In recent years, there have been a number of studies that have explored public attitudes towards the use of patient data and data sharing, which have shown that people generally know relatively little about how data is used, but are broadly supportive of the NHS accessing and using patient data.
However, whilst there is a large amount of research covering how people feel about patient data generally, there is less information available about specific use cases. Many public attitudes studies cover views on secondary uses of data (i.e. uses other than your own individual health investigations and treatment) either only cover specific things like attitudes towards sharing data with third parties or solely data for research, or ask about all secondary uses of data without distinguishing between research, planning or population health. The OneLondon public deliberation in 2020 covered the use of data for planning but highlighted a notable lack in public attitudes research on this and recommended further research into views on the use of data for planning.
We commissioned Qa Research, a research agency, to carry out this work. They captured the views of members of the public across the UK through a two-part research project: one nationally representative online survey of 1,000 people, followed by a series of four workshops guided by a semi-structured discussion guide and stimulus presentation.
Surveys are useful because the size of the sample is likely to generate insights that are representative of the general UK population, whereas workshops generate more in-depth insights into individuals’ views and add nuance. This is why we thought it would be helpful to do both. It is important to note that whilst we recruited a diverse sample for the workshops, the sample size for the workshops is too small to be statistically significant or representative. There are also some groups within the survey sample which were too small to make statistically significant observations.
Our research explored:
For ease of understanding, we will use the term “NHS” to refer to the publicly funded health system in each nation of the UK. During the research, “NHS” was substituted with “HSC” for participants in Northern Ireland.
Read more about the key findings from our research below. To find out more about each section including relevant survey results, click to expand.
When asked how much they knew about how the NHS uses patient data, 20% of survey respondents said they knew nothing at all, 34% said they knew a little, 33% said they knew some and 8% said they knew a lot.
Participants were more likely to identify individual care (also known as direct care) as something the NHS uses patient data for (86%), compared to research (71%), population health (68%) and planning (66%). Only 51% of participants correctly identified that the NHS uses patient data for all purposes.
In workshops, very few participants could say how the NHS uses their data, with examples primarily revolving around individual care and research. Workshop participants also found it difficult to define the terms ‘planning’ and ‘population health’, but once explained, they understood the concepts easily.
When asked how familiar they were with different examples of how the NHS uses data for planning and population health, survey respondents were most familiar with examples relating to population health management and risk stratification, and least familiar with examples relating to service evaluation and managing demand for services.
People from older groups, ABC1 social grades and a white ethnic background were more likely to correctly identify the different use cases (individual care, research, planning and population health). However, when asked about specific examples of data being used for planning and population health, those aged 18-34 and people from a minority ethnic background were more likely to say they knew at least a little about each example.
When asked about specific uses of patient data for planning and population health, survey respondents reported high levels of support for all examples, with each one receiving around 90% net support (‘strongly support’ or ‘somewhat support’) and around half of respondents strongly supporting each example. 78% of respondents said they would strongly or somewhat support their data being used for all planning and population health examples.
84% of survey respondents agreed with the statement ‘Generally speaking, I am supportive of the use of patient data in the NHS.’ This could suggest either that planning and population health uses of data are particularly popular, or that people are slightly more likely to lend their support when presented with specific examples of how the NHS uses patient data than when asked about the use of data in the abstract.
Workshops found unanimous support for the same planning and population health uses of data presented to survey respondents. However, a notable minority of participants raised concerns about these uses of data not translating into the expected outcomes in terms of improved NHS services.
When presented with a list of examples of how data can be used for planning and population health, respondents reported similar levels of support across all uses, but there were some differences.
“Analysing Accidents and Emergency (A&E) admission data to ensure the hospital has the right amount of staff and resources available to deliver care” and “Monitoring and evaluating outcomes to understand which services are performing well and which ones need additional support, ensuring patients receive effective care” both received 94% net support.
The example that garnered the lowest level of support was “Analysing data to identify cohorts of people with similar lifestyles or risk factors that may contribute to health conditions and targeting groups of people with different health interventions,” though it is important to note that this example still received 89% net support.
Workshop participants showed more enthusiasm towards the specific examples of how data can be used for planning and population health than the abstract concepts themselves, and examples that participants had first-hand experience of, or that were perceived to be close to them personally, were more likely to be chosen as an important use of patient data. This suggests people respond more positively to the use of data for planning and population health when given tangible examples that are relatable to their personal experiences and needs.
The majority of survey participants (54%) said they strongly or somewhat prefer the use of de-identified data for planning and population health purposes, whereas 29% said they had no preference. Only 13% of people said they strongly or somewhat preferred the use of identifiable data for planning and population health. People who reported lower levels of support for the use of data for these purposes were more likely to prefer the use of de-identified data.
The preference for de-identified data was echoed in workshops where it was felt this conveyed fewer risks and a more secure use of data. However, participants were generally comfortable with the use of identifiable data on the basis that the NHS would only use identifiable data when necessary and that safeguards were in place to prevent misuse of identifiable data.
The majority of survey respondents believed the NHS should have automatic access to patient data for planning and population health purposes, with 35% of people saying it should only have access to de-identified data and 30% saying it should have access to both de-identified and identifiable data. 19% said the NHS should not have automatic access to their data for planning and population health. In workshops, the majority of participants were in favour of an opt-out system where the NHS has access to data for planning and population health purposes automatically, but where people have an element of choice as this was considered a core principle.
Awareness of the National Data Opt-Out in England was low among both survey respondents and workshop participants. When presented with different data use cases, respondents were more likely to say that they had a choice in whether data was used for individual care (41%) than planning and population health (32% and 34%, respectively). This suggests low levels of awareness of the policy as patients in England can generally opt out of their data being used for the latter, but not for individual care. This was echoed in workshops where participants were generally unaware of the English Opt-Out policy.
Generally, the research did not find large differences between the views of people from different nations. However, there were a few differences. Please note we have only included differences that were statistically significant here.
When asked whether the benefits of using patient data for planning and population health outweigh the risk, respondents in England were more likely to answer 'no' (9%) compared to the sample as a whole (8%), and respondents in Scotland were significantly more likely to say 'yes' (88% compared to 78% overall).
Respondents in Scotland and Northern Ireland were also more likely to prefer the use of de-identified data to identifiable data for planning and population health, with 67% and 73% of respondents saying they 'strongly' or 'somewhat' preferred the use of de-identified data, respectively. This can be compared to 54% of the sample as a whole.
If you are interested in seeing the full data tables from the surveys in .xls format, please contact us.
UPD will be sharing the findings from this research widely with people working on the use of health data for planning and population health, including policymakers in the NHS and civil servants across the four nations of the UK, to influence the way patient data is used for these purposes. We will also be sharing it with healthcare organisations, members of the wider data community, and other policy stakeholders such as patient advocacy groups and think tanks.
We think it has particular relevance for:
If you have any questions about the report, please email us at hello@understandingpatientdata.org.uk.