Appendix A

Introduction	Part 2	Part 3	Appendices
Part 1	Introduction to Part 2	Introduction to Part 3	Appendix A
Introduction to Part 1	Chapter 6	Chapter 9	Appendix B
Chapter 1	Chapter 7	Chapter 10	Appendix C
Chapter 2	Chapter 8	Chapter 11	Appendix D
Chapter 3	Conclusions to Part 2	Thesis Conclusions	Appendix E
Chapter 4			Appendix F
Chapter 5			Bibliography
Conclusions to Part 1			Books and Articles
			Legal Cases

Appendix A: Studies on the misdiagnosis of PVS

In recent years a number of studies have attempted to quantify the level of misdiagnosis occurring in the diagnosis of PVS. This appendix attempts to summarise the main results of these studies with particular emphasis on:

(i) the rates of misdiagnosis found.

(ii) the medical conditions which should have been diagnosed. And

(iii) the conceptual distinctions between these alternative conditions and PVS, which would have permitted a correct diagnosis.

An examination of the methodologies adopted in the studies is not undertaken as such an examination would bring us too far from our original purpose which is to estimate the extent of misdiagnosis and to show that misdiagnosis is of such a magnitude as to suggest that the supposed lack of consciousness of patients diagnosed as PVS is open to reasonable doubt. Although the various studies differ slightly in their definitions of PVS, they all accept that any patient who exhibits even a minimal level of awareness cannot be classified as PVS.

The studies which are examined in this appendix are listed in the following tables. Table A-1 lists those studies which are directly relevant to gauging the extent of misdiagnosis; Table A-2 identifies studies which indirectly deal with the misdiagnosis of PVS.

Author	Date	Country	Title
Donald Tresch et al	1990	US	‘Clinical Characteristics of Patients in the Persistent Vegetative State.’ Archives of Internal Medicine. 151, May 1991, 930 -2.
Nancy Childs et al	1992	US, Canada	‘Accuracy of Diagnosis of Persistent Vegetative State.’ Neurology 43, August 1993, 1465-7.
Keith Andrews	1993	UK	‘Recovery of patients after four months in the persistent vegetative state.’ British Medical Journal. 306, 12 June 1993, 1597-1599.
Keith Andrews et al	1996	UK	‘Misdiagnosis of the vegetative state: retrospective study in a rehabilitation unit’. British Medical Journal, 313, 6 July 1996), 13-16.

Table A-1: Studies which directly consider the misdiagnosis of PVS.

Author	Title
Keith Andrews (1992)	‘Managing the persistent vegetative state.’ British Medical Journal Vol. 305, (29 August 1992), 486-7.
* Andrew Grubb et al. (1996a)	‘Survey of British clinicians’ views on management of patients in persistent vegetative state.’ The Lancet 348, (6 July 1996), 35-40.
* Andrew Grubb et al. (1997)	Doctors’ views on the management of Patients in Persistent Vegetative State: A UK Study. London: The Centre of Medical Law and Ethics, King’s College, 1997.
Andrew Grubb et al.	Doctors’ views on the management of Patients in Persistent Vegetative State: A European Study. London: The Centre of Medical Law and Ethics, King’s College, (unpublished draft copy).
Andrew Grubb et al. (1996b)	Doctors’ views on the management of Patients in Persistent Vegetative State: A report of a small survey in Ireland. London: The Centre of Medical Law and Ethics, King’s College, 1996.
Ronald Cranford (1996)	‘Misdiagnosing the persistent vegetative state’. Editorial, British Medical Journal Vol. 313, 6 July 1996, 5-6.

* the first of these sources is a précis of the second.

Table A-2: Studies which indirectly consider the misdiagnosis of PVS.

The main studies[i]

The Tresch Study (1990)

Introduction

From a population of 1611 patients in four nursing homes in Wisconsin, 62 had been diagnosed by nursing home personnel - both nurses and physicians - as being in a PVS.[ii] This research project involved a re-examination of these patients to determine the correctness of the original diagnosis.

Rate of Misdiagnosis

The study found that of the 62 patients diagnosed as PVS, 11 had some level of awareness of their environment with some of these capable of demonstrating volitional movements. Given the criteria for PVS it follows that these 11 were misdiagnosed; this represents a misdiagnosis rate of 18%.

Comments

The authors expressed surprise at the large number of patients (25%) in a PVS who were diagnosed as having end stage dementia because:

“Patients with dementia rarely develop the profound and complete loss of awareness characteristic of PVS, and it is important to distinguish between PVS and the usual dementia state.” [iii]

The high percentage of PVS patients (27%) who were being treated with antipsychotic agents, also occasioned surprise as, in the circumstances, such treatment appeared inappropriate.[iv]

The Childs Study (1992)

Introduction

This study compared the pre-admission with the post-admission diagnoses for 193 patients suffering from severe brain injuries who had been referred[v] to a neurorehabilitation unit over a period of five years with a diagnosis of either PVS or coma.

