Pica in individuals with developmental disabilities

Pica is a rare but serious and potentially life-threatening behavior disorder which isquite difficult to treat in individuals with autism and intellectual disabilities ID.Until recently t

Series Editor: Johnny L Matson

Autism and Child Psychopathology Series

Peter Sturmey

Don E. Williams

Pica in

Individuals with Developmental Disabilities

Autism and Child Psychopathology Series

Series editor

Johnny L Matson, Baton Rouge, LA, USA

Peter Sturmey • Don E Williams

Pica in Individuals with Developmental Disabilities

123

ISSN 2192-922X ISSN 2192-9238 (electronic)

Autism and Child Psychopathology Series

ISBN 978-3-319-30796-1 ISBN 978-3-319-30798-5 (eBook)

DOI 10.1007/978-3-319-30798-5

Library of Congress Control Number: 2016932858

© Springer International Publishing Switzerland 2016

This work is subject to copyright All rights are reserved by the Publisher, whether the whole or part

of the material is concerned, speci fically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on micro films or in any other physical way, and transmission

or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed.

The use of general descriptive names, registered names, trademarks, service marks, etc in this publication does not imply, even in the absence of a speci fic statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.

The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made.

Printed on acid-free paper

This Springer imprint is published by Springer Nature

The registered company is Springer International Publishing AG Switzerland

To individuals with pica and their families.

could find little research to guide us in assessing and treating you It took us a long time to provide what you needed Some died prematurely perhaps, and we know you have suffered We apologize for our shortcomings This book represents our latest attempt to improve your clinical treatment and your

recounts a program for people with pica established over 25 years ago at one facility Although an article was published in 2009 describing the program for people with pica, this article has been ignored by some, and criticized by some, but we launched the pro- gram for you and your families and we think

for research, but now we think parents and staff have the right to know what we have done We are sure more criticism will come, but that should not cause you to suffer We

difference between those who acted and those who did not After all,

it was B.F Skinner who said

“caring is…a matter of action.”

One of the most severely challenging aspects of raising an individual with autism isdiscovering the limited number of people who know how to effectively help yourchild The professionals, including pediatricians and doctors, that a parent typicallyturns to for support in a crisis are often ill-equipped to address the deficits of autismand chart a clear path for the parent who is asking:“What do we do next?” Due to

an increased national focus on autism, however, there has been a correspondingexpansion of dedicated research toward treatment and etiology There is now a

evidence-based treatment utilizing applied behavior analysis (ABA) Nevertheless,for individuals with autism who experience severe behavior problems, includingaggression and self-injury, parents mayfind themselves overwhelmed by the risk ofinjury to themselves, siblings, or the child with problem behaviors Ingesting items

of little or no-obvious nutritional value is among the most serious self-injuriousbehaviors with consequences including infections, choking, intestinal blockage, andpossibly death In the presence of low-incidence behaviors that pose a high risk tothe client or staff, parents frequentlyfind the door to help closed with the prospect

of placement outside the home looming large

The possibility offinding an effective treatment for autism and other mental disabilities is more real today than ever—thanks to practitioners of behavioranalysis who continue to undertake the work of isolating and documenting tech-niques that are effective when applied consistently by trained educators and parents.Much of the field’s work in the past 20 years has focused on documenting anddisseminating basic treatment information to ameliorate the key symptoms ofautism seen across a broad swath of the spectrum, including deficits in language,social interaction, self-care, and academics Major cities typically have at least ahalf-dozen or more site-based treatment programs as well as practitioners consultingwith families and school districts to provide treatment that is increasingly subsi-dized by insurance or public funding As a result, many families today have access

develop-to professionals who are familiar with the basic methods of treating the mostcommon deficits Nevertheless, the number of behavior analysts with a depth of

vii

experience treating pica is small, and with the closure of institutional settings, theresponsibility for addressing severe problem behaviors today falls ever moresquarely on families.

This work by Sturmey and Williams is among those that represent the next step

in the behavior analytic treatment literature for autism: works that shed light on howthe field can better address problem behaviors, like pica, that lie at the extremes

of the spectrum The authors gather what is known about past successes and failures

in the treatment of pica and provide direction for researchers and practitioners whomust start on the same page in order to collaborate to effectively treat behavior thatposes severe, often life-threatening danger to clients’ physical and emotional safety

An effective treatment of all problem behaviors, including pica, requires sistency across environments with participation by family members and profes-sionals Because of the limited number of pica cases in most treatment locations,

con-it is essential that behavior analysts in geographically dispersed locales have acommon base of knowledge so that they may jointly move the treatment forwardand collaborate with families for consistency The authors have laid the groundworkfor that collaboration They begin with descriptive information on pica and follow

up with a comprehensive review of the existing literature on studies of pica inautism, many of them single-subject design They also review severalmeta-analyses of the literature This informs the assessment of the function of mostpica as automatic positive reinforcement as well as a hierarchy of the efficacy oftreatment protocols

As parents of individuals with autism, we have identified ABA as the fieldoffering the greatest promise for treating our children with autism We have alsospent inordinate amounts of time learning the concepts and language of ABA tomore effectively implement recommended procedures Nevertheless, ours are likemost families that remain heavily reliant on public schools, therapists, and dayprograms to carry out treatment protocols We are acutely aware of the issues thatmost programs face, including lack of training resources, high client-to-staff ratios,and inadequate supervision by board-certified behavior analysts, all of whichcombine to slow or eliminate our children’s progress—whether as students oradults In addition to recommendations for designing and conducting a treatmentprogram, the authors acknowledge and provide guidance on some of the practicalissues surrounding staffing, staff training, and creating pica-safe environments in asection of the book that describes a 12-year-long pica program with 41 clients in aninstitutional setting The information is highly useful for professionals as well asfamilies evaluating and executing programs for a child or an adult with pica.Over the past 20 years, much scientific research has emphasized the importance

of early intervention for ASD As with all facets of behavioral treatment of autism,early identification and intervention in the treatment of pica behaviors shouldenhance long-term outcomes Unfortunately, the existence of pica behaviors is notalways acknowledged or addressed before it becomes a life-threatening or anendemic issue that isolates the child from environments that support integratedlearning experiences, according to the authors Amidst a range of deficits, parentsand professionals are not attuned to pica as a problem until health suffers or a child

is hospitalized for swallowing a dangerous object Laying a foundation for treatingpica behaviors and teaching safe adaptive behaviors early and in the context ofcommunity and family settings is a significant need highlighted by this volume.The more rapidly research is advanced and disseminated, the quicker targetedtreatment protocols will be widely available to families and service providers Webelieve, as do the authors of this book, that a solid foundation exists for researchingand disseminating the best treatment protocols for pica This volume will be anoutstanding resource in moving those efforts forward.

Lisa Hill Sostack, MBA, Co-founderAmy M Wood, Pharm.D., PresidentFamilies for Effective Autism Treatment (FEAT)—Houston

Pica is a rare but serious and potentially life-threatening behavior disorder which isquite difficult to treat in individuals with autism and intellectual disabilities (ID).Until recently there was little evidence to guide treatment Rather, individualsreceived no treatment or tokenistic and ineffective treatment, or their problembehavior was prevented as best as could be managed by restrictive practices such asrestraints, including fencing masks worn permanently, one-on-one staffing, andlocked in barren environments with no treatment in place The results of suchtreatments were miserable and restricted lives with no positive outcomes andcontinued risks to clients.

