Occupational Medicine 2006;56:329337 Published online 17 July 2006 doi:10.

1093/occmed/kql059

Symptoms, ill-health and quality of life in a support group of Porton Down veterans
Steven Allender1, Noreen Maconochie2, Thomas Keegan1, Claire Brooks1, Tony Fletcher2, Mark J. Nieuwenhuijsen3, Pat Doyle2, Lucy M. Carpenter1 and Katherine M. Venables1
...................................................................................................................................................................................

Background There has been a Human Volunteer Programme at the British chemical weapons research facility at Porton Down since the First World War, in which some of the participants were exposed to chemical warfare agents.
...................................................................................................................................................................................

Aim Methods

To identify any striking specic morbidity patterns in members of the Porton Down Veterans Support Group (PDVSG). A self-completed postal questionnaire was prepared including health immediately after the visits to Porton Down, subsequent diagnoses and hospital admissions, symptoms in, and after, the rst 5 years after the visits, fatigue symptoms and current quality of life, measured using the SF-36. Responses were received from 289 of 436 (66%). Results reported here relate to 269 male respondents of mean age 66.8 years. Sixty-six per cent reported their rst visit to Porton Down in the 1950s. The most common diagnoses or hospital admissions reported were diseases of the circulatory system. In the rst 5 years after their visits the most common symptoms were headache, irritability or outbursts of anger and feeling un-refreshed after sleep. In the later period, most common symptoms were fatigue, feeling un-refreshed after sleep and sleeping difculties. Sixty-ve per cent met the denition for a case of fatigue. Current quality of life dimensions were consistently lower than age-specic estimates from general population samples.

...................................................................................................................................................................................

...................................................................................................................................................................................

Results

...................................................................................................................................................................................

Conclusions Members of the PDVSG responding to this survey reported poorer quality of life than the general population. Despite there being no clear pattern of specic morbidities, we cannot rule out ill-health being potentially associated with past experience at Porton Down.
...................................................................................................................................................................................

Key words

Chemical warfare agents; quality of life; symptoms; veterans.

...................................................................................................................................................................................

Introduction
A chemical weapons research facility was established in the United Kingdom at Porton Down during the First World War. Since its inception in 1916, at least 30 000 British servicemen have taken part in experiments within the Human Volunteer Programme which has studied the potential effects on military capability of exposure to chemical warfare agents and, in particular, the effectiveness of protective measures against them. Between 1939 and 1989, up to 20 000 members of the armed forces
1 2 3

University of Oxford, Oxford OX3 7LF, UK. London School of Hygiene and Tropical Medicine, London WC1E 7HT, UK. Imperial College of Science, Technology and Medicine, London SW7 2AZ, UK.

Correspondence to: Katherine M. Venables, Department of Public Health, University of Oxford, Old Road Campus, Oxford OX3 7LF, UK. Tel: 144 1865 227034; fax: 144 1865 226720; e-mail: kate.venables@dphpc.ox.ac.uk

have taken part in tests involving a wide range of chemicals including nerve agents (e.g. sarin), blister agents (e.g. mustard), riot control agents (e.g. CS) and central nervous system incapacitants (e.g. LSD). It is estimated that 7000 of these veterans are still alive and a number of them have recently expressed concerns about long-term health impairments they attribute to taking part in tests at Porton Down. A review of possible long-term effects of several agents was published in 2000 [1] and a review specically relating to mustard gas and lewisite exposure was published by the US Institute of Medicine in 1993 [2]. In 2002, the Medical Research Council (MRC) commissioned a large programme of research on the health of the Porton Down veterans. The main part of this work, which is currently underway, involves a cohort study of mortality and cancer incidence in up to 40 000 veterans: those who went to Porton Down between 1939 and 1989 and a similar group of veterans who did not. An initial

The Author 2006. Published by Oxford University Press on behalf of the Society of Occupational Medicine. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

330 OCCUPATIONAL MEDICINE

part of the research programme included a small exploratory health survey of a specic group: all of the 436 members of the Porton Down Veterans Support Group (PDVSG). The purpose of the health survey was to describe the morbidity experienced by the PDVSG members and identify any striking specic morbidities which might have been unsuspected previously. Wherever possible, standard questions were used to enable comparisons with similar data collected from general population surveys. Ideally, the study would have included a representative sample of all Porton Down veterans with a comparison group of veterans who did not go to Porton Down but this was not considered practicable by the scientic advisors to the funders, given the level of funding.

