DHSC statistics: our quality management approach

Overview

This document explains how the Department of Health and Social Care (DHSC) manages the quality of the statistics it produces.

Our aim is to ensure our statistics are trustworthy, high quality and useful to the public.

Our approach is informed by the Code of Practice for Statistics and the AQuA Book. This document outlines the high-level principles behind the department’s approach to quality management. Specific quality assurance processes for statistics publications will vary by publication.

Our commitment to quality

We want users to have confidence in our statistics. To do this, we make sure our data and methods are:

• accurate
• reliable
• transparent
• clearly explained
• appropriate for their intended use

We review our processes regularly to maintain and improve quality.

How we manage quality

Understanding data sources

We check that the data we use is suitable for producing reliable statistics. We assess:

• how the data is collected
• whether it is complete and consistent
• whether it may contain errors or biases
• how often it is updated

We explain any known issues in our publications.

Using sound methods

We use methods that are consistent with recognised statistical standards.
We:

• choose appropriate techniques for processing and analysing data
• document our methods
• review our approaches when data, systems or user needs change

Quality assurance

We quality assure data at each stage of production. This includes:

• checking input data
• checking code and calculations
• reviewing charts, tables and commentary
• carrying out independent checks by colleagues
• checking reproducibility when our statistics are produced using code

We take a proportionate approach based on the complexity and importance of the output.

Being open about quality

We provide clear information in methodology notes or similar documents, such as data quality statements, about:

• strengths and limitations of the data
• any uncertainty in the statistics
• how revisions are made
• methods used and how they differ from previous releases

This helps users understand how the statistics can be used.

Improving quality over time

We make continuous improvements by:

• reviewing our products and methods
• adopting new tools and technologies where they improve quality, or where they improve efficiency and maintain quality
• improving consistency across related outputs
• updating supporting information and guidance
• being aware of and sharing latest best practice (see roles and responsibilities section below)

We publish revisions transparently and explain why they were made. See our revisions and corrections policy for more information.

Roles and responsibilities

DHSC statisticians are responsible for ensuring that the statistics they produce meet quality standards.

Teams follow the Code of Practice for Statistics and DHSC internal guidance on quality assurance.

Indicator Methodology Review Group (IMRG)

With limited exceptions, indicators published by the Office for Health Improvement and Disparities (OHID) are published only after review and approval by IMRG.

A proposed indicator is assigned a reviewer, separate to the publication team, who assesses the methodology and ensures consistency across existing indicators. The reviewer will present and discuss their findings with the rest of IMRG, where the indicator will be approved for publication or be redeveloped in response to the review.

Quality Assurance Working Group

We ensure best practice and knowledge is shared across teams and professions by ensuring all areas of the department are represented on our Quality Assurance Working Group.

The group develops and delivers tools, training and guidance to support the delivery of best practice standards for quality assurance within the department.

Cross-government groups

We keep up to date with best practice and developments by contributing to cross-government analytical groups. These include:

• Government Statistical Service Quality Champions
• Statistics Heads of Profession Group
• Cross-government Quality Assurance Working Group
• UK Health Statistics Steering Group
• Health and Care Statistics Leadership Forum

Statistics head of profession

While the responsibility for managing the quality of statistics publications lies with a publication’s lead analyst, the department’s head of profession for statistics has overall responsibility for the department’s portfolio of statistics publications, including their quality. More information on the head of profession for statistics role can be found on the Analysis Function website.

Contact and feedback

We welcome feedback on all our statistical outputs, policies and practices.

Contact: statistics@dhsc.gov.uk.

Alternatively please contact the responsible statistician or team named on each publication for feedback or information specific to that statistical output.