Health and social care - data strategy: consultation analysis

Part 2 – Empowering Those Delivering Health and Social Care Services

Part 2 of the consultation examined how people involved in delivering health and social care services can be empowered to use data. Questions included the need to develop data skills, and the technology, infrastructure and frameworks required to ensure high quality data can be collected, stored, accessed and used effectively to deliver services.

Data skills and training

6A. Considering skills and training opportunities for those delivering health and social care services, what are the top skills and training gaps relating to data in Scotland's health and social care sector?

At Q6A, respondents were presented with a list of six data skills and asked which were most in need of additional skills and training in the sector. Respondents ranked the skills from the most to least in need of additional training. An 'other' option was also available with respondents asked to elaborate on their answer.

The table below presents the outcome of the ranking exercise. In this instance, the data is presented among all respondents i.e. it includes those who did not answer the question.

Skill	Top 3	1st	2nd	3rd	4th	5th	6th	7th
Understanding of what data exists and where to find it	46%	25%	14%	7%	6%	6%	3%	1%
Knowledge of how to access data	38%	4%	15%	19%	11%	9%	3%	2%
Understanding of governance	30%	9%	10%	11%	10%	10%	8%	4%
Confidence in using data	30%	8%	9%	13%	12%	10%	9%	0%
Understanding / use of management information by managers	18%	6%	6%	6%	11%	14%	14%	4%
Data visualisation	16%	4%	7%	5%	8%	12%	19%	6%
Other	10%	6%	2%	2%	2%	1%	2%	22%
Not answered	-	38%	36%	37%	39%	39%	42%	60%

Respondents identified the understanding of what data exists and where to find it as the key skill and training gap. One quarter (25%) of all respondents selected this as the top skill gap, with almost half (46%) ranking it as one of the top three areas in need of additional training. While knowledge of how to access data was only selected by 4% as the top priority, over one third (38%) placed this in the top three.

Understanding of governance and confidence in using data were similarly ranked; just under one in ten selected each of these as the top skills gap, with three in ten including each area in their top three. The remaining skills of understanding and using management information and data visualisation were noted by some, but only featured in the top three of around one in six respondents.

Free text comments were invited from respondents who indicated there were 'other' skills and training gaps. The most common response focused on a need for training in digital literacy to help health and social care workers analyse data and use it to benefit their service and service users. This theme is related to the 'confidence in using data' option in the list of data skills presented in this question.

"After consultation with our members, the RCOT believes that there is a training gap regarding being able to understand and analyse data. Some members are unsure on what conclusions can be drawn from data and how this may help health and social care practices." – Royal College of Occupational Therapists

Some respondents provided comments connected with the 'understanding governance' option. The importance of ensuring that data is collected, processed and stored securely and in line with relevant legislation was highlighted in these responses. A few referred to issues around sharing data with other organisations and ensuring that members of the public can access data held about them.

Another theme was training in skills for collecting data in an accurate and standardised way. A few made comments about the importance of understanding equalities issues when collecting and analysing data. Other suggestions for training, each made by one respondent, included: systems thinking; process science; using data for service evaluation; data science; and the use of artificial intelligence and wearable devices in data collection.

6B. How do you believe skills and training gaps about data should be addressed?

Many respondents to Q6B noted that training in the categories mentioned in Q6A would be the best way to address the skills gap. In order from most respondents to least, these included: training in data literacy and analytics, specifically understanding the value of data collection and the benefits it would bring to health and social care work; governance; training for managers; and training in systems and platforms.

Resourcing

The need for further resources to address skills and training opportunities was mentioned by several respondents. Paid training time was mentioned by some, while others highlighted the need to invest in updated technology and more trained staff.

"Support and investment in skills, workers and infrastructure are needed to make this strategy a reality. It is imperative that staff are supported to feel comfortable handling data and using new practices. For example, one factor for improving data in Scotland is the standardisation of data across health boards that will allow for its more effective use. Health professionals must be engaged in the need for this standardisation and how it can best support patient care, whilst not creating excessive burden on health professionals." – British Heart Foundation Scotland

Standardisation

Several respondents mentioned a standard, national approach to training to ensure there was a uniform approach to data collection, coding, analysis, access and use. This is examined in more detail in the analysis of Q8A and Q8B.

"A combination of in house and consistent regional/national training programmes (e.g. Scottish Digital Academy) are essential to build capacity and capability and would support public organisations to utilise data effectively for drawing insight, trend analysis and modelling. – Renfrewshire HSCP

Include in curriculum

It was noted by some respondents that data science should be a requisite part of medical training and education courses. One health board noted that data standards will soon be a core competency of many positions and should therefore be reflected in the curriculum.

Accessible training

Continued professional development (CPD) was mentioned by some as needed to keep the current workforce up-to-date with technological advancement. Others argued that training and assessment should be part of job descriptions to ensure staff are undertaking mandatory training modules. A few mentioned using accreditation or accredited programmes, such as a master's degree, to encourage staff participation. Some mentioned tailored training for different positions. For example, one organisation recommended The Data Lab^[2] as a resource to facilitate programmes.

