Lessons from Lucia

��

��

��

��

��

��

��

�� !�� "� ��

��#�� $��

��

� ��

��%

��

��

�� &��

��'��(��

��!� ��$��

��

��

��)*��

�

��+�� +��$��,-�� .�� )��-�� /��*�� 011$��

Lessons from Lucia

[email protected]://www.math.leidenuniv.nl/~gill

mailto:[email protected]

http://www.math.leidenuniv.nl/~gill

Overview of Lecture

• Background

• Theory (statistical paradigms)

• Lucia

• Conclusions

Background• Serial killer nurses: is there an epidemic?

• Victorino Chua (UK)

• Daniela Poggiali (It)

• Nils H. (Germany)

• Ben Geen (UK)

• …

• But perhaps also an epidemic of falsely convicted innocent nurses !

Academic (?) research

• Katherine Ramsland (2007) Inside the minds of health care serial killers: why they kill

• Elizabeth Yardley and David Wilson (2014) In Search of the ‘Angels of Death’: Conceptualising the Contemporary Nurse Healthcare Serial Killer

“Red flag check-list”: inspired (in part) by the case of Lucia de Berk …from the time when everyone knew she was guilty

Sources: newspaper reports and prosecution documents

was necessary. For example, through discussion, we interpreted regularly as meaning a pat-tern of employment that was out of the ordinary, and which might have seen the nurse set-tling at a job for only a few months and then moving on. However, it is also clear that atsome later stage this checklist should be further refined and scrutinised, especially as it

Table 14. HSK and red flag score

HSK Total score

Charles Cullen 11Kimberley Saenz 9Kristen Gilbert 8Robert Rubane Diaz 8Sonia Caleffi 8Beverley Allitt 7Cecile Bombeek 7Vickie Dawn Jackson 7Aino Nykopp-Koski 6Orville Lynn Majors 6Benjamin Geen 5Petr Zalenka 5Christine Malevre 4Irene Becker 4Stephan Letter 4Colin Norris 2

Table 15. Prevalence of items on ‘red flag’ checklist

Item N% of cases in which this

item was present

1. Moves from one hospital to another 6 382. Secretive/difficult personal relationships 7 433. History of mental instability/depression 10 634. Predicts when someone will die 4 255. Makes odd comments/claims to be ‘jinxed’ 3 196. Likes to talk about death/odd behaviours when someone dies 1 67. Higher incidences of death on his/her shift 15 948. Seems inordinately enthused about his/her skills 6 389. Makes inconsistent statements when challenged about deaths 3 1910. Prefers nightshifts—fewer colleagues about 6 3811. Associated with incidents at other hospitals 6 3812. Been involved with other criminal activities 7 4313. Makes colleagues anxious/suspicious 9 5614. Craves attention 7 4315. Tries to prevent others checking on his/her patients 0 –16. Hangs around during investigations of deaths 0 –17. In possession of drugs at home/in locker 8 5018. Lied about personal information 0 –19. In possession of books about poison/serial murder 1 620. Has had disciplinary problems 4 2521. Appears to have a personality disorder 8 5022. Has a substance abuse problem 3 19

Note: % ≠ 100 as all cases had at least one checklist item

In search of the ‘Angels of Death’

Copyright © 2014 John Wiley & Sons, Ltd. J. Investig. Psych. Offender Profil. (2014)

DOI: 10.1002/jip

In Search of the ‘Angels of Death’: Conceptualising theContemporary Nurse Healthcare Serial Killer

ELIZABETH YARDLEY* and DAVID WILSONBirmingham City University, Centre for Applied Criminology, City North Campus, Franchise Street,

Perry Barr, Birmingham, B42 2SU, United Kingdom

Abstract

Focusing specifically upon nurses who commit serial murder within a hospital setting, thispaper aims to establish insights into this particular subcategory of healthcare serial killer.In addition, the paper aims to test the usefulness of an existing checklist of behavioursamong this group of serial murderers. Drawing upon existing lists of healthcare serialkillers produced by other scholars as well as legal records and an online news archive,we identified and researched healthcare serial killer nurses, collating socio demographicand criminological data and applying the aforementioned checklist to each case. Our find-ings suggest that to date, the label ‘healthcare serial killer’ has been applied in too loose amanner, making the understanding of this phenomenon problematic. In further refining thedefinition and identifying the socio-demographic and criminological characteristics of thevictims, perpetrators and crimes, we have developed more specific and therefore usefulinsights for practitioners and identified a potentially useful checklist which, with revisions,could contribute towards preventative strategies and interventions. Copyright © 2014John Wiley & Sons, Ltd.

