Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit

Principles of Health Data Analytics Assignment

Deadline for submission: April12th 2024, 10:00

Provisional mark and feedback release date: June 1 2024

Length: 2,500 words (excluding bibliographies, tables, appendices, pictures and graphs and associated captions).

Aim:  In this assignment you are provided with a ‘Case Study’ comprising a short description of a problem faced by a walk-in clinic and a set of data about the clinic. You are asked to compare two proposed changes to the clinic staffing and assess the likely impact of each on specific operational outcomes. You have access to a software tool that you can use, along with the synthetic data downloadable from the Moodle page and the information provided in this briefing note, to model the clinic and the impact of changes to aspects of the clinic. Output from the model should allow the proposed changes to be assessed. Your submission should be in the form of a report presenting your results to a key decision-maker who is not a data analyst, including appropriate recommendations and an explanation of any assumptions and limitations of the work.

The assignment aims to assess your ability to: structure a problem, identify a model and assess its validity and reasonableness, and to analyse and present data and model output in a way tailored to inform decisions made by managers who are not specialists in data analysis.

Case Study: The Borchester Urgent Care Centre

A local commissioning group is reviewing the operation of a “24 hr walk-in” clinic for minor injuries / GP services. Managers have concerns about the way the clinic functions given the large fluctuations in demand: big queues can build up and sometimes people leave before being seen, likely attending the AE of the local hospital, while at other times doctors are left sitting idle. The commissioning group’s stated aim is to reduce queues but without too much clinician idle time. The two options identified are:

1) to add another member of staff to the team assessing and treating patients

2) to add a member of staff who would triage patients, so that some patients can be discharged more quickly.

You are asked to assess the impact of these on waiting times and staff utilisation in the clinic. Your model will need to include the following values, which can be obtained from the data and information provided:

i) number of available staff

ii) mean rate at which new patients arrive over the period concerned iii) mean duration of consultations

You should assess the impact of the two proposed interventions on:

(a) mean wait time

(b) mean queue length

(c) proportion of time each member of staff is engaged in assessing or treating patients (including completion of notes).

You will also need to consider how to model triage, e.g. how long does the triage process take? What proportion of patients is discharged or transferred at triage?

You may choose to restrict your attention to the performance of the clinic at peak times.

Briefing

The Borchester Urgent Care Centre (UCC) is based in a former hospital building and is used to deliver a variety of community services. Borchester UCC provides treatment for any illnesses or injuries which aren’t life threatening, but still need treating quickly such as: minor head injuries, suspected broken bones and fractures, sprains, cuts and scrapes, bites, eye problems and rising temperatures. Staff at Borchester UCC are able to request X-rays and some other imaging investigations.

Staffing levels

The clinical staffing at Borchester UCC consists of a GP and an emergency nurse practitioner at all times. The Emergency Nurse Practitioner works autonomously and can see a wide range of conditions. Consultant cover is provided by Borchester A&E department but because of other commitments is rarely physically based in the UCC.

Activity levels

Activity levels at the unit are, overall, lower than was anticipated when the unit was set up. Nevertheless there have been complaints that at certain times patients are faced with long queues and choose to leave before being seen, often attending at the local A&E or making use of other out-of-hours services.

Triage

For options that incorporate use of a triage nurse, patients with cuts, grazes and bruises would be dealt with by the triage nurse and sent home and patients with circulatory or obstetric  problems would  be  directed  immediately  to the  local  emergency  department. Triage that does not involve any treatment is estimated to take five minutes on average.

Data

You are provided with a dataset containing information about the patients attending the clinic, including arrival time, time they are seen and discharged, problem, service time and any supplementary notes for the year 2023. The management also provides you with the monthly number of patient attending the clinic for the years 2019-2022 (note that in this world COVID did not occur).

Report

Your submission should be in the form of a formal report of no more than 2,500 words in length and should include the following elements:

.       Executive Summary ~250 words

This should be a short summary of the key findings of the report

.       Introduction ~500 words

The introduction should summarise the context of the work and justify the approach taken

.       Method ~750 words

The methods section should include a detailed account of how the data were used. A detailed account of the assumptions  made and of any values that  had to  be estimated should be part of this section.

.       Results ~500 words

Results should be summarised clearly in words and, where appropriate, in a table or diagram. Think about what form of presentation is most appropriate. Be careful not to include so much data or model output that your key messages are obscured.

.       Discussion ~300 words

The discussion section should include a clear statement of what the limitations of the work are and of the uncertainty associated with the conclusion proposed in the recommendation section.

.       Recommendations ~200 words

The recommendations a set of recommendation clearly  stating which option is recommended and state any risk associated.

Use of AI for the assessment

The assessment for PHDA falls in category 2 of the use of AI tools in assessment (the description of the different categories can be found here). You can use AI in an assistive role (e.g. proofreading, testing code), but the work and text that you submit must be your own. If you make use of AI tools, this should be declared in your report just as you will do with any other tool you use, e.g. Excel, Tableau, ...