Surprisingly, there are at least four previous books on pica Cooper (1957)presented a review of the cultural and medical literature on pica Cooper alsoreported an empirical study of pica in low-income families in Baltimore in the1950s, whose children were at risk for lead poisoning and pica, which focusedmostly on environmental risk factors, such as lead paint, child nutrition, poverty,and lack of child supervision at home This volume mentioned autism and intel-lectual disabilities only in passing Bicknell (1975) published a similar book butwhich did focus on pica in children with autism and intellectual disabilities LikeCooper, she too presented a descriptive longitudinal survey of the characteristics of

15 children with autism and ID and pica The survey searched for potential chodynamic risk factors, but could only conclude that these risk factors wereheterogeneous In some cases child development appeared typical and pica mayhave been the cause of developmental delay due to ingestion of lead One notablefeature of this volume is the mention in passing of the possibility of behavioraltreatment, but with little data available to discuss, there was little to say about it.More recently, Young (2011) published a comprehensive review of pica mostlyfrom a cross-cultural and nutritional perspective, making little mention of pica inindividuals with autism or intellectual disabilities This volume provided a fairlycomprehensive review of the history of pica, pica in literature, and epidemiologicalstudies in pregnant women, children, and certain cultural groups that engage in pica

psy-as part of culture-specific practices Young also attempted to integrate this mass of

xi

cross-cultural data into an integrated theory of pica Finally, Conner (2013)published a brief self-help Kindle edition book on how to treat your own pica,which provides everyday advice on how to treat your own pica, but now which isnot research based.

Both the authors of the present volume were involved in treatment of pica ininstitutional settings and were faced with the challenge of delivering effectivetreatment to groups of individuals almost all of whom were adults with severe andprofound intellectual disabilities scattered across multiple residential settings One

of us (Don Williams) led a team that developed and evaluated a program for

41 individuals with pica over a 9-year period that addressed both reducing picabehavior and safely eliminating restraint and reducing medical risks, such assurgeries for pica

This volume brings together the research literature and our own clinical rience in treating pica Since the publication of Bicknell’s volume on pica inindividuals with autism and intellectual disabilities, behavioral research hasincreased apace In the 1970s and 1980s research developed and evaluated effectiveprocedures using only positive punishment procedures Influenced by the work ofCarr (1977) and Iwata et al (1982/1994), the conceptual framework and relatedbehavioral technology of functional assessment and analysis was subsequentlyextended to assess pica and develop function-based, ideographic behavioral treat-ments In addition, a smaller quantity of research, including a small number ofexperiments, supports the use of dietary interventions for some individuals withpredetermined nutritional deficiencies There are a small number of uncontrolledstudies of various psychotropic medications and other interventions that researchershave not yet evaluated with well-controlled studies Thus, over the past 30 years agroup of evidence-based practices has been developed that may safely reduce andperhaps eliminated pica in some individuals with autism and intellectual disabilities.Over the past 15 years, institutional settings have reduced in number and size inmany (but not all) countries and there is a greater pressure to provide education andadult services in integrated settings Doubtless, many individuals with autism andintellectual disabilities have benefited in many ways because of these changes Yet,geographically dispersed, multi-agency services face many significant challengesand oversight is often very weak leading to unregulated and unsafe school and adultcommunity services that too often rely on restrictive and sometimes unsafe behaviormanagement practices (Sturmey 2015) This review of pica identifies two majorgaps in the research literature and indeed practice First, we lack models of effectiveand safe treatment of pica in contemporary school and adult service settings This is

expe-a serious gexpe-ap in the literexpe-ature expe-and in services expe-as the problem of picexpe-a remexpe-ains expe-aserious and life-threatening one in community services Second, little is knownabout prevention of pica It might not be too difficult in principle to identifyindividuals at risk for pica, such as infants and children with severe and profoundintellectual disabilities and high rates of mouthing and other oral-stimulatingbehavior present over unusually extended periods of time It would be interesting toknow if generic early intervention for young children with severe and profoundintellectual disabilities and autism would be effective in preventing pica merely

through promoting better skills development and increasing the range of socialreinforcers available to young children or whether pica-specific interventions areneeded, for example, to increase alternate adaptive behavior, increase social rein-forcers, reduce potential oral, and/or feeding behavior that might precede thedevelopment of pica and reduce the reinforcement value of engaging in pica at anearly age.

This book comprises three parts Thefirst part, Foundational Issues, addressesbasic concepts such as definition and diagnosis, epidemiology, and theories of pica.This section includes material on pica in a wide range of populations who engage inpica, thereby providing an opportunity to examine the similarities and differencesbetween pica in pregnant women, pica as a cultural phenomenon, pica in individ-uals with psychiatric disorders, and individuals with developmental disabilities Thesecond part, Treatment, focuses on treatment of pica using applied behavior anal-ysis and individuals with ID/autism spectrum disorders (ASD) This sectionaddresses functional assessment and analysis of pica, behavioral interventions forindividuals, and a description of a group program that the second author wasinvolved in for many years Consideration is also given to nonbehavioral treat-ments, such as nutritional interventions, psychotropic medication, cognitivebehavior therapy, and other treatments Thefinal part, Emerging Issues, consists ofone chapter that highlights the future directions for both research and services,especially in the context of contemporary community services

