Maybe having a single summary such as life expectacy or standardised
mortality ratio is less useful than a profile of age-specific death
rates.
For one thing, life expectancy is (I believe) calculated as if the current
age specific death rates would apply in the future to those born today.
Michael, below has highlighted the problem of end effects, other comments
have referred to the unreliability of numbers for small areas and the use
of synthetic estimates (rather than counts).
Maybe a review/overview article for Radical Statistics on the MEASUREMENT
of health inequalities is in the offing ... would anyone be interested in
writing it? It could be a joint effort from the various contributors to
this discussion.
Beat wishes
Jane Galbraith
co-editor Radical Statistics
> As John said: concentrating on extremes is not really helpful. The
> methodology used to calculate life expectancy (at least the ones used by
> ONS
> [http://www.statistics.gov.uk/downloads/theme_other/GSSMethodology_No_33.pdf]
> and the health observatories
> [http://www.sepho.org.uk/download.aspx?urlid=9847&urlt=1]) for such
> small areas introduces quite a few artefacts into the estimates. These
> predominantly manifest themselves in the form of over-estimates, because
> there is no upper bound placed on the length of survival in the final age
> group. I wrote a conference paper on this about a year ago which I hope to
> have published---you can read the conference proceedings version of it
> online
> [http://epc2010.princeton.edu/abstractViewer.aspx?submissionId=100690].
> There probably is a real, substantive gap in life expectancy, but 16.8
> years is most likely an over-estimate. A more conservative estimate would
> be provided by taking the inter-quartile range, or some other measure
> which drops the outliers.
>
> As for smaller areas, well, we'd all love to have them, but much smaller
> scale than the ward-level and the numbers become so tiny that it really
> is an exercise in getting blood out of a stone. After all, "life
> expectancy" is a property of populations, not individuals, so a
> reasonable level of aggregation has to be maintained for the figures to
> be meaningful.
>
> I hope this is helpful information.
>
> Best wishes,
> Michael.
>
>
> On Wed, 2010-11-10 at 08:56 +0000, Martin Rathfelder wrote:
>> I know about these thank you. What I want are ward level stats. Or
>> smaller, I suppose, but the small area stats I have seen were very
>> difficult to use because they didn't relate to any identifiable places.
>>
>> On 10/11/10 08:39, Potter, Lesley wrote:
>> > http://www.statistics.gov.uk/statbase/Product.asp?vlnk=8841
>> >
>> > There is a collection of resources on life expectancy here, produced
>> by
>> > the Office for National Statistics,
>> > Hope this is helpful,
>> > Lesley
>> >
>> >
>> > -----Original Message-----
>> > From: email list for Radical Statistics
>> [mailto:[log in to unmask]]
>> > On Behalf Of Martin Rathfelder
>> > Sent: 10 November 2010 08:32
>> > To: [log in to unmask]
>> > Subject: Health inequality:
>> >
>> > There is, according to the local PCT, a 16.8 year gap in life
>> expectancy
>> >
>> > experienced by female residents who live in the Macclesfield Town
>> > Tytherington area compared to those living in Crewe's Central and
>> Valley
>> >
>> > area
>> >
>> > This sort of local inforamtion is very useful politically. Does
>> anyone
>> > know of, or is anyone capable of producing, a convenient useful source
>> > of such stuff? Ideally showing the most extreme disparities in close
>> > proximity.
>> >
>> > I presume that the smaller the areas the greater the disparities. The
>> > DPH in Huddersfield used to produce some good graphs showing the
>> > contrasts between the two sides of the same road.
>> >
>> > Martin Rathfelder
>> > Director
>> > Socialist Health Association
>> > 22 Blair Road
>> > Manchester
>> > M16 8NS
>> > 0161 286 1926
>> > www.sochealth.co.uk
>> >
>> > If you do not wish to be on our mailing list please let us know and we
>> > will remove you
>> >
******************************************************
Please note that if you press the 'Reply' button your
message will go only to the sender of this message.
If you want to reply to the whole list, use your mailer's
'Reply-to-All' button to send your message automatically
to [log in to unmask]
Disclaimer: The messages sent to this list are the views of the sender and cannot be assumed to be representative of the range of views held by subscribers to the Radical Statistics Group. To find out more about Radical Statistics and its aims and activities and read current and past issues of our newsletter you are invited to visit our web site www.radstats.org.uk.
*******************************************************
|