Optimal intervention for an epidemic model under parameter uncertainty

We will be concerned with optimal intervention policies for a continuous-time stochastic SIR (susceptible-->infective-->removed) model for the spread of infection through a closed population. In previous work on such optimal policies, it is common to assume that model parameter values are known; in reality, uncertainty over parameter values exists. We shall consider the effect upon the optimal policy of changes in parameter estimates, and of explicitly taking into account parameter uncertainty via a Bayesian decision-theoretic framework. We consider policies allowing for (i) the isolation of any number of infectives, or (ii) the immunisation of all susceptibles (total immunisation). Numerical examples are given to illustrate our results.     

Interaction of maturation delay and nonlinear birth in population and epidemic models

 A population with birth rate function B(N) N and linear death rate for the adult stage is assumed to have a maturation delay T>0. Thus the growth equation N′(t)=B(N(t−T)) N(t−T) e⁻ d ₁ T−dN(t) governs the adult population, with the death rate in previous life stages d ₁≧0. Standard assumptions are made on B(N) so that a unique equilibrium N _e exists. When B(N) N is not monotone, the delay T can qualitatively change the dynamics. For some fixed values of the parameters with d ₁>0, as T increases the equilibrium N _e can switch from being stable to unstable (with numerically observed periodic solutions) and then back to stable. When disease that does not cause death is introduced into the population, a threshold parameter R ₀ is identified. When R ₀<1, the disease dies out; when R ₀>1, the disease remains endemic, either tending to an equilibrium value or oscillating about this value. Numerical simulations indicate that oscillations can also be induced by disease related death in a model with maturation delay. Received: 2 November 1998 / Revised version: 26 February 1999     

Analysis of an SEIRS epidemic model with two delays

 A disease transmission model of SEIRS type with exponential demographic structure is formulated. All newborns are assumed susceptible, there is a natural death rate constant, and an excess death rate constant for infective individuals. Latent and immune periods are assumed to be constants, and the force of infection is assumed to be of the standard form, namely proportional to I(t)/N(t) where N(t) is the total (variable) population size and I(t) is the size of the infective population. The model consists of a set of integro-differential equations. Stability of the disease free proportion equilibrium, and existence, uniqueness, and stability of an endemic proportion equilibrium, are investigated. The stability results are stated in terms of a key threshold parameter. More detailed analyses are given for two cases, the SEIS model (with no immune period), and the SIRS model (with no latent period). Several threshold parameters quantify the two ways that the disease can be controlled, by forcing the number or the proportion of infectives to zero. Received 8 May 1995; received in revised form 7 November 1995     

Optimal control of the chemotherapy of HIV

 Using an existing ordinary differential equation model which describes the interaction of the immune system with the human immunodeficiency virus (HIV), we introduce chemotherapy in an early treatment setting through a dynamic treatment and then solve for an optimal chemotherapy strategy. The control represents the percentage of effect the chemotherapy has on the viral production. Using an objective function based on a combination of maximizing benefit based on T cell counts and minimizing the systemic cost of chemotherapy (based on high drug dose/strength), we solve for the optimal control in the optimality system composed of four ordinary differential equations and four adjoint ordinary differential equations. Received 5 July 1995; received in revised form 3 June 1996     

Two SIS epidemiologic models with delays

 The SIS epidemiologic models have a delay corresponding to the infectious period, and disease-related deaths, so that the population size is variable. The population dynamics structures are either logistic or recruitment with natural deaths. Here the thresholds and equilibria are determined, and stabilities are examined. In a similar SIS model with exponential population dynamics, the delay destabilized the endemic equilibrium and led to periodic solutions. In the model with logistic dynamics, periodic solutions in the infectious fraction can occur as the population approaches extinction for a small set of parameter values. Received: 10 January 1997 / 18 November 1997     

