- Legal Medicine

Legal Medicine 11 (2009) S76–S79
Contents lists available at ScienceDirect
Legal Medicine
journal homepage: www.elsevier.com/locate/legalmed
Review Article
Medico-legal autopsies as a source of information to improve patient safety
Burkhard Madea *
Institute of Forensic Medicine, University of Bonn, Stiftsplatz 12, 53111 Bonn, D, Germany
a r t i c l e
i n f o
Article history:
Received 13 January 2009
Accepted 14 January 2009
Available online 16 March 2009
Keywords:
Medical malpractice
Autopsies
Lethal outcome
Patient safety
a b s t r a c t
Clear data on the epidemiology of medical malpractice are missing, at least for Germany. Especially data
on the frequency of malpractice claims in penal and civil law are not available. However, from epidemiological and healthcare research some data on the frequency of adverse events (AE), preventable adverse
events (PAE) and negligent adverse events (NAE) are available. According to a review of the German Alliance of Patient Safety adverse events can be expected in 5–10%, preventable adverse events in 2–4%, negligent adverse events in about 1% and lethal outcome in about 0.1% of hospitalized patients. However,
these data are not compatible with figures from civil and penal law and there seems to be a great ‘‘litigation gap”.
Data on malpractice cases are available from the files of the arbitration committees on medical malpractice, files of liability insurance companies and files of the institutes of forensic medicine. However,
these files are not complete and data sources may have some interferences.
The arbitration committees are, however, mainly dealing with living patients. Lethal cases are a special
subgroup and the best available data source are the files of the institutes of forensic medicine. This subgroup is of special importance since death is the most severe outcome of medical malpractice and the
reproach to have caused the death of a patient by medical negligence is the most severe malpractice
claim.
From a retrospective multicentre study on autopsies performed in cases of medical malpractice claims
(altogether 101.358 autopsies, 4450 due to medical malpractice claims) the most important data will be
presented (disciplines concerned, cause of accusation, classification of accusation, causes of preliminary
proceedings, occupational group and number of approved medical malpractice, outcome of medical malpractice proceedings). However, these data are not only of a descriptive value but have utmost importance also for risk analysis and to increase patient safety. A thorough evaluation of serious incidences,
although less sophisticated than a root cause analysis, produces far more information than the usual hospital reporting system. Identification and evaluation of errors as well as reporting of errors may also contribute to the prevention of errors which is among other medical disciplines also a task of forensic
medicine.
Ó 2009 Elsevier Ireland Ltd. All rights reserved.
1. Introduction
Clear data on the epidemiology of medical malpractice are missing, at least for Germany [7,9,10,15,16]. Furthermore data on the
frequency of medical malpractice claims – both in penal and civil
law – are not available [23]. Quite detailed data are available from
the arbitration committees of the medical councils which are published once a year on a national basis (MERS: Medical Error Reporting System).
Based on epidemiological- and health care research studies figures on the frequency of adverse events (AE), preventable adverse
events (PAE) and negligent adverse events (NAE) were calculated
by the German Alliance of Patient Safety [26–28]. According to
* Tel.: +49 228 73 83 15; fax: +49 228 73 83 68.
E-mail address: [email protected]
1344-6223/$ - see front matter Ó 2009 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.legalmed.2009.01.108
the systematic review of the German Alliance of Patient Safety
AE can be expected in 5–10%, PAE in 2–4%, NAE in about 1% and
lethal outcome in about 0.1% of hospitalized patients. Concerning
17.5 million hospitalized patients in Germany therefore 880,000–
1750,000 adverse events, 350,000–700,000 PAE, 175,000 NAE and
17,500 deaths could be expected every year. However, these data
are not compatible with figures from civil and penal law and they
were questioned after their publication. Nevertheless they are in
accordance with other international studies: e.g. the Institute of
Medicine stated that up to 98,000 patients die of preventable medical errors in American hospitals each year [11]. Among 850,000
individuals dying in US hospitals per year a major diagnosis remains clinically undetected in at least 8.4% (71,400 deaths). Furthermore 34,850 patients would have survived until discharge
had misdiagnoses not occurred [29,30]. One of the Institutes Of
Medicine recommendations called therefore for a mandatory
B. Madea / Legal Medicine 11 (2009) S76–S79
reporting system for deaths and serious injuries. Medico-legal
death investigation systems are one component of this approach
[11].
