ArticlePDF Available

A game theoretic approach to repeated foreign body ingestion

A game theoretic approach to repeated foreign
body ingestion
Between January 2010 and February 2014,
a 39-year-old asthmatic woman with
borderline personality disorder, post-
traumatic stress disorder, and a history of
childhood sexual trauma was admitted to
our hospital system 61 times for foreign
body ingestion. During these hospitaliza-
tions, she underwent 62 upper endos-
copies. She also required two exploratory
laparotomies for surgical retrieval of a
pencil fragment and a battery that had be-
come obstructed in the small bowel. She
has swallowed pencils, portions of dispos-
able razors, knitting needles, metal nail
files, toothbrush handles, plastic pen frag-
ments, hair barrettes, disposable cutlery,
and AA batteries. All her admissions oc-
curred late in the evening or on weekends.
During the past 4-year time period, she
also underwent at least 28 additional
upper endoscopies at outside hospitals for
intentional foreign body ingestion. The
patient was assessed as being at high risk
for repeated self-harm. She engaged in
cognitive behavioral therapy, dialectical
behavioral therapy, and electroconvulsive
therapy, but her pattern of repetitive swal-
lowing remained unchanged.
A minority of patients is responsible for
the majority of intentional foreign body
ingestions with substantial economic
burden to hospitals [1, 2]. The interaction
between hospital and patient can be
phrased in terms of a nonzero sum game
between two adversaries [3, 4]. Two strate-
gic options are available to both the hospi-
tal and the patient. In
"Fig. 1 a, the four
outcomes associated with the two options
available to each adversary are arranged in
a two-by-two matrix. Each of the four en-
tries in the matrix contains two numbers.
The first number represents the prefer-
ence ranking of the hospital, and the sec-
ond number represents the preference
ranking of the patient. For instance, in
"Fig. 1 a, the upper left entry 2, + 2 re-
flects the (negative) loss to the hospital
through endoscopy, and the psychological
gain experienced by the patient through
foreign body ingestion and subsequent
medical treatment. The entry 3, + 1 in
the lower left cell represents a worse out-
come for the hospital, if the patient is not
managed immediately, and higher costs
accrue through additional medical adverse
events. The adverse events also diminish
the patients gain associated with a more
risky and complex medical encounter. As
reflected by the upper and lower cells in
the right column of
"Fig. 1 a, without
foreign body ingestion, no costs or benefits
accrue to hospital and the patient. From
the patients perspective, better outcomes
are always achieved by choosing the
strategy of foreign body ingestion irre-
spective of what the hospital subsequently
decides. From the hospitals perspective,
endoscopy always represents the lesser
evil. Therefore, the game between hospital
and patient always ends with the outcome
of the upper left cell marked in red.
The hospital might consider spending
some money on providing the patient
with an incentive not to swallow foreign
objects. For instance, the hospital may
provide the patient with a paid hospital
position to alleviate her economic hard-
ship and social isolation, contingent on
the condition that she stops swallowing
foreign objects. Such a hypothetical sce-
nario is depicted by
"Fig. 1 b with its
changed lower row. If the patient contin-
ues to ingest foreign bodies in spite of the
incentive, the hospital would lose money
from the endoscopy and the wasted in-
centive. The patient may still experience
some psychological benefit from the
medical encounter, but would lose her
monetary reward. Such outcome is reflect-
ed by the 3, 1 entry of the lower left cell
in the revised game matrix shown in
"Fig. 1 b. If the patient accepts the in-
centive as being equivalent or even better
than the endoscopy encounters, the hos-
pital loses some money, but less than with
repetitive endoscopy (dependent on the
expenditure and type of incentive). Such
outcome is reflected by the 1, + 3 entry of
the lower right cell in
"Fig. 1 b.
From the patients perspective, overall
better outcomes are now associated with
a strategy of no foreign body ingestion,
provided that the hospital also adheres to
its strategy of offering an incentive. The re-
vised game between hospital and patient
results in a new steady state marked in
red by the outcome of the lower right cell.
Using game theory and the principles of
contingency management, patients with
repetitive foreign body ingestion could be
offered an incentive to keep them from
swallowing foreign objects [4,5].
Competing interests: None
Sarah J. Diamond, Amnon Sonnenberg
Portland VA Medical Center and Division
of Gastroenterology/Hepatology, Oregon
Health & Science University, Portland,
Oregon, USA
1Frei-Lanter CM,Vavricka SR,Kruger TH et al.
