Annex 1
HOW THE
NRAC FORMULA WORKS
IN PRACTICE
Health Board target shares are calculated by
adjusting the population of each Health Board area for three factors
that are known to influence healthcare utilisation:
1. the age-sex profile of the population (age-sex
cost weights),
2. the additional needs of the population due
to Morbidity and Life Circumstances (MLC weights), and
3. the unavoidable excess costs of supplying
services (excess cost weights).
The four main components within the formula
(population, age-sex costs weights, MLC weights and excess cost
weights) are generated for datazones (which are key small-area
statistical geographies in Scotland introduced by the Scottish
Government for use in Scottish Neighbourhood Statistics),
intermediate datazone (IDZ) and GP practice level as appropriate.
1. POPULATION
Population is the primary component of the formula.
1.1 Hospital and Community Health Services
(HCHS)
The population figures used within the formula
are re-based General Register Office for Scotland (GROS) population
projections. Re-based population projections are a simple adjustment
made to the GROS population projections, by updating them using
the latest mid-year population estimates (MYEs) that have been
published since the Health Board level projections were published.
For example, to re-base the 2004 based projection of 2008 using
2006 MYE the calculation is:
(2006 MYE) + [(2004-based projection of
2008)—(2004-based projection of 2006)].
This calculation takes account of any over or
under-estimation of the projection in the years to 2006, and this
adjustment is applied to the projection between 2004 and
2008.
1.2. GP Prescribing
The population figures used are based on the
Community Health Index (CHI) population. The CHI population count
is deflated at GP Practice level to match the re-based Scotland
population projection used for HCHS (for more information see
Technical Addendum B—Population—15 August 2007 document
on the NHSScotland Resource Allocation Committee "NRAC"
web site).
2. AGE-SEX
The formula adjusts for the age-sex profile
of the population to take account of the effect of age differences
on the cost of delivering different NHS services. On the whole,
older people tend to consume greater resources and the costs can
rise steeply with age.
Calculation of the age-sex cost weight starts
with the age-sex breakdown for the population of each datazone.
This gives a population structure for each area to which the national
average cost per head of population (by age group) can be applied.
These costs are specific to each of the care programmes (acute,
care of the elderly, mental health and learning difficulties,
maternity and community) analysed in the formula. The total "cost"
associated with each care group within each age-sex band is obtained
by multiplying the number of individuals by the national average
cost per head appropriate for that age-sex group.
These "costs" are then totalled across
all age-sex bands for each care programme. This total is then
divided by the population of the datazone to get a datazone cost
per head for the care programme across all ages and sexes. This
is then compared to the Scottish average cost per head for the
care programme to produce a care programme index (by datazone).
This shows the amount by which the expected costs for the datazone
are above or below the national average for each care programme.
All theses indices are combined using care programme
weightings (obtained from the Scottish Health Service Costs Book—see
table 1) to produce the final age-sex index (for more information
see Technical Addendum C—age sex—19 September 2007 document
on the NHSScotland Resource Allocation Committee "NRAC"
web site).
Table 1
CARE PROGRAMME WEIGHTS (YEAR ENDED 31 MARCH
2006)
|
| Acute | Care of
the Elderly
| Mental Health
& learning
Difficulties
| Maternity | Community
Travel-based
| Community
Clinic-based
| Overall
HCHS |
GP
Prescribing |
|
| 49.4% | 3.9%
| 12.7% | 3.5%
| 10.1% | 5.0%
| 84.6% | 15.4%
|
|
3. ADDITIONAL NEEDS
(MLC)—DUE TO
MORBIDITY, LIFE
CIRCUMSTANCES AND
OTHER FACTORS
In general, people who are less healthy and/or more deprived
have a greater need for healthcare so this index directs relatively
greater resources towards Boards with higher premature death rates
and greater socioeconomic deprivation.
The factors that best explained the variation in need for
each care programme were identified using statistical regression.
For example, for the acute care programme the variables were identified
as the mortality rate for under 75s and the limiting long-term
illness rate. A combination of these two variables were used to
calculate the MLC index for each IDZ. This represents the needs
over and above those explained by the age-sex structure (for more
information see Technical Addendum D—Morbidity and Life
Circumstances—19 September 2007 document on the NHSScotland
Resource Allocation Committee "NRAC" web site).
4. UNAVOIDABLE
EXCESS COSTS
OF SUPPLY
This index takes account of the excess costs of supplying
health services in different urban-rural areas and gives greater
weights to remote and rural areas where hospitals and clinics
serve smaller populations and where dispersed populations mean
greater travelling distances for staff. There are four components
of the unavoidable excess cost factor: hospital services, community
clinic based services, community travel based services and GP
prescribing.
The unavoidable excess cost index for hospital services is
developed at datazone level based on the ratio of local to national
average costs for the 10 Scottish Executive Urban-Rural Categories
(SEURC) in which the datazone lies. The GP prescribing index is
set to one for all areas as prescriptions are reimbursed at national
fixed prices. The community care programme index has two elements;
clinic based services and travel based services. Both community
indices are calculated at datazone level and represent the excess
costs of providing these services to residents of the datazone.
The overall unavoidable excess cost index for each datazone
is obtained by combining all the hospital and community excess
costs indices using care programmes weightings (table 1). (For
more information see Technical Addendum E1, E2 & E3 on
NHSScotland Resource Allocation Committee "NRAC"
web site).
5. OVERALL INDICES
The indices for HCHS and GP Prescribing parts of the formula
are calculated separately. These indices are then aggregated up
to Health Board level for each care programme and each element
of the formula. For example, the acute age-sex indices for every
datazone in a Health Board are averaged (weighted by population)
to give an acute age-sex index for that Health Board. Similarly,
the additional needs indices are averaged using populations adjusted
for age and sex as weights, and the calculations of the Health
Board level excess costs indices use populations adjusted for
age-sex and additional needs as weights.
With these small area "building blocks" the figures
can be split in any number of ways eg to give Health Board level
indices for individual factors (eg an additional needs index for
Greater Glasgow & Clyde); an index for a particular care programme
(eg distributions for maternity); or an index for a different
geography eg CHP.
6. HEALTH BOARD
SHARES
The small areas indices are amalgamated to Health Board level,
and applied to the population share to give a final value that
is used to inform the target share of funding for each Health
Board. (for more information see NRAC—Final Report on NHSScotland
Resource Allocation Committee "NRAC" web site).
| 
