Domain: Research
Forecasting assistive technology needs in aged and ageing populations
In our ageing world, assistive technology (AT) needs will increase. Yet there is little understanding about how and when access to AT will change as populations get older. Jamie Danemayer is a PhD student, co-supervised by UCL and the London School of Hygiene and Tropical Medicine, who is working to maximise sparse data in this field and build a model that will forecast future AT needs.
Data about AT is collected to varying degrees across the world, by organisations and governments through surveys and censuses. Yet there is no global consistency on the type of data collected, or how frequently it is collected. Some surveys collect information on a few types of AT, such as eyeglasses or wheelchairs. Others will simply ask if AT is required, but not the type of AT. Overall this makes much existing data on AT difficult to compare. Furthermore, much of the information that is collected about AT sits within local organisations or governments. It is typically not collated or reviewed on a larger scale, in a way that would help policymakers consider the bigger picture of AT needs.
The lack of longitudinal data (data that are collected regularly and consistently, for example in multiple waves of a survey) is also problematic in terms of planning the provision of AT. However, our knowledge about population trends and demography may help to fill in some gaps. The trajectory of population ageing is different across the world, but the general trend is that the proportion of older people in the population is increasing in most settings. Along with this, the need for AT will increase, as we know the prevalence of functional limitations tends to increase with older ages. But without a longitudinal perspective on AT need, it is virtually impossible for policymakers and governments to adequately plan for meeting the demand as or when it arises
Gathering AT data
In response to this, Jamie Danemayer is focusing on collecting the data that exists on AT, and is using it to create a model for predicting future AT needs.
“There is a huge issue with how lack of data looks like a lack of demand,” said Danemayer. “And it's very hard to achieve political prioritisation for something that you don't have.”
Danemayer’s background is in public health and demography. Through her work as a research assistant at the GDI Hub, she noticed how there was a good amount of survey data about AT, but little longitudinal understanding about AT access. By adapting a demographic tool called a life table with machine learning, she wants to use the available AT data and demographic information on ageing trends to build a model to forecast future AT needs. This will enable governments to understand what type of AT they will need to purchase and when a population’s AT needs are likely to balloon, as well as existing inequities in AT access that will widen as individuals age and their needs increase.
To start this process, Danemayer has sourced secondary data that already exists about AT. She honed in on population cohorts and multi-wave cross sectional surveys – such as censuses or UN health surveys that are routinely conducted and ask questions about AT.
Often, surveys had to be excluded because they ask about functioning limitations even when using AT, but do not separate out AT use as an individual question, making it impossible to discern the number of individuals using AT. Danemayer has also found that existing cohort studies often exclude people who have pre-existing conditions, including cognitive functional limitations, making them less representative of older adults who age with disabilities. Getting data that is representative of this population as well will be crucial for the future.
Next steps
Now that Danemayer has identified the data that are useable for this project, the next step will be to look for demographic characteristics that are strong predictors of having a met or unmet need for AT in cohort studies. This will be used in conjunction with data on general population demographics to extrapolate the unmet need for AT at the population-level.
Beyond that, Danemayer plans to test the predictors for AT needs that the data identify, and undertake a modelling study, through which she will test and validate the forecasts. Further work will assess how applicable the model is to situations in data-poor settings, or for use in countries with different patterns of ageing. In these cases, there will be country-specific factors to consider. For example, countries that are post-civil conflict may have a much higher need for specific types of AT, such as prosthetics.
As well as working on her PhD, Danemayer works on the data repository for the GDI Hub. The secondary data sources she is collecting for her PhD are also going into the repository. Eventually, the repository will be a go-to resource to understand which countries ask questions about AT, and it will help to advise policymakers which AT questions they should be incorporating into routine data collection. Danemayer is also encouraging more organisations to use the repository at the GDI Hub, as well as to use World Health Organisation (WHO) designed survey methodologies to measure AT access. This produces globally comparable data and supports future longitudinal analysis work.
“Many high income countries are already extremely aged,” said Danemayer. “It’s an urgent problem, but we still don’t have an understanding of the role AT access plays to support healthy ageing. There are issues outside of healthcare that governments are having to react to with older populations, such as pensions in the UK and the social security imbalance in the US. We're at a very unique point where we can take our evidence, including our experience of what doesn’t work to support this population, and hopefully make it useful to help other countries to avoid the same mistakes and instead be prepared before their populations age as significantly as ours. And with 2020 to 2030 being the UN Decade of Healthy Ageing, now is the time to mobilise political will and generate holistic evidence to support these efforts.”
Funded by: AT2030