Peter SturmeyDon E Williams

1 Definition and Diagnosis 3

1.1 A Brief History of Pica 3

1.1.1 Some Common Factors 4

1.2 Diagnosis 6

1.2.1 Pica Terms 7

1.2.2 Differential Diagnosis 8

1.3 Culturally Normative Pica 9

1.3.1 Culturally Normative Forms of Pica 9

1.3.2 Pica During Pregnancy 10

1.4 Pica in Clinical Populations 11

1.4.1 Developmental Disabilities 11

1.4.2 Psychiatric Disorders 14

1.5 Pica in Non-humans 15

1.6 Risks 16

1.7 Summary 18

2 Epidemiology 19

2.1 General Populations 19

2.1.1 Clinical Populations 20

2.1.2 Summary 20

2.2 Developmental Disabilities 20

2.2.1 Institutional Settings 20

2.2.2 Total Population Samples 22

2.2.3 Risk Factors 23

2.2.4 Discussion 25

2.3 Other Clinical Populations 26

2.3.1 Sickle Cell Anemia 26

2.3.2 Children with Lead Poisoning 27

2.4 Summary 27

xv

3 Etiology 29

3.1 Learning Models 30

3.2 Neuroanatomical Models 32

3.3 A Dopaminergic Model 32

3.4 Vitamin/Nutrient Deficiency Models 33

3.5 Gastrointestinal Protection Hypothesis 34

3.6 Hunger Hypothesis 35

3.7 Genetic Models 35

3.8 Young’s Biocultural Approach 36

3.9 Behavioral Equivalent of Depression Hypothesis 36

3.10 Psychoanalytic Theory 37

3.11 Summary 37

Part II Treatment 4 Behavioral Assessment and Treatment: An Overview 41

4.1 General Features of Behavioral Approaches 41

4.2 Ethics of Treatment 43

4.3 Behavioral Assessment of Pica 47

4.3.1 Measurement of Pica 48

4.4 Behavioral Interventions Used with Pica 49

4.4.1 Treatment Goals for Behavioral Interventions 49

4.4.2 Risk Assessment 51

4.4.3 Behavioral Interventions for Pica 51

4.5 Summary 63

5 Functional Assessment and Analysis and Function-Based Interventions 65

5.1 Functional Assessment of Pica 65

5.1.1 Observational Functional Assessments of Pica 66

5.1.2 Psychometric Measures of Function 68

5.2 Functional Analysis of Pica 69

5.2.1 Functional Analyses of Pica 70

5.2.2 Response Effort 75

5.2.3 Functional Analysis of Response-blocking Parameters 77

5.2.4 Stimulus Control 77

5.3 Function-based Treatment of Pica 78

5.4 Summary 79

6 Behavioral Interventions: Non-function-based Treatments 81

6.1 Non-function-based Reinforcement-Based Procedures 81

6.2 Antecedent-based Procedures 83

6.3 Positive Punishment 85

6.3.1 Aversive Stimuli 87

6.3.2 Restraint 88

6.3.3 Response Blocking and Response Interruption 89

6.3.4 Overcorrection 90

6.3.5 Negative Practice 93

6.3.6 Abbreviated Habit Reversal 94

6.3.7 Visual Screening 94

6.3.8 Reprimands 95

6.3.9 Idiosyncratic Aversive Stimuli 96

6.3.10 Comparative Studies 97

6.3.11 Food Aversion 99

6.3.12 Summary 100

6.4 Summary 101

7 Behavioral Treatment: Treatment Programs and Outcome Data 103

7.1 Case Series 103

7.1.1 Williams et al (2009) 104

7.1.2 Call et al (2015) 105

7.1.3 Commentary 106

7.2 Systematic Reviews and Meta-analyses 107

7.2.1 Bell and Stein (1992) 107

7.2.2 McAdam et al (2004) 108

7.2.3 Hagopian et al (2011) 109

7.2.4 McAdam et al (2012) 110

7.2.5 Comments 112

7.3 Summary of Evidence 113

8 Prevention, Treatment, and Management of Pica 115

8.1 Prevention Using Crisis Intervention 115

8.1.1 One-to-One Staffing to Temporarily Prevent Pica 116

8.1.2 Continuous Restraint 117

8.1.3 Continuous Restraint Plus One-to-One Staffing 118

8.1.4 Summary 118

8.2 The Program Methodology 119

8.2.1 Pica Survey and Participants 119

8.2.2 Client Rights and Protections 121

8.2.3 Behavior Plans 121

8.2.4 Measurement of Pica 123

8.2.5 Pica Prevention by Environmental Systems Management 123

8.2.6 Administrative Priority Requiring Leadership 126

8.2.7 Determining Risk 129

8.2.8 Outcome Data 130

8.3 Summary 131

9 Biomedical and Other Treatments 133

9.1 Nutritional Interventions 133

9.1.1 Controlled Studies 134

9.1.2 Summary 136

9.2 Psychotropic Medications 136

9.2.1 Summary 138

9.3 Exposure Therapy 138

9.4 Other Psychosocial Treatments 139

9.5 Summary 139

Part III Emerging Issues 10 Future Directions for Research and Services 143

10.1 Future Directions for Research 144

10.1.1 Community-Based Research 144

10.1.2 Early Development and Pica 145

10.1.3 Applied Behavior Analysis 146

10.1.4 Integrating Research Across Populations 148

10.1.5 Integrating Biomedical and Behavioral Research 148

10.1.6 Summary 148

10.2 Treatment Implications 149

10.2.1 Individual Cases 149

10.3 Implications for Contemporary Services 153

10.3.1 Statewide and Regional Planning 158

10.4 Advice for Families 161

10.5 Conclusions and Recommendations 162

References 165

Index 179

About the Authors

Peter Sturmey Ph.D is a Professor of Psychology at Queens College and TheGraduate Center, City University of New York He has published more than 200papers, 50 chapters, and 20 authored and edited books on developmental disabil-ities, behavior analysis, clinical psychology, and evidence-based practice

Don E Williams Ph.D., BCBA-D is a consultant with over 30 years of practiceand research experience with severe behavior disorders, staff training and super-vision, and developmental disabilities He has published extensively on restraintelimination, and the prevention and treatment of self-injurious behavior (picaespecially) and other severe behavior problems

xix

Part I Foundational Issues

Chapter 1

As long as humanity has existed, people have eaten strange things of little or noobvious nutritional value Clarke (2001) presented evidence that Homo Habilis atedirt or clay some two million years ago Many people are familiar with pregnantwomen experiencing food cravings and eating odd substances to satisfy thosecravings Many of us learn in history classes that at times of famine or war, peopleeat wood, tree bark, dirt, sawdust, and other minimally nutritious substances andincorporate them into existing food, such as bread, apparently to stave off hunger.Such practices sometimes make the news when very poor people, such as those inparts of contemporary Haiti, eat dirt cakes or bon bon terres, perhaps because oflack of availability of alternate food sources, to avoid hunger, and to provide someminimal nutrition from the content of the clay

Less widely known are culture-specific religious practices For example, forsome Catholic and other religious groups, eating sacred earth is said to be imbuedwith healing, magical or quasi-magical properties This occurs as part of the cult ofOur Lord of Esquipulas, Guatemala, where believers take sacred earth, rub it onthemselves, consume it in water, or take it back home This practice has spreadfrom Guatemala as far north as Chimayo, New Mexico, where over 30,000 pilgrimsconsume 25–30 tons of dirt a year, sometimes walking 90 miles from Albuquerque

to do so The church in Chimayo might have been built on a location where TewaIndians used a sacred spring ascribed with healing powers, so perhaps such culturalpractices are specific examples of Christian colonists adapting indigenous practices

to Catholicism

Similar culture-specific forms of pica occur when there are outbreaks of fads forpica, such as consumption of large quantities of solid starch, ice, or clay amongcertain groups within a society (Cooper 1957; Young 2011) Today, some indi-viduals with pica for starch have now become YouTube starlets, pleading for helpwhile eating starch from boxes for all to see Such examples of pica are interesting

© Springer International Publishing Switzerland 2016

P Sturmey and D.E Williams, Pica in Individuals with

Developmental Disabilities, Autism and Child Psychopathology Series,

DOI 10.1007/978-3-319-30798-5_1

3

behavioral phenomena that require description and explanation Two previousvolumes on pica (Cooper 1957; Young 2011) have described them fully.

Researchers have also reported individual cases of pica among individuals withpsychiatric diagnostic criteria such as dementia, obsessive–compulsive disorder,and schizophrenia These case studies are usually more dramatic and are often farmore serious and dangerous forms of pica than the preceding examples (Formalpsychiatric diagnostic criteria for pica, such as those from DSM and ICD, usuallyexclude these culturally appropriate forms of pica.)

This book, however, will focus on more serious forms of pica which are foundamong individuals with developmental disabilities, such as autism spectrum dis-abilities (ASD) and intellectual disabilities (ID) Only, Bicknell’s book from themid-1970s has addressed pica in this population This was published a long timeago before most of the research on assessment and effective treatment of pica hadbeen conducted

These clinical forms of pica share at least four features with other non-clinical forms

of pica Afirst similarity is that sometimes the person with pica appears or reports to

be obsessed with eating certain substances, perhaps excluding other substancesfrom their diet Thus, their preferences for the pica items are often highly specificand highly motivated Individuals who eat clay only eat certain kinds of clay andwould never eat dirt or other forms of non-preferred types of clay! Some peoplewho engage in pica with ice report only craving certain kinds of ice and will travelmiles to purchase the preferred kinds or even purchase expensive ice machines ofthe preferred brand, even when other non-preferred types of ice are readily andmore conveniently available People who eat solid starch have strong preferencesfor specific brands of starch and for specific aspects of physical texture and taste.They would never eat a non-preferred brand of starch or drink starch dissolved inwater! This strong preference for specific pica items is similar to the strong andspecific preferences that individuals with ASD and/or ID and some psychiatricpatients with pica have for specific items (Piazza et al 1996) For example, anindividual with ASD, ID, and pica might search for certain kinds of string and rejectall other apparently similar items, but would never eat a cigarette, whereas othersonly eat cigarettes and would never eat string Some psychiatric patients with picaseek out certain kinds of metallic items, but would never eat string or cigarettes

A second similarity is that it seems that certain physical or sensory properties ofthe pica items are highly important to the person Young (2011) noted that manycommonly consumed pica items have a somewhat bland or neutralflavor or leave asomewhat metallic or mildly acidicflavor in the mouth after consumption such asmight be experienced by those who eat cornstarch, baby powder, chalk, or certainkinds of clay The physical texture of the pica items also seems highly character-istic, sometimes grainy, as in certain kinds of clay, or with specific crunchy or other

mechanical properties Young also emphasizes that many people engage in picasecretively and away from other people, perhaps out of shame, again suggestingthat pica is more of a nonsocial/sensory activity, although the Internet now providesample social networks for almost all minority interests, including people whoengage in amylophagy (starch pica) (Young 2011) In a similar vein, functionalanalyses of pica behavior in individuals with ID and/or ASD indicate almost uni-formly that pica behavior is insensitive to social consequences In contrast, othermaladaptive behaviors, such as self-injurious behavior (SIB), are maintained by avariety of social and nonsocial consequences, whereas pica behavior is almostalways maintained by automatic positive reinforcement, i.e., the reinforcing sensoryconsequences of consuming the pica item.