These figures on the frequency of AE, NAE, PAE and deaths – this
has to be kept in mind – were calculated from epidemiologicaland health care research studies.
Data on malpractice cases are available from the files of the
arbitration committees on medical malpractice [1], files of liability
insurance companies and files of the institutes of forensic medicine
[17,20]. The arbitration committees are, however, mainly dealing
with living patients [1,12]. Lethal cases are a special subgroup
and they are found within the material of the arbitration committees only in 2.7% of all cases (Fig. 1).
The best available data source on lethal cases is in the files of
the institutes of forensic medicine. This subgroup is of special
importance since death is the most severe outcome of medical
malpractice and the reproach to have caused the death of a patient
by medical negligence is the most severe malpractice claim.
2. Retrospective analysis of medical malpractice claims
On behalf of the German Ministry of Health we carried out a
standardized retrospective analysis on medical malpractice claims
in lethal cases. The multicentre analysis was based on the data of
17 German institutes of forensic medicine and covered the time
period 1990–2000. Altogether 101,358 autopsy reports were evaluated. Autopsies due to medical malpractice were carried out in
4450 cases (4.4%). Data were made completely anonymous. Details
of the study can be found in [24], further information in
[4,5,17,19,20,25]. In 2863 cases the autopsy reports, in 2587 cases
further comprehensive written expert evidence was available for
analysis.
Altogether autopsies due to medical malpractice were carried
out in 4.4% of all autopsies, but the rate varied between different
institutes from 1.4% to 20%. The number of cases increased over
the evaluated period from 300 to 600 a year. Today we can expect
for Germany about 1000–1500 medico-legal autopsies which are
carried out for suspicion of medical malpractice per year.
Concerning the occupational group preliminary proceedings
were mostly against hospital doctors (2811 cases), doctors in private practice were much less concerned (901). Surgeons are – as
in other statistics – at the top of disciplines charged for medical
malpractice. This is mainly due to the fact that in surgery in terms
of criminology the questions who, where and when are known (Table 1).
Although mostly surgical disciplines were concerned, cause of
accusation was predominantly conservative therapy (n = 2604) followed by surgical therapy (n = 1737), endoscopy (n = 232), inten-
S77
Table 1
Medical disciplines concerned.
Medical disciplines
Cases
Surgical disciplines
Internal medicine
Practitioners
Anaesthesiologists
Gynaecologists and obstetricians
Orthopaedic surgeons
Psychiatrists/neurologists
Pediatricians
ENT (ear, nose and throat) specialists
Urologists
General practitioners
1272
699
434
156
151
126
117
86
72
66
20
sive care (n = 88), naturopathic treatment/alternative treatment
(n = 18).
There are different classifications and typings of medical malpractice and mistakes, e.g. concerning medical treatment, information, documentation, medication errors, machine/medical product,
organization etc. We used the following classification (Table 2).
Most accusations were due to negligence (48.5%) followed by
complications within surgical therapy (33.1%), wrong therapy
(17.2%), medication error, adverse event due to drug therapy
(12.5%), mistakes in care (7.2%) and accusation not specified
(3.4%). A really interesting point is that of the way a case comes
to the attention of the public prosecutor. In over 40% of cases the
classification of the manner of death in the death certificate as unclear or unnatural brought the case to the attention of the public
prosecutor (Table 3) [14,16]. The often heard verdict against medical doctors that physicians frequently make errors in their diagnosis and treatment and when discovering them they usually do their
best to hide the malpractice is therefore not right in this strict
sense.
Of special importance is the correlation of the confirmation of
medical malpractice with the occupational group (Table 4).
Although clinicians are mostly concerned with medical malpractice claims, medical malpractice is confirmed only in 7.8%, however, in doctors in private practice in 14.7% and in nursing staff
in 20.3%. These data are of special importance, since all epidemiological and health care research studies on adverse events focus for
methodological reasons on hospitalized patients but the more serious sector seems to be the ambulant sector of health care.