Key words: healthcare serial killers; nurses; hospitals

INTRODUCTION

Healthcare serial killers (HSKs), sometimes also known as medical murderers (Hickey, 2010),have emerged as a common concern within popular culture and true crime, with various books,documentaries and films all eager to understand what might have motivated, for example,Beverly Allitt, Harold Shipman or Charles Cullen to have taken the lives of their patients(Davies, 1993; Graeber, 2013; Peters, 2005). However, whilst the HSK has been emerginginto popular consciousness, academic, criminological research about this type of offenderremains somewhat underdeveloped. Even so, a relatively small body of work has begun to iden-tify, map and describe cases of healthcare serial murder (see, for example, Ramsland, 2007;

*Correspondence to: Elizabeth Yardley, Birmingham City University, Centre for Applied Criminology, CityNorth Campus, Franchise Street, Perry Barr, Birmingham B42 2SU, UK.E-mail: [email protected]

Journal of Investigative Psychology and Offender ProfilingJ. Investig. Psych. Offender Profil. (2014)Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jip.1434

Copyright © 2014 John Wiley & Sons, Ltd.

Theory: statistical paradigms

• Bayes (one person statistics)

• Frequentist (two person, collaborative statistics)

• Likelihood (avoiding the issue)

NB two paradigms of probability: “subjective” (Bayesian), “objective” (frequentist)

Bayes’ rule

• Posterior odds

= prior odds * likelihood ratio

• Likelihood ratio

= Prob( data | HP ) : Prob( data | HD )

Bayesian/frequentist peaceful coexistence theorem:{Decision theoretic admissible} = {Bayesian (for some prior)}

Current research: use of Bayes net

(aka graphical model)• Bayesian model of (probabilistic) causality

• Bayesian means “probability as degree of belief” (epistemological, not ontological)

• Statistical correlations “explained” by causal dependence on past events

• Some of those events are known, others unknown

• Computations: with GeNIe, HUGIN Lite, or in R

and is not attuned to the need to double the prior probability, one mightargue, as a pragmatic solution, that the “correct” likelihood ratio, 1/(2P ),should be doubled, so as to build this correction in automatically. . . whichwould bring us back to the value 1/P .

5 Complex patterns of evidence

The di�culties of assessing a single item of evidence are compounded whenwe want to account for the complex inter-relationships between the manyitems of evidence in a case. To organise the evidence it is then helpful toconstruct a diagrammatic representation all the evidence and hypotheses inthe problem, and the relationships between them. This idea was first sug-gested by Wigmore (1937): see Anderson et al. (2005) for an introduction tothe “Wigmore chart” method. More recently the methods of graphical mod-elling and Bayesian networks — also known as probabilistic expert systems(Cowell et al. 1999) — have been applied. Such a network contains a nodefor each variable in the problem, with arrows between nodes to denote prob-abilistic dependence of a “child” node on all its “parents”. To complete thedescription we need the numerical or algebraic specification of the associatedconditional probabilities.

5.1 Example

Dawid and Evett (1997) consider a fictitious burglary case, described asfollows:

An unknown number of o↵enders entered commercial premiseslate at night through a hole which they cut in a metal grille. In-side, they were confronted by a security guard who was able to seto↵ an alarm before one of the intruders punched him in the face,causing his nose to bleed.

The intruders left from the front of the building just as a policepatrol car was arriving and they dispersed on foot, their getawaycar having made o↵ at the first sound of the alarm. The securityguard said that there were four men but the light was too poor forhim to describe them and he was confused because of the blow hehad received. The police in the patrol car saw the o↵enders onlyfrom a considerable distance away. They searched the surroundingarea and, about 10 minutes later, one of them found the suspecttrying to “hot wire” a car in an alley about a quarter of a milefrom the incident.

10

Example: Dawid and Evett (1997)

Example (ctd) : Dawid and Evett (1997)At the scene, a tuft of red fibres was found on the jagged end of

one of the cut edges of the grille. Blood samples were taken fromthe guard and the suspect. The suspect denied having anything todo with the o↵ence. He was wearing a jumper and jeans whichwere taken for examination.