A third similarity between pica in the general population and pica in individualswith ID/ASD is that the pica items appear to be very powerfully positively rein-forcing for a range of pica-related behavior Thus, individuals with pica in thegeneral population may spend large amounts of time and money on pica-relatedbehavior They may think“obsessively” about their favorite pica item, read about it,and search for information from friends, family, and the Web People who engage

in pagophagy (ice pica) spend much of their day thinking about ice, planning trips

to the store to buy ice, and spending money every day on ice They eat ice eventhough they injure themselves by wearing down their teeth We sometimes describethis by saying the person is “obsessed” with the pica item and that they areapparently“compelled” to engage in pica In a similar manner, a person with ASDand/or ID is often described as“obsessed” with cigarettes, twigs, or eating threadsfrom clothes They may be placed on one-to-one staffing, or restrained mechani-cally in an attempt to prevent pica-related injuries, such as choking or gastroin-testinal damage These strategies, which deprive the person of the opportunities toengage in pica, only appear to increase the individual’s motivation to engage inpica The moment the one-to-one staff turns their back or the moment the personcan work their way out of mechanical restraint, the individual may bolt for thenearest location where a pica item may be and consume it immediately Thus, anindividual with ASD/ID and pica engages in the behavior despite some immediateunpleasant consequences

Afinal similarity is that both groups of people engage in pica for its short-termbenefits despite its long-term harms Thus, some pregnant women eat dirt andexpose their fetuses to parasites, others eat starch and put on weight with long-termhealth costs, and others knowingly damage their teeth when eating ice Individualswith ASD/ID engage in pica but struggle with staff to obtain the item and injurethemselves when attempting to obtain the item and its long-term harm

Although similarities exist across pica in typically developing people and thosewith ASD and/or ID, important differences do indeed exist Some forms ofculture-specific practices, such as pica-related to religion, pregnancy, social conta-gion, such as groups interested in amylophagy, appears to be different to pica inindividuals with ASD and/or ID That is, these forms of pica are probably highly

influenced by socially mediated learning processes, such as modeling andrule-governed behavior These people may imitate other people’s pica behavior; for

example, they may observe family members and friends engaging in religious-relatedpica or observe other pregnant women engaging in pica They may receive directinstruction from others to engage in pica,“go ahead try it, you might like it,” anotherpregnant woman might say They may also receive indirect forms of instruction.For example, a religious person may here that“True Catholics believe in the power ofOur Lord of Esquipulas” or “Eat this dirt I brought you back from Chimayo It is said

to be very powerful.” In such examples, the person never comes in direct contactwith contingencies—they never consume the dirt and feel better—but their previoushistory of reinforcement for following instructions and rules related to religiouspractices influences a more general class of religious rule-governed behavior Clearly,not only the topography and pica items are quite different between these populations,but also the nature of pica is quite different Thus, other people are important in theacquisition and maintenance of these forms of pica in the general population, whereas

in other forms of pica in individuals with ASD/ID, this may be less true

it coexists with other disorders, such as ID and/or ASD, and then it must warrant itsown attention The ICD-10 diagnostic criteria for pica are broadly similar, and bothcriteria are applicable to both adults and children

There were relatively few changes in DSM criteria for pica when the APA revisedDSM-IV-TR to make DSM-5 (Hartmann et al 2012) For example, the AmericanPsychiatric Association (2013) classified pica as a feeding and eating disorder,alongside binge eating disorder, anorexia nervosa, and bulimia nervosa, whereas inDSM-IV-TR it had been classified as a disorder usually first diagnosed in infancy,childhood, or adolescence Thus, the revisions to DSM-IV criteria appear relativelyminor in that they only clarified that pica could be diagnosed at any age

The new DSM-5 criteria for pica require a reassessment of the diagnosis of picafor several reasons First, both children and adults can now be diagnosed with pica.Formally, pica was only diagnosed as a disorder which usually onsets in childhoodand adolescence Onset of pica can occur in adulthood, for example, both in someindividual with ASD and/or ID and in various other forms of pica; the new DSM-5

definition now permits such diagnoses with adult onset Thus, there may be adultswho should now be diagnosed with pica

Young (2010) discussed various forms of eating unusual items She noted thatalthough DSM diagnostic criteria were more operationalized than previous defini-tions, they failed to capture what for her is one of the key features of pica, namelyintense craving for the pica item She argued that when people eat earth and clay attimes of famine, they do so out of necessity and because of lack of availability ofalternate foods, but without craving for the substance Similarly, some culturallyappropriate forms of pica, such as religious-based pica, are due to local social cus-toms and practices, rather than intense cravings for specific pica items She arguedthat both of these forms of unusual eating are not true pica because the forcefulcraving for the item, akin to a drug addiction, is absent Therefore, she proposed thatpica is “the craving and purposeful consumption of substances that the consumerdoes not define as food for >1 month” (p 405) This refinement is interesting because

it not only captures something about pica in the general population that is missingfrom DSM and other definitions, but also captures something of the quasi-obsessionalaspects of pica in individuals with ASD/ID

Potential problems with the reliability and validity of different diagnostic criteriafor pica come from data published by Cooper et al (2007) Although this study was

an epidemiological study rather than a formal study of the reliability of the diagnosis

of pica, some of theirfindings hint at this problem In a large-scale study conducted inGlasgow of 1023 individuals with ID aged over 16 years, they reported that theprevalence of pica ranged from 0.0 to 2.0 % depending upon the diagnosis andmethods used While these differences are small with respect to the absolute differ-ences in magnitude, the relative differences are large Given the difficulties inestablishing reliability of low-frequency diagnoses, these data suggest that this mattershould receive more attention in formal reliability and validity studies

Cooper (1957) listed a number of historical terms that have been used to refer topica These include citta, malacia, mal d’estomache, erdessen, Cachexia Africana,allotriophagia, and geophagy, and others have used terms such as parorexia(Ruddock 1924) There are several terms that combine a prefix to indicate the type

of pica item consumed with the suffix -phagy Thus, McAdam et al (2012) dividedthe physical classes of materials associated with pica into six classes These were(1) biologic secretions including copraphagia (feces), vomit (emetophagia), blood(hematophagia), mucous (mucophagia), and urine (urophagia); (2) biologic solidsincluding dermatophagia (skin), fingernails (onychophagia), bone (osteophagia),and hair (trichophagia); (3) chemicals including cuprophagia (copper), pharma-cophagia (pharmaceuticals), and lead chips (plumbophagia); (4) food stuffsincluding geomelophagia (potatoes), gooberphagia (peanuts), lectophagia (lettuce),and oophagia (eggs); (5) organic materials including amylophagia (laundry starch),bibliophagia (book pages), coniophagia (dust), foliophagia (acorns, grass, pinecones, leaves), geophagia (dust, sand, clay), and pagophagia (ice, freezer frost); and

(6) physically damaging materials including acuphagia (sharp items), reiophagia (matches), hyalophagia (glass), lignophagia (bark, twigs), tobaccophagia(cigarette butts), and xylophagia (wood) Others have unnecessarily gilded thedictionary with the terms sapophagia (soap; Saddichha et al 2012) and, perhapsfacetiously, jumperphagia (Jumpers!; Nash et al 2003).

cautopy-The meaning and functional value of such distinctions—such as those betweeneating leaves rather than twigs—are unclear and may represent little more thanpseudoacademic neologia nervosa or perhaps even logophagia! A more importantfunctional distinction might be between those forms of pica where some aspect oftexture and/or taste might be the important automatic reinforcer maintaining picabehavior and those forms of pica where a substance, such as caffeine or nicotine,might be the important consequence, as this distinction has important implicationsfor indicated and contraindicated behavioral treatments

An accurate and swift diagnosis of pica is essential in order to avoid delay intreatment As some forms of pica are highly dangerous and indeed lethal, failure tomake accurate diagnoses and provide evidence-based treatment is unethical (seeBox 1.1) Failure to make an accurate diagnosis also places people who areunderdiagnosed at a higher risk of harm because they will receive no treatment planfor pica, thereby unnecessarily exposing them to preventable risks, which isincompatible with professionals’ ethical obligation of beneficence to their clients.Thus, failure to diagnose pica or deliberate misdiagnosis (i.e., a false negative)would be a serious oversight

Box 1.1 Ethics and Accurate Diagnosis of Pica

“We do not have any cases of Pica here” stated the medical director andnursing staff at a residential school and center for individuals with ASDand/or ID If the matter had been left there, nothing would have happened;indeed, a review of the medical and nursing records would not have revealedany diagnoses of pica

