This project used the real-world data to address a problem faced by the National Health Service (NHS). I used Python to explore the available data, create visualisations to identify trends, and extract meaningful insights to make inform decision-making.
I am part of a team of data analysts that was contracted by the National Health Services (NHS), a publicly funded healthcare system in England. The team has been provided with internal and external data and a number of high-level business questions concerning the utilisation of services, missed appointments, and the potential value of using external data sources such as Twitter (now rebranded as X) . My role is to refine the business questions to actionable analytic questions based on the review of the available data. The exploratory data analysis has been done and presented the insights to the respective stakeholders.
The NHS must expand its infrastructure and resources to match its increasing population capacity. For this, it needs to budget correctly. When deciding on budget allotment, the NHS must understand the utilisation trends of each component in its network.
Some stakeholders feel that the NHS’s capacity should be increased while others feel that, based on current trends in utilisation, the current capacity is adequate and that efforts to make better use of existing infrastructure and resources are sufficient. I have been tasked to explore the available data and offer suggestions and recommendations based on the observations.
Therefore, reducing or eliminating missed appointments would be beneficial financially as well as socially. The government needs a data-informed approach to decide how best to handle this problem. At this stage of the project, the two main questions posed by the NHS are:
- Has there been adequate staff and capacity in the networks?
- What was the actual utilisation of resources?
Files included as below:
- actual_duration.csv – Details of appointments made by patients. For example, the regional information, date, duration, and number of appointments pertaining to a certain class.
- appointments_regional.csv – Details on the type of appointments made by patients. For example, regional information, the month of appointment, appointment status, healthcare professional, appointment mode, the time between booking and the appointment, as well as the number of appointments pertaining to a certain class.
- national_categories.xlsx – Details of the national categories of appointments made by patients. For example, the regional information, date of appointment, service setting, type of context, national category, and the number of appointments pertaining to a certain class.
- metadata_nhs.txt – Details of the data set, data quality, and reference.
- tweets.csv – Data related to healthcare in the UK scraped from Twitter.