From abailey@hackney.gov.uk Tue May 11 06:59:31 1999 Received: from mxu4.u.washington.edu (mxu4.u.washington.edu [140.142.33.8]) by lists.u.washington.edu (8.9.3+UW99.02/8.9.3+UW99.01) with ESMTP id GAA18832 for ; Tue, 11 May 1999 06:59:31 -0700 Received: from ms.hackney.gov.uk (ms.hackney.gov.uk [156.61.8.6]) by mxu4.u.washington.edu (8.9.3+UW99.02/8.9.3+UW99.01) with ESMTP id GAA26645 for ; Tue, 11 May 1999 06:59:26 -0700 Received: from betty.hackney.gov.uk (unverified [156.61.8.5]) by ms.hackney.gov.uk (Integralis SMTPRS 2.0.15) with ESMTP id for ; Tue, 11 May 1999 15:02:58 +0100 Received: from betty ([156.61.45.184] (may be forged)) by betty.hackney.gov.uk (2.5 Build 2639 (Berkeley 8.8.6)/8.8.4) with SMTP id OAA00708 for ; Tue, 11 May 1999 14:53:57 +0100 Message-Id: <006c01be9bb6$7ac2ea20$b82d3d9c@hackney.gov.uk> From: "Andy" To: References: Subject: Re: WAPHGIS: geostatstics Date: Tue, 11 May 1999 14:59:22 +0100 MIME-Version: 1.0 X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.00.2014.211 X-Mimeole: Produced By Microsoft MimeOLE V5.00.2014.211 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit I'm currently working on a health initiative a local council in the UK. i'm presently compiling sources of data from a wide range of sources that will aid us in appreciating reasons for bad health and trying to combat problems laid out the UK government paper 'our healthier nation'. Some of the departmental names will be obviously different elsewhere but i'd like to hear any comments based upon the choice and the availability of information. andrew bailey gis officer for london borough of hackney Individual Factors Diet Physical Activity Sexual Behaviour Smoking Alcohol Drugs Government: provide information on how to avoid osteoporosis so that accidents don't lead to broken bones; run public safety campaigns, ensure strategies are coordinated across governmental bodies; provide information on ways to avoid accidents. Local Players: ensure those who are in need have aids to prevent accidents e.g. child carseats, 'work for whole' school approaches to health and safety, target accident prevention at most risk. Diet Dispersal of good, affordable food. (GIS dept.) Physical Activity %. Pop (age) that attend fitness centres. Fitness centre customer levels. (Sport Centres/Swimming Pools (Service Area)) % of Pop with extreme Cholesterol levels & Obesity levels. (Presently unknown) Sexual Behaviour No. transmitted disease cases. (Presently unknown) (HIV/AIDs (Service Area)) Smoking No. hospital admittances re. respiratory problems. (Presently unknown) Mortality rates re. cancer and heart disease. (Presently unknown) Alcohol No. attendances at centres. (Alcohol Counseling (Service Area)) Drugs No. registered addicts. (Drugs (Service Area)) No. drug related arrests. (GIS dept.) Socio-Economic Factors Poverty Unemployment Social Exclusion Crime Government: Improve areas of deprivation through urban regeneration, tackle social exclusion and joblessness. Local Players: tackle social exclusion and joblessness in the community. Social Exclusion: the definition covers all aspects of deprivation, crime and unemployment however the Prime Minister has indicated three aspects which are to be investigated immediately: truancy levels, street living and bad estates. Poverty Index of Local Conditions Index of Local Deprivation. (GIS dept.) Unemployment % of Pop (age) that are unemployed. (Employment Service (Service Area) Social Exclusion Truancy levels (Education (Service Area)) Street Living (Housing Advice Centre (Service Area)) Bad Estates Estates with ILC & ILD (GIS dept.) Crime Crime statistics. (GIS dept.) Access to Services Education Transport Social Services NHS Government: Encourage health professionals to give appropriate advice, ensure professionals are trained in accident prevention. Local Players: Provide appropriate treatment to high-risk groups to prevent osteoporosis. Education Distribution of health information centres. (Presently unknown) Transport Distribution of rail stations Distribution of bus stops (GIS dept. & London Transport) Social Services Distribution of service points (Social Services Commissioning Team (Service Area)) NHS Distribution of hospitals (inc. Accident & Emergency Wards) Distribution of General Practitioners (GIS dept.) Environment Air Quality Water Quality - optimum levels of fluoridation leads to 50% less tooth decay Housing Government: Improve safety of roads, ensure compliance with seatbelt requirements and other road traffic laws, help set standards for products and appliances, promote higher standards of safety management. Local Players: Improve facilities for pedestrians and cycle