As part of our March 2023 update, we focused on developing a portfolio of lifestyle data – such as Body Mass Index (BMI) measurements and smoking status – in a format that can be more easily utilised by researchers. Through the example of alcohol-consumption data, we explain some of the new possibilities, as well as the difficulties encountered and resolved along the way.
The importance of understanding lifestyle factors
The influence of lifestyle on an individual’s health and wellbeing is already widely recognised. Alcohol is one of several prominent lifestyle risk-factors for health, alongside smoking, poor diet, and physical inactivity. These are all associated with preventable ill health and shorter life expectancy. These risk-factors can also be socioeconomically related, meaning they contribute significantly to widening health inequalities. Therefore, research continues to be necessary to better understand the diverse nature of these impacts, which requires a foundation of high-quality and reliable data.
The challenges for the DataLoch service and researchers
Prior to undertaking this work, we did not know the complexity involved in answering the simple question: how much alcohol do people drink?
There are hundreds of codes used by GPs to capture an individual’s alcohol consumption in different ways, such as intake-levels, symptoms, treatment, and history. In reviewing these codes, it became apparent to us that these codes are not all used consistently by different health care professionals and preferences on which codes to use change over time. Also, some of the codes are more helpful for research than others, since they provide an estimation of the level of alcohol consumption within the code itself.
In combination, these factors make it extremely complicated for researchers to identify the relevant data for their projects.
Improving the data available for research
From our exploration of GP data, out of the hundreds of codes relating to alcohol, we found that the most reliable way to understand alcohol consumption is through eight principal Read Codes that capture daily alcohol units. In March 2023, these data cover 42% of records from the Lothian cohort (which may be a source of bias for population-wide studies focused on alcohol consumption). We highlight these specific Read Codes within the new Observations table in our Metadata Catalogue for researchers to request as part of their applications, alongside other lifestyle data like BMI and smoking. If you have a research project and would like to find out more, you can access our Metadata Catalogue from the application portal.
Benefits for the application process
Through developing the Observations table, we have opened up several key possibilities.
Firstly, the table highlights opportunities for researchers to request lifestyle-related data that they may not have previously considered being a factor within their analytical approach.
Secondly, returning to the idea of societal inequalities, linking data from the Observations table with Scottish Index of Multiple Deprivation data (also available through our Metadata Catalogue) is an easier process.
Thirdly, distinguishing the most insightful lifestyle data and highlighting these for researchers, makes the data-selection process much simpler. Previously, researchers would have had to guess which Read Codes to select or selected all of them and grappled with the data-quality issues themselves.
Further opportunities through DataLoch
For public health researchers and others with specific interests in the effects of alcohol, smoking, and BMI on health and wellbeing, we can now offer improved support and advice through greater knowledge of which codes should be of greatest benefit.
In terms of Observations data, we continue to build on the lifestyle data summaries that we have developed to date, which will further streamline the application process and enhance the possibilities for research.
For a summary of the types of data that are hosted by our service, visit the About the Data page.