Some respondents highlighted the importance of creating training packages that encourage participation through ease of access. Others emphasised the need for training to be online, mobile-friendly, and free. Recommended educational materials also included leaflets, manuals, glossaries, dictionaries, and course catalogues. In contrast, a few suggested that courses be held face-to-face.

The response from Skills Development Scotland provided detail around the data skills training provision available in Scotland, but noted the demand for data skills across all sectors and that health and social care will be competing with the private sector for talent.

Less commonly mentioned themes

Other suggestions mentioned by small numbers of respondents included the following:

• Define career pathways for those with data training and experience to enter the health and care services.
• Create clear expectations and guidance regarding data standards.
• Assess the training needs for staff already in the sector.
• Engage in knowledge exchange with academia, private and third sector organisations, specifically regarding successful training programmes.
• Use innovative new technology, particularly developments in artificial intelligence.
• Adjust job descriptions to include necessary technological and data qualifications.
• Hire staff with protected characteristics to lead training sessions.
• Ensure that collection of protected characteristic data is central to new training and focus on how the lack of data can negatively impact treatment practices. For instance, the Coalition for Racial Equality and Rights (CRER) highlighted that "a lack of adequate data to monitor the health outcomes of BME groups impacts the ability of health services to provide for minority ethnic needs."

A few respondents suggested training tools that are already available, including Turas and LearnPro training software. Others recommended tools such as Microsoft Teams to create drop-in sessions. Scottish Care recommended using the Digital, Data and Technology (DdaT) Capability Framework to support the planning of any training programmes.

6C. What actions must be taken as a priority to ensure that the public have access to health and social care data that they can understand and use?

Create practical and simple pathways

The most common theme in the 115 responses to Q6C was that health and social care data, whether that was personal, related to NHS systems or general medical information, was easy to find using intuitive interfaces. For example, respondents suggested platforms that provided data visualisation tools. Others noted the importance of simple, jargon-free language and a few mentioned digital banking apps as an example of how to share complex data with the public in an easily accessible and comprehensible manner.

Public consultation

The second most prevalent theme, mostly raised by organisations, focused on the importance of getting input from service users about their wants and needs regarding data access. Some recommended a patient-centred approach including user research that attempts to understand what types of information users want and how they plan on using it, before designing a platform. A few recommended using a co-design approach.

"A pilot programme could identify what support and education people need to understand their healthcare record and what would be required to support this." – ENRICH Scotland

Other respondents focused on data management on potential platforms, noting that data should be co-produced or co-managed between service providers and service users, allowing the public to add and edit their own information.

"We need to move beyond the 'access' paradigm into a 'co-manage' one. This question supposes the main value is in reading and reuse of data, when in many scenarios citizens or their trusted people need to be able to write into the record as well. Some of the Digital Health & Care Innovation Centre's most successful recent service transformations are based on citizens writing directly into clinical systems / records." – The Digital Health & Care Innovation Centre

Clear communication and Guidance

Several respondents mentioned the need for clear communication and guidance with the public throughout the process of researching, creating and launching any new data access interfaces. This included making presentations – covering the planned tools, safety and security measures and uses of data – available in multiple formats, such as digital communication and print pamphlets, and through public health channels. In line with responses at Q3C, some suggested that raising public awareness about the diverse uses of data and the resulting benefits would be essential to establish support for the project.

Standardisation and Interoperability

Confusion about terminology and measurements, inconsistent data capture and repeating medical histories were highlighted by several individuals and organisations as barriers to accessible and comprehensible health and social care data. Respondents noted support for interoperable systems and national standards which they felt would simplify the system for service users and improve care, specifically for instances of geographical relocation.

"Consistently capturing, managing and sharing the same sets of data regardless of geographical location. This will help with portability across locations where service users/patients move between different HSCPs/Health Boards and also aids us to compare services across locations." – Renfrewshire HSCP

Digital exclusion and data literacy

Concerns about digital exclusion were mentioned by some respondents, who stressed that those without digital access to their data should not be further disadvantaged as a result. They recommended providing access to any necessary equipment and to training and support for those who need help using technology. While digital exclusion was considered important, a few noted access to data will only be effective if service users also have a level of health literacy. They argued individuals should be helped to understand their health and social care data as well as access it.