paths, develop safer routes for schools, adopt traffic calming and other engineering measures and make roads safer, work for healthier and safer workplaces, make playgrounds safe, provide child pedestrian and cycle training. Air Quality Point set of monitoring stations (Pollution (Service Area)) Water Quality Non-spatial data set of water breakdown (Pollution (Service Area)) Housing Property Stock Assessment Surveys (Strategy & Commissioning, Regulatory Services (Service Area)) Traffic Incidents Traffic Incidents (accidents, seatbelt violations, speeding violations) (Police) ----- Original Message ----- From: Lance Waller To: Sent: Tuesday, May 11, 1999 2:46 PM Subject: Re: WAPHGIS: geostatstics > > As a biostatistics faculty member whose research and teaching involves > spatial statistics for public health data, I'm not sure I agree with > the statement on there not being much better than geostatistics for public > health "until we invent it". Here are a few references: > > Elliott, P., Martuzzi, and Shaddick, G. (1995) Spatial statistical methods > in environmental epidemiology: a critique. Statistical Methods in Medical > Research 4, 137-159. > > Elliott, P., Cuzick, J., English, D., and Stern, R. (eds) (1992) > Geographical and environmental epidemiology: methods for small area > studies. Oxford: Oxford University Press. > > Marshall, R. (1991) A review of methods for the statistical analysis > of spatial patterns of disease. Journal of the Royal Statistical Society, > Series A. 154: 421-441. > > Lawson, A. and Waller, L.A. (1996) A review of point pattern methods for > spatial modelling of events around sources of pollution. Environmetrics > 7, 471-487. > > Vine, M.F., Degnan, D., and Hanchette, C. (1997) Geographic Information > Systems: their use in environmental epidemiologic research. Environmental > Health Perspectives 105, 598-605 (related story, ``Mapping the > Environment'' pp. 594-596). > > In addition, four special issues of the journal Statistics in Medicine > have been devoted to various spatial public health aspects: > > 1993, Volume 12, No 19/20: Statistics and Computing in Disease Clustering > > 1995, Volume 14 (beginning on page 697), Topic: Mapping and Geographic > Display of Data > > 1995, Volume 14, No 21/22, Topic: Spatial Disease Patterns > > 1996, Volume 15, No 7/8/9, Topic: Statistics and Computing in Disease > Clustering > > There are some applications of geostatistics (kriging) to disease rates > (Carrat and Valleron 1992 American Journal of Epidemiology), but as a > statistician, I'm not sure it is such a good idea. The primary problem is > kriging assumes constant variance of all measurements but the variance of > disease rates depends on the number of people at risk in each rate (which > can vary widely from tract to tract, etc). The surface generally looks > great, but the prediction intervals are not very accurate. The issue to > me is that a great map of bad data is much more "believable" than a bad > map of good data. That is we are tempted to mentally link the accuracy of > the geographic data in a GIS map to the accuracy of any associated > attribute values, even though the attribute data may be very inaccuracte. > Mark Monmonier's books "How to Lie with Maps" and "Cartographies of > Danger" (University of Chicago Press) include very nice descriptions of > this and other related issues. > > Geostatistical techniques seem much more appropriate (to me) for > predicting maps of environmental exposures, etc. which are more in line > with the types of data where geostatistics originally developed. > > I hope these references help point out some related work in the > statistical literature. I'm not saying that these provide all the > answers, nor that statisticians always get it right, rather > I've found that working in GIS/spatial public > health can be frustrating since one has to search through so many > different fields' literature (GIS, geography, epidemiology, environmental > sciences, statistics, public policy, etc). I always appreciate when > someone provides me with an entree into another set of literature, so I > hope the list of review papers and special issues helps. > > Lance Waller > > On Sun, 9 May 1999, Richard E. Hoskins wrote: > > > Check out http://www.geovariances.fr/ for information and courses on spatial stats. > Its all oriented towards geostatistics but right now, us public health people do not > have much better until we invent it. > > > > > For spatial stats for "the rest of us" try "Interactive Spatial Analysis" by Bailey, Gatrell, > Longman, 1995, ISBN 0-470-23502-0. I sent this recently, got a couple of requests to re-send it. > > > > Dick Hoskins > > > > > .