A spray pattern of blood was found on the front and right sleeveof the suspect’s jumper. The blood type was di↵erent from that ofthe suspect, but the same as that from the security guard. The tuftfrom the scene was found to be red acrylic. The suspect’s jumperwas red acrylic. The tuft was indistinguishable from the fibres ofthe jumper by eye, microspectrofluorimetry and thin layer chro-matography (TLC). The jumper was well worn and had severalholes, though none could clearly be said to be a possible origin forthe tuft.

In this example there are three general kinds of evidence: eye-witness,blood, and fibre; and for each kind a variety of individual evidential items.We can summarise the salient features of the evidence against the suspect asfollows:

EYEWITNESS

G : The evidence of the security guard

W : The evidence of the police o�cer who arrested the suspect

BLOOD

R : The bloodstain in the form of a spray on the suspect’s jumper

X1: Suspect’s blood type

X2: Guard’s blood type

Y2: Blood type of blood spray on jumper

FIBRES

X3: Properties of the suspect’s jumper

Y1: Properties of fibre tuft

The uncertain hypotheses and variables that enter are:

HYPOTHESES

C: Whether the suspect was or was not one of the o↵enders

11

J. Mortera A. P. Dawid (2006), Probability and Evidence, Research Report No. 264, Department of Statistical Science, University College London.

Example: Dawid and Evett (1997)

Squares = observed = evidence; circles = not observed; C = hypothesis of interest

Dawid and Evett (1997)

At the scene, a tuft of red fibres was found on the jagged end ofone of the cut edges of the grille. Blood samples were taken fromthe guard and the suspect. The suspect denied having anything todo with the o↵ence. He was wearing a jumper and jeans whichwere taken for examination.



EYEWITNESS



BLOOD





FIBRES




HYPOTHESES


11

At the scene, a tuft of red fibres was found on the jagged end ofone of the cut edges of the grille. Blood samples were taken fromthe guard and the suspect. The suspect denied having anything todo with the o↵ence. He was wearing a jumper and jeans whichwere taken for examination.



EYEWITNESS



BLOOD





FIBRES




HYPOTHESES


11A: The identity of the person who left the fibres on the grille

B: The identity of the person who punched the guard

N : The number of o↵enders

Of these the specific charge before the court is C = true; the others areincluded to provide a complete account of the problem.

Figure 1: Bayesian network for burglary example

Figure 1 shows a graphical representation of the problem as a Bayesiannetwork. The evidence items are shown as squares, and the hypotheses ascircles. Variable Y2, the measurement of the blood type of the spray on thejumper is dependent on X1, the suspect’s blood type (because it might be aself stain) and the guard’s blood type X2. But information is also providedby R, the variable which describes the shape of the stain, because that shedslight on whether or not it might be a self stain. In turn, the shape of the stainis influenced by the way in which the guard was punched, G2, and B, theidentity of the person who did it; while B is in turn influenced by whetheror not the suspect was one of the o↵enders, variable C, and also the numberof o↵enders, N .

Dawid and Evett (1997) describe how the graph can be used to read o↵implicit properties of independence: for example, to show that, conditionallyon knowing A and N , the pair of variables (B, R) is independent of the

12

A: The identity of the person who left the fibres on the grille

B: The identity of the person who punched the guard

N : The number of o↵enders

Of these the specific charge before the court is C = true; the others areincluded to provide a complete account of the problem.

Figure 1: Bayesian network for burglary example

Figure 1 shows a graphical representation of the problem as a Bayesiannetwork. The evidence items are shown as squares, and the hypotheses ascircles. Variable Y2, the measurement of the blood type of the spray on thejumper is dependent on X1, the suspect’s blood type (because it might be aself stain) and the guard’s blood type X2. But information is also providedby R, the variable which describes the shape of the stain, because that shedslight on whether or not it might be a self stain. In turn, the shape of the stainis influenced by the way in which the guard was punched, G2, and B, theidentity of the person who did it; while B is in turn influenced by whetheror not the suspect was one of the o↵enders, variable C, and also the numberof o↵enders, N .

Dawid and Evett (1997) describe how the graph can be used to read o↵implicit properties of independence: for example, to show that, conditionallyon knowing A and N , the pair of variables (B, R) is independent of the

12

Dawid and Evett (1997)

J. Mortera A. P. Dawid (2006), Probability and Evidence, Research Report No. 264, Department of Statistical Science, University College London.