"Enabling individuals to have "control" of their data must ensure that inequalities in control are not inadvertently created. It will be important to ensure that there are other access methods available to citizens who are not fully digitally enabled or not digitally enabled at all. There remain fundamental issues with regard to internet access and the capacity to speedily access electronic records particularly in some rural areas and this must be addressed for benefits to be gained across all of Scotland." – NHS Scotland Board Chief Executives' Group

Investment in infrastructure

Some respondents mentioned investment in infrastructure as the first step in improving access for service users. This included investment in hardware that allowed for standard and easy data collection as well as tools to allow people without necessary devices to access their data. Other respondents mentioned investment in platforms and interfaces as necessary to provide clear access pathways for service users. Infrastructure is also explored in more detail under the analysis of Q13A

Concerns

Respondents raised a few concerns that did directly answer the question but highlighted some issues they saw as important to service user accessibility:

• Clear data protection and storage procedures were mentioned by some as necessary before granting public access. This included concerns about transparent communication, data ownership, and ensuring anonymity of service users.
• Respondents noted that data had to be a certain quality before it warranted sharing with the public.
• Staffing resources i.e., hiring trained staff or upskilling current staff was considered necessary to create accessible systems and programmes.
• A very small number of respondents said they did not believe any change to user accessibility regarding data was necessary.

Technology and infrastructure

7A. Thinking about improving the quality of data that is used by health and social care services, what three things are needed to improve quality and accessibility?

Interoperability

The most common theme in response to Q7A was the importance of interoperability, shared systems, and linked data to improving data quality and accessibility. Respondents highlighted the benefits of an interoperable system, such as a 'tell-it-once' or a 'once for Scotland' approach to data collection which allows data shared by users to be included in a Scottish-wide system, reducing the repetition of work by both service users and providers. Others highlighted how data silos or safe havens can be a barrier to creating a unified system for health and social care data.

"Interoperability – a great deal of research and statistical work involves lengthy and imperfect data linkage work to connect health records of individuals across pathways of care and measure outcomes. Therefore silos in health data systems not only affect joined up patient care but can also hamper research and innovation." – Research Data Scotland

Standardisation

The second most prevalent theme focused on standardising data collection. Respondents recommended reducing duplication, using standard data collection methods, terminology, and coding practices, and requested standardising the types of data gathered and the forms used for data collection. A few noted the importance of agencies and bodies using the same programmes, software or platforms so they work together seamlessly. A few suggested using open-source software to connect different systems to one user platform.

"If the NDP [National Data Platform] was allowed access to structured records across the NHS today, they would find inconsistent, variably coded and duplicated data for many of the high impact data items that systems and citizens want access to… the structured data needed by systems and citizens will only be routinely collected at any kind of scale and efficiency through root and branch redesign of clinical workflows at all levels and tiers." – The Digital Health & Care Innovation Centre

Training and support

Training, targeted recruiting and technical support was mentioned by many. This included training staff in data literacy and data security including GDPR, including an improved understanding of the importance of collecting accurate and timely data. Others recommended hiring and training specific teams, for example, IT teams to create new platforms, frameworks and programmes, or data management teams with medical knowledge to maintain health and social care records.

User-friendly platforms

The necessity of an easy-to-use platform was raised by many respondents, often giving the example of a 'digital front door' where health and social care data from different agencies and bodies is collated into one accessible place. There were suggestions for an interface where service users could ask questions, clean and clarify their personal data, and access information about their conditions or healthcare plans. Several also highlighted the importance of a user-friendly platform for service providers, noting that gathering quality data would only be possible if practitioners found it easy to use the designated platforms. Respondents recommended including mechanisms to capture digital copies of paper notes and investing in hardware that could immediately capture data electronically.

"Practitioners need to ensure accurate entry of data and completion of all fields etc. They need to see systems as an enabler rather than something that gets in their way." – Aberdeen HSCP

"It is therefore crucial that any new data strategy does not add yet more administrative burdens and pressures on social workers. The system needs to be intuitive and simple to use. Social workers should be able to see the positive impact that the system is having on their work. It should be time saving for them and make their work easier to do. For example, have easy-to-use templates that can be used to record and analyse data across departments and professions." – Scottish Association of Social Work

Infrastructure improvements

Many respondents noted that infrastructure improvements, both hardware and software, would be needed to improve the quality of data collection and accessibility. Suggestions for improvements included methods for efficient data entry including hardware for easy data capturing, or tools, forms or platforms that enable the easy sharing of data. Some called for a mechanism that enables real-time data sharing. A few called for tools that enabled clear data visualisation, both for service providers and service users.

Guidance and standards

Stressing that data quality cannot be improved through technology alone, several suggested clear guidance and expectations for data collection for staff. They suggested that as well as governance and legal frameworks, guidance should provide clarity about the data being collected and shared and who will have access. One respondent felt clear guidance would allow organisations to feel more confident about correctly sharing user information. Some respondents highlighted a need to audit and validate data regularly.

"There is a need for ongoing, centralised, routine maintenance and enhancement of quality conducted on data sets. Currently, there is a tendency to leave data accumulating and only assess its quality as and when it is required. A cultural shift is required to move away from spending long periods of time cleaning data towards proactively monitoring datasets before they become out of control. This would require investment but would be worthwhile by improving the overall quality of data" – Royal Society of Edinburgh

Less commonly mentioned themes

A few respondents each noted the need for the following: further research, including pilot programmes and user research; inclusion of private and third sector service providers in data collecting and sharing; a person-centred approach to data collection; and standards that mandate the collection of equalities data which is useful in analysing differential impacts. One felt it was important to distinguish between health data and social care data so that social care models are not overshadowed by medical models. Another mentioned the need to consider accessibility for people who may be less technologically literate.

