Making data work for you: the basics of data collection and handling for patient groups
In the second week of July, we hosted a three-day online workshop that focused on data use, collection and storage. The aim of the workshop was to provide guidance and practical advice to patient groups seeking to better understand their data collection practices and how to get the most out of what they collect. The workshop delved into why data should be collected, the key regulations and best practise for approaching data storage and use, and ways that data can be utilised to inform key projects and strategies.
We broke the workshop topic down into three two-hour-long sessions, day one ‘What counts as data and why have we collected it’ took us back to the basics of what data fundamentally is and why patient groups need to pay special attention to the data they are collecting. Day two ‘The rules and regulations of data collection’ focused on collecting and handling data guidelines and legality. On our third and final day, we looked at ‘How to collect data efficiently and how to use it once it is collected’ where we heard case studies from patient groups discussing how they have used data to help their community.
Read our day-by-day breakdown!
11:00am - 1:00pm, Wednesday 7th July
Rick Thompson, Findacure
Day one was kicked off by Findacure CEO, Rick Thompson, whose presentation took us back to the basic principles of what data is and encouraged participants to think about their approach to data.
Rick began by outlining the overall takeaway of his talk which was ‘think like a scientist’. He meant that when patient groups are collecting data they should think more critically, plan ahead and be prepared to understand the strengths and weaknesses of the data. Thinking like this will allow them to save time and draw better, and more accurate conclusions when it comes to processing their data.
Rick explained that one of the first things patient groups should do is ensure they are collecting good quality data. To do this they should consider what they are trying to do with the data and assess if the data they are collecting will allow them to do it. He went on to outline three key considerations patient groups need to think about to ensure best practice for data collection. The first of these is bias, it is likely that there will be some biases in the data patient groups collect. These could be towards a certain age group, ethnicity or geographical region. It is important to identify any biases and understand what these could mean for their interpretation. The second key thing to consider is sample size – the larger the sample size the better analysis you can deliver. The third is question structure, poor question structure can lead to confusion, it is important to keep any questions clear and neutral. Rick finished his presentation by sharing his golden rule which is you can’t always do it perfectly, aim to do your best and make sure you recognise the limitations of your approach and account for these in your analysis. This will give you the best data you can collect and the most sensible interpretation of your results, meaning the actions you go on to take will be evidence-based and logical.
And don’t forget – think like a scientist!
Conan Donnelly, International Niemann-Pick Disease Registry
Next up we heard from Conan Donnelly the CEO of the International Niemann-Pick Disease Registry. Conan is an Epidemiologist and health researcher; the aim of his presentation was to provide us with a scientific perspective of data collection. Conan began by outlining the importance of planning when undertaking any data collection project. He recommended talking with a variety of stakeholders (experts, patients, families) throughout the data collection process to help identify any biases and avoid mistakes. He also explained how qualitative data can play an important role in the early stages of data collection, to help you decide what you want to collect and where you should start.
Conan ended his presentation by highlighting the benefits of collecting high-quality data. Collecting better data means patient groups can manage their support services better, better understand the needs of their users and therefore be better advocates of their community. It also allows patient groups to deliver their services more efficiently, leading to better representation of their community overall. His closing point was to emphasise that all data on rare diseases is rare and therefore, valuable and important! Data patient groups collect can often become a launching point for other larger data collection processes, for example, registries or health technology assessments.
Rick and Conan gave us an excellent introduction to the topic of data, overall it was a great first day!
11:00am - 1:00pm, Thursday 8th July
Tim Musson, Computer Law Training LTD
Day two of our workshop was focused on the key rules and regulations of collecting and handling data as well as informing on best practise to avoid issues, ensure efficient procedures and minimise the risk of breaches. We had expert trainer Tim Musson lead the whole session to shed light on these important topics.
Tim began by outlining the purpose of data protection. He explained that when handling/processing personal data you are dealing with information about people, therefore, if you don’t treat the data correctly you are not treating the people correctly. This is especially true when dealing with ‘special category data’ (particularly sensitive personal data, for example, health data, ethnicity etc) which has additional requirements.
Before you begin to collect personal data, you must decide on your purpose for processing it, your ‘legal basis’, how long you intend to keep it, who else might hold or process it and what countries you might transfer the data to. It is vital that you provide the data subjects with all of this information and that they are aware of what you plan to do with their data. When deciding what you want to do with the data (your purpose) Tim explained three things you need to consider for best practice: your purpose for processing must be specific, explicit and legitimate. Once you have decided on these parameters you cannot further process the data in a way that is incompatible with those purposes.
Tim went on to discuss the big risks surrounding data collection, these can come from people e.g. hackers, human error or complaints. There is also the chance of personal data breaches, these can occur when data is available to unauthorised people, when data is unavailable i.e. it’s lost or has been accidentally deleted or when data is compromised. Tim outlined the most likely sources of personal data breaches which are phishing attacks, email fraud and portable media (laptops, phones etc). He emphasised the importance of using secure passwords and, where possible, two step authentications.
