I’ve found myself thinking a lot about this paper by Herold et al. over the past few days. It describes the far-reaching implications of climate change for health and agriculture across different regions within Australia.
In particular I keep thinking about the implications of these two graphs:
The first graph shows heatwave frequency and the second one shows heatwave amplitude for different Australian cities for the recent past (blue), near-future (green) and far-future (red). Bottom and top of boxes indicate the 25th and 75th percentiles.
NB: For those not familiar with it, heatwave amplitude is a way of measuring and modelling the hottest day of the hottest heatwave within a year. This is different from the other common way of measuring heatwave intensity– magnitude–that looks at the average temperature across all heatwave days within a year. °C2 is a heatwave unit of measurement and isn’t the same thing as degrees Celsius.
There are many impacts and consequences associated with this model, but the frequency and scale of near- and far-future heatwaves alone should terrify us.
One of Mike Lynch’s bot creation is the Glossatory, which generates definitions of words based a recurrent neural network. Mike illustrates some of the more outlandish ones at @email@example.com
The Glossatory is based on a recurrent neural network, which is the type of artificial neural network that’s typically used for machine translation. Mike used a training set of 82,115 definitions that were taken from WordNet. The results are often amusing, frequently incomprehensible, but mostly they’re just odd.
On one hand these type of mHealth platforms allow data collection that’s meaningful to people, as well as services and systems. In developing country contexts this has often never been collected before except through specific population health surveys.
On the other hand, lots of health systems and major philanthropic funders are talking about using this data for linkage and data mining. They’re also talking about developing algorithms to guide care and risk stratification. Again, this is mostly being done with a focus on care in developing countries, but also developed ones too.
I’m ambivalent about this. The potential to actually capture data and to harmonise/minimise variation in care could be quite profound and far-reaching.
I just don’t trust algorithms though, largely because they reproduce our biases. I’m worried that investment might be in device-driven care rather than developing local workforces.
We identified (1) computerised decision support, (2) feedback and benchmarking on diabetes care quality, (3) culturally tailored programs (usually delivered by community health workers), and (4) enhanced practice nurse involvement in diabetes care as interventions that had a positive impact across several outcome measures. These outcomes included:
clinical outcomes (blood pressure, and body mass index, or weight)
biochemical outcomes (glycosylated haemoglobin [HbA1c], lipid profile, or renal function)
psychological outcomes (anxiety, depression, diabetes-related distress, perceived seriousness and vulnerability, self-efficacy, or self-care)
health-related quality of life measures from participant self-report.
The evidence about the use of decision aids (e.g. when to commence statins) was equivocal in terms of their impact on HbA1c, lipid profile and renal function. One high quality integrated care study on a virtual clinic, which brought together nurses, diabetologists and GPs, showed improvements in blood pressure but not renal function. No single intervention improved all the outcome measures that were considered.
Faruqi, N. et al. (2019) ‘Primary health care provider–focused interventions for improving outcomes for people with type 2 diabetes: a rapid review’, Public Health Research and Practice, 29(4), p. e29121903. doi: 10.17061/phrp29121903.
Mark Harris, Andrew Knight and I published a paper in the Medical Journal of Australia. It’s an update of a review of the management of chronic conditions in Australia, ten years on. What was surprising to me is that we haven’t made too much progress has in the obvious areas – integrated care, information sharing, multiple provider care plans. Many of the promising activities that have emerged over the past decade have been at the individual and practice level. For example, using peer navigators, and tailoring care to people’s literacy and activation levels.
The table below suggests some potential solutions, but these are only some of them.
Every year Sydney Local Health District holds EquityFest, an event that showcases work being done to address health equity within the District and to consider what else needs to be done. The theme of this year’s EquityFest was “investing in the future – leaving no one behind”. It was an impressive showcase of the diversity of the District, and the work that they’re undertaking to respond to their communities’ needs.
More than that though, it provided a clear statement that equity matters to them as an organisation, something that we almost never see in contemporary Australian social institutions.
There’s a need to invest meaningful time in understanding lived experience, not just giving lip-service to it through a narrow focus on things like patient experience measures.
Reflective and interrogative practices are required, personally and organisationally, to identify biases and to think about who’s excluded in, and through, our activities.
We need to think about “soft infrastructure” more meaningfully and systematically in planning (referring to services, networks, and community social assets), and sharing this through networks/backbone organisations.
There is a groundswell of interest in equity, but the challenge is developing a shared vision of how to achieve it, being accountable for change, and the related challenge of meaningful engagement.
There was a welcome from the Chief Executuve Dr Teresa Anderson and a warm welcome to country by Uncle Allen Madden.
[bctt tweet=”Lessons from #EquityFest: People living in public housing, people who are homeless, people with substance use disorders, sex workers, and people in custody all experience extremely high levels levels of social exclusion” username=”ben_hr”]
A/Prof Jane Lloyd from the Health Equity Research and Development Unit gave an overview of the importance of social inclusion, and the health harms of exclusion. She spoke about who’s excluded in Australia, based on data from the Brotherhood of St Laurence Social Exclusion Monitor. This highlighted that people living in public housing, people who are homeless, people with substance use disorders, sex workers, and people in custody all experience extremely high levels levels of social exclusion.