A few respondents provided examples or recommendations for improved programmes. These included the use of natural language processing, using new data sets like Patient Reported Outcome Measures or patient activity data, and compliance with WCAG 2.1 AA+ accessibility standards as a minimum. The Digital Health & Care Innovation Centre recommended starting with small-scale projects, such as centralised vaccine records data, which could build momentum towards larger-scale change. Mydex CIC noted that the Digital Identity Scotland project is already working toward this goal.

Concerns were raised by small numbers. ABPI noted that existing data would need to be decluttered and cleaned to fix incorrect and incomplete information, and that the timescales needed to implement change may impact Scotland's competitiveness compared to other countries. Another noted legal barriers to data sharing and one highlighted that staff often do not have the time to collect and collate high quality data.

7B. If you are responding on behalf of an organisation, what role do you believe your organisation has to play in improving accessibility and quality of health and social care data?

Implementing standards

Of the 78 respondents who answered Q7B, the most common theme was the role organisations could take in complying with standards themselves, or assisting members, providers or service users in complying with new standards. Others highlighted the success of their own standards or codes, specifically regarding governance, confidentiality, and data protection, which highlighted their experience in supporting and implementing standards. For example, the Office for Statistics Regulation mentioned their Code of Practice for Statistics, the sharing of which could help those producing and analysing data.

Management of data

Several respondents described their data management experience, outlining the type and volume of data they store, how they collect, store and access that data, and their governance strategies. While not all these respondents spoke explicitly about what they would do with this data under a new data strategy, a few noted their interest in data sharing to benefit their members, patients or service users, and others mentioned an openness to sharing their data. This data ranged from health and social care details gathered from patients to public health sources, as well as the knowledge and experience of their service providers. As an example, Research Data Scotland Ltd mentioned they could assist by bringing together multiple datasets and facilitate the creation of new data catalogues with metadata which describes or gives information about the data itself.

"We need to facilitate the sharing of data but also sharing of our knowledge and expertise of methodologies for collecting, coding, analysing, and interpreting data internally and with colleagues in other local and national organisations." – NHS Fife

Strategy implementation

While some generally expressed support for partnership working, some others specifically suggested how their organisation could help implement the new strategy. This included coordinating and communicating with stakeholders and colleagues across sectors, undertaking user research, working with individuals and professionals to co-design strategies, integrating impact assessments, advocating for members, leveraging clinical leadership to improve data quality, policy campaigning, and advising on ethics.

"Since the introduction of the Public Bodies Joint Working (Scotland) Act 2014 there has been considerable discussion across Health and Social Care Partnerships (HSCPs) on sharing information but unfortunately to date there has been limited progress. COSLA's hope is that the development of a data strategy for health and social care will enable better collaborative working across health and social care services, in turn improving service delivery and satisfaction" – COSLA

Innovation

Developing technology, systems or platform architecture was mentioned by some respondents as a way their organisation could contribute to improved accessibility and quality. This included a few private organisations providing technological services, as well as public bodies or third sector organisations who have already adopted successful platforms and frameworks. Scottish Care mentioned that they had experience with practice-based innovation that improved data accessibility and quality.

Training

Some respondents suggested their role would be in staff training and promoting digital and data literacy with their workforce or members. A few mentioned the need for a commitment to resources, technological tools and time to facilitate training in accurate data collection and coding. A small number mentioned the importance of including data literacy and competency in curricula or national skills development programmes.

Review of current systems

Auditing and analysis of data gaps was highlighted by some respondents as the way in which their organisation could improve data quality, whether that was lobbying for research or conducting the research themselves.

"Our audit programme for 2022/23 includes a roundtable initiative on data gaps. This will engage leaders from across the public sector in what needs to be done in getting the basics rights with data so that its value can be realised. Data is an area that we will continue to monitor and consider as we plan our audit work programme." – Audit Scotland

Less commonly mentioned themes

Other roles mentioned by small numbers included: promoting the utility of data, communicating how data is collected, stored and used to service users, and data analysis. A few respondents highlighted concerns about existing accessibility barriers and suggested that national guidance was needed to ensure all partners are working to the same principles and creating consistent and cohesive data sets.

7C. What data, that is generated outside of the health and social care sector, do you think could be made available to health and social care professionals to improve health and social care outcomes in Scotland?

Types of data

Among the 109 respondents who answered Q7C, there was support for using a range of externally sourced data sets. Respondents noted the value this data could provide in building a fuller profile of service users and improving patient outcomes.

"In Scotland, health is defined and influenced by more than clinical data; developing a health data ecosystem that can receive, link and allow health boards and researchers to access a diverse range of data sets beyond the health sphere would add richness and extend opportunities for quality improvement and research." – British Heart Foundation Scotland

Many respondents highlighted the value of sharing housing data and information to improve health and social care outcomes. Housing data – defined broadly as relating to housing tenure, quality of housing, and care homes, but also including health and social care data related to care in the home such as telecare and ambient assisted living^[3] - was mentioned as a key way to track poverty and inequality by some respondents.

