Pressure and force meets space and time

Put simply, '- what really matters is who does the kicking and who is being kicked.' Seen in this light, the real purpose of NHS reforms e.g., independent treatment centres, patient choice, payment by results, practice based commissioning, incentives to general practitioners, foundation trusts, targets etc, is to undermine the power of professionals and the unions to make medicine a 'creature of the state'.

One may well ask, "Given that's the goal, where do we fit in?" What place do mathematical models have in a world of rhetoric and spin? At the recent Royal Statistical Society conference in Belfast, so competently organised by Adele Marshall, Prof Adrian Raftery a statistical expert in weather forecasting, stated 'because of the complexity' it is impossible to forecast weather more than 23 days in advance. Are we better than that?

In this issue we are privileged to have two complementary Professorial contributions which throw light on the complex interactions between supply and demand.

Using a simple queueing model, Prof Michael Pidd, from the University of Lancaster explains why NHS care Trusts may have extreme difficulty in meeting their 2008 waiting list target of 13 weeks from referral to treatment. Most Trusts have succeeded so far, from 65 weeks to 26 weeks for outpatient appointment, probably by tackling long waits. But, further gains will be harder to achieve because straight lines are meeting curves.

On a similar vein, Prof Glenn Schmidt from the University of Utah uses the OM triangle to explain the interplay between inventory, capacity, and information and the 'curse of variability'. He also uses the triangle to explain choices in purchasing a MRI scanner. In the next issue he discusses management options in three different clinical scenarios.

Without a coherent long-term plan, can anything be achieved? Today (28th Sept 2006) The UK Prime Minister, Tony Blair, in his last year has thirty new projects he wants to introduce before he goes: further modernisation of the health and social care system, reforming education, solving world crises etc. Our ambitions are much simpler; a conference in Ireland in 2008, a revamped newsletter in 2007 and further development of the nosokinetics.org website

Understanding diminishing returns in the reduction of waiting times, Professor Michael Pidd, Lancaster

"...Queuing theory has been widely applied in understanding detailed aspects of health care provision.... In particular, simple queuing models clearly demonstrate that the laws of diminishing returns apply to systems in which service capacity is close to the demand for that capacity. Hence, reductions in long waiting times may be fairly easy to achieve but further reductions get harder and harder..."

Applicability of the OM Trangle to Health Care, Professor Glen Schmidt, University of Utah, Salt Lake City

"If you are a pilot, you know about the Bermuda triangle. If you study relationships, you have heard of a love triangle. If you enjoy recreational mathematics, you are probably intrigued by Pascal's triangle. But if you read Nosokinetics News, the triangle you should definitely know about is the OM triangle!..."

Measuring Productivity in Health Service Delivery an Australian Government Initiative.

Contributions are being arranged via State Treasury Departments although non-Government researchers may wish to approach the Commission directly. The expected release date of the project is Nov 2006. Contact: Paul Gretton (02) 6240 3252, Assistant Commissioner, Trade and Economic Studies. 

http://www.pc.gov.au/researchproject/2005/051102.html

Readmission

Uses multivariate statistical analysis to model the chance of readmission, based on a 10% sample of English HES data (1999-2004) specified conditions known to be associated with predictable readmission (e.g. congestive heart failure, diabetes, COPD, sickle cell disease). Algorithm score 0-100 based on 21 most powerful - statistically significant - variables. Tested model on 10% sample. Key variables were age, sex, ethnicity and number of previous admissions. At a score of 50, 54.3% of readmissions identified correctly, 34.7% flagged incorrectly. At 70 and 80, 22.6% and 15.7% flagged incorrectly. Paper also describes regional differences in readmissions.

Failure time analysis

Uses failure time analysis methods to study turnaround times in pathology. The model views a laboratory specimen like a living patient. When the specimen enters the laboratory, the time is analogous to the time of diagnosis for a patient. When the specimen's analysis is completed, the event is analogous to a patient who has died. Shows that the Kaplan-Meier plotting method, the log-rank test, and the Cox model can all be applied to turnaround times and provide useful results.