A review of injury and incident data, restraint records, informal interviewswith staff of individuals who had been restrained extensively, however,quickly revealed a number of individuals who had been restrained for months

or longer to prevent them engaging in pica None had treatment plans toaddress pica, even though some had choking incidents and other medicalcomplications from pica incidents

When the interviewer confronted medical and nursing staff with these factsthey merely looked away Later some explained quietly that if the diagnosis

of pica was“not on the books,” treatment was unnecessary

Pica may be confused with some other DSM-5 disorders which may have sometopographical and functional similarities For example, hand mouthing and chronicmouthing of objects have some topographical similarities that might be confusedwith pica Hand mouthing and chronic mouthing of objects also have some func-tional similarities in that they are likely to be behaviors that are nonsocial behaviorsthat are automatically positively reinforced in many cases (Goh et al 1995; Piazza

et al 1996)

Pica, however, can be distinguished from hand mouthing and chronic mouthing

of objects because in the case of pica (a) there is a clear focus on specific objectsrather than the hands and mouth, and (b) the items are often consumed rapidlyrather than repeatedly mouthed Hartmann et al.’s (2012) discussion of DSM-5criteria for pica noted that pica might also overlap and/or be confused withavoidant/restrictive food intake disorder (ARFD) in that ARFD might involveintense focus on a few food items This is perhaps not a common confusion as picainvolves non-nutritive substances, whereas ARFD involves restricted interest inactual food items

As noted earlier, some forms of pica are viewed as culturally typical behavior.These include both long-term cultural practices that may have gone on for gener-ations and more recent social contagions in which many people within a societybegin practicing pica such as outbreaks of consumption of clay, starch, ice, or othernon-nutritive substances, among certain populations in specific geographic areas Infact, pica as a cultural practice has been recognized historically for over twothousand years

Both Cooper (1957) and Young (2011) provide numerous such examples InCooper’s history of pica, it is notable that children and individuals with ASD and/or

ID and individuals with mental health problems feature little until the late teenth century Rather, many older historical accounts focus on unusual eatinghabits in pregnant women and cravings for unusual or strong tasting foods Duringthe period of European colonization of the Americas, Africa, and Asia, Europeanswrite accounts of cultural practices that were unfamiliar to Europeans, such aseating clay near river areas in South America and Africa, and eating dirt in AfricanAmerican slaves and poor Whites in Roanoke It was only in the late nineteenth andearly twentieth centuries that quasi-systematic surveys and medical reportsappeared on pica in children, sometimes in relation to lead poisoning or other healthrisks Presciently, Verga (1849 cited in Cooper 1957) reported a case of pica in a

child with ID and commented that “idiots and mental patients eat bizarre thingsbecause they don’t know what else to do with the things they pick up” (Cooper

1957, p 34) Such comments foreshadow the rationale for behavior analytictreatments such as teaching how to engage in alternate behavior with pica items,such as turning in pieces of string or putting them in a trash can and discriminatingbetween food and non-food items (see Chap.4, this volume)

Young (2011) provides numerous examples of cultural practices related to picaand has assembled a pica database of 367 reports of geophagy and related phe-nomena spread all over the world For example, in the USA, several commercialproducts, such as starch, clay, and earth from graveyards, are commerciallyavailable, inter alia, for consumption The use of starch in the form of laundrystarch or cornstarch power is relatively well known by the manufacturers of theseproducts Periodically, they modify the packaging to add“not recommended forfood use” or change the product’s form, so it is less appealing to consume.Nevertheless, it is relatively easy tofind material on YouTube on this form of picaand tofind information on cornstarch pica support groups on the Internet.There are also several religious groups that ingest clay and dirt forquasi-religious purposes including Christian, Muslim, Hindu, and other religiousgroups For example, in Hoodoo, a Voodoo-like religion, dirt mixed in hot water isused to hasten childbirth, soothe a troubled mind, and offer general protection.Graveyard dirt is sometimes used to punish adultery or bring bad luck to an enemy

Pregnant women often report strong cravings for specific odd foods or non-nutritivesubstances and may eat dirt, coal, clay, etc Sometimes they also report relatedolfactory cravings Such phenomena have been known for at least two thousandyears (Cooper 1957)

Young (2011, Appendix B) reviewed 47 studies of the prevalence of pica duringpregnancy The most commonly craved items were earth, starch, and ice, but alsoincluded charcoals and paper, chalk, and ash Prevalencefigures varied widely from

8 to 71 %, presumably reflecting factors such as population studies and ology used A recent systematic review of micronutrient deficiencies and picaduring pregnancy (Miao et al 2015) revealed that pregnant women who wereanemic had an increased risk of pica (OR = 1.92), although other biologic andsociocultural factors are probably also important in determining pica duringpregnancy

method-Such practices are generally time-limited and usually carry moderate risks,although sometimes the risks to mother and fetus may be more significant Risksinclude (a) parasite infections from consumption of soil and clay; (b) heavy metaltoxicity, if pica involves soil or other items with lead, copper, or other heavy metals

present and depending on the pica items consumed; and (c) risks of gastrointestinaldamage It is uncertain whether such behavior may be physiologically adaptive insome ways, although on some occasions such risks may be significant both to themother’s and to the child’s health (see below).

Some people with ASD and/or ID may engage in pica Unlike the culturallyappropriate forms of pica discussed above, pica in this population does not focus onthe selection of specific forms of clay, ice, or starch Rather, it often involvesspecific items, such as cigarette butts, grass, leaves, twigs, string, threads, paper,and specific small objects, such as buttons, dice, paper clips, and trash Forexample, Matson and Bamburg (1999) identified 45 individuals with pica in aninstitutional setting The pica items identified were cigarette butts (N = 28), paper(N = 11), clothing (N = 3), fecal materials (N = 2), and linen/towels (N = 1)

It appears that pica is more common among individuals with more severe ID andwho are nonverbal (Ashworth et al 2008) For example, in a large sample of over

2200 individuals with ID in a geographically defined area, Smith et al (1996) foundthat the prevalence of pica increased monotonically with degree of ID Theyobserved that the prevalence of pica was 0.0, 0.4, 0.1, and 3.2 % in individuals withmild, moderate, and severe/profound ID (p < 0.001) Tewari et al (1995) found that

of 25 individuals who displayed pica in an institutional setting, none had mild ID, 4had moderate, but 21 had severe/profound ID

Tewari et al described that their individuals with pica did not display nalizing behavior problems, such as aggression and tantrums Rather they describedthem as “withdrawn, amenable, and submissive,” perhaps reflecting the lack ofsocial skills described more systematically in studies below Matson and Bamburg’s(1999) reported similar data in a case series of 45 individuals, which included

exter-30 men and 15 women of whom 39 had been identified with profound and 6 withsevere ID (it is possible that there were individuals with mild ID in communitysettings that were not in this institutional sample) Matson and Bamburg alsoreported that 9 of these individuals also had autistic disorder and 15 had stereo-typical movement disorder Matson and Bamburg compared individuals with picawith individuals without pica on the MESSIER, a measure of social skills Theyfound that individuals with pica had fewer positive social skills than those withoutpica, but found no differences in terms of overall negative social behaviors betweenthe two groups Matson et al (2012) reported similar results when they comparedthe social skills of three groups of individuals with ID: those with ID alone(N = 22), those with ID and ASD (N = 22), and those with ID, ASD, and pica(N = 15) They found that the group with ID, ASD, and pica had fewest positive andmost negative social skills and behaviors Thus, pica tends to be more common inthose individuals with ID and/or ASD with more severe disabilities

An interesting study of the social correlates of pica in individuals with ID ininstitutional settings was published by Ashworth et al (2009) who conducted asurvey of pica in 1008 institutionalized individuals with ID in Ontario, Canada, andits social correlates Using multivariate analyses to control for potential con-founding variables, such as level of adaptive behavior, they found that individualswith pica were approximately 1.5–2.3 more likely to lack family contact and socialcontact and do not participate in day programming compared to individuals withoutpica; however, they were not more likely to experience interpersonal conflict thanthose without pica Again, these observations are consonant with earlier studiesshowing lack of skills generally and social skills in particular as correlates of pica.The authors noted that emphasis should be placed not only on the health correlates

of pica, but also on its social consequences Tewari et al (1995) also described thattheir participants with pica had little daytime activity in an institutional setting andonly 2 of 25 had any behavior management plan of any kind (they did not comment

on the content or quality of the two plans)