Peter Jack from Sydney Local Health District spoke about his experiences of working with people in Redfern and Waterloo about health concerns. In discussion with Jane he reflected on life, growing up as an Aboriginal person and the challenges of working with people who are socially excluded in Sydney and Adelaide. The lessons I took from from Peter were the importance of investing time to develop credibility and trust with excluded people, before we try to do anything else.
Zione Walker-Nthenda spoke about the challenges of inclusiion in justice and domestic violence. She emphasised that self-interrogation is important, both organisationally and institutionally, to identify biases and barriers to inclusion. Essentially, that we need to constantly check that we’re not kidding ourselves about how inclusive we’re being.
Liz Harris from HERDU presented a framework that emphasised the need to consider place as the dynamic interplay between people and space – meaning the norms, use, services, and culture that make up the bulk of community life. Her critique was that in planning in human services often focuses on “need”, constrained to understanding people in atomised ways, and rarely considers place assets and issues in meaningful ways. She further suggested that more work is required to develop coherent program logic for these activities.
[bctt tweet=”Lessons from #EquityFest: Planning in human services often focuses on “need”, constrained to understanding people in atomised ways, and rarely considers place assets and issues in meaningful ways” username=”ben_hr”]
Hal Pawson from the UNSW Faculty of the Built Environment spoke about the factors that drive locational disadvantage. The drivers he emphasised were:
pushing people without resources to the fringes, or less accessible parts of cities
the diminishing pool of low cost housing in acccessible areas
a political unwillingness to use planning powers to reverse these trends.
Pam Garrett, Director of Planning within SLHD, spoke about the scale and pace of development within the inner west, ranging from Green Square, through to Waterloo/Redfern, Parramatta Road, and the Bankstown line. This represents a challenge in general, but also presents specific challenges to the District to make sure that services are responsive and that health inequalities aren’t exacerbated.
Geoff Turnbull from Inner Sydney Voice and REDwatch emphasised the need to get back to genuine community development, and the need to work in genuine partnership with communities. He emphasised the need to create organisations that bring together interests to have a voice in changes that would affect their suburbs and communities.
I wrote a piece for Croakey on the first of the 2016 Boyer Lectures, which focus on the social determinants of health. The four Boyer Lectures air weekly from Saturday 3 September 2016 at 1pm on Radio National, or you can subscribe to the podcast.
“One thing leads to another.”
This was the secret of life imparted to Michael Marmot by a patient, and recounted as part of the opening 2016 Boyer Lecture Fair Australia – Social Justice and the Health Gap. It could also be the theme of this year’s lectures.
In his first lecture, Marmot set out a compelling argument for the links between the circumstances of one’s life and work and health outcomes. These circumstances are socially determined and not equally shared, as shown by his foundational research on the social gradient in health.
This is familiar territory for most Croakey readers, and the information wasn’t new. It was well-crafted however, and drew on evidence, comparisons and anecdotes. Marmot led the audience, inexorably, from the problem of health inequalities, to their causes, then to the changes that are required to address them.
One thing leads to another.
The points in the lecture that resonated most with me were:
Health status is a better indicator of wealth than GDP
Marmot cited the example of Costa Rica, who have invested heavily in education. He outlined the benefits this has brought to their health and national development, in ways that aren’t captured by GDP and conventional economic measures alone.
This reminder that health is a meaningful and under-recognised indicator of wealth and assets is timely given the publication of Clare Bambra’s excellent Health Divides: Where you live can kill this week. It’s destined to become a seminal work on the political geography of health, but it also puts forward several creative ways of presenting data on health inequalities.
My favourite idea is a European Health Championship, modelled on the European Football Championship. Health League
In a sporting nation like Australia, this sort of approach could get attention.
Imagine if we competed internationally to have the best health. Imagine if successfully reducing the gap in life expectancy between Indigenous Australians and non-Indigenous Australians became a source of national pride.
Marmot emphasised the role that empowerment, or lack of empowerment, plays in determining people’s health. He drew on the framework for empowerment developed for the WHO Commission on the Social Determinants of Health. This emphasises the need for material, psychosocial, and political empowerment if people are to take control of their health and have the same opportunity for good health.
Using data to drive change
Marmot gave the example of Coventry in the UK, who describe themselves as a “Marmot City”. They have established monitoring and reporting frameworks to drive activity on health inequalities. Accountability has enabled sustained action and a better understanding of the issues.
In Australia, many services have been working on the same issues. South Eastern Sydney Local Health District has incorporated a set of comprehensive indicators of health inequalities into its Equity Strategy. Data enables action to move beyond rhetoric.
A different audience
The value of the Boyer Lectures is not their ability to preach to the converted. It’s about exposing an engaged audience to new ways of thinking about Australia’s future, and in that task last night’s lecture succeeded.
