Policy break: maternal mortality


One of the most encouraging things to happen in recent American political discourse is the new and heightened focus on racial disparities in maternal mortality rates. Black women are three to four times more likely to die in childbirth than white women – and this is a scandal, and needs to be seen as such. It’s a cruel and senseless world in which creating new life can carry a death sentence, and this does not need to be happening. Every preventable death that takes place in a maternity ward – and up to 60% of these are preventable – is a woman who was, objectively, murdered by a social order that fails to allow the essential needs of human life to be met. It has to end. How?

One popular approach comes from Senator Kamala Harris, also running for the US Presidency. Her Maternal CARE Act explicitly aims to eliminate this racial disparity through three proposals: providing funds to ‘incentivise’ healthcare providers to deliver ‘integrated health care services to pregnant women,’ providing competitive grants to encourage medical and nursing schools to introduce implicit bias training, and directing the National Academy of Medicine to make recommendations on a further rollout of implicit bias training as part of medical education. Of these three, the proposals concerning implicit bias have received by far the most emphasis, from both Harris and the media. It’s a popular policy, and it’s already being woven into Harris’s Presidential campaign.

It’s the other proposal, however, that has the greatest chance of offering a potential solution. The racial pregnancy outcomes gap is not fixed or universal: in most of the United States, the gap is widening – but one US state, North Carolina, has managed to almost entirely close the gap. Black women died during childbirth at a rate of 24.3 per 100,000 in 2013, down from nearly 60 in the early 2000s; white women at a rate of 24.2. Some of this narrowing is accounted for by a rise in white mortality, which more than doubled in the same time period. I don’t think there should be any question that it would be far better, if it were the only option, to reduce the total number of preventable deaths while maintaining a racial disparity (North Carolina is 71% white). But the rise in white mortality is in line with a nationwide collapse in quality of life for white working-class individuals (the national rate climbed, while the decline in black mortality, in both relative and absolute terms, is unique. One significant factor is the state’s Pregnancy Medical Home programme, which uses the existing Medicaid system to deliver state funds that promote early intervention for high-risk pregnancies. The programme is expansive, addressing not only strictly medical issues but factors such as homelessness or food insecurity that strongly correlate with deaths during childbirth. It shows concretely that policy aimed at improving the lives of the working class can massively alter racial disparities. The most shocking and deadly effects of racism really can simply vanish once an effort is made to redress inequality in general.

The programme is, of course, deeply insufficient. It’s brought mortality rates for black women in North Carolina down to around the level of the national average, which is still monstrously high. But it shows the kind of outcomes that could emerge out of more radical intervention. Currently, the programme offers women advice and assistance dealing with food insecurity and homelessness – what if there were a serious redistributive programme to eliminate these factors altogether? In New York City, 63% of white patients give birth in the safest hospitals in the city; for black women, it’s 23%. What if no hospitals were unsafe? This is why the question of race and childbirth mortality is so crucial: as soon as you get really serious about solving it, you start dealing with the totality of oppression in general. After all, isn’t the question, at its root, that of life itself?

Senator Harris is seemingly not interested in confronting that question. It proposes a demonstration project, in which ten states would, for a limited period, mimic the South Carolina model. When compared with more ambitious policies, such as Medicare for All, it’s simply not enough. But the flagship proposal has nothing o do with increasing the quantity of care available: the radical element, the part that stands out, is the implicit bias training.

Implicit bias theorises that behaviour is influenced by unconscious stereotypes – that, for instance, even an avowed and conscientious anti-racist might hold racist attitudes and adhere to stereotypes, even as they explicitly reject them. In this context, the implication is that the unconscious biases of medical workers lead them to deliver a worse standard of care to black patients – because black suffering is simply not valued as much as white suffering. Implicit bias training aims to overcome this effect. First, trainees typically take an electronic implicit bias test, in which they’re asked to associate names or terms with the categories ‘white or pleasant,’ or ‘black or unpleasant,’ or ‘white or unpleasant’ or ‘black or pleasant.’ Their response times are measured. Typically 70% of participants (including nearly 50% of black Americans) have a harder time associating positive terms with the ‘black or pleasant’ category than the white. This gives a numerically quantifiable indicator of the test subject’s unconscious racism. They’re then trained to recognise this bias, confront it, own up to it, and overcome it. Then, the test is administered again, to see if they’ve improved.