Pragmatism: OED Philos. The doctrine that the whole meaning of a conception expresses itself in practical consequences 1898.

Ormerod R. History and ideas of pragmatism Journal of the Operational Research Society 2006;57:892-909

Lists 12 reasons for thinking that pragmatism could serve OR practitioners well:

1. Pragmatism is what we do, how practitioners behave
2. Pragmatism supports an empirical (in other words scientific) approach
3. Pragmatism emphasizes the uncertainty and changing nature of findings
4. Pragmatism recognizes the individual psychological nature of meaning
5. Pragmatism holds that inquiry is social, as is knowledge
6. Pragmatism supports a theory of learning based on experience, experimentation and action
7. Pragmatism addresses morality, social interests and politics
8. Pragmatism places theory in the service of knowledge
9. Many OR approaches can find support in the philosophy
10. Pragmatism's stance on many things seems surprisingly modern
11. Pragmatism's biological approach should stand it in good stead to adapt to new science
12. Pragmatism is flexible enough to accommodate other philosophical positions.

Baffled by statistics: try this

Thierry Chaussalet drew our attention to this paper. Here you will find all you need to know. Given a 'statistically significant' result, does it have clinical or biological significance? How do you tell the good from the bad? As poor study design, chance, confounding variables and bias in observation and conduct can influence the results. Whatever your role whether manager, clinician or researcher, this paper (freely available on the web) is a classic.

Academic Positions in Statistics and Operational Research

Readership in Statistics / Operational Research (Full-time) (£41,133 to £52,106) per annum)

Lecturerships (2) in Statistics / Operational Research (Full-time) (£28,849 to £42,367)

CenSSOR is a vibrant research centre, whose members have nurtured a strong research environment and team spirit. Current research interests: Survival Analysis, Bayesian Networks, Markov Modelling and Stochastic Models. One focus of research application is the development of healthcare research and medical SOR.

Successful candidates will be expected to develop their own research plans, further enhancing the research in the Centre, while demonstrating a commitment to teaching excellence at all levels in SOR.

Closing date: Friday 29th October 2006.

Further information and to download application pack: www.qub.ac.uk/jobs

University of Westminster, London

Research Associate: 1 Year, Fixed Term

To undertake a 12-month study describing the care and administrative processes involved in the hospital care for older patients with hip fracture. Funded by SPARC (Strategic Promotion of Ageing Research Capacity) commencing November 2006.

Salary Dept Location £23,742 p.a. (Inc. LWA) HSCS Research Harrow Campus

Closing date: 9th October 2006; Interviews likely to be held on: 20th October

Informal enquiries about the position to Dr Christos Vasilakis by e-mailing C.M.Vasilakis@westminster.ac.uk or by telephoning 020 7911 5000 ext. 4029.

Forthcoming Conferences

5th IMA QUANTITATIVE MODELLING IN THE MANAGEMENT OF HEALTH CARE

Goodenough College, Central London on 2nd to 4th April 2007 Conference website / or the IMA website http://www.ima.org.uk/

Abstracts of 300-500 words to Lucy Nye at Lucy.Nye@ima.org.uk by 1 DECEMBER 2006. Authors of accepted abstracts will be notified by 1 January 2007.

Selected papers presented at the conference (whether orally or as a poster) will be published in the Springer journal Health Care Management Science.

Dr. T.J. Chaussalet, Reader, Department of Information Systems, University of Westminster, 115 New Cavendish Street, London W1W 6UW. Email: chausst@wmin.ac.uk

Nosokinetics News is mailed to supporters and collaborators interested in developing a scientifically valid approach to measuring and modeling health and social care systems. To be added to / removed from the mailing list email Prof Peter Millard nosokinetics@tiscali.co.uk

For earlier editions http://www2.wmin.ac.uk/coiec/nosokinetics.htm

The on line version is at:
http://www.iol.ie/~rjtechne/millard/index0.htm.
http://www.nosokinetics.org/

Some navigational notes:

Copyright (c)Roy Johnston, Ray Millard, 2005, for e-version; content is author's copyright,