A demand‐induced overtreatment model with heterogeneous experts

• Author: Marie‐Hélène Jeanneret‐Crettez,Régis Deloche,Bertrand Crettez
• Published date: 01 September 2020
• Date: 01 September 2020
• DOI: http://doi.org/10.1111/jpet.12465

J Public Econ Theory. 2020;22:1713–1733. wileyonlinelibrary.com/journal/jpet © 2020 Wiley Periodicals LLC

|

1713

Received: 18 April 2019

|

Accepted: 5 July 2020

DOI: 10.1111/jpet.12465

ORIGINAL ARTICLE

A demand‐induced overtreatment model with

heterogeneous experts

Bertrand Crettez

1

|Régis Deloche

2

|

Marie‐Hélène Jeanneret‐Crettez

2

1

C.R.E.D., Université Panthéon‐Assas, Paris

II, Paris, France

2

L.I.R.A.E.S., Université de Paris, Faculté

Sociétés et Humanités, Paris, France

Correspondence

Régis Deloche, L.I.R.A.E.S., Université de

Paris, Faculté Sociétés et Humanités, Paris,

France.

Email: regis.deloche@parisdescartes.fr and

regis.deloche@gmail.com

Abstract

We propose an original game‐theoretic demand‐

induced supply model of overtreatment with hetero-

geneous experts. We show that experts, whatever

their abilities, must give the treatment if customers

expect them to do so. We show how launching

awareness campaigns (on the prudent use of treat-

ment) disseminating information to both the general

public and prescribers may, by undermining custo-

mers' expectations, reduce overtreatment. Introdu-

cing a diagnostic test may have the same effect, but

not if its use is costly, especially for good experts. We

rely on our model to account for antibiotics

overprescription.

KEYWORDS

antibiotics, awareness campaigns, credence good, demand‐induced

supply, diagnostic tests, heterogeneous experts, overtreatment

JEL CLASSIFICATION

C72; D82; I11; I12; I18; L15

1|INTRODUCTION

It is common knowledge that antibiotics work against bacterial infections but are not effective

against viral infections. Yet, for years patients have been demanding (and doctors have been

prescribing) antibiotics for viral respiratory infections such as colds and flu. Regarding patterns

of antibiotic consumption (AC) over time and across countries, the results of the study made by

Klein et al. (2018) are clear‐cut. Assessing global trends in 76 countries for both AC expressed in

defined daily doses (DDD) and antibiotic consumption rate (ACR) expressed in DDD per 1,000

inhabitants per day, this report reveals global increase and geographic convergence in AC

between 2000 and 2015. This ever‐growing use of antibiotics, joint to the fact that “the usual

figures show that one out of three prescriptions of antibiotics is inappropriate”(Pulcini

et al., 2017, p. xix), exacerbates the risk of potential resistance to antibiotics.

Antibiotics overuse is an instance of overtreatment or oversupply of a credence good.

Credence goods are goods for which services quality is difficult or impossible for the customer

to ascertain even ex‐post, as opposed to search goods (goods, like clothes, whose quality is easily

evaluated before purchase) or experience goods (goods, like wines, whose quality is difficult to

observe in advance, but can be ascertained upon consumption). To account for the overuse of

certain credence goods (such as medical treatment for patients with respiratory disease or repair

services for cars), we propose an original visit game with heterogenous experts (experts' ability

can be low or high). A distinctive feature of our approach is that expert's prices are the same

and are of secondary importance for both experts and customers. What primarily matters for

expert is customer's loyalty and what primarily matters for customers is the quality of the

treatment, and not its price (the latter being often reimbursed). This is in contrast with many

contributions to the study of credence goods.

We argue that both experts heterogeneity and wrong customers' beliefs are key to account for

overtreatment. To wit, customers expect to receive a treatment whenever the expert has a high

diagnosis ability, and accordingly threaten to switch experts if they do not receive the treatment.

More precisely we show that our visit game has a perfect Bayesian equilibrium in which both high

and low‐ability experts always give a treatment (that equilibrium is sustained by the out‐of‐

equilibrium belief that whenever a customer does not receive a treatment, he believes that his

expert is a low‐ability one). We also show that there is no way high‐ability experts can single out.

That is, there is no pure‐strategy equilibrium in which high‐ability experts never give useless

treatment.

1

In our view, this result goes in long way to explain the rise in antibiotics overuse.

We also consider stewardship programs for managing better use of credence goods. Doing

that, we generalize what is now called “antimicrobial stewardship”in the health care sector,

where this term is “referring to programs and interventions that aim to optimize antimicrobial

use”(see, e.g., Dyar, Huttner, Schouten, Pulcini, & Pulcini, 2017; Pulcini, 2017). It has been well

documented (see, e.g., Pulcini et al., 2017) that two main barriers to optimal antimicrobial use

are (a) the poor knowledge and inappropriate attitudes of both the prescribing doctor and the

patient (patients, for instance, do not know that antibiotics are not useful for viral infections,

and are unaware that misuse of antibiotics could lead to resistance), and (b) the uncertainty

prescribers face when making prescribing decisions (infections commonly have nonspecific

presentations, resembling those of multiple noninfectious diseases, and understanding the

source of an infection is not easy). Two key antimicrobial stewardship actions are aimed at

solving these problems: (a) the repeated launching of public awareness campaigns (on the

responsible use of antibiotics) by national and international health authorities

2

and (b) the ever‐

increasing efforts of these authorities to promote the use by doctors of rapid diagnostic tests.

3

1

Nor is there an equilibrium in which all low‐ability experts never give a treatment.

2

Since the beginning of the 2000s, many countries around the world (see, e.g., chapter 19 in Pulcini et al., 2017) and

major international institutions, such as the World Health Organization (with the World Antibiotic Awareness Week—

WAAW—taking place every November), have launched awareness campaigns aiming both to educate primary care

prescribers and the public about appropriate antibiotic use and to curb unnecessary antibiotic use. For an up‐to‐date

evaluation of these antibiotic awareness campaigns conducted on a national or regional level, see Huttner et al. (2019).

3

For a review of rapid diagnostic tests used by antimicrobial stewardship programs, see Bauer, Perez, Forrest, and Goff

(2014) and chapter 6 in Pulcini et al. (2017).

1714

|

CRETTEZ ET AL.