Simple models for excitable and oscillatory neural networks

 Chains of coupled oscillators of simple “rotator” type have been used to model the central pattern generator (CPG) for locomotion in lamprey, among numerous applications in biology and elsewhere. In this paper, motivated by experiments on lamprey CPG with brainstem attached, we investigate a simple oscillator model with internal structure which captures both excitable and bursting dynamics. This model, and that for the coupling functions, is inspired by the Hodgkin–Huxley equations and two-variable simplifications thereof. We analyse pairs of coupled oscillators with both excitatory and inhibitory coupling. We also study traveling wave patterns arising from chains of oscillators, including simulations of “body shapes” generated by a double chain of oscillators providing input to a kinematic musculature model of lamprey.. Received: 25 November 1996 / Revised version: 9 December 1997     

Optimization models for the force and energy in competitive running

 In [2] the author has developed an optimization model for the force and energy in competitive running. In this paper the energy processes in the muscle were described by a three-compartment hydraulic model. Here this is reviewed briefly and applied to the current world records in order to determine the key parameters, maximal force, energy reserves and oxygen uptake. These values agree well with those given in the literature and those obtained by other means. The velocity profiles for 100 m sprints are described equally well. The model is then applied to older world records to deduce a relation between the force and energy by linear regression. Finally the fully parameterized model is used to compute the effects of adverse wind and altitude. Inasmuch as there are data available, there is a good agreement. Received 19 July 1995; received in revised form 27 February 1996     

Convergence in almost periodic Fisher and Kolmogorov models

 We study convergence of positive solutions for almost periodic reaction diffusion equations of Fisher or Kolmogorov type. It is proved that under suitable conditions every positive solution is asymptotically almost periodic. Moreover, all positive almost periodic solutions are harmonic and uniformly stable, and if one of them is spatially homogeneous, then so are others. The existence of an almost periodic global attractor is also discussed. Received: 11 November 1996 / Revised version: 8 January 1998     

On the formulation and analysis of general deterministic structured population models I. Linear Theory

—We define a linear physiologically structured population model by two rules, one for reproduction and one for “movement” and survival. We use these ingredients to give a constructive definition of next-population-state operators. For the autonomous case we define the basic reproduction ratio R ₀ and the Malthusian parameter r and we compute the resolvent in terms of the Laplace transform of the ingredients. A key feature of our approach is that unbounded operators are avoided throughout. This will facilitate the treatment of nonlinear models as a next step. Received 26 July 1996; received in revised form 3 September 1997     

Analysis of a model for macroparasitic infection with variable aggregation and clumped infections

 A model for macroparasitic infection with variable aggregation is considered. The starting point is an immigration-and-death process for parasites within a host, as in [3]; it is assumed however that infections will normally occur with several larvae at the same time. Starting from here, a four-dimensional, where free-living larvae are explicitly considered, and a three-dimensional model are obtained with same methods used in [26]. The equilibria of these models are found, their stability is discussed, as well as some qualitative features. It has been found that the assumption of “clumped” infections may have dramatic effects on the aggregation exhibited by these models. Infections with several larvae at the same time also increases the stability of the endemic equilibria of these models, and makes the occurrence of subcritical bifurcations (and consequently multiple equilibria) slightly more likely. The results of the low-dimensional model have also been compared to numerical simulations of the infinite system that describes the immigration-and-death process. It appears that the results of the systems are, by and large, in close correspondence, except for a parameter region where the four-dimensional model exhibits unusual properties, such as the occurrence of multiple disease-free equilibria, that do not appear to be shared by the infinite system. Received 28 October 1996; in revised form 11 April 1997     

Stochastic host-parasite interaction models

 We contribute to the discussion of causes and effects of aggregation (overdispersion) of macroparasite counts, focussing particularly upon the effects of clumped infections and parasite-induced host mortality. The simple nonlinear stochastic model for the evolution of the parasite load of a single host, investigated in Isham (1995), is extended to allow three parasite stages (larval, mature and offspring), and to allow durations of these stages to be non-exponentially distributed. As in the earlier work, exact algebraic results are possible, providing insight into the aggregation mechanisms, as long as the only source of interaction between host and parasites is an excess host mortality linearly related to the parasite load. Results are obtained on the distribution of parasite lad and on host survival. In particular, although parasite-induced host mortality is usually thought of as a process that reduces parasite aggregation (Anderson and Gordon 1982), it is shown that, for this model, parasite-induced host mortality cannot cause the index of dispersion to fall below unity. Host heterogeneity and disease control are also discussed. An approximation based on moment assumptions appropriate to a specially-constructed multivariate negative binomial distribution is proposed. This approximation, which is applicable to other processes, and an alternative based on the multivariate normal distribution are compared with exact results. Received: 17 December 1998 / Revised version: 2 June 1999     