Although surgical disciplines are at the top of all disciplines
concerned for medical malpractice, malpractice is confirmed only
in 6.8%, the causality for death only in 3.1%. Obviously the surgeons
have learned their lesson from malpractice claims (Table 5). The
medico-legal death investigation system is very effective: regarding the results of our study medical malpractice was negated in
Table 2
Different types of medical malpractice.
Group 1 – negligence
(omitting the necessary
treatment)
Group 2 – complications at and/
or after surgery
Group 3 – wrong treatment
Group 4 – mistake in care
Group 5 – adverse drug event,
medication errors
Fig. 1. Classification of bodily damage Arbitration committee north-Rhine.
- Insufficient diagnostics
- For instance no CT after craniocerebral
trauma, no ECG in a cardiac emergency
- Transfusion reaction (control of ABOcompatibility omitted)
- Retained instruments
- Unsufficient prophylaxis of decubital
ulcers
- Wrong positioning during operation
- Wrong drug
- Wrong dose
- Wrong application/administration
- Disregarding drug allergy
S78
B. Madea / Legal Medicine 11 (2009) S76–S79
Types of medical malpractice in drug therapy seen in our retrospective study were e.g. [19]:
Table 3
Causes of preliminary proceedings according to analysed documents.
Cause of preliminary proceedings
Number of
cases (%)
Manner of death ‘‘unclear” or ‘‘unnatural” in death certificate
Cause of proceedings is unknown
Complaint of the offence by relatives (including friends and
carers)
Complaint of a co-treating or post-treating physician
Complaint of offence by relatives as well as type of death unclear
or unnatural in death certificate
No formal preliminary proceeding by the prosecutor, but
proceeding to clear cause and manner of death
Self complaint of the physician(s)
Complaint by the patient himself before death
Complaint by staff (especially nursing staff)
Anonymous complaint of offence
Other
1715 (38.5)
1303 (29.3)
831 (18.7)
271 (6.1)
190 (4.3)
intrathecal administration of Vincristin when simultaneously
different antineoplastic drugs are given,
mix-up of electrolyte solutions (KCl instead of NaCl),
disregarding drug allergy,
inadequate substitution of drug addicts,
wrong dose, especially in renal insufficiency, when antineoplastic drugs are given (especially Methotrexate).
73 (1.6)
21 (0.5)
18 (0.4)
10 (0.2)
9 (0.2)
9 (0.2)
Without autopsy and toxicological investigations these cases
would never have been cleared and also clinicians who work on
adverse drug events admit: ‘‘Thus, without having access to autopsy data and drug analysis data clinicians are ‘‘walking in the
fog” as far as detecting fatal ADEs is concerned” [19].
3. Conclusions
Table 4
Occupational groups concerned by accusations of medical malpractice, number of
approved errors in treatment without consideration of causality for death.
Occupational group
Accusations/occupational group
Clinician
Doctor in private practice
Nursing staff
Emergency service doctor
A group of doctors
Ambulance officer
Practitioner with cottage-hospital affiliation
Alternative practitioner
2809
877
172
253
50
23
14
12
2873 cases, in 1971 cases already immediately after autopsy. Altogether 64.3% of cases were clarified already by autopsy and the
preliminary investigation by the prosecutor was terminated.
In all epidemiological studies nosocomial infections and medication errors are at the top of AE during health care. They are neither seen in our retrospective analysis [15,19] nor in the files of
other malpractice registers [13], e.g. the arbitration committees,
in this high frequency. A Scandinavian prospective study on patients in internal medicine revealed an incidence of lethal Adverse
Drug Reactions (ADR) of 0.95%, 49.6% of these lethal ADRs were
preventable. If those Scandinavian data were transferred to Germany taking into account 6 million hospitalizations in internal
medicine, 57,000 lethal cases of ADR would be expected, 28,000
fatal outcomes would be classified as preventable. These expectations on the frequency of lethal ADRs are, however, not mirrored
in the files of the different malpractice registers.