Endoscopy for repeatedly ingested sharp
foreign bodies in patients with borderline
personality disorder: an international sur-
vey. Eur J Gastroenterol Hepatol 2012; 24:
793 797
2Huang BL,Rich HG,Simundson SE et al. In-
tentional swallowing of foreign bodies is a
recurrent and costly problem that rarely
causes endoscopy complications. Clin Gas-
troenterol Hepatol 2010; 8: 941 946
3Straffin PD. Game theory and strategy.
Washington, DC: The Mathematical Associa-
tion of America; 1993
4Sonnenberg A. Personal view: the paradox of
runaway competitions in gastroenterology.
–2, +2 0,0
–2, +2 0,0
–1, +3–3, –1
Fig. 1 Game matrix of the interaction be-
tween hospital (row player) and patient (col-
umn player); abaseline scenario; bscenario
after introduction of incentive.
Cases and Techniques Library (CTL) E79
Diamond Sarah J, Sonnenberg A. Foreign body ingestion Endoscopy 2015; 47: E79E80
Downloaded by: IP-Proxy CONSORTIUM:Harrass (Oregon HealthScience Univ), Oregon Health and Science University. Copyrighted material.
Aliment Pharmacol Ther 2006; 23: 871
5Dutra L,Stathopoulou G,Basden SL et al. A
meta-analytic review of psychosocial inter-
ventions for substance use disorders. Am J
Psychiatry 2008; 165: 179 187
Endoscopy 2015; 47: E79E80
© Georg Thieme Verlag KG
Stuttgart · New York
ISSN 0013-726X
Corresponding author
Amnon Sonnenberg, MD, MSc
Portland VA Medical Center P3-GI
3710 SW US Veterans Hospital Road
OR 97239
Fax: +1-503-220-3426
Diamond Sarah J, Sonnenberg A. Foreign body ingestion Endoscopy 2015; 47: E79E80
Cases and Techniques Library (CTL)E80
Downloaded by: IP-Proxy CONSORTIUM:Harrass (Oregon HealthScience Univ), Oregon Health and Science University. Copyrighted material.
Background and aims: To expedite a consult resolution, referring physicians sometimes inflate the urgency and need for endoscopic workup. The aim of the present decision analysis was to study the impact of inflationary indication on the expected benefits to gastroenterologists and referring physicians. Methods: The study aims were pursued in terms of game theory and medical decision analysis using decision trees. Different outcomes associated with true versus false urgent indication in immediate versus delayed endoscopy were ranked according to different preference schemes of gastroenterologists versus referring physicians. Results: The decision analysis shows that inflating the urgency of indication for endoscopy reduces the benefit from the perspective of gastroenterologists and referring physicians alike. Raising the level of false urgent indications results in a lost opportunity for immediate endoscopy among patients with true urgent indications and, thus, diminishes the overall benefit of endoscopy. By comparison, all other influences play only a marginal role. For referring physicians, the small benefit of expediting nonurgent endoscopies by exaggerated claims does not compensate for the concomitant loss of truly needed endoscopy slots. For gastroenterologists, a small benefit derived from delaying endoscopies in patients with false urgent endoscopies rapidly wears off as inflationary indications become common practice. Conclusion: An underlying communication problem between referring physicians and gastroenterologists needs to be resolved by educating referring physicians about the operative exigencies of endoscopy units and about the true appearance of alarm symptoms in common digestive diseases.
Full-text available
Despite significant advances in psychosocial treatments for substance use disorders, the relative success of these approaches has not been well documented. In this meta-analysis, the authors provide effect sizes for various types of psychosocial treatments, as well as abstinence and treatment-retention rates for cannabis, cocaine, opiate, and polysubstance abuse and dependence treatment trials. With a comprehensive series of literature searches, the authors identified a total of 34 well-controlled treatment conditions-five for cannabis, nine for cocaine, seven for opiate, and 13 for polysubstance users-representing the treatment of 2,340 patients. Psychosocial treatments evaluated included contingency management, relapse prevention, general cognitive behavior therapy, and treatments combining cognitive behavior therapy and contingency management. Overall, controlled trial data suggest that psychosocial treatments provide benefits reflecting a moderate effect size according to Cohen's standards. These interventions were most efficacious for cannabis use and least efficacious for polysubstance use. The strongest effect was found for contingency management interventions. Approximately one-third of participants across all psychosocial treatments dropped out before treatment completion compared to 44.6% for the control conditions. Effect sizes for psychosocial treatments for illicit drugs ranged from the low-moderate to high-moderate range, depending on the substance disorder and treatment under study. Given the long-term social, emotional, and cognitive impairments associated with substance use disorders, these effect sizes are noteworthy and comparable to those for other efficacious treatments in psychiatry.