Optimal allocation between nutrient uptake and growth in a microbial trichome

 A microbial trichome grows by assimilating nutrients from its environment, and converting these into catalytic macro-molecular machinery. This machinery may be divided into assimilatory machinery and proliferative machinery. The former type is involved in nutrient uptake, whereas the latter type enables the trichome to grow. The cells in the trichome are faced with an allocation problem: given the availability of nutrients in the environment, how many macro-molecular building blocks should be allocated to the synthesis of assimilatory machinery, and how many to the synthesis of proliferative machinery? We answer this question for a particular model, which is a generalization of the Droop quota model. We formulate a two-dimensional non-linear optimal control problem, corresponding to this model. An optimal allocation regime with a singular segment is derived, based on Pontryagin’s maximum principle. We give a direct proof of optimality. We discuss how actual biological cells might implement this optimal regime. Received: 16 December 1996 / Revised version: 14 September 1997     

Population dynamics and competition in chemostat models with adaptive nutrient uptake

 The standard Monod model for microbial population dynamics in the chemostat is modified to take into consideration that cells can adapt to the change of nutrient concentration in the chemostat by switching between fast and slow nutrient uptake and growing modes with asymmetric thresholds for transition from one mode to another. This is a generalization of a modified Monod model which considers adaptation by transition between active growing and quiescent cells. Global analysis of the model equations is obtained using the theory of asymptotically autonomous systems. Transient oscillatory population density and hysteresis growth pattern observed experimentally, which do not occur for the standard Monod model, can be explained by such adaptive mechanism of the cells. Competition between two species that can switch between fast and slow nutrient uptake and growing modes is also considered. It is shown that generically there is no coexistence steady state, and only one steady state, corresponding to the survival of at most one species in the chemostat, is a local attractor. Numerical simulations reproduce the qualitative feature of some experimental data which show that the population density of the winning species approaches a positive steady state via transient oscillations while that of the losing species approaches the zero steady state monotonically. Received 4 August 1995; received in revised form 15 December 1995     

Optimal harvesting of stochastically fluctuating populations

 We obtain the optimal harvesting plan to maximize the expected discounted number of individuals harvested over an infinite future horizon, under the most common (Verhulst-Pearl) logistic model for a stochastically fluctuating population. We also solve the problem for the standard variants of the model where there are constraints on the admissible harvesting rates. We use stochastic calculus to derive the optimal population threshold at which individuals are harvested as well as the overall value of the population in the sense of the model. We show that except under extreme conditions, the population is never depleted in finite time, but remains in a stationary distribution which we find explicitly. Needless to say, our results prove that any strategy which totally depletes the population is sub-optimal. These results are much more precise than those previously obtained for this problem. Received 24 June 1996; received in revised form 7 April 1997     

Discrete chiasma formation models and their associated high order interference

. We introduce some special chiasma formation processes. First a family of discrete chiasma formation processes is introduced and we determine the nature of higher order interference associated with those processes. Secondly we consider a two-stage chiasma formation process, where the associated recombination frequency between two markers depends not only on their map distance but also on their location along the chromosomes. We characterise under this process, in some cases, the nature of interference between two segments. Received: 22 January 1996 / Revised version: 17 September 1997     

Periodic solutions: a robust numerical method for an S-I-R model of epidemics

 We describe and analyze a numerical method for an S-I-R type epidemic model. We prove that it is unconditionally convergent and that solutions it produces share many qualitative and quantitative properties of the solution of the differential problem being approximated. Finally, we establish explicit sufficient conditions for the unique endemic steady state of the system to be unstable and we use our numerical algorithm to approximate the solution in such cases and discover that it can be periodic, just as suggested by the instability of the endemic steady state. Received: 1 September 1995 / Revised version: 30 April 1997     