Several respondents each mentioned the following types of data, listed from most to least frequently mentioned. These included both data about an individual, and about their wider circumstances, environment or community:

• Information from apps or wearable technology, tracking data such as exercise, regular heartrate and sleeping patterns.
• Other biopsychosocial indicators of heath and health behaviours, such as educational attainment, socio-economic level, and HMRC information.
• Interactions with other services, including social services, counselling and therapy, private health services, dental, pharmacy, optometry, third sector organisations, community services and social prescribing, and emergency services.
• Demographic data, including employment status.
• Population indicators such as geographic information including poverty and deprivation levels and environmental information, including adverse climate, water quality and pollution levels.

Some respondents each mentioned the following types of data:

• Public health and health research^[4].
• Equalities and protected characteristics.
• Transport, travel, and tourism.

A small number of respondents each mentioned the following data categories:

• Immigration status.
• Search engine data.

A few respondents suggested that certain data be made available to service providers and users for reference, such as data from A Local Information System for Scotland (ALISS)^[5].

Concerns

While the question concerned the use of data from outside the sector, some respondents opted to raise concerns about privacy and the excessive sharing of health and social care data outside the sector. The Royal College of Physicians and Surgeons of Glasgow and Turning Point Scotland cited concerns about sharing some medical information with organisations or departments without a health remit, who could misuse or misinterpret the information. However, Social Work Scotland noted there are existing protocols for sharing information with service providers outside health and social care. Others approved of wider sharing in theory but highlighted the importance of a clear plan covering what data could be shared and the safeguards in place to protect individuals. A few called for no additional sharing of data beyond what is already shared, stating a fear that an all-seeing state may misuse the data.

A small number raised concerns about the practicality of data collection and quality control. Community Pharmacy Scotland noted that there are gaps in accessibility within health and social care that need to be addressed, for instance local pharmacies are not able to access information about prescriptions histories, such as those filled in hospitals, that could help in ensuring patient safety.

"We need to recognise that 'social care data' will be held not just by 32 Local Authorities, but by several thousand social care service providers across the third, private and public sector. The strategy should recognise that the level of complexity is an order of magnitude greater in the Social Care sector than Health. Quality, consistency and access are all major issues created by the complexity of the landscape and the strategy needs to articulate how these fundamental issues will be addressed." – NHS Education for Scotland (NES)

Data standards and interoperability

8A. We have heard that a more consistent approach to data standards will help improve insight and outcomes for individuals. To what extent do you agree with the proposal that Scottish Government should mandate standards for gathering, storing, and accessing data at a national level?

8B. What data standards should we introduce?

Support for standards

There was widespread support for mandating standards among those who answered Q8A – 88% agreed, 2% disagreed and 9% were unsure. Though high among both, agreement was higher among organisations (94% agreed, 6% unsure) than individuals (83% agreed, 5% disagreed, 12% unsure). While organisations all recorded high agreement with mandating standards, there was some variation. Nine out of ten (91%) public bodies agreed, but this ranged from 100% among HSCP to 90% among other health public bodies and 80% of other public bodies. The third sector (93% agreed), representative / membership bodies (88%) and academia (83%) were also in support.

Many of the 118 respondents who answered Q8B either reiterated their support for mandating standards or expressed their support for the standards outlined in the consultation paper. This included general comments in support of consistent data collection and storage, which would provide a foundation for interoperability and ensure high quality data capture. Respondents supported standards for coding, governance and data protection, and for access and transparency about data usage.

"Common data standards are important in maximising the value of data and joining up of services. They can help foster collaborative working across services and sectors." – Audit Scotland

"We agree for mandated national standards, but this is not the same as mandating the actual data sharing activity itself. These standards should allow for a range of privacy protecting methods to be used, and encode citizen ownership, control and choice at their heart." – The Digital Health & Care Innovation Centre

Existing standards, strategies, and frameworks

Existing standards, codes and guidance were suggested by several respondents. Some recommended general frameworks, such as the Health and Social Care Alliance's Core Principles from the 'My World, My Health'^[6] project, and others mentioned general standards set out by the UK Records Standards Body (PRSB), the Office for National Statistics (ONS) and Scotland's Artificial Intelligence Strategy. Others provided very specific examples, including:

• Standards such as Fast Healthcare Interoperability Resources (FHIR), OpenEHR, and Observational Medical Outcomes Partnership (OMOP) Common Data Model.
• Coding systems, such as Systemised Nomenclature of Medical Clinical Terms (SNOMED), the Dictionary of Medicines and Devices (dm+d), Classification of Everyday Living (COEL), and International Classification of Disease, particularly the ICD10 and ICD11 codes.
• Other tools, such as AHP Operational Measures Project, the Caldicott principles, NHS research governance frameworks, European Data Format, use of HDR UK's metadata specification and the Data Utility Framework, and FAIR Principles.
• A few noted the importance of aligning new standards with legislation in existence, such as GDPR, Data Protection Act, and Public Records Act.