One of the more alarming problems with implicit bias training is that it doesn’t work. Studies of the literature have found that the correlation between implicit bias test scores and actual discrimination outcomes is ‘close to zero.’ Systemic racism is not the same as the aggregate of millions of unconscious ideas, and the unconscious mind moves in stranger ways than causing you to hesitate on a timed computer test. Worse still, it’s been suggested that implicit bias trainings can have an effect – in the wrong direction. An exhaustive training in the persistence of racial biases can, it seem, have a mimetic effect. The sessions might encourage, not alleviate, racial stereotypes.

This is of minor importance when it comes to implicit bias training in universities or the corporate sector – even if it really is counter-productive, that doesn’t affect its primary purpose as a PR fig-leaf. But if you believe, as Senator Harris appears to, that the disparity in health outcomes is caused to some degree by unconscious bias, the consequences here are potentially monstrous. Outside of the ten states selected for the Pregnancy Medical Home demonstration project, her proposal could directly lead to a widening of the racial disparity, and more black women dying during childbirth.

* * *

All this assumes, of course, a certain vision of what policy is: we have a society that’s mostly good, but which has some problems, and after reviewing the evidence we can decide to do things that might fix those problems and help society function better and more equitably. I happen to have another view, and try to be as resistant to facts and evidence as possible. The sphere of potential is vast – and policy is a dream we have about ourselves, the kind of people we think we are, the kind of world we think we live in. This is why the argument that Trump’s wall wouldn’t be very effective at keeping out undocumented migrants is itself so singularly ineffective: Trump’s base don’t want a wall because they’re convinced it will lead to desirable objective outcomes. They want a wall; they want to live in a country that’s fortified.

But Harris and her ideological kin are very much wedded to the utilitarian and technocratic approach. See, for instance, her most notorious policy innovation: her practice, as a California District Attorney, of throwing the parents of truant children in jail. This is, as critics have pointed out, a profoundly unpleasant thing to be doing – but her campaign defended it in explicitly technocratic terms. ‘A critical way to keep kids out of jail when they’re older,’ a spokesperson said, ‘is to keep them in school when they’re young.’ Her contention is that the policy worked – school attendance rose in San Francisco during her tenure as DA – and there’s therefore nothing to complain about. The ends (kids in education) justifies the means (intensified police surveillance and discipline of the working classes) – so long as it’s effective. So why, then, is she now proposing policies which are so profoundly unlikely to advance their stated aims?

The Maternal CARE Act is incomprehensible when evaluated according to her own criteria. Under a different set, it makes a lot more sense. The findings that this procedure fails to achieve its intended result shouldn’t really be counter-intuitive: DARE doesn’t stop kids taking drugs either, and few social problems are attributable to people not being berated or lectured to enough. If these procedures have proliferated, to the extent that elements of the State now want to introduce them in legislation, it’s because their actual purpose is something very different. These are mandatory sessions in which workers are castigated for their shortcomings, told they’re responsible for some of the worst evils of the world, and subjected to hyper-surveillance and discipline as a corrective measure. It’s an upwards redistribution of power, a Taylorism for the reflexes, the assimilation of not just the conscious self but of hazy unconscious attitudes to the sovereignty of the administrative class. If the central question of policy is that of the kind of world we want to live in, the image painted here is bleak. A world of faulty machines. A world in which people are constantly being dragged down by their own evil natures, and have to be improved by an enlightened elite with its dictatorship of prods and nudges. A world in which the solution to what causes us to suffer isn’t shared struggle based around shared needs, but the same atomised self-negation that constitutes much of that suffering.

That Harris and her supporters so badly want implicit bias to be the problem, and this mode of surveillance and control to be the solution, is instructive. The desire is far stronger than their fetish for rationality or evidence; technocracy has far more to do with power itself than efficiency, outcomes, or the actual expertise of the knowledge-monopolising classes. (In the first wave of Taylorism, the savings made by firms through increased industrial efficiency were entirely swallowed up by the costs incurred by the new administrative classes.) This example can, I think, shed some further light on Harris’s truancy policy. The point wasn’t to improve school attendance by any means necessary – it was to impose state discipline, using any excuse available. It should be clear that the anti-racism in these purely managerial articulations of anti-racist politics is hollowed out and infinitely deployable. After all, Senator Harris seems willing to let black women die, if it means she gets to tell other people how it’s all their fault.