The use of taurine analogues to investigate taurine functions and their potential therapeutic applications

Summary.  Despite the multitude of evidence for the beneficial effects of taurine supplementation in a variety of disease, the underlying modifying action of taurine with respect to either molecular or biochemical mechanisms is almost totally unknown. We have assessed the development of taurine analogues, particularly where there has been substitution at the suphonate or amine group. Such substitutions allow the investigator to probe the relationship between structure and function of the taurine molecule. In addition such studies should help to ascertain taurine's point of interaction with the effector molecule. These results will prepare the way for the development of the second generation of taurine analogues. Received January 2, 2002 Accepted January 28, 2002 Published online August 30, 2002 Acknowledgements This research has been funded by the COST Chemistry programmes COST D8 “Chemistry of Metals in Medicine” and D-13 “New Molecules for Human Health Care”. All of the authors are members of the Working Group D13/0011/00 “Investigation of mechanisms underlying the pharmacological actions of taurine upon cell apoptosis and calcium homeostasis”. Authors' address: Dr. R.J. Ward, Unite de Biochimie, Catholic Universite de Louvain, Place Louis Pasteur 1, B-1348 Louvain-la-Neuve, Belgium, E-mail: ward@bioc.ucl.ac.be     

A unifying framework for chaos and stochastic stability in discrete population models

 In this paper we propose a general framework for discrete time one-dimensional Markov population models which is based on two fundamental premises in population dynamics. We show that this framework incorporates both earlier population models, like the Ricker and Hassell models, and experimental observations concerning the structure of density dependence. The two fundamental premises of population dynamics are sufficient to guarantee that the model will exhibit chaotic behaviour for high values of the natural growth and the density-dependent feedback, and this observation is independent of the particular structure of the model. We also study these models when the environment of the population varies stochastically and address the question under what conditions we can find an invariant probability distribution for the population under consideration. The sufficient conditions for this stochastic stability that we derive are of some interest, since studying certain statistical characteristics of these stochastic population processes may only be possible if the process converges to such an invariant distribution. Received 15 May 1995; received in revised form 17 April 1996     

Linear stability criteria for population models with periodically perturbed delays

We examine some simple population models that incorporate a time delay which is not a constant but is instead a known periodic function of time. We examine what effect this periodic variation has on the linear stability of the equilibrium states of scalar population models and of a simple predator prey system. The case when the delay differs from a constant by a small amplitude periodic perturbation can be treated analytically by using two-timing methods. Of particular interest is the case when the system is initially marginally stable. The introduction of variation in the delay can then have either a stabilising effect or a destabilizing one, depending on the frequency of the periodic perturbation. The case when the periodic perturbation has large amplitude is studied numerically. If the fluctuation is large enough the effect can be stabilising.     

Overcompensatory recruitment and generation delay in discrete age-structured population models

 The effect of overcompensatory recruitment and the combined effect of overcompensatory recruitment and generation delay in discrete nonlinear age-structured population models is studied. Considering overcompensatory recruitment alone, we present formal proofs of the supercritical nature of bifurcations (both flip and Hopf) as well as an extensive analysis of dynamics in unstable parameter regions. One important finding here is that in case of small and moderate year to year survival probabilities there are large regions in parameter space where the qualitative behaviour found in a general n+1 dimensional model is retained already in a one-dimensional model. Another result is that the dynamics at or near the boundary of parameter space may be very complicated. Generally, generation delay is found to act as a destabilizing effect but its effect on dynamics is by no means unique. The most profound effect occurs in the n-generation delay cases. In these cases there is no stable equilibrium X ^* at all, but whenever X ^* small, a stable cycle of period n+1 where the periodic points in the cycle are on a very special form. In other cases generation delay does not alter the dynamics in any substantial way. Received 25 April 1995; received in revised form 21 November 1995