Eva Cox’s 1995 Boyer Lectures on A Truly Civil Society were instrumental in sparking a wave of interest in social capital in the design and evaluation of social support in Australia during the mid ‘90s. Eva herself would say the tide has gone out on social capital, but its impact is still felt.
The research program I manage includes a stream on place-based interventions to address health inequities. As part of this we have collected neighbourhood-level data on trust and social capital in disadvantaged areas in South Western Sydney since 1998. This research can be directly traced back to the agenda set by the 1995 Boyer Lectures, and informs and guides our work to this day.
It’s my hope that this year’s Boyer Lectures may have a similar catalyst effect.
The remaining lectures will focus on the importance of early childhood, our work and living conditions, and the action required by researchers, civil society, and government.
There is a series of articles at The Conversation on the themes covered by this year’s Boyer Lectures, featuring contributions from Sharon Friel and Fran Baum.
Hopefully this will lead to a broader appreciation of the social gradient in health in Australia, and a shared understanding of the case for action.
One thing leads to another.
To coincide with the Boyer Lectures, Croakey has also released a collection of articles on the social determinants of health. Includes contributions from Fran Baum, El Gibbs, Liz Harris, Sharon Friel, Tim Senior, Marie McInerney, Marilyn Wise, Peter Sainsbury, Denis Raphael, Fiona Haigh and many others, including me. Recommended reading.
Last week I attended the final day of a Learning by Doing health impact assessment (HIA) training program. Learning by Doing is a structured six-day training program run by the Centre for Health Equity Training, Research and Evaluation (CHETRE), where I work. The training is broken up into distinct stages. People learn about HIA, go away and do the steps they’ve learned about, then come back to reflect and learn about the remaining steps.
Some of the stand-out lessons from the Learning by Doing sites were:
A lot of the value of HIA lies in highlighting and clarifying assumptions made during planning.
A planned engagement approach is useful as part of all HIAs.
Conceptual learning remains an important outcome of HIAs, in particular learning about health equity.
Involving consumers in HIAs has multiple practical benefits, such as identifying alternatives and providing an understanding of context and history.
Scoping remains as critical as always – being rigorous but not biting off more than you can chew (it also reminded me of this paper about scoping in EIA).
Not all barriers can be overcome; you need to be realistic about what you can achieve within the limits of an HIA.
More on Learning by Doing
For more on CHETRE’s work on health impact assessment go to HIA Connect. The Learning by Doing approach is described in greater detail in the chapter below:
Harris E, Harris-Roxas B, Harris P, Kemp L. “Learning by Doing”: Building Workforce Capacity to undertake HIA – An Australian case study, in O’Mullane M (ed) Integrating Health Impact Assessment into the Policy Process: Lessons and Experiences from around the World, Oxford University Press: Oxford, 2013, p 99-108. ISBN 9 7801 9963 9960 Google Books link
I’d be interested in any feedback. In particular I’d appreciate any responses to what I think remains unknown/future directions for HIA evaluation research:
How does HIA change perceptions at individual and organisational levels and how can we better account for this through research design?
How can we better evaluate learning that can occur through HIAs, at individual and organisational levels?
Social (Glasbergen, 1999)
How can we account for different forms of bias in evaluations of HIA?
Narrative fallacy (Harris-Roxas et al 2014)
How can we develop a more nuanced approach to scoping HIAs to consider the determinants of health inequities, as distinct from the determinants of health?
Source: Harris and Harris-Roxas 2010
“What is clear here is that impact assessment is beginning to be seen not just as a tool for informing and inﬂuencing decision-makers, but as a process which changes the views and attitudes of stakeholders who engage with the process such that their own attitudes and practices change outside of the immediate decision making context. That is, the inﬂuence of impact assessment processes may extend well beyond the narrow decision window in which they operate. There is also recognition that such learning operates on an institutional and social level as well as on an individual level.” Bond & Pope (2012:4)
Bond A, Pope J (2012) The State of the Art of Impact Assessment in 2012, Impact Assessment and Project Appraisal, 30(1): 1-4. Download PDF
Glasbergen P (1999) Learning to Manage the Environment in Democracy and the Environment: Problems and Prospects (Eds Lafferty W and Meadowcroft J), Edward Elgar: Cheltenham, 175-193.
Harris E, Harris-Roxas B (2010) Health in All Policies: A pathway for thinking about our broader societal goals, Public Health Bulletin South Australia, 7(2): 43-46.
Harris-Roxas B, Haigh F, Travaglia J, Kemp L (2014) Evaluating the impact of equity focused health impact assessment on health service planning: Three case studies, BMC Health Services Research, 14(371), doi:10.1186/1472-6963-14-371. www.biomedcentral.com/1472-6963/14/371
HIAs provide an opportunity to advance equity. Practitioners often struggle with how to effectively communicate about equity in HIA, and strategically communicate about this core value of HIA. An effective approach to communication is fundamental to ensuring that HIAs can impact policies and support change.
This guide intends to aid HIA practitioners in their efforts to communicate about equity as an essential step towards advancing equity.