Tim did an excellent job of condensing such a large topic into two informative and accessible hours!
11:00am - 1:00pm, Friday 9th July
We wanted to use day three of our workshop to highlight examples of existing data projects in the rare disease space to provide insight and inspiration to our attendees.
Amy Hunter, Genetic Alliance UK
First, we heard from Amy Hunter the Director of Research at GAUK who discussed a study they conducted looking into the effect living with a rare condition has on mental health. Amy’s presentation gave us an insight into what a mixed-method data collection study involves and highlighted some of the advantages and challenges of this type of study design.
To start off she explained that a mixed methods study collects both qualitative (verbal and written responses) and quantitative (numerical) data. This allowed them to collect an in-depth data set about peoples experiences alongside hard numbers which they could present to policymakers. Amy went on to discuss the six stages which made up the study, these included a literature review, forming an advisory board, applying for ethics approvals, interviews, a survey and ended with a multi-stakeholder workshop.
By using this mixed-method approach each stage of the study can be used to inform the next and results can be directly compared and contrasted to each other. These are some of the benefits of this type of data collection. Amy concluded her presentation by sharing some of the challenges of the study, for example, not being a part of an academic team at a university meant they did not have access to certain resources or support that would have been useful. They combatted this by ensuring they had academics on their advisory board!
What a great example of a data project in the rare disease space! If you are interested in finding out more about GAUK’s Mental Health Report, follow this link to view the findings!
Jess Hobart, The UK Mastocytosis Support Group
Next up we heard from Jess Hobart Co-Chair of The UK Mastocytosis Support Group. Jess spoke about how her organisation has collected and used data to help their community. She discussed a survey they conducted to gather information about patient’s reactions to the Covid-19 vaccinations. Jess reiterated what we heard on day one especially the idea of thinking critically and ‘like a scientist’ when approaching a data project. Jess explained how they spent time carefully considering what question they could answer with the survey (and which they couldn’t) and identifying who could help them (she suggested contacting young researchers who need to get published). Jess gave us examples of how they tried to minimise the chance of bias by not using leading questions, asking patients to provide descriptions of their symptoms and using tick boxes.
She left us with some top tips for patient groups: don’t be scared to undertake data projects especially if there is information your community needs! Ask for help, especially around GDPR and ethics approval. Ensure you have consent for the data you are collecting, that the data is anonymised where possible and that you are only collecting the data you need! Keep surveys reasonable short to avoid survey fatigue, define questions clearly, think ahead about what you want to do with your data and test your survey on your community. Thanks, Jess!
Karen Harrison, Alex TLC
In our third presentation of the day, Karen Harrison from Alex TLC provided us with a great example of a patient group identifying a need within their community and using data to help find a solution. Alex TLC knows that the condition they support is likely to be prevalent in ethnic minority communities, however, they did not feel they were reaching these individuals. In order to ensure their community was inclusive and representative, and that they are supporting all families affected by the condition, they decided to conduct an audit of their database. They asked their community to re-submit their registration info through a form that now included a question on ethnicity. They teamed up with Breaking Down Barriers on this project, who shared their expertise in interacting with ethnic minorities and helped them choose appropriate language, Karen could not emphasise enough how beneficial the partnership was. Karen explained that not only was the data they collect through this project beneficial for the organisation’s ability to connect more effectively with ethnic minorities, it can also be used to inform further projects. For example, Alex TLC is launching a patient registry which this data will help build.
As Conan said on day one, all data on rare diseases is beneficial and you never know where your data could take you!
Nina Knight and Karen Dolman from Acrodysostosis Support and Research
Last but not least we heard from Nina Knight and Karen Dolman from Acrodysostosis Support who spoke about their experience of launching a registry. Registries are often at the top of the agenda for patient groups and, like so many others in the rare disease community, Acrodysostosis Support decided to launch theirs as they realised they knew more about the condition than the experts.
Karen explained that the first thing they thought about was how to collect data that is as accurate as possible, and cited the phrase often used in the scientific community which is ‘garbage in, garbage out’. She gave us an overview of the steps they took to ensure this, to start with they focussed the survey on issues specific to their patients. They did this by compiling a list of all the symptoms patients have experienced from their Facebook group and directly from individuals. Therefore, they were able to tailor the questionnaire on issues that have already been reported on and provide an extensive tick list for patients which made the survey simple to complete. The survey included multiple terms and definitions for specific medical issues so people were more likely to recognise them. They also broke the questionnaire down into different parts to minimise the chance of survey fatigue.
It is clear that Karen and Nina have been thinking critically about their project and have provided a great example to other patient groups who are considering starting a registry!
After three jampacked days tackling the challenging and extensive topic that is data, we hope that attendees left the workshop with a clearer understanding of the benefits and procedures that surround data collection and are inspired by what is already being done by patient groups in this area.