Example: prior to entering evidence

Example: posterior to entering evidence

��

��

��

��

��

��

��

�� !�� "� ��

��#�� $��

��

� ��

��%

��

��

�� &��

��'��(��

��!� ��$��

��

��

��)*��

�

��+�� +��$��,-�� .�� )��-�� /��*�� 011$��

Lucia

Shifts Court dataCourt dataCourt data Corrected dataCorrected dataCorrected dataCorrected data

JKZ MCU-1 incidentincident incidentincident Oct ’00 – Sept ’01 with without with without

Luciawith 9 b133 b7 b135

Luciawithout 0 b887 b4 b883

RKZ-42 Aug – Nov ’97

Luciawith b6 b52 b5 b53

Luciawithout b9 272 10 273

RKZ-41Aug – Nov ’97

Luciawith 1 bb0 1 bb2

Luciawithout 4 361 4 359

Lucia: the data

Lucia

de B.

Reconstructie vaneen gerechtelijke dwaling

Lucia de B.Reconstructie van een gerechtelijke dwaling

Ton Derksen

Lucia de B. zou zeven patiënten hebben gedood in de tijd dat zijals verpleegster werkte. Ze kreeg de zwaarst mogelijke straf,levenslang met TBS, een straf die niet eerder was toegekend inNederland en die inmiddels door de Hoge Raad als ‘juridischonmogelijk’ is bestempeld.

Ton Derksen, hoogleraar wetenschapsfilosofie, laat zien dat nietalleen de straf ondeugdelijk is. Ook het bewijs is flinterdun. Hetgerechtshof in Den Haag heeft zich laten leiden door negatievebeeldvorming, ondeugdelijk gebruik van statistiek, onjuistgeïnterpreteerde medische gegevens en een kwaadwillende lezingvan het dagboek van Lucia de B. Feitelijk is er een dader gezochtvoor overlijdensgevallen die eerder door artsen als ‘natuurlijkedood’ waren aangemerkt. Met hulp van artsen, juristen en andereexperts laat Derksen zien hoe een kwalijke bewijszucht van hetgerechtshof de waarheidsvinding dramatisch in de weg heeftgestaan.

Een spraakmakend boek over de blunders van een gerechtshof!

Ton Derksen is emeritus hoogleraar wetenschapsfilosofieen cognitiefilosofie aan de Radboud UniversiteitNijmegen en emeritushoogleraar algemene kennis- enwetenschapsleer aan de Universiteit van Tilburg. Hijschreef vele artikelen en verscheidene boeken, waaronderhet bekende Wetenschap en willekeur, overwaarschijnlijkheid, kans, wetenschap en rationaliteit.

Lucia de B.R

econstructie van een gerechtelijke dwaling

Veen Magazines

Veen MagazinesLaser Proof

Ton Derksen

Ton Derksen

cover.lucia.def 17-05-2006 18:17 Pagina 1

Lucia: time-line• Sept. 4, 2001, “unexpected” death of Amber

• 2003: life sentence for 4 murders and 2 attempts; proof: statistical

• 2004: life sentence of 7 murders and 3 attempts; proof: medical

• 2006: confirmed by supreme court

• 2006: publication of book by Ton Derksen (philosopher of science)

• 2006: case submitted to special committee for review of exceptional possibly unsafe convictions

• 2008: CEAS reports death of Amber natural, recommends reopening

• 2008: “advocate-general” to supreme court admits there is no “novum”, commissions further investigations

• 2009: AG recommends case is reopened (with “novum” if required: former key pathologist agrees with new findings – he had less information at his disposal

• 2009: supreme court accepts, case is reopened

• 2010: not-guilty verdict (all deaths natural; nurses behaviour exemplary; medical errors)

Lucia: likelihood ratio

• Hypothesis of the prosecution: (most of the) Lucia incidents are murders or attempted murders

• Hypothesis of the defence: the events are natural and would have happened anyway

• Prob(data|HP):Prob(data|HD)=1:1

Lucia: the original statistical analysis

• Frequentist approach; hypothesis test; null hypothesis = “balls in vases”

• For each of three data sets, court’s statistician computed the “p-value” P(as extreme as Lucia or more | balls in vases model)

• For JKZ MCU-I, he multiplied by 26 (= # nurses worked on the ward that year)

• Product of three p-values = 1 in 342 million

Lucia: the defense• Judge: “what is the probability the coincidence is due

to chance?”

