Reducing variation

Reducing Variation
Variation is an important concept in health and healthcare and one that rightlyreceives a lot of attention; but it is a problem that presents some significantchallenges. For example it can refer to differences in access to healthcare services ortreatments or it could be differences in health outcomes such as the outcome oftreatments, in health status, or in rates of hospitalization or deaths.
Such differences will arise naturally due to differences in population characteristics.
For example we can expect both a higher utilization rate of healthcare services and ahigher death rate to occur naturally in communities that have a higher proportion ofelderly residents. However, variation in health outcomes and healthcare persist evenafter such demographic differences are taken into account. It is this ‘unwarranted’variation that is of particular interest. Where there is evidence of systematic variationby virtue of particular group characteristics – such as gender, socioeconomiccircumstances or ethnic origin – then the unwarranted variation is regarded as ahealth inequality and a further question of equity arises.
Muir Gray and Philip DaSilva in their introduction to ‘The NHS Atlas of Variation inHealthcare’ (2010) state: “To generate widespread change, the need to identify and reduce unwarrantedvariation must be placed at the centre of commissioning decision-making, and alsoneeds to be a priority for clinicians and patients”. Opportunities
It can be helpful to consider unwarranted variation in healthcare in three ways: Effective care is where a treatment confers significant benefit and the ‘right’ rateof treatment is 100% and anything markedly less than this represents under-useSupply sensitive care is where treatment rates are largely determined by localcapacityPreference sensitive care is where more than one effective care option isavailable and variation can be accounted for by informed patient choice orprofessional opinion The first two categories are of particular relevance to improving the effectiveness andcost-effectiveness of healthcare provision; the third can represent ‘good’ variation reflecting person-centred care. In the case of supply sensitive care it is not necessarilytrue that more means better. Where health outcomes are equivalent (such asmortality or quality measures) then high rates of treatment may indicate over-useand can represent poor value for money for the local health economy. Thisemphasises the importance of considering both treatment rates and heath outcomesin combination. It is also important to consider where local practice sits within anational picture of variation as well as taking account of variation within the city.
In terms of Sheffield’s position in countrywide variation a good starting point is ‘The
NHS Atlas of Variation in Healthcare’ available at www.rightcare.nhs.uk/atlas/ 1.
The Atlas sets out geographic variations in key indicators of effective health
treatments and outcomes at Primary Care Trust (PCT), Local Authority and Strategic
Health Authority population levels.
Of the 36 indicators of effective care at PCT level, Sheffield is in the most effectivefifth of PCTs for: A low rate of emergency admission for epilepsy (57.2 per 100,000 population in2006/7 to 2008/9, England range 40.2 – 129.7)Low rates of ezetimibe prescribing – an expensive cholesterol lowering drug(£31 per unit of need in July 2009 – September 2009, England range £15.3 – £87.3) However areas for improvement include indicators where Sheffield appears in theleast effective fifth of PCTs: Very high hospital bed day usage for cancer admissions (66.4 per 1000population in January to March 2008/09 – September to December 2009/10,England range 29.7 – 72.4), although expenditure on cancer services ranks inlowest fifth (£27,254 per 1000 population in the same period, England range£19,746 – £42,767)Chronic Obstructive Pulmonary Disease (COPD) bed day usage (664 per 1000population in 2008, England range 271 – 1069)Outlier for the completeness of diabetes treatments although this measure isaffected by incomplete data Often the key to tackling issues in unwarranted variation nationally lies in tacklingvariation within the local health community. By improving the local outliers withinthe City therefore we can raise the overall average.
Our local analysis of unwarranted variation includes key measures of diagnosis andtreatment effectiveness for the main disease categories making up the long term conditions (LTCs) for patients registered with each of the 91 GP practices in Sheffield.
In this analysis we look at the number of GP diagnosed cases of patients who haveCOPD, heart disease, diabetes and hypertension (high blood pressure), and thenumber who are receiving key treatments for each of these conditions and/or theexpected outcomes from the treatment. We then compare this with the number ofcases we would expect to find in each GP practice population from the demographicprofile. From this we are able to identify where there is likely under-diagnosis,under-treatment and poor outcomes and therefore where we can target supportmeasures and identify opportunities to refine the clinical pathway.
Challenges
The challenges in reducing unwarranted variation in healthcare are considerable,even where our analysis identifies condition-specific contributions to variation in themanagement of long term conditions. Ultimately the problem of unwarrantedvariation boils down to the adoption by the health care professions of best practicebased on sound clinical evidence.
Recommendations
Once established, the new Clinical Commissioning Group for Sheffield should: Maintain a focus on identifying and managing variation as a guiding principlein its commissioning and clinical development workUndertake further work to understand and tackle unwarranted variation incancer treatment and outcomes relative to other areas of the Country.
Maintain a focus on diabetes and COPD with the aim of reducing unwarrantedvariation by supporting improved detection and outcomes in the areas ofSheffield where these are below average 1. The ‘NHS Atlas of Variation’ represents a series of maps charting the pattern ofvariation across the country on a range of health indicators that are deemed to beclinically important in relation to improving health and tackling health inequalities

Source: http://www.publichealthsheffield2011.nhs.uk/wp-content/uploads/2011/08/Reducing-Variation.pdf