"Standards should be agreed on what data should be collected, agreeing a minimum core data that everyone has committed to, to ensure a core quality across Scotland. For example, across health boards male/female is currently recorded differently. There needs to be consistency across boards. Core fields that are mandatory regardless of the condition and then condition specific fields should be agreed. These should not be set by the strategy but should be agreed and endorsed by all health boards. Standardised data fields reduce the admin burden and allow for greater interoperability. Fragmented and disconnected data is a barrier to research as well as clinical practice, it would be helpful if data standards were also included for modernising data architecture for interoperability that could incorporate research readiness e.g., structured data store using common data model. Simplifying existing processes and structures should be front of mind (rather than creating new ones) at all times." – ABPI

Concerns

Several respondents did not suggest specific standards but raised concerns about the implementation of standards on a national level. These included a few each mentioning:

• Resourcing challenges, specifically training and staffing capabilities. A few argued that forcing data collection tasks onto an already burdened workforce could negatively impact care and data quality.

• Fears about use of data and access. Cancer Medicines Outcomes Programme (CMOP) specifically mentioned limiting access to for-profit firms unless appropriate scrutiny is in place.
• Innovation and the possibility that national standards could exclude suppliers and third-party tools that meet global standards and integrate with legacy systems.
• Failures of past attempts, for example eCARE^[7] and the National Clinical Datasets Development Programme.
• Concerns about adopting high-level or ideological standards rather than standards that focus on practical application.
• Cost of national data sharing standards, which may outpace the potential benefits.
• Responsiveness to local practice as national standards may remove autonomy from health boards.

Other recommendations

A few respondents each mentioned suggested that standards should:

• Ensure a user-friendly output.
• Consider equalities and protected characteristics for data collection purposes.
• Create a human-rights or person-centred approach to data collection, especially regarding the types of data important to service users.
• Ensure the data standards are auditable and are being adhered to through enforcement or punitive policies.

While not directly related to types of data standards, a small number provided recommendations for implementation, such as:

• Improve infrastructure to ensure that all necessary hardware and software is available to all service providers participating in data collection and management.
• Use a meta-layer of infrastructure to solve interoperability problems.
• A few respondents including Aberdeen HSCP, Coalition of Care and Support Providers in Scotland saw the creation of the National Care Service as an opportunity to embed digital in social care, improve data collection quality and address existing information governance and data sharing barriers.

Management information

9A. When considering the sharing of data across Scotland's health and social care system, do you agree with the idea that greater sharing of an individual's health and social care data between the organisations in the health and social care sector will lead to better quality services?

Three quarters (76%) of those answering Q9A agreed that greater sharing of data in the sector will lead to better quality services; 7% (all individuals) disagreed, and 17% were unsure. Organisations were more likely to agree than individuals (85% compared to 67% respectively). Agreement varied by sector. Nine in ten (91%) of all public bodies agreed, including 89% of HSCP and other health public bodies. Agreement fell from 83% of academic respondents to 71% of representative / membership bodies and 64% of those from the third sector.

9B. If you are a clinician – how could we improve patient safety through better sharing of data and information?

The 80 respondents who answered question 9B interpreted it in two ways. Many described positive outcomes which could result from better sharing of data, building on their response to Q9A. Several others, however, provided technical solutions and implementation strategies that they deemed necessary to improve safety through data sharing. It should also be noted that a few respondents highlighted that the question wording referred only to health setting rather than care, suggesting 'practitioner' should have been used instead of 'clinician' and 'person' instead of 'patient'.

Improved safety and services

In line with Q9A, many respondents described ways in which sharing data and information would improve service user safety and lead to better quality services.

"Ensuring that primary, secondary, and social care information can be seen across the different agencies should in itself (assuming the system actually works, and can be easily viewed) reduce patient harm, both directly and indirectly." – NHS Tayside Directorate of Public Health

Several respondents believed data sharing would enable health and care practitioners to meet the needs of service users more fully. These needs included not repeating histories, timely referrals to appropriate services, and earlier intervention for those nearing crisis points. One organisation noted the value of shared data in alerting primary care providers of potential problems, for example monitoring collection of regular prescriptions. Others mentioned the need for medical and social care records to be available across health board boundaries for complete care to be provided to those who may be more mobile, such as university students. The importance of sharing information to smooth the transition of young people from children to adult services was also noted.

A few suggested shared data would provide enhanced analysis of patient experience and treatment successes, such as Patient Reported Outcome Measurements, and could allow researchers to spot trends and target prevention programmes at a population level.

The challenge of primary care providers getting data from social care and vice versa was mentioned by a few, as was the slow process of sharing data between primary and secondary care providers. Other care providers, such as the ambulance service, do not have access to full patient records that would allow them to help service users at home.