Rate of Misdiagnosis

The study found:

“… behaviour indicative of cognitive responsiveness in 37% of the patients, thereby invalidating the pre-admission diagnosis of PVS or coma.” [vi]

Comments, including reasons for the misdiagnosis

The authors suggested two reasons for the misdiagnosis: firstly and most importantly, a confusion over terminology; secondly, a lack of diligence in the medical staff in their observation of patients for possible signs of awareness.

Terminology

The study found that:

“Referral sources used the terms ‘coma’ and ‘PVS’ inconsistently and interchangeably.” [vii]

The authors expressed surprise at such a fundamental error because:

“Simply put, coma is an eyes-closed state of unresponsiveness whereas in PVS the eyes are open.” [viii]

The main conclusion of the report was that:

“There is a need for consensus in description and definition of PVS. If we do not adhere to a strict and precise definition of PVS then confusion and misdiagnosis will follow.” [ix]

The link between inappropriate terminology and misdiagnosis was also noted:

“We suspect that misdiagnosis in most of these patients was due to the confusion in the terminology used to describe alterations in states of consciousness in the brain-injured. ... Additional confusion arises over appropriate terminology for patients demonstrating early awareness.” [x]

Observation

The report noted that because changes in patient behaviour may be subtle or occur over an extended period, they may initially appear to be random and thus be easily overlooked by the physician on his ward rounds. The report recommends that observations by the patient’s family should not be casually dismissed as the family often are in a better position to observe such changes than are medical staff.[xi]

The Andrews 1993 Study[xii]

This study was a retrospective review of the case notes of 43 patients admitted to a unit specialising in the rehabilitation of patients in PVS . The patients had not been preselected on the expectation that they would recover but it “... is impossible to say whether they were a representative sample of vegetative patients.” [xiii]

Rate of Misdiagnosis

Of the 43 patients in the study, 11 regained awareness though one of these patients was unable to communicate; if this one patient is included then the misdiagnosis rate is 26%; if not included, the rate is 23%.

The Andrews 1996 Study[xiv]

This was a retrospective study of 40 patients admitted between 1992 and 1995 to a rehabilitation unit dealing with profoundly brain damaged patients. All but one of the patients had been referred by a hospital consultant and the diagnosis of PVS had been made by a neurologist or neurosurgeon or rehabilitation specialist “... all of whom could have been expected to have experience of vegetative state.” [xv]

The study sought - by means of sensory stimulation of the patients in conjunction with sustained observation for possible responses - to identify mechanisms which would permit communication with the patient to be established. One patient, for example, was found to have a slight shoulder shrug which could be used to achieve a morse-like communication.[xvi] The main methods used to establish patient awareness were the “... ability to follow a simple command to press a buzzer switch or look at a named object.” [xvii]

Rate of Misdiagnosis

The study found a misdiagnosis rate of 43%. The survey method - in that it required patients to respond to verbal commands - implicitly assumed that none of the patients was deaf. In so far as 62% of these misdiagnosed patients were blind or severely visually impaired,[xviii] this assumption implies that the misdiagnosis rate could - if deafness was included - have been considerably in excess of 43%; accordingly this percentage should be interpreted as a minimum figure.

The subsidiary sources[xix]

Andrews (1992)

This article draws on Andrews’ considerable clinical experience of PVS and, though not directly concerned with misdiagnosis, touches on two issues which have great relevance to it. These issues are terminology and standard of care.

Terminology

Andrews contends that PVS is one of the “... least understood conditions in rehabilitative medicine.” [xx] As such there is a great danger that issues of terminology will exercise a dominance over issues of substance; therefore, it is highly important that an appropriate terminology - i.e. one without unjustified connotations - be used.

Andrews considers the terms ‘persistent’ and ‘vegetative’ to be unfortunate. ‘Persistent’ is understood by many to mean ‘permanent’ - that is a statement of final outcome rather than a comment on an existing state;[xxi] ‘vegetative’ - in its association with ‘vegetable-like’ - accentuates this tendency. Such misunderstandings hinder a positive attitude to treatment with the result that the diagnosis of PVS becomes “A self-fulfilling prophecy - the prognosis is poor, therefore no treatment is given, therefore the prognosis is poor.” [xxii]

Standard of Care

Andrews considers that appropriate stimulation programmes - of touch, taste, and sound - can affect patient responses. He states:

“ ... [in]10 years experience of training relatives of more than 250 patients in a persistent vegetative state to use a programme of stimulation ... only 4% of patients did not improve; one third became functionally independent.” [xxiii]

Cranford (1996)

This article was a critique of the Andrews 1996 study.[xxiv] Cranford questioned the method used to identify awareness - viewing ‘the buzzer system with some scepticism’ [xxv] - but this objection seems disingenuous.