Earlier studies of the prevalence of pica show that the majority have severe orprofound ID (Griffin et al 1986), which may offer another explanation for some ofthe above research Individuals with severe or profound ID generally learn at amuch slower pace than those with moderate and mild ID The study by Griffin et al.was assessed on a statewide basis with a population of approximately 10,000individuals with ID/ASD The fact that pica exists in community settings indicatesthat it is not an artifact of institutions, although it existed there at higher rates in the1980s and 1990s

Several factor analytic studies of different versions of the Behavior ProblemInventory (BPI; Rojahn et al 2012) have reported the association of pica with otherchallenging behaviors For example, Matson et al (2012) found that pica loadedmoderately onto the self-injury rather than stereotypy or aggression/destructionscales Thus, pica was associated with other SIBs, such as self-scratching,head-hitting, and hair pulling rather than stereotypical behaviors such as handmovements or aggressive/destructive behaviors such as hitting or scratching others.Other factor analyses of the BPI have reported similar data (Mascitelli et al 2015;Rojahn et al 2012; Sturmey et al 1993, 1995)

The severity of pica varies from a mild to severe/life-threatening problem.Sometimes pica is relatively low risk and can be managed with routine supervision,redirection, some skills training, and routine behavior support plans Other times itcan be life-threatening and may involve intrusive and sometimes risky methods ofmanagement, such as one-to-one staffing; mechanical restraints, including armsplints, and fencing masks; and restrictive behavior management interventionsincluding positive punishment

Some cases are treatment responsive, although they may require continuing andextended behavioral or other treatment, some cases may be highly treatmentresistant and may only respond to restrictive programming, and some may be

unresponsive to restrictive programming, despite ongoing efforts to eliminate suchpractices Because the frequency of pica is relatively low, especially for severeforms of pica (see below), clinicians and services often have limited experience andskills in managing the more severe forms of pica, thereby placing some individuals

at risk of continued restrictive management practices, injury and injury and death.One empirical study on the severity of pica comes from Jacobson (1982b) who,

as part of a larger study on the epidemiology of challenging behavior, includingpica (see Chap 2), asked a sample of 56 doctoral-level psychologists workingwithin the New York State Developmental Disabilities agency to rate the severity of

26 challenging behaviors“in terms of the extent to which they would pose a barrier

to future community placement for a person living in a developmental center”(p 374) Their ratings were converted to a 10-point scale where 10 represents themost severe challenging behavior

The most severe items were physical assault andfire setting or attempts to setfires were rated 10 and property destruction was rated 9 Least problematic werelack of appropriate affect and perseveration (both rated 2) and echolalia (rated 1).Pica, along with substance abuse and tantrums, was rated 5, indicating that onaverage, it is a moderately ranked barrier to community placement Such ratings ofcourse hide large individual differences

A second study comes from Williams (2015) who developed an 5-point SeverityIndex for Pica This was used in clinical settings for determining risk Figure 1.1

describes the Severity Index for Pica in detail The second author has found thisscale clinically useful, although there are no data evaluating its usefulness sys-tematically at this time

1 Mild Mouths objects and has swallowed small pieces of paper or strings

without and passed with no difficulty known

2 Moderate Mouths objects and has swallowed small pieces of paper or strings or other items considered non-dangerous in small amounts Has experienced one or two incidents of coughing up items

3 Severe Mouths objects and has swallowed small pieces of paper or strings or

other items considerednon-dangerous in small amounts Has experienced one or two incidents of choking and coughing up items Has also had X-rays to rule out pica on more than one occasion

4 Dangerous Ingests foreign object during probes at least weekly History shows several X-rays and documented ingestion of foreign objects considered dangerous (screws,bolts,jewelry,metalcoins)

5 Life-threatening Has had one or more surgeries for the removal of foreign objects and continues to engage in pica at least once every 30–90 days during probes

Fig 1.1 Severity index for Pica (Don E Williams, Ph.D., BCBA-D @ 2015) Note This is

one tool that may be helpful but should not be the sole determinat or of stafing, programs,

or dangerousness

1.4.1.2 Summary

Among individuals with developmental disabilities, pica is typically seen amongindividuals with severe or profound ID who have poor social skills and who arewithdrawn, isolated individuals Generally, pica is a moderately intense manage-ment issue, but varies considerably from relatively easily managed to very chal-lenging and life-threatening

Pica in individuals with serious mental problems and dementia is reported rarely.For example, Tracy et al (1996) found that only 4 % of 400 in patients withschizophrenia displayed pica As with individuals with ASD and/or ID, in aminority of cases, pica may be highly dangerous and life-threatening or has to bemanaged using restrictive behavior management practices For example, Nash et al.(1987) reported treatment of pica in a 70-year-old man with organic brain syndromeand schizophrenia using contingent restraint

Sometimes pica is described as being similar to obsessive–compulsive disorder(OCD) For example, Baheretibeb et al (2008) described a case study in which picabehavior appeared to follow recurrent, unwanted intrusive images of mud andconsuming mud and that engaging in pica was followed by decreases in anxiety.This pattern appears to be broadly similar to that in more typical cases of OCD.Aksoy et al (2014) also reported a case study of pica for clay and ice in a54-year-old woman with a 35-year history of pica She also presented withchecking and washing and iron deficiency

Similarly, Zeitlin and Polivy (1995) reported assessment and treatment of anindividual of above-average intelligence (full-scale IQ = 112) with coprophagyconceptualized as a case of OCD This individual had a history of eating dirt androcks as a young child and a history of severe abuse in which his father, whoreportedly had sexually assaulted him, repeatedly forced him to eat feces as pun-ishment His mother died when he was young, his father subsequently committedsuicide, and his foster mother also died of a heart attack On several occasions, hehad been told that he was responsible for these deaths, that he was bad, and that hewas responsible for harm to others through his carelessness He began to ruminate

on thoughts of these events and at periods of intense stress ate his own feces whichresulted in reduction in anxiety Six months before referral, the frequency of these

problems increased and he put away any cues that reminded him of his family andrestricted his eating, especially of bran, to reduce the likelihood of coprophagia Theauthors diagnosed him with OCD and subsequently treated his coprophagia usingbehavioral methods for OCD (see Chap.9 for details).

Commenting on diagnostic issues in this case, Luiselli (1996) reported that hehad observed two clients with severe/profound ID who are nonverbal and whoengaged in pica involving ingestion of paper, plastic, cloth, wood, and metalobjects Although they had received several behavioral treatments, which aresupported by research (see Chap.7), including response interruption and preven-tion, positive reinforcement, and free access to food, they did not respond to thesebehavioral treatments Luiselli described that they were “‘hypervigilant’ to theirsurroundings, touching surfaces, and moving rapidly without purpose… seemed

‘anxious’ and ‘overaroused’ to most practitioners who worked with them …appeared to‘perseverate’ on their physical surroundings and were easily distressedwith environmental change” (p 195) Thus, Luiselli made the case that someindividuals with ID/ASD and pica might also be diagnosed with OCD and mightperhaps respond to medical treatments, such as specific serotonin reuptake inhibi-tors (SSRIs) (They might also benefit from behavioral treatments for OCD, such asgraded exposure to anxiety-provoking triggers See Chap.9) Thus, several authorshave noted the parallels between pica and OCD in some cases, both in individualswith ID/ASD and in psychiatric patients

Domestic animals also engage in problematic behavior related to eating orattempting to eat unusual substances For example, cows may lick, chew, andeventually eat nails and wire Such behavior may be dangerous to the animals,costly to the owners, and may represent an adaptive response to nutritional

deficiencies

1.6 Risks

Pica in the general population carries some risks which are usually moderate Theseinclude risks of infections from eating soil and clay, possible teeth damage fromeating ice, and some mild to moderate digestive problems that might arise fromeating excessive quantities of starch or other unusual substances

