Ben Torsney

Working backs Scotland - A public and professional health education campaign for back pain

Study Design. Public and professional health education campaign. Objective. To change public beliefs about the management of back pain. Summary of Background Data. Within the past decade, there has been a reversal in the strategy of management of back pain, from rest to staying active. There is only one previous public health education campaign on back pain, in a workers compensation setting in Australia. Methods. A multimedia campaign was based around 1777 radio advertisements, which were heard by 60% of adults. Information leaflets were prepared for people with back pain, for all health professionals who treat back pain, and for employers. A Web site was set up: www.workingbacksscotland.com. Structured monthly samples of 1000 adults were surveyed on their beliefs about rest or staying active, professional help sought and advice received for 2 months before the campaign and over the following 3 years. Royal Mail sickness absence rates and new awards of social security benefits for back pain were compared in Scotland versus the rest of the United Kingdom, before and after the campaign. Results. There was a significant (P < 0.001) change in the balance of beliefs, from about 55% rest versus 40% staying active to about 30% rest versus 60% staying active. This occurred within 1 month of the launch and was maintained over 3 years. There was a comparable change in professional advice. There was no change in advice about work or the number who said they stayed off work. There was no effect on sickness absence or new awards of social security benefits for back pain. Conclusion. There was a major shift in public beliefs and professional advice but no change in work-related outcomes.

An algorithm for optimal designs on a design space

Two variations of a simple monotunic algorithm for computing optimal designs on a finite design space are presented. Various properties are listed. Comparisons witn other algorithms are made.

An optimal design problem in rhythmometry.

A trigonometric regression model is assumed for a problem involving circadian rhythm exhibited by peak expiratory flow. Experimental designs are sought with a view to estimating a particular nonlinear function of the parameters. Both optimal and nonoptimal, but more practicable, designs are derived and their relative efficiencies are established.

A Moment Inequality and Monotonicity of an Algorithm

Monotonicity of a proposed algorithm for a class of constrained optimisation problem is investigated. A moment lemma is proved which yields a condition sufficient for monotonicity. This condition is seen to be satisfied in one example. The problem is defined in section 2, while examples are revealed and optimality conditions are outlined in sections 3,4. Algorithms are discussed generally in section 5, with the proposed algorithm appearing in the next section. The moment lemma and sufficient conditions are presented in section 7 and empirical results form a concluding section.

D-Optimal Designs for Generalized Linear Models

This paper develops some simple methods for obtaining D-optimal designs for generalized linear models with multiple design variables. In some important cases the numerical complexity can be reduced to that of the two parameter case regardless of the original dimension. The form and properties of the obtained D-optimal designs are illustrated and discussed through a few interesting examples.

Construction of Constrained Optimal Designs

We consider the problem of finding an ‘approximate’ design maximising a criterion subject to an equality constraint. Initially the Lagrangian is formulated but the Lagrange parameter is removed through a substitution, using linear equation theory, in an approach which transforms the constrained optimisation problem to a problem of maximising two functions of the design weights simultaneously. They have a common maximum of zero which is simultaneously attained at the constrained optimal design weights. This means that established algorithms for finding optimising distributions can be considered. The approach can easily be extended to the case of several constraints, raising the ‘prospect’ of solving an expanded class of problem.

Construction of marginally and conditionally restricted designs using multiplicative algorithms

The problem of constructing optimal designs when some of the factors are not under the control of the experimenters is considered. Their values can be known or unknown before the experiment is carried out. Several criteria are taken into consideration to find optimal conditional designs given some prior information on the factors. In order to determine these optimal conditional designs a class of multiplicative algorithms is provided. Optimal designs are computed for illustrative, but simplistic, examples. Two real life problems in production models and a physical test for predicting morbidity in lung cancer surgery motivate the procedures provided.

The effect of deprivation and HbA1c on admission to hospital for diabetic ketoacidosis in type 1 diabetes

Aims/hypothesisDiabetic ketoacidosis is a potentially life-threatening complication of diabetes and has a strong relationship with HbA1c. We examined how socioeconomic group affects the likelihood of admission to hospital for diabetic ketoacidosis.MethodsThe Scottish Care Information – Diabetes Collaboration (SCI-DC), a dynamic national register of all cases of diagnosed diabetes in Scotland, was linked to national data on hospital admissions. We identified 24,750 people with type 1 diabetes between January 2005 and December 2007. We assessed the relationship between HbA1c and quintiles of deprivation with hospital admissions for diabetic ketoacidosis in people with type 1 diabetes adjusting for patient characteristics.ResultsWe identified 23,479 people with type 1 diabetes who had complete recording of covariates. Deprivation had a substantial effect on odds of admission to hospital for diabetic ketoacidosis (OR 4.51, 95% CI 3.73, 5.46 in the most deprived quintile compared with the least deprived). This effect persisted after the inclusion of HbA1c and other risk factors (OR 2.81, 95% CI 2.32, 3.39). Men had a reduced risk of admission to hospital for diabetic ketoacidosis (OR 0.71, 95% CI 0.63, 0.79) and those with a history of smoking had increased odds of admission to hospital for diabetic ketoacidosis by a factor of 1.55 (95% CI 1.36, 1.78).Conclusions/interpretationWomen, smokers, those with high HbA1c and those living in more deprived areas have an increased risk of admission to hospital for diabetic ketoacidosis. The effect of deprivation was present even after inclusion of other risk factors. This work highlights that those in poorer areas of the community with high HbA1c represent a group who might be usefully supported to try to reduce hospital admissions.

Multiplicative Algorithms for Constructing Optimizing Distributions : Further Developments

We consider a class of optimization problems in which the aim is to find an optimizing distribution. In order to determine the optimizing distribution, a class of multiplicative algorithms, indexed by a function f(·) which depends on the derivatives of the criterion function, is considered. The function f(·) is positive, strictly increasing and may depend on one or more free parameters. In an attempt to improve convergence, we consider an objective approach to choosing f(·) which allows the criterion function to have negative derivatives. We consider objective choices of the class of functions f(·) for the proposed approach. The approach enjoys considerable improvements in convergence of the algorithm.

W-Iterations and Ripples Therefrom

The focus of this contribution is on algorithms for constructing optimal designs. Henry Wynn, one of the earliest contributors in this field, inspired David Silvey to point me in the direction of further algorithmic developments. I first heard of this topic at a seminar David gave at University College London in the autumn of 1970 in which he spoke of Henrys work (he had been Henrys external examiner). Henry also attended this seminar! The rest is history. In this chapter, Henrys work and ripples therefrom will be explored.