Methods
Veterans invited to take part in the health survey were members of the PDVSG, a voluntary mutual support group which circulates a newsletter to its membership and maintains a web page. In addition to the veterans, members also include some spouses and relatives of deceased veterans. A self-completed postal questionnaire was designed containing sections relating to personal characteristics of the veteran; their experiences of tests, including chemical exposures and information provided to them at the time; current state of health and quality of life; health immediately after the tests; subsequent clinical diagnoses and hospital admissions for at least 24 h; a symptoms checklist relating to two time periods (within 5 years and .5 years after the tests); symptoms associated with fatigue; reproductive health; alcohol and tobacco consumption and open questions on the perceived effects on subsequent health of taking part in tests at Porton Down. Questions regarding exposure were derived from a questionnaire used in the Ministry of Defences voluntary Medical Assessment Programme (MAP) for Porton Down veterans [3]. Current quality of life was assessed using version 2 of the SF-36 questionnaire, one of the most widely used generic measures of subjective health status [4,5]. The symptoms check-list was derived from that used in a study of British soldiers in the rst Gulf War [6] with the modication that respondents were asked to compare their symptoms to people of comparable age. The fatigue scale was derived from an established set of questions [7] with a case of fatigue dened as someone scoring 4 or above [8]. Reproductive history questions were adapted from a study of reproduction and child health among British Gulf War veterans [9]. Smoking questions were taken from the 1986 MRC respiratory symptoms questionnaire [10]. A draft version of the questionnaire, an accompanying letter, information sheet and consent form were piloted

by six members of the PDVSG in November 2002 and modied accordingly. The questionnaire materials were mailed in July 2003, after a delay at the request of the PDVSG. Two further mailings were made to non-respondents or, in the case of overseas residents, one further mailing to ensure that questionnaires were completed at a comparable time. Researchers set up and answered queries from a telephone helpline for respondents during JulyOctober 2003. All questionnaires were entered in duplicate to minimize data entry errors. Reported clinical diagnoses and reasons for hospital admissions were coded according to the Tenth Revision of the International Classication of Diseases (ICD-10) [11]. Decisions on grouping the codes were made prior to examination of the data. SF36 scores were compared with those from four studies which provided general population data from surveys including older people: the 1996 Health Survey for England; the Ofce for National Statistics omnibus survey in Great Britain and two Oxford Healthy Life Surveys for 199192 and 1997 [12,13]. Reported exposures received during tests were derived by combining responses to an open question and a check-list. Owing to project resource constraints, responses to other open questions have not been coded. Descriptive statistics and graphs were prepared using Stata 7 and SPSS for Windows 12.0.0. Ethical approval was obtained from the South-East Multicentre Research Ethics Committee (02/01/115) and the Defence Medical Services Clinical Research Committee.

Results
Survey materials were sent to all 436 current PDVSG members. Five were returned undelivered because the address was incorrect. Of the remaining 431, 289 (67%) returned completed questionnaires, 44 (10%) declined to participate and there was no response from 98 (23%). Nineteen questionnaires were completed by a surrogate respondent on behalf of a deceased male veteran and one questionnaire was returned by a living female veteran. PDVSG members who were spouses or relatives of deceased Porton Down veterans were less likely to return a completed questionnaire (47% of 43) than PDVSG members who were living veterans (68% of 393). Unless stated otherwise, the results presented in this paper relate to the 269 living male veterans who either responded personally (266) or for whom a surrogate responded (3). The 269 male PDVSG respondents ranged in age from 32 to 88 years, with a mean of 66.8 years (SD 8.3) (Table 1). All but six veterans (98%) reported ever being married and 86% were currently living with a wife or partner. Over half (60%) had left school before taking O levels or