In some children, risks from pica are more significant, such as those that mayarise from eating lead-based paints and chronic geohelminth infections; however, inclinical populations, the risks may be much greater For example, in individualswith developmental disabilities, pica may carry risks of infections, transmission ofhepatitis, damage to the gastrointestinal system from ingested objects, chokingincidents, and surgeries to remove intestinal blocks (which themselves may carrysignificant risk of death in some individuals) and objects lodged in the gastroin-testinal system Sometimes phytobezoars, consisting of seeds and other organicmatter, and trichobezoars, consisting of hair, may develop which cause gastroin-testinal blockage These may have to be removed from the person’s gut for theirsafety, although the surgery itself comes with risks (see below)

Pica carries with it the risk of heavy metal poisoning, the most common form ofwhich is lead poisoning This risk has been long recognized mostly from earlysurveys of children from low-income families who engaged in pica and wereexposed to lead paintflakes (Cooper 1957; Ruddock 1924) and other sources ofindigestible lead, such as earth contaminated from mining waste of otherlead-containing pollutants, paper, chalk, glazed pottery, and powder (Young 2011).The risk of lead poisoning is also shown by high rates of pica among individualswith lead poisoning (Young 2011) Pica for lead items in pregnant mothers can alsoresult in medical problems in the pregnant mother (Trivedi et al 2005) and leadtoxicity in the newborn baby (Estrine 2013; Hamilton et al 2001) Although longrecognized as a risk of pica, research continues to this day on the effects of pica incontemporary at risk populations, such as children in mining areas or those exposed

to new environmental sources of lead Other forms of pica-related poisoning havealso been reported For example, Kupiec et al (2004) reported a case of a boy taken

to hospital because of an unexpected seizure He had engaged in pica with rock saltleading to sodium toxicity Some cases of cigarette pica have also involved toxicityfrom the nicotine (Evans 1989)

Sometimes pica causes medical problems, such as general gastrointestinal tress, constipation, abdominal pain, and gastrointestinal injuries and non-specificbehavioral and medical problems (Al-Busairi and Ali 2003; Rashid et al 2010) Inindividuals who are nonverbal, these medical problems may be difficult to identifyand it may be difficult to observe low-frequency events, such as consumption ofpica articles, making recognition of pica-related medical problems challenging.They may require medical treatment such as removal of the pica items from the gut(Gulia et al 2007; Halleran et al 2015) or placement of tubes to permit feedingwhen the intestine is blocked and the blockage cannot be removed (Miyakawa et al.2011)

dis-Surgeries and choking incidents carry a risk of death, and there have been anumber of cases of pica-related deaths due to choking and surgeries McLoughlin(1988) reported three pica-related deaths in men with severe/profound ID In their10-year follow-up of treatment of pica with overcorrection, Foxx and Livesay(1984) reported that 3 of the 4 children had died whereas all four other childrentreated with overcorrection without pica were still living Jancar and Spellar (1994)reported other cases and noted that of those who died, few had previous historiesthat obviously indicated the severity of the problem, in that few had had previoussurgeries and their final illness was usually only observed during 24 h prior todeath Byard (2001) reported the case of a nine-year-old boy, who, after a shorthistory of diarrhea and vomiting, collapsed and died as a result of bowel obstructiondue to pica Kamal et al (1999) reportedfive cases of pica in which the adults with

ID had consumed vinyl gloves In four cases, surgery was required to remove thegloves which could not be removed endoscopically, and one individual died fromthe effects of consuming the vinyl glove Pica has also resulted in deaths in seniorswith pica Dumaguing et al (2003) reported three such deaths in individuals withchronic schizophrenia, in which one of whom also had ID In all three cases, picaonsets late in life and was reported to fail to respond to behavioral interventions,although such interventions were probably relatively informal

Surgeries, such as laparotomies (surgeries with incision through the abdomen),

to remove pica items and/or to treat blocked bowels are also dangerous andlife-threatening For example, Decker (1993) reported a case series of 35 patients in

a large institution treated for pica on 56 occasions between the years 1976 and

1991 Seventy-five percent required surgery including 34 laparotomies Deathoccurred on 11 % of occasions Finally, Kahlid and Al-Salamah (2006) reported asimilar case series of 19 men and three women with ID consecutively operated onfor acute abdominal problems over a 5-year period Thirty-six percent of thepatients had a history of pica, 33 % had complications from the surgery, and 25 %died Thus, surgery for pica is also highly dangerous and life-threatening Hence,interventions that reduce pica and the associated risks of surgeries have high socialvalidity (see, e.g., Williams et al 2009 and Chap.8, this volume)

Infections, which are sometimes serious, may also result from pica For example,Hsueh et al (2013) reported a case of a 15-year-old boy with moderate ID and picawho, after two weeks of progressive illness, presented at the emergency room withmeningitis which probably due to consumption of unknown pica items Pica mightalso play a part in maintaining high rates of population-wide infections, such aspinworm in communal-living settings (Lohiya et al 2000)

Finally, cigarette pica has characteristic risks Ingestion of cigarette butts thatcontain a large quantity of nicotine can result in vomiting, nausea, lethargy, paleand flushed appearance, and gagging over a 12-h period (Lewander et al 1997).Ingestion of used tobacco butts may expose the person to saliva-borne diseases orother dirt on discarded cigarette butts Chronic ingestion of tobacco may result inoral cancer, poor gum health, and periodontal disease (Goh et al 1999)

1.7 Summary

Pica refers to persistent consumption of non-nutritive substances that is mentally and culturally inappropriate Common pica items include starch, dirt, clay,and ice It is often a nonsocial behavior where the pica item(s) is/are highly specificand apparently highly reinforcing, perhaps due to their taste or texture It should bediagnosed accurately and distinguished from hand mouthing and other feedingdisorders in order to ensure that individuals with pica get effective treatmentpromptly

develop-Pica occurs relatively frequently in the general population and in specific ulations such as pregnant women and members of certain cultural groups as areligious or other cultural phenomenon It also occurs at times of food shortage such

pop-as extreme poverty, war, and famine Pica also occurs in clinical populations such

as a minority of people with ID, ASD, dementia, and schizophrenia where the risksare sometimes much more severe Pica also occurs in domestic and wild animals.Again, sometimes it is associated with lack of nutrition and may be a significantmanagement issue Among individuals with ID and/or ASD, it is more frequent inindividuals with more severe intellectual and social impairments

For many individuals, the consequences of pica are mild and/or can be managedreadily with intuitive strategies and regular services For a minority of individuals,

life-threatening medical issues such as infections, poisoning, choking and trointestinal problems, and life-threatening surgeries It also has negative socialconsequences

gas-This chapter has outlined some of the characteristics of pica The next chaptergoes on to look at the prevalence of pica, especially among individuals withASD/ID, and the next chapter also looks at various theories that might explain thedevelopment and maintenance of pica and their implications for treatment

Chapter 2

Epidemiology

This chapter reviews the epidemiology of pica Thefirst part briefly reviews theepidemiology of pica in individuals without developmental disabilities The secondpart proceeds to review the epidemiology of pica in individuals with developmentaldisabilities including risk factors and their implications for treatment

There are several studies reviewed by Young (2011) on the prevalence of pica inthe general population, including pregnant women and children Pica is well knownamong pregnant women Young (2012, Appendix B) reported some 47 such studiescoming from many different parts of the world and published between 1950 and

2010 These studies reported estimates of the prevalences of pica that varied from0.02 to 76.5 % Studies that reported very high prevalences tended to have smallsamples—perhaps less than 100 participants and in some cases as low as only 40participants These studies also tended to study specific groups of pregnant womenand to use personal interviews prospectively to collect data In contrast, studies withlow prevalences tended to have very large sample sizes—for example, in one study,there were 70,000 participants They also tended to use retrospective, passivemethods of data collection, such as reviews of clinical notes