Implementation recommendations

Several respondents suggested changes to data processes that are needed for patient safety to be improved. A few respondents suggested that efficient collection of data was needed, including rapidly updating systems with complete information.

"Timing and completeness are key to improving clinical safety. Having immediate access to the most up to date data on a service user is an imperative to ensuring frontline staff can make the best decisions with confidence. And as an extension they need all the available information relevant to their role and the service they are providing. All relevant information should be accessible at the point and time of demand." – NHS Education for Scotland (NES)

A few respondents noted this could be accomplished through aggregated systems, such as a single patient platform or a single electronic health and care record. Community Pharmacy Scotland mentioned it should include secondary care provider records, such as pharmacy records and prescription histories. However, a few others highlighted the importance of data sharing guidance, security and governance, noting that linking data could cause harm if data is shared without consent. Respondents mentioned different guidance currently in use, such as the Caldicott Principles or Role Based Access Controls (RBAC), that could be integrated into larger security protocols. It was noted by a few respondents that the data would need to be of better quality and regularly updated or it would not be useful on a shared platform.

Concerns about misinterpretation and misuse

Some of the respondents who disagreed with or were unsure of Q9A were concerned about the misuse, misunderstanding or misinterpretation of data, whether by patient or other medical and care professionals.

10A. Thinking about the actions needed to improve the quality of management information and internal reporting data across health and social care, what are the priority pieces of management information needed (that are not currently available) to provide better health and social care services?

Operational insight data

Of the 75 respondents to Q10A, the most common theme was a recommendation to collect operational data. Suggestions included Demand, Capacity, Activity and Queue (DCAQ) information, dynamic risk assessments, workforce size data and future service plans. A few mentioned an interest in collating this across health boards for comparison.

A few respondents also mentioned financial data including cost effectiveness to support commissioning decisions. NHS Education for Scotland (NES) also suggested gathering data on the channels of associated funding and resources.

"Relevant management information would include staffing levels, service need, patient and client numbers, individual measures of patient/client dependency and details of any additional services involved in delivering care. Many of the pieces of management information needed to improve health and social care services are available but can be limited to a particular service or system and may not be available in a timely manner to allow services to respond to current need. Aligning systems across organisations and consistent coding of information would support use of data across sectors to inform service provision." – The Innovative Healthcare Delivery Programme (IHDP)

Service user outcomes

Service user outcomes, either reported by the user or the measured by the provider, were mentioned by several as an important piece of management information. These included monitoring effectiveness, service user experiences and consistency of care. This was not seen as limited to primary care, but included outcomes from secondary, social, community and intermediate care.

"The single health care record for the patient should also be single record of all the citizen's interactions with health and social care – that is a record of the care provided by whom, what, where and when. Ensuring this is recorded consistently, accurately and in real time (or as close to real time as is practical) will substantially improve the quality of management information and internal reporting data. If we ensure we get the clinical coding correct upfront and the input of it is of a consistently high standard, then the management information should be able to flow easily from this." – Royal Pharmaceutical Society

Impact on inequalities

A few respondents mentioned the importance of gathering data on protected characteristics, particularly how they intersect with poverty data, to track the impact of social inequalities. This is examined more under the analysis of Q3A and Q3B.

Less frequently mentioned types of data

Other improvements to management information mentioned by small numbers included the needed to define and track Key Performance Indicators, and to include metadata which tracks who created the data and who has shared or seen the data.

Implementation

While many mentioned the management information above, several respondents used this question to reiterate best practices for implementation. This included noting the importance of integrated systems for sharing data between service providers, and a few who mentioned data governance and privacy, and efficient systems that improve visualisation and interpretation of data while reducing duplication.

10B. What is needed to develop an end-to-end system for providing business intelligence for health and social care organisations in Scotland?

Removing barriers to data sharing

Of the 81 responses to Q10B, many respondents highlighted that removing barriers to data sharing was necessary to develop an end-to-end system. Respondents suggested the system should be a national, interoperable, central hub with a single access point or a digital front door. A few respondents suggested it needs to integrate old legacy systems as well as use interfaces between applications. The importance of using a single identifier for service users, such as the CHI number, to enable information to be linked across organisations was raised by a small number. A few others also noted that anxiety around information governance remains a barrier to sharing data between services.

"A key barrier to this is that data and systems to support planning and performance are not joined up and organisations are not always willing to share data. For an end-to-end system of business intelligence to exist there is a need to: address those parts of the system where there is a lack of data and intelligence – primary and community care and social care; make improvements to data sharing; use common data standards across the system; and address any legacy systems that exist." – Audit Scotland

"Data should be freely shared within the NHS (including GP-controlled data) and Local Authority care services to allow them to be used to improve the delivery of integrated care. Concerns around information governance can prevent effective sharing of information between different parts of NHS and HSCPs, and it would be helpful to address this." – Healthcare Improvement Scotland.