S. ALLENDER ET AL.: SYMPTOMS, ILL-HEALTH AND QUALITY OF LIFE IN THE PDVSG 331

Table 1. Characteristics of male PDVSG respondents Total, n (%) Age at 31 July 2003 (years) ,55 5559 6064 6569 7074 $75 Total Marital status Married or living with partner Never married Separated, divorced or widowed Total Highest level of education Left school before O levels/GCSE O level/GSCE A levels/highers Degree Total Current employment Working full time Working part time Unemployed, seeking work Not working due to ill-health Retired Total Ethnicity White Black/other Total Alcohol (units/week) None None, decreased in the past 5 years 13 410 1120 2130 .30 Total Smoking Never Ever Light (,20 per day) Heavy ($20 per day) Current Ex Total Age at rst visit ,20 2024 2529 3034 $35 Total Year of rst visit Pre-1950 195054 195559 196064 196569 1970 or later Total

26 19 31 92 78 23 269

(10) (7) (12) (34) (29) (9) (100)

228 (86) 6 (2) 32 (12) 266 (100) 148 52 29 19 248 41 19 2 33 172 267 (60) (21) (12) (7) (100) (15) (7) (1) (12) (65) (100)

266 (100) 1 (0) 267 (100) 34 15 61 71 43 27 12 263 67 201 93 104 55 146 268 (13) (6) (23) (26) (16) (10) (5) (100) (25) (75) (35) (39) (20) (54) (100)

99 (38) 139 (54) 17 (7) 2 (1) 2 (1) 259 (100) 21 (8) 98 (38) 73 (28) 21 (8) 23 (9) 23 (9) 259 (100)

GCSEs and all but one were white (Table 1). Although none were currently serving in the armed forces, 60 (22%) were in employment at the time of the survey. Most men (81%) reported being current drinkers of alcohol at the time of survey, 15% drinking 21 units/week or more (Table 1). The majority (75%) reported smoking, either currently (20%) or in the past, on average smoking a pack of (20) cigarettes a day for 33.1 (SD 24.4) years. Two-thirds of respondents reported having made their rst visit to Porton Down during the 1950s (66% of 259) (Table 1). The earliest reported visit was made in 1939 and the latest in 1990. Respondents were, on average, aged 21 years (SD 2.8) at their rst visit. Most (93% of 259) men reported having visited Porton Down in only one calendar year. Tests involving nerve agents were reported most frequently (59%), with 51% reporting exposure to a blister agent and 46% reporting tests involving other substances (Table 2). The most frequently reported circumstances of the tests involved a chamber (80%); tests involving a pill, tablet or drink were recalled least (21%) (Table 2). The majority (83%) of respondents rated the quality of information provided at the time of the tests about the purpose or nature of tests at Porton Down as poor or very poor. This percentage fell from 87 for respondents tested pre-1960 to 71 for those tested after 1969 (Appendix Table 1, available as supplementary data at Occupational Medicine Online). Two-thirds of male PDVSG respondents (68%) recalled experiencing a physical reaction within the rst few days of a test at Porton Down and 36% recalled a health problem within the rst month. The majority (92%) reported at least one doctor-diagnosed illness or hospital admission for .24 h between the last visit to Porton Down and the date of survey. As might be expected, this percentage tended to increase with age (Table 3). There was no apparent relationship between type of reported exposure and doctor-diagnosed illness or hospital admission overall (Table 3). Specic doctor-diagnosed illnesses or hospital admissions, as coded according to ICD-10 (lifetime, since Porton Down visit), are presented in Table 4. Illnesses and admissions are counted as many times as they were reported. The results are ordered according to the magnitude of the prevalence estimates for each of the major coding groups. Reported diseases of the circulatory, digestive and musculoskeletal systems were the most prevalent. The prevalence estimates of reported symptoms were derived from the question asking compared to people of your age at the time, have you experienced more of the following [symptoms]?. These related to two time periods: the rst 5 years following the visits to Porton Down and after the rst 5 years (Table 5). Fatigue, feeling unrefreshed after sleep and sleeping difculties were in the top ve for both time periods. The median number of

