Prevalence data in the QOF

Please click on one of the links below to take you to the relevant section -

Overview

Use of QOF prevalence data in points and payments calculations

National prevalence estimates

Warnings about the use of QOF prevalence figures for other purposes

Overview

Prevalence is a measure of the burden of a disease or health condition in a population at a particular point in time (and is different to incidence, which is a measure of the number of newly diagnosed cases within a particular time period).  

Prevalence data within the QOF are collected in the form of practice "registers".  A QOF register may count patients with one specific disease or condition, or it may include multiple conditions.  There may also be other criteria for inclusion on a QOF register, such as age or recency of diagnosis. 

A QOF prevalence rate is simply the total number of patients on the register, expressed as a proportion or percentage of the total number of patients registered with the practice.  QOF prevalence rates are very crude because they are not adjusted to account for patient age distribution or other factors that may differ between general practices.  Furthermore, although registers may be restricted (e.g. to only include persons over a specified age) the QOF prevalence rate is based on the total number of persons registered with the practice (the practice list size) at one point in time.

Use of QOF prevalence data in points and payments calculations

Prevalence data are used within QOF to calculate points and payments within each of the clinical indicator groups.  Specifically:-

• Points can only be awarded to a practice for a given clinical indicator group if the practice can produce a register of patients with that condition or group of conditions.
• The number of pounds per point in each clinical domain area is adjusted up or down according to each practice's prevalence rate for the relevant QOF register, relative to the estimated national prevalence rate from the same register.

What follows is a brief summary on the use of prevalence data in QOF calculations.  Further details can be found:

• within the  General Medical Services Statement of Financial Entitlements for 2006 onwards 
• on the Scottish Executive's Pay Modernisation website, within chapter 3 of the document Implementing the nGMS contract in Scotland

The aim of the prevalence adjustments in each of the clinical indicator groups is to deliver a more equitable distribution of payments in the light of different workloads that practices face in achieving the same number of points.  Practices with a high prevalence of a specific condition or group of conditions will receive more pounds perpoint for the relevant indicator group than practices with a low prevalence of the same condition or group of conditions.  However, the calculations are set such that even practices with very low prevalence from a given register still receive a minimum payment for providing appropriate services for it.  This is because, even if there are only a small number of patients on the register, practices still have significant costs in identifying morbidity or other healthcare needs and establishing systems to support those patients effectively.

During calculation of QOF payments, the baseline number of pounds per point (£124.64 in 2007/08) is adjusted up or down within each clinical domain area according to each practice's prevalence for that disease or condition, relative to the estimated national prevalence (see below).  The amount by which the pounds are is adjusted up or down is known as the Adjusted Disease Prevalence Factor (ADPF).  For example, if a practice has an ADPF of 1.20 then the adjusted pounds per point for asthma = £124.64 x 1.20 = £149.57.

The calculation of the ADPF uses, but is not based purely on, the practice's raw disease prevalence for a given condition.  The raw disease prevalence for each practice is calculated by dividing the number of patients on the relevant disease register by the number of patients the practice has on its registered list.  The prevalence data published here are shown as rates per 100 patients.  For example:

A practice has 104 patients on its asthma disease register Its total list size at 1st January 2008 was 2000 The raw prevalence estimate (per 100 patients) equals: (104 / 2000) x 100 = 5.2

A fuller explanation of the method used to calculate ADPFs is contained in chapter 3 of the document  Implementing the nGMS contract in Scotland.

National prevalence estimates

For payment purposes, a practice's adjusted disease prevalence factor (ADPF) for each group of clinical indicators is calculated by comparing that practice's prevalence rate from each register with a national estimate.  The national estimate used in the payment calculations is based on prevalence data recorded in the payment calculation system (QMAS - see Key stages in the QOF process) as at a date referred to in QOF as "National Prevalence Day".  National Prevalence Day was 14th February 2008 for 2007/08 QOF data, and 14th February 2005, 2006 and 2007 for 2004/05, 2005/06 and 2006/07 QOF data. Practices that could not or did not provide prevalence data for National Prevalence Day (e.g. because their clinical IT systems could not at the time be connected into the QMAS payment calculation system) were not included in the National Prevalence Day estimates.

Within these web pages, two types of Scotland-level prevalence estimates are published:

1. Scotland-level estimates based on all prevalence data available in QMAS by the time QOF data were extracted for publication (date of extraction is noted in each data table). This includes those practices whose register data were amended after National Prevalence Day.  These are the prevalence data that are likely to be of greater interest to most people.
2. Scotland-level estimates calculated on the basis of National Prevalence Day (14th February) data, and used in the calculations of Adjusted Disease Prevalence Factors for each practice.  These estimates are the ones shown for example on the  2007/08 - Average Prevalence Rates for Scotland page and the equivalent page for 2006/07.

The above is a brief summary on the use of prevalence data in QOF calculations.  Further details, including a fuller explanation of the method used to calculate adjusted disease prevalence factors (ADPFs) for each practice can be found within chapter 3 of the document Implementing the nGMS contract in Scotland . Although the examples given relate to the original QOF for 2004/05, the basic principles of the calculations still apply for subsequent years.

Warnings about the use of QOF prevalence figures for other purposes

Data on the prevalence of specific diseases or health conditions are an important element of the QOF and are of interest to many people. They can potentially be used to examine variations in the prevalence of the chronic diseases included in the clinical domains, but they should be interpreted with caution.  The main points to note are as follows:-

• QOF prevalence rates are what is known as "raw" or "crude" rates - which means that they take no account of differences between practice populations in terms of their age or gender profiles, or other factors that influence the prevalence of health conditions.  A QOF prevalence rate is simply the total number of patients on the register, expressed as a proportion or percentage of the total number of patients registered with the practice at one point in time.  This could mean for example that one practice with an older population might appear to have higher prevalence rates for age-related conditions such as cancer or stroke than another practice with a younger population.
  
• Some QOF registers are restricted to include only persons over a specific age.   However, the QOF prevalence rates use as their denominator the total (all ages) number of patients registered to the practice at one point in time.  Diabetes registers are based on patients aged 17 and over; epilepsy, chronic kidney disease and learning disabilities registers are based on patients aged 18 and over; and obesity registers are based on patients aged 16 and over.  This means that for these conditions the QOF-reported prevalence will appear lower than would be the case if the age restriction was also applied to the population denominator.
  
• Prevalence figures based on QOF registers may also differ from prevalence figures from other sources because of coding or definitional issues.   For example, to be on the QOF diabetes register, patients need to be aged 17 or over and the type of diabetes (type 1 or type 2) must be specified by the practice.  If the type is not specified the patient will not be counted in the register.  Information on diabetes as reported elsewhere may not be subject to these restrictions.
  
• Year-on-year changes in the size of QOF registers are difficult to interpret for various reasons including:- changes in epidemiological factors (such as an ageing population); improvements in case finding by practices; changes over time in the definition of the registers.  This point is addressed in more detail within our page on 2007/08 - Average Prevalence Rates for Scotland.

QOF prevalence rates can also be affected by other factors such as:

• health care seeking behaviour - people differ in the readiness with which they seek health care when they are not well;
• access to services - people are more likely to consult for a condition if services are readily accessible;
• diagnostic practice - it is impossible to completely standardise the methods clinicians use to make diagnoses;
• data recording - there may be variations in the completeness and accuracy of practice records.

These explanations all need to be considered when examining apparent variations in prevalence.