Trained staff

Several respondents mentioned trained staff were needed to develop a new system, i.e., upskilling current workers or hiring those with expertise in delivering end-to-end systems, health and social care data, hardware and software infrastructures and governance and data ethics. Others recommended creating specific employment pathways for employees, such as clinical practitioner informatics roles to work alongside data analysts, or pathways for clinicians with an interest in data collection or IT.

"There is a need and opportunity to invest in digital maturity for the health and social care sector. This should build skills at every level, establish infrastructure that enables real time analysis, and be informed by an understanding about how data reflects people's lived experience." – East Ayrshire HSCP

Consistency and standardisation

Standardising collection, coding and storage procedures was raised by several respondents as an issue that must be addressed before an end-to-end business intelligence system can be considered. Respondents noted that disparate workflows and systems across Scotland create different types of data that may not currently be compatible into one system.

Comments about guidance on governance were made by some. A few mentioned that understanding what guidance and standards is needed requires an audit and analysis of the data gaps in the current system to provide more detailed recommendations and a thorough understanding of each sectors' needs. This is examined more under Q13B.

Collaborations

Some respondents suggested collaboration to assist in the creation of an end-to-end system. These included collaborations with both larger-scale consulting experts as well as organisations with local business intelligence who could provide strategic direction. Others noted the importance of seeking input from key stakeholders in health and social care. A few, including Audit Scotland, recommended research among service users and service providers to understand the most relevant types or fields of data. Two others stressed the importance of collecting and analysing Patient Reported Outcome Measures (PROMs).

Infrastructure improvements

Investment in updated tools, both systems and hardware, were mentioned as necessary by some respondents. When considering the new technological needs, a few respondents noted the importance of scalability to allow for growth, and of flexibility that allows local systems to be easily migrated and integrated. Other recommendations included using artificial intelligence, agile working overcome challenges in integrating disparate systems, and utilising public computing.

"This ambition requires transformative solutions, likely only to be realised by fully leveraging the power of digital technology, public computing, and the extensive growing data infrastructure underpinning them." – Microsoft

The Coalition of Care and Support Providers in Scotland called for the strategy to recognise the role of commissioning and procurement in shaping the system. They argued that providers should be supported and enabled to deliver the strategy, partly by ensuring social care contracts include funding for digital to deliver safe, reliable infrastructure, future development costs and staff training.

Population health data

11A. Thinking about improving the quality and ability to reuse data sets across health and social care setting and for innovation & research, what key data sets and data points do you think should be routinely reused across health and social care to reduce duplication of effort and stop people having to re-tell their story multiple times?

Medical or clinical records

Many of the 99 responses to Q11A suggested data from medical or clinical records which could be routinely reused. These included: full Emergency Medical Records, key diagnoses, allergies, vaccinations, medication lists, previous admissions, outpatient care records, and imaging records. One respondent advocated open access between NHS GP records and private healthcare providers.

"We believe that information on diagnosis, treatment (what, where and when and by whom) and outcome is a key data set for each citizen that can be reused not only to ensure patients only have to tell their story once but also for innovation and research with appropriate data control and consideration of ethics." – Royal Pharmaceutical Society

"The GP record needs to be the anchor dataset that is then linked to a variety of other datasets e.g. SMR01^[8], HEPMA and PIS. There is considerable movement of people across UK borders so the ability to share comparable data across UK nations would be advantageous." – University of Edinburgh

As described throughout Part 1 of this report, not all respondents believed all medical data should be shared. A few specifically noted there should be variable access to medical records, only making them available to service providers who need to see them.

Personal data

Personal demographic data was mentioned by several respondents. This included date of birth, address, employment status, marital status, dependents, and how service users prefer to be addressed. Often respondents considered this to be the bare minimum of the types of data that should be available for use across health and social care.

Treatment choices

Some respondents noted the importance of treatment choices, such as the Recommended Summary Plan for Emergency Care and Treatment or the Anticipatory Care Plans. They also highlighted the importance of including next of kin details or information about any guardianship or Power of Attorney in place; this is covered in more detail at Q2B.

Less frequently mentioned data sets or records

A few respondents each mentioned the following types of data: Social care notes and pathways; functional abilities, such as communication, walking, cooking and personal care needs; protected characteristic data as defined by the Equality Act; Patient Reported Outcome Measures (PROMs); and self-generated personal data, such as data gathered from wearable technology.

Other themes

Some did not explicitly answer the question, but instead noted specific benefits of reusing data and reducing duplication. These included assisting service users whose clinical or personal histories evoke trauma, or those who move and change primary care providers. Others noted the benefit to service providers, especially those who work across health and social care, such as occupational therapists or pharmacists.

Some respondents highlighted concerns. These included worries about data privacy and the extent to which medical information can be shared under GDPR. A few mentioned the importance of attaching personal identifiers, such as CHI, to health and social care data. While this is not a data set, respondents pointed out that it is needed to make the sharing of data practical. Mydex CIC and The Health and Social Care Alliance Scotland suggested the best next step would be to focus on personal data stores that individuals could share with their care providers, rather than building government platforms.