widespread acceptance that health inequality was real and had something to do with social conditions (although some health economists continued to argue against this, even to this day). And the explanations that Black et al. accepted were (1) "behavioural cultural", i.e. there was something in the culture of the social groups with worse paid, more arduous jobs that encouraged unhealthy behavour (2) "material" explanations, which focused on the actual biological effects of poverty, poor housing, arduous working conditions and stress.

But I know what many people are interested in is the reception of the Black Report. . It was published soon after the election of a Tory government. Only 260 copies were produced, in a kind of cyclostyled typescript form not even properly printed, released on the Friday before an August Bank Holiday.  It is a huge doorstep of a thing, and those of us who managed to get our hands on a copy still treasure it greatly. There was little fanfare. I don't remember Sir Douglas going on the TV though I hardly watched any in those days. A great account of the emergence of the Black Report can be read in the Introduction to

P Townsend, N Davidson , M Whitehead "Inequalities in Health" (Penguin 1988)

A good summary is:

D Blane "An assessment of the Black Report's 'explanations of health inequality'" Sociology of Health and Illness 1985 7; 3: 423-445.

The subsequent story of the inter relationship between research and policy in this area is documented in :
K Smith "Beyond Evidence Based Policy in Public Health :The Interplay of Ideas" (Palgrave, 2013)

The Black Report itself had a lot more influence on research than on policy. Especailly after it was updated as Margaret Whitehead's "The Health Divide". But unfortunately health inequality then became a kind of bandwagon where anything that seemed to show the importance of health behaviours of or "Selection" got published in high impact journals and other papers showing a bigger influence of material factors, often more methodologically complex due to defensiveness, went unremarked.

Partly as a result of this, all the well meaning hype that followed the Health Divide  and the subsequent Acheson Report just fizzled out into "lifestyle drift", that is, in the end, more preaching to poorer people about their "behaviour".

This was where I started from early this morning, as I noticed (& tweeted) that this time round only a single journalist no one ever heard of had written that the recent plateau in life expectancy (with falls in the poorest women) was due to their fact that they are too fat and in any case live quite long enough. It has become quite unrespectable, both for academics and for informed opinion leaders, to blame health inequality on behaviour. I was impressed by this and thought, well, maybe those of us who have worked on this stuff for the last 40 years didnt totally waste our time.

The functionalist theory of health inequality

It took me a long time to realize that the implicit theory underlying a lot of work on health inequality was the classic American idea of Structural-Functionalism, developed by people like Talcott Parsons and Robert Merton. This was dozy of me, as decades ago Gordon Marshall had pointed out that the Registrar-General's Social Class schema used in research on health inequality in UK was based on, as he saw it, outdated functionalist and eugenic notions.
When the national statistics office for England & Wales adopted a new measure of social class for the 2001 Census and all other official statistics from then onward, I did notice that people had trouble using it. There seemed to be a constant attempt to drag the meaning of the measure away from "employment relations and conditions", which was its theoretical basis, back toward something like "manual/ non-manual". I have already written about this all over the place, including in previous blogs. So I won't bind on about it too much more here.

But recent exhaustive international comparative research has given the idea new legs. It turns out that over the last 30-40 years, Italy and Spain have stubbornly retained the smallest differences in mortality and life expectancy between those in the most and the least advantaged social circumstances, whether these were measured by income, education or social class. I well remember the total shock 25 years ago when this was first discovered. Italy & Spain don't have the lowest income inequality for example, or the most egalitarian welfare states. In fact, health inequality in most studies is higher in the egalitarianNordic nations. For many years the standard explanation was that the smoking epidemic was delayed in these Mediterranean nations: smoking was slow to become concentrated among people with lower income and less advantaged occupations.
However, as the years rolled by, it began to seem less likely that the narrower health gaps in the Mediterranean nations could be purely due to smoking. So an alternative explanation began to emerge.
According to this idea, the size of the differences in health & life expectancy between more & less socially advantaged groups might be due to the ways in which their home nations allocate people into these groups. Accroding the the Functionalist theory, societies like the Norway & Sweden are more meritocratic. Education is available to everyone up to an advanced level. This helps to make sure that the fittest & most able people are channeled into those jobs that are most essential for the "functioning" of society, like senior management, judges, military and political leaders.  Regardless of origin family, the fittest and most intelligent people will be channelled through the schools and universities into these important positions. In order to motivate the fittest people to aspire to these destinations, salaries & status are high. As fitness for high position is only partly determined by genetic inheritance from the parents, this process is important to make sure that "good functioning" is ensured. A society cannot just rely on allocating the sons (and it would be sons) of the powerful into powerful positions themselves. There will have to be a turnover such that the less fit sons of more advantaged families in one generation are filtered out by the education system and replaced by the fitter sons of the less advantaged. So the less fit fall down the social ladder and contribute to worse health in the less advantaged social groups, and vice versa. Michael Young wrote about this many decades ago in his book "The Rise of the Meritocracy", depicting an eventual dystopia in which society was divided into extremes of health, intelligence, and income.
There is some evidence in favour of this idea

 https://academic.oup.com/eurpub/article/23/6/1010/439677

In addition, there were large increases in health inequality in England and Wales duriing the 1950s-1980s, a time when the numbers of middle class jobs increased enormously, with an associated increase in social mobility (although this mobility took place during the work career, not through the educaiton system). One might argue that the increase in mortality among older working men in unskilled manual jobs (which was the main source of the rising inequality) resulted from an unfit group being left behind.

https://tinyurl.com/u4vazyg

On the other hand, extensive sociological research has documented a pretty hefty role for the social class, income and education level of the parents in determining those of the children, even after taking coognitive variables ("intelligence") into account. If anything, the expansion of higher education in the UK, for example, has mainly benefited middle class children. Almost 100% of middle class children now go to University and get degrees, which must therefore cover a wide spectrum of the ability range.

So there is a certain amlunt of evidence out there that can be used to test this idea further. But a lot remains to be done.