Similar trends can be found in Young’s (2012, Appendix C) review of lence of pica among children which identified only 11 such studies publishedbetween 1942 and 2004 The reported prevalences varied from 1.7 to 74.4 % Thestudy that produced the lowest prevalence of 1.7 % was based on a representativesample of 659 children aged 1–10 years from two upstate New York counties,whereas studies which reported higher prevalences often used clinic samples ofpassive forms of data collection such mail surveys

preva-© Springer International Publishing Switzerland 2016

P Sturmey and D.E Williams, Pica in Individuals with

Developmental Disabilities, Autism and Child Psychopathology Series,

DOI 10.1007/978-3-319-30798-5_2

19

2.1.1 Clinical Populations

Among certain clinical groups such as individuals who are anemic (Beyan et al.2009), including anemic pregnant women (Kettaneh et al 2005) and individualswith sickle cell anemia (Ivascu et al 2001), and children with lead poisoning (Riva

et al 2012; Ruddock 1924), rates of pica may be higher than other referencegroups For example, Young (2012, Appendix E) reviewed 28 studies on theprevalence of pica in populations with iron deficiency and/or anemia These paperswere published between 1962 and 2010 and reported prevalences of pica ranging ashigh as 76.5 % in a sample of 281 pregnant low-income women from Prairie View,

TX Figures varied considerably as some studies reported specific forms of pica,such as ice and starch, and very specific geographical/clinical populations, butmany studies reportedfigures in the 20–40 % range

Studies that have attempted to estimate the prevalence of pica have producedwidely differing estimates, although certain subpopulations, such as pregnantwomen and people with anemia, do have higher rates of pica than other groups.These studies have produced a very wide range offigures reflecting problems inmeasuring pica, population definition and sampling, and other methodologicalproblems Consequently, no accurate answer can be given as to the question“What

is the prevalence of pica?”

Ali (2001) reviewed several studies of pica in individuals with ID/ASD in bothinstitutional and community settings Ali concluded that the prevalence in institu-tions ranged from 9 to 25 % and in community settings ranged from 0.3 to 14.4 %.Table2.1updates and extends Ali’s review and finds broadly similar findings Forthe purpose of this chapter, these reviews were divided into the types of populationssampled, namely institutional populations, studies that sampled total population orspecified geographical catchment areas and studies with ad hoc samples

2.2.1 Institutional Settings

Several studies have reported the prevalence of pica in institutional settings Thesestudies tend to be published in the 1980s through the early 2000s reflecting the

growth of behavioral services in American institutions at that time during tional reform, downsizing, and closure For example, Danford and Huber (1982)found that the prevalence of pica was 25.8 % in a sample of 991 individuals.Girffin et al (1982) surveyed all Texas institutions and found a prevalence of

institu-Table 2.1 A summary of prevalence studies in developmental disabilities

(1982)

(1986)

5.8 % (based on diagnoses in

records)

and ASD

Melville et al (2008)

2009)

(2004)

a Included 9 samples of children and adults in community and institutional settings (N = 1122)

b Included 2 samples of adults in day and residential services (N = 232)

c Figure varied according to the diagnostic criteria used

d Hardan and Sahl ’s (1997) sample or 233 individuals referred to a community clinic sample included 63 individuals with average intelligence and 36 individuals with “borderline” ID The figure the authors reported was 4.3 % based on 10 cases of pica in the entire sample (10/233 = 4.29 %) If the figure is recalculated to include only individuals with borderline through profound ID, then the figure becomes 10/170 = 5.9 %

e Sample was an ad hoc sample of adults attending a day habilitation program

13.7 % Matson and Bamberg (1999) surveyed pica at a large residential facility inLouisiana and found a prevalence of 6 % As shown in Table2.1, surveys of pica ininstitutional settings have produced a wide range offigures which are consistentlyhigher than surveys of entire community samples.

There have been fewer epidemiological studies which estimate the prevalence ofpica in complete geographical samples of children and/or adults with ID/ASD.These studies typically analyze existing state registers of adult disability servicesthat include surveys of challenging behavior completed at admission and/or annualstaffing or sometimes use surveys that are designed specifically for the study andconducted prospectively Such studies are important because they are not limited bysampling bias inherent in institutional studies where placement of individuals withthe most severe challenging behavior and changing patterns of service provisionover time may produce wildly varying prevalencefigures Here, we consider thesepapers

Jacobson (1982a) published a notable early study based on the New York stateregister of individuals with developmental disabilities which included 30,578 chil-dren and adults with developmental disabilities A standardized measure of chal-lenging behavior was completed on each individual, which included an item related

to pica As noted in Table2.1, the overall prevalence of pica was 1.9 %, but thisvaried substantially as function of degree of cognitive impairment, age, psychiatricdisability, and setting For example, in individuals aged under 22 years, the preva-lence of pica was 0.2, 0.6, 2.4, and 3.4 %, respectively, in individuals with mild,moderate, severe, and profound ID, but among individuals aged over 22 years was0.2, 0.2, 0.8, and 4.8 %, respectively, among individuals with mild, moderate,severe, and profound ID Hence, there was a 5- to 24-fold increase in the prevalence

if pica with increasing cognitive disability, but little systematic effect of age Amongindividuals without psychiatric/behavioral disorders, the rates of pica were 1.9 % forchildren and adults, but for those with dual diagnosis, the rates were 3.0 and 2.1 %for children and adults, respectively Finally, the prevalence of pica was substantiallyaffected by the setting in which the person lived Among children, the rates of picawere 0.0, 1.1, 1.0, 1.7, and 3.9 % among individuals living independently, withparents, in family care, in community residences, and developmental centers,respectively Among adults, the corresponding rates were 0.0, 0.2, 0.2, 0.3, and3.6 %, respectively Hence, the rates of pica were substantially below or similar tothe average of 1.9 % for all settings other than the developmental settings.Some 14 years later, Cooper et al (2007) assessed 1023 individuals living inGlasgow aged 16 years and older, using a standard battery of screening questions,psychometric instruments, and semi-structured psychiatric interviews As noted inChap 1, they found an overall prevalence of 0.0–2.0 %, depending upon thediagnostic criteria used The prevalence was correlated with gender and degree of

cognitive disability The prevalences were 2.5 and 1.3 % for men and women,respectively Among individuals with mild ID, the prevalences were 0.0 and 0.5 %among men and women, and among individuals with moderate through profound

ID, the prevalences were 3.9 and 1.9 %, respectively (This study did not report databroken down by setting.)

identified 311 adults with ID living in 20 counties of the west coast of Norway ofwhom approximately 40 % were identified as having mild ID, approximately 43 %moderate ID and 17 % severe/profound ID There were 2.9 % who were identifiedwith pica There was an increasing trend such that those with more severe ID weremore likely to show pica Thus, 0.8, 3.8, and 5.9 %, respectively, of individualswith mild, moderate, and severe/profound ID showed pica but this difference wasnot statistically significant Finally, individuals with ASD were more likely to showpica than those without pica (p = 0.0002) The prevalence of pica was notably lowerthan that reported in almost all other studies Presumably, this was in part because itincluded both community and institutionalized participants, but perhaps also due tounderreporting when collected data using interview and psychometric measures.Finally, Smith et al (1996) prospectively surveyed 2202 adults with ID in a totalpopulation sample using trained interviewers and a standardized psychometricmeasure of challenging behavior They found that the prevalence of pica was 0.0,0.4, 0.1, and 3.2 % of individuals with mild, moderate, severe, and profound ID,respectively (They did not report an overall prevalence in this paper.)

These total population studies consistently report that pica is a relativelylow-frequency problem which occurs in 1–4 % of individuals with ID Very fewindividuals with mild, moderate, and severe ID show pica and around 3–5 % ofindividuals with profound ID show pica

As noted earlier gender, the variable that has the strongest correlation to risk of pica

is degree of disability; that is, most individual with ID and pica have profound andsome have severe ID, but few have borderline through moderate ID Despite thisrelatively strong association, it is unclear why it exists and why some individualswith profound ID show pica and some do not Some studies have found additionalcorrelations between lack of sociability and social skills, but again how these tworisk factors result in the development and maintenance of pica is unclear Some, butnot all, studies have found a somewhat higher prevalence of pica among males thanfemales Again, why this is so is unclear

Settings also have large influences on the rate of pica Namely, there are sistentfindings that pica is more common in institutional than other settings Thedifferences between institutional and other settings are quite large (see Table2.1).Again, the reasons for this difference are unclear, as it may reflect some causative