Cranford’s distinction between the ‘late recovery’ and ‘late discovery’ of consciousness is of importance:

“Patients who start regaining consciousness several months after the injury (late recoveries) should not be confused with patients who may have been conscious for some time before discovery (late discoveries).” [xxvi]

However, Cranford’s most important point is his questioning of Andrews’ assertion that none of the misdiagnosed patients were vegetative at the time of admission.[xxvii]

The Grubb Reports (1996, 1997)

In 1994 Professor Grubb of The Centre of Medical Law and Ethics at King’s College was commissioned by the EU (under its BIOMED programme) to investigate the moral, legal and medical issues which arose in the management of PVS patients. A comparative study amongst the various EU countries was undertaken to examine both the legal structures affecting the treatment of PVS patients, and the moral attitudes to such treatment. In order to determine the attitudes of the medical profession to such questions a survey of doctors involved in the management of PVS patients was conducted by means of a postal questionnaire. Various reports analysing the results of these questionnaires were published; those for Ireland, the UK and the European Comparative study have been consulted in the writing of this thesis.[xxviii]

These reports did not directly address the question of misdiagnosis but were of indirect relevance in that:

(i) the level of confidence of doctors in making a ‘prediction of outcome’ for PVS patients was examined.

(ii) questions of terminology - and in particular the meaning of the term ‘recovery’ - were considered.

Medical Confidence in making a diagnosis of PVS

The consensus in the medical literature is that the vegetative state may be considered permanent after one year,[xxix] yet an appreciable percentage of practitioners who responded to the survey, appeared ‘not at all confident’ in predicting the outcome at this stage.[xxx] The Irish Report expressed surprise at this incongruity;[xxxi] the reason, however, may lie in the fact that the surveys omitted to establish the confidence level of doctors in making the original diagnosis of PVS. Because the only opportunity offered in the questionnaire for expressing lack of confidence was in relation to the ‘prediction of outcome’, it seems not unreasonable to suggest that a low level of confidence in the predicted ultimate outcome is related to wider concerns.

‘not at all confident’ in predicting outcome after:	Ireland	UK	Europe*
7-12 months	13%	9%	4% - 14%
1 year	7%	6%	3% - 16%

* tabulated by countries; minimum and maximum shown

Table A-3: Lack of confidence in predicting outcome after 6 months.[xxxii]

Terminology

The Grubb Reports were insistent on the necessity for clear terminology. They regretted the developments whereby ‘persistent’ has, to some ears, come to mean ‘permanent’ and they noted that “... confusion about the word ‘persistent’ has extremely important implications.” [xxxiii]

Some insights into the meaning of the term ‘recovery’ can be drawn from the report.

Recovery

The survey sought to relate the ‘predicted outcome’ to the decision to withdraw treatment. Three possibilities were offered to the respondents; the first was that the patient would continue in a vegetative state and this possibility achieved a nearly unanimous acceptance,[xxxiv] but it is the responses to the second and third possible outcomes[xxxv] that are of most interest; these results are summarised in the Table A-4:

	IRELAND				UK				EUROPE*
	Not treat acute infections		Withdraw ANH		Not treat acute infections		Withdraw ANH		Not treat acute infections		Withdraw ANH
	A.	B.	A.	B.	A.	B.	A.	B.	A.	B.	A.	B.
Yes	13%	-	0%	-	16%	32%	12%	22%	4% - 18%	15% - 39%	0% -14%	13% - 29%
No	53%	-	50%	-	64%	47%	68 %	52%	46% -82%	26% -68%	50% -82%	25% - 78%
Other	34%	-	50%	-	19%	21%	20 %	25%	13% -36%	13% -34%	13% -50%	9% - 63%

* tabulated by countries; minimum and maximum shown

Table A-4 : The willingness to withdraw treatment, if the predicted outcome from treatment continuance was severe disability with (A) with the ability to speak, or (B) with the ability to communicate but without speech.

It seems clear that such outcomes would not be classified as ‘recoveries’ by the doctors who were willing to withdraw treatment. This is of importance as it shows that the medical use of the term ‘recovery’ is not wholly amenable to, and resolvable by, scientific methods; it has an irreducible core of meaning which is amenable only to an ethical analysis.