• Defence 1. There are so many different probability models, you cannot compute a probability

• Defence 2. Multiplying p-values is wrong (reductio ad absurdam)

• Judges: “we are not here to do thought experiments, but to determine facts”

• Judges: “The verdict of the court does not depend on a statistical computation of probabilities”

Lucia: the defense• Judge: “what is the probability the coincidence is due

to chance?”

• Defence 1. There are so many different probability models, you cannot compute a probability

• Defence 2. Multiplying p-values is wrong (reductio ad absurdam)

• Judges: “we are not here to do thought experiments, but to determine facts”

• Judges: “The verdict of the court does not depend on a statistical computation of probabilities”

No one checked the data!

• Three children responsible for multiple identical events, some in Lucia’s shifts, some not

• No consistent definition of “incident”

• No consistent definition of “time of incident”

• The data suggested the hypothesis

• No-one studied the “normal” situation (clusters of events, clusters of shifts are normal)

Shifts Court dataCourt dataCourt data Corrected dataCorrected dataCorrected dataCorrected data

JKZ MCU-1 incidentincident incidentincident Oct ’00 – Sept ’01 with without with without

Luciawith 9 b133 b7 b135

Luciawithout 0 b887 b4 b883

RKZ-42 Aug – Nov ’97

Luciawith b6 b52 b5 b53

Luciawithout b9 272 10 273

RKZ-41Aug – Nov ’97

Luciawith 1 bb0 1 bb2

Luciawithout 4 361 4 359

1

Some p-values

• Cochran-Mantel-Haenszel test & Elffers’ post-hoc correction 1 in 916

• Ultimate stratification 11 days at JKZ with both incident & Lucia on duty 1 in 25

• Gamma(1) heterogeneity over Poisson intensity JKZ, RKZ pooled 1 in 25

Aftermath

• Since 2010, no more media interest

• The legal system got the blame, the taxpayer paid the bill

• There have been reforms, improvements, communication between legal and scientific communities

• Medical community is silent

Interview with president Council for Justice

• “The system worked fine”

• “Murderers who escape conviction usually confess on their deathbed”

What really happened?

• In Dutch hospitals: 2000 deaths per year due to avoidable medical errors; culture of denial; frequent communication failures

• During 9 months up to 4 Sept. 2001, there was gossip about Lucia among nurses and specialists

• Medical errors by specialists were being associated with Lucia

• Director and top medical staff (but not all), under oath: there was no suspicion till 4 Sept. 2001

• No suspicion at all till 4 September, 2001? Director Paul Smits reported 10 unnatural deaths and suspicious reanimations, over last year, within 15 minutes of being informed of death of Amber, and on the very same day

• Strange fact: these 10 “incidents” were also reported to Health Inspectorate. Conclusion: nothing wrong.

• 4 medical specialists, it appears, have lied to police and to courts (and to one another) concerning the treatment of their own patients

What really happened?

Key case: baby Amber• Baby Amber did not die of digoxin poisoning

• In fact the circumstances of her death are completely consistent with a “natural” process

• Lucia did not have opportunity (doctors were with the baby at the time when the court had determined she must have acted)

• It seems there might have been digoxin in the body, but it did not play any role in her death, and there are many innocent explanations for how it got there … if it was there at all

HCSK’s

• Once a hospital has “identified” a HCSK, the suspect has no chance any more

• Lucia got accused through a combination of unlucky coincidences

• She got exonerated through another combination of lucky coincidences … and a lot of very hard work of very many “outsiders”

��

��

��

��

��

��

��

�� !�� "� ��

��#�� $��

��

� ��

��%

��

��

�� &��

��'��(��

��!� ��$��

��

��

��)*��

�

��+�� +��$��,-�� .�� )��-�� /��*�� 011$��

[email protected]://www.math.leidenuniv.nl/~gill

Conclusions• Forensic statistics is in its infancy

• It requires non-standard paradigms and will need new methodology

• Multiparty statistics

• Nuisance parameters

• Model the forensic investigation process

• Communication of statistical ideas to non-statisticians is going to be the bottle-neck

mailto:[email protected]

http://www.math.leidenuniv.nl/~gill