Of course different data sources on medical malpractice are not
identical. Files of the arbitration committees have huge and well
documented material on minor injuries. In forensic medicine, however, the possibly most dramatic cases – death due to medical malpractice – are seen. Therefore the files of the institutes of forensic
medicine are indispensable for the epidemiology of adverse events
and improvement of patient safety. They should be part of medical
malpractice registers [13]. Another important point is that most
studies on adverse events focus on hospitalized patients; the files
of the institutes of forensic medicine cover also the ambulant sector of health care. For doctors in private practice and nursing staff
the rate of approved medical malpractice is with 14.7% and 20.3%,
respectively, much higher than for clinicians. Although surgery is
the discipline with the most accusations of medical malpractice,
the rate of confirmed medical malpractice with approved causality
is with 3.1% low. Obviously the surgeons have learned their lessons
from the high rate of medical malpractice claims.
Autopsy is a very efficient method of clarifying medical malpractice claims, in 64.3% the cases were already clarified after autopsy. Immediate clarification means that in two third of cases
already after autopsy the preliminary investigation can be terminated. This is of course also in the interest of doctors charged for
malpractice [2,18,21].
Autopsy based studies on medical malpractice have furthermore the advantage that the cause of death is classified on an
objective basis, forensic pathologists being familiar with giving expert evidence in penal law, especially concerning causal connection. Furthermore forensic pathologists are not embarrassed in
Table 5
Results of expert opinions in the different occupational groups and clarification of causality of the malpractice for death (in brackets percentage of total number of cases of each
group).
Occupational group
Number of cases
Medical malpractice confirmed
Causality negated
Causality approved
Surgery
Internal medicine
Unknown
Family doctor/practitioner
Emergency service doctor
Other disciplines
Anesthesiology
Orthopaedics
Emergency doctor
Gynaecology
Paediatrics
ENT specialists
Psychiatry
Urology
1266
698
534
434
254
195
157
127
108
88
87
74
68
67
86 (6.8)
72 (10.3)
42 (7.9)
64 (14.8)
31 (12.2)
40 (20.5)
19 (12.1)
4 (3.2)
10 (9.3)
8 (9.1)
13 (14.9)
5 (6.7)
6 (8.8)
2 (3.0)
47
37
29
43
20
21
9
1
5
4
10
1
0
0
39
35
13
21
11
19
10
3
5
4
3
4
6
2
(3.7)
(5.3)
(5.3)
(9.9)
(7.9)
(10.8)
(5.7)
(0.8)
(4.6)
(4.5)
(11.5)
(1.4)
(3.1)
(5.0)
(2.4)
(4.8)
(4.3)
(9.7)
(6.4)
(2.4)
(4.6)
(4.5)
(3.4)
(5.4)
(8.8)
(3.0)
B. Madea / Legal Medicine 11 (2009) S76–S79
any field of medicine. In comparison to epidemiological studies a
disadvantage is of course that the data are not representative since
a correlation with a well defined reference sample is missing.
The data of medical malpractice claims have beside the medicolegal purposes also to be used for the prevention of errors [3]. The
World Alliance for Patient Safety has already recommended some
years ago alternative sources of information for patient safety:
‘‘Analysis of claims, for example, has identified the factors that increase the probability of a foreign body being retained following
surgery. The limitation of malpractice claims is their non-representativeness. However, they do provide data on events that are significant – serious injuries – as well as data that are typically much
more comprehensive than provided to most reporting systems.”
Therefore we in forensic medicine should contribute by identification of errors, an evaluation of errors, reporting of errors also to the
prevention of errors. ‘‘Even today old-fashioned low-technologies
can uncover many important diagnoses missed by modern hightechnology medicine” [22]. Own case reports on wrong site injection of anteneoplastic drugs when simultaneously different antineoplastic drugs were given were noticed by the drug
commission of the German Medical Council and were reason to
publish warnings against mixing up of drugs in these cases [6].
Compared to epidemiological data on the frequency of adverse
events or death as consequence of negligent adverse events malpractice proceedings are underreported [23]; there is – obviously
– a great number of unreported cases or a ‘‘litigation gap”. The estimated factor of about 30 is realistic. By comparison of cases which
come to our attention with the number in epidemiological studies
areas with a high number of unreported cases can be identified.