The general guidelines on the management of ingested foreign bodies (FBs) do not address specific aspects raised by psychiatric patients, particularly in patients with borderline personality disorders (BPD) who repeatedly ingest FBs. The aim of this survey was to collect data on experience and opinions on the management of FBs in psychiatric patients with BPD and to review the relevant literature. A survey focusing on the indication and the timing of endoscopy for sharp FB removal in patients with BPD was e-mailed to 215 gastroenterologists, psychiatrists, and surgeons in Switzerland, Germany and Austria, discussing this clinical problem using a specific case vignette. Responses were received from 63 of 215 (29%) contacted physicians. Two-thirds of the respondents knew patients with BPD who had swallowed FBs repeatedly; 86% recommended removing sharp FBs endoscopically even in the case of repeated FB ingestion and 14% of respondents argued against emergent endoscopic FB removal in the case of repeated ingestions. Different specialities expressed partially divergent opinions regarding the management of these patients. Repeated FB ingestions can be a problem in patients with BPD. Although published data show that the perforation risk of unremoved FBs is low, most clinicians support repeated endoscopies also in the case of repeated FB ingestions. Nevertheless, in selected cases, repeated endoscopies need to be discussed and an interdisciplinary consensus and/or the involvement of an ethical committee is advised.
Guidelines support endoscopic removal of certain gastric foreign bodies (FBs) and all FBs lodged in the esophagus. We studied the poorly understood group of patients who intentionally ingest FBs. Cases of intentional ingestion of FBs (n = 305) were identified, retrospectively, from an electronic endoscopy database and followed. Cases occurred among 33 different patients, who underwent endoscopy from October 1, 2001, to July 31, 2009 (39.0 cases/year); 79% were diagnosed with a psychiatric disorder. Financial cost analysis was performed using hospital billing and cost systems. Commonly ingested FBs included pens (23.6%), batteries (9.2%), knives (7.2%), and razor blades (6.9%). Most endoscopic procedures were performed under general anesthesia. FBs were commonly retrieved by snares (58.0%), rat-toothed forceps (14.4%), and nets (11.5%), assisted sometimes by use of overtubes (10.8%), and hoods (4.6%). FB extraction was unsuccessful at the initial endoscopy in only 20 cases; 2 cases eventually required surgical extraction. Minor complications occurred in 11 endoscopies. There were no deaths or perforations. The total estimated costs were $2,018,073 ($1,500,627 in hospital costs, $240,640 in physician fees, and $276,806 for security services). Costs were significantly higher for inpatients. Major payers were Medicare (48%) and Medicaid (31%). Intentional FB ingestion occurs among a relatively small number of patients with psychiatric disorders and is costly. Endoscopic retrieval is relatively effective and safe, but often requires general anesthesia. These cases utilize significant hospital and fiscal resources. Attention should be focused on preventing these recurrent and costly episodes.
Occasionally, gastroenterologists harbour reservations about the appropriateness of various procedures and management options, yet find themselves drawn into a competition and forced by outside exigencies to provide such services. The aim of the present analysis was to describe patterns of runaway competitions and factors that influence their occurrence in gastroenterology. The interactions between two physicians or physician and patient are modelled as non-zero-sum games of two players. The outcomes associated with two behavioural strategies of two players are arranged in a two-by-two game matrix. Many scenarios of clinical gastroenterology and social interactions among gastroenterologists are characterized by the underlying game of prisoner's dilemma, where two players can choose to co-operate or compete with each other. Although from their joint perspective co-operation results in a higher joint utility, the perspective of each individual favours competition. Following the inescapable logic of their individual perspective, the players end up with a worse outcome than through co-operation. A runaway competition ensues if the game is played repeatedly with players successively increasing their stakes. Ultimately, each player tries to outdo the opponent while heading in the wrong direction. Recognizing the hallmarks of the game of prisoner's dilemma may help gastroenterologists to avoid such competitions.