These areas are
Medication errors, adverse drug reactions [8].
Infections, especially hospital infections.
Identification and evaluation of errors as well as reporting of errors may also contribute to the prevention of errors which is
among other medical disciplines also a task of forensic medicine
to improve patient safety [17,20].
Conflict of interest
The author states that there is no conflict of interest.
References
[1] Berner B. Tätigkeit der Gutachterkommissionen und Schlichtungsstellen in
Deutschland. In: Madea B, Dettmeyer R (Hrsg.). Medizinschadensfälle und
Patientensicherheit. Häufigkeit, Begutachtung, Prophylaxe. Köln: Deutscher
Ärzte-Verlag; 2007. p. 33–7.
[2] Bove KE, Iery C. The role of the autopsy in medical malpractice cases. I. A
review of 99 appeals court decisions. Arch Pathol Lab Med 2004;126:1023–31.
[3] Council of Europe, Committee of Ministers. Recommendation REC7 of the
Committee of Ministers to Member States on Medical Management of Patient
Safety and Prevention of Adverse Events in Health Care; 2006. <http://
www.coe.int/T/E/socialcohesion/health>.
[4] Dettmeyer R, Preuß J, Madea B. Rechtsmedizinische Behandlungsfehlerbegutachtung. Konsequenzen aus der im Auftrag des Bundesministeriums
für Gesundheit und Soziales durchgeführten mulizentrischen Studie.
Rechtsmedizin 2006;16:389–96.
S79
[5] Dettmeyer R, Preuß J, Madea B. Zur Häufigkeit begutachteter letaler
Behandlungsfehler in der Rechtsmedizin. In: Madea B, Dettmeyer R (Hrsg.).
Medizinschadensfälle und Patientensicherheit. Häufigkeit, Begutachtung,
Prophylaxe. Köln: Deutscher Ärzte-Verlag; 2007. p. 65–86.
[6] Dettmeyer R, Driever F, Becker A, Wiestler O, Madea B. Fatal
myeloencephalopathy due to accidental intrathecal vincristin administration
– a report of two cases. For Sci Int 2001;122:60–4.
[7] Deutscher Bundestag 15. Wahlperiode, Drucksache 15/530. Gutachten 2003
des Sachverständigenrates für die Konzertierte Aktion im Gesundheitswesen.
[8] Erikssen J. Deaths associated with drug therapy. 1. Deutscher Kongress für
Patientsicherheit bei medikamentöser Therapie, Saarbrücken, 19.4.2005; 2005.
[9] Hansis ML, Hansis DE. Der ärztliche Behandlungsfehler. 2. Aufl. Ecomed
Landsberg; 2001.
[10] Hansis ML, Hart D. Medizinische Behandlungsfehler in Deutschland.
Gesundheitsberichterstattung des Bundes 2001;4:1–15.
[11] Kohn LT, Corrigan JM, Donaldson MS, editors. To err is human. Building a safer
health system. Washington, DC: National Academy Press; 2001.
[12] Laum HD, Beck L. Großes Interesse im Ausland an außergerichtlicher
Schlichtung – Gutachterkommission für ärztliche Behandlungsfehler bei der
Ärztekammer Nordrhein zieht positive Bilanz des Berichtszeitraums 2001/
2002 – Kürzere Dauer der Verfahren. Rheinisches Ärzteblatt 2003;57:19–20.
[13] Lauterberg J. Diversität und Kerndatensatz – Bericht aus der Arbeitsgruppe
Behandlungsfehlerregister.
In:
Madea
B,
Dettmeyer
R
(Hrsg.).
Medizinschadensfälle und Patientensicherheit. Häufigkeit, Begutachtung,
Prophylaxe. Köln: Deutscher Ärzte-Verlag; 2007. p. 269–76.
[14] Madea B. Die ärztliche Leichenschau. Rechtsgrundlagen, praktische
Durchführung, Problemlösungen. 2. Aufl. Berlin, Heidelberg, New York,
Tokyo: Springer; 2006.
[15] Madea B, Henssge C, Lignitz E. Fahrlässige Tötung durch medikamentöse
Therapie. Rechtsmedizin 1993;4:123–31.
[16] Madea B, Dettmeyer R. Ärztliche Leichenschau und Todesbescheinigung. Dt
Ärztebl 2003;100A:3161–79.
[17] Madea B, Dettmeyer R. Behandlungsfehler und Medizinschadensfälle – nicht
nur ein Thema der Rechtsmedizin. Editorial. Rechtsmedizin 2006;16:253–4.
[18] Madea B, Preuß J, Vennedey C, Dettmeyer R. Ausgang strafrechtlicher
Ermittlungsverfahren
gegen
Ärzte
wegen
Verdachts
eines
Behandlungsfehlers. DMW 2006;131:2073–8.
[19] Madea B, Preuß J, Musshoff F, Dettmeyer R. Behandlungsfehlervorwürfe bei
Arzneimitteltherapie – Gutachterliche Aspekte. In: Kauert G, Mebs D, Schmidt
P (Hrsg.). Kausalität. Forensische Medizin, Toxikologie, Biologie, Biomechanik
und Recht. Hansjürgen Bratzke zum 60. Geburtstag. Berlin: Berliner
Wissenschafts-Verlag; 2006. p. 77–99.
[20] Madea B, Dettmeyer R. Medizinschadensfälle und Patientensicherheit.
Häufigkeit, Begutachtung, Prophylaxe. Köln: Deutscher Ärzte-Verlag; 2007.
[21] Madea B, Dammeyer Wiehe de Gomez B, Dettmeyer R. Zur Reliabilität von
Leichenschaudiagnosen bei fraglich iatrogenen Todesfällen. Kriminalistik
2007;12:767–73.
[22] Nichols L, Aronica P, Babe C. Are autopsies obsolete? Am J Clin Pathol
1998;110:210–8.
[23] Orben ST. Rechtliche Verantwortung für Behandlungsfehler. Hallesche
Schriften zum Recht. Band 19. Köln: Carl Heymanns Verlag KG; 2004.
[24] Preuß J, Dettmeyer R, Madea B. Begutachtung behaupteter letaler und nicht
letaler
Behandlungsfehler
im
Fach
Rechtsmedizin.
Bundesweite
multizentrische Studie im Auftrag des Bundesministeriums für Gesundheit
und Soziales (BMGS); 2005. <http://www.bmgs.de/deu/gra/publikationen/
pforschung.php>.
[25] Preuß J, Dettmeyer R, Madea B. Begutachtung behaupteter letaler
Behandlungsfehler im Fach Rechtsmedizin. Bundesweite Mulicenterstudie.
Rechtsmedizin 2006;16:367–82.
[26] Schrappe
M
et
al.
Aktionsbündnis
Patientensicherheit:
Agenda
Patientensicherheit. Witten: Selbstverlag; 2007.
[27] Schrappe
M
et
al.
Aktionsbündnis
Patientensicherheit:
Agenda
Patientensicherheit. Witten: Selbstverlag; 2008.
[28] Schrappe M, Lessing C. Zur Häufigkeit von Medizinschadensfällen. In: Madea B,
Dettmeyer R (Hrsg.). Medizinschadensfälle und Patientensicherheit. Häufigkeit,
Begutachtung, Prophylaxe. Köln: Deutscher Ärzte-Verlag; 2007. p. 21–32.
[29] Shojania K, Burton E, McDonald K, et al. The autopsy as an outcome and
performance measure. Evidence report/Technology Assessment Number 58
(prepared by the University of California at San Francisco-Stanford, Evidencebased practice centre under contract no. 290-97-0013). AHRQ Publication for
Health Care Research and Quality, October 2002.
[30] Shojania K, Burton E, McDonald K, Goldman MM. Changes in rates of autopsydetected diagnostic errors over time. JAMA 2003;289:2849–56.
ID
104055
Title
Medico-legalautopsiesasasourceofinformationtoimprovepatientsafety
http://fulltext.study/journal/66
http://FullText.Study
Pages
4