332 OCCUPATIONAL MEDICINE

Table 2. PDVSG self-reported test substance by test type Blister agent (% of 269) Tests involving A chamber Drops placed on your arm The medical incident room An injection A pill, tablet or drink Blood being taken Your eyes being examined Protective clothing Total
a

Nerve agent (% of 269)

Other substance (% of 269)

Responding yesa (% of 269)

115 125 37 41 32 60 70 74 137

(43) (46) (14) (15) (12) (22) (26) (27) (51)

149 97 41 56 29 83 103 70 158

(55) (36) (15) (21) (11) (31) (38) (26) (59)

105 84 47 55 46 64 72 75 124

(39) (31) (17) (20) (17) (24) (27) (28) (46)

214 169 65 84 57 120 131 115

(80) (63) (24) (31) (21) (45) (49) (43)

Subjects may appear in more than one category.

Table 3. Doctor-diagnosed illnesses and hospital admissions for >24 h at any time since the Porton Down visit(s) Total At least one doctordiagnosed illness or admission to hospital for .24 h at any time since the Porton Down visit(s), n (prevalence %) 247 (92) 66 87 72 22 (87) (95) (92) (96)

mentary data at Occupational Medicine Online). Of those ever attempting to father a pregnancy, 25% (62 of 249) had fathered at least one which ended in a miscarriage, stillbirth, other adverse pregnancy outcome or neonatal death. Ten per cent reported that they, or a partner, had consulted a doctor because of problems with getting pregnant.

Discussion
This morbidity survey collected information on medical histories, symptoms and quality of life from a group of the veterans who took part in experiments at Porton Down between 1939 and 1990. One notable nding from this survey is that the current self-reported quality of life of respondents was worse than that of men in the general population of a similar age; other ndings relating to medical histories and symptoms are discussed below. Notwithstanding issues discussed below relating to selection, measuring current quality of life using the SF-36 questionnaire enabled us to compare results with those of men in the general population of similar age. We identied four general population samples from the United Kingdom which included older age groups [12,13]. For each age group, and for each dimension of this measure, current quality of life in these male PDVSG respondents was lower than in the general population. This nding adds information to the recent study of 111 Porton Down veterans who attended the Ministry of Defences voluntary MAP which did not collect information on quality of life and apparently found no evidence of excess morbidity [3]. These veterans were members of an active support group and therefore inherently different from a representative sample of Porton Down veterans obtained by random sampling. The PDVSG are a small fraction (6%) of the total number (estimated at 7000) of Porton Down veterans from 193989 who are thought to be still alive. Respondent members of the PDVSG inevitably carried with them not only factors associated with their experience

All men alive at survey Age at survey (years) ,65 6569 7074 $75 Reported exposure at Porton Blister agent Nerve agent Other
a

269 76 92 78 23 Downa 137 158 124

127 (93) 147 (93) 112 (90)

Individuals may appear in more than one category.

symptoms reported was three (out of the total of 48) in the rst 5 years and 14 (out of 48) after the rst 5 years (Appendix Figure 1, available as supplementary data at Occupational Medicine Online). Figure 1 shows mean quality of life scores derived from the SF-36 questionnaire, which may range from 0 (worst possible quality of life) to 100 (best possible quality of life). The results are stratied by age and the age-specic mean scores from the four normative datasets specied earlier are presented. It can be seen that the mean quality of life scores for the male PDVSG respondents were lower than scores from the general population in each age group and for each dimension. Reporting on symptoms in the last few weeks, 160 (65%) respondents met the denition for a case of fatigue. Most (93%) respondents had fathered a pregnancy, or tried to father a pregnancy, with a median of two liveborn children (Appendix Table 2, available as supple-

S. ALLENDER ET AL.: SYMPTOMS, ILL-HEALTH AND QUALITY OF LIFE IN THE PDVSG 333

Table 4. Doctor-diagnosed illnesses and hospital admissions for >24 hours at any time since the Porton Down visit(s) reported by 269 male PDVSG respondents alive at time of survey: ordered by prevalence of major coding groups Diagnosis or reason for hospital admission (ICD-10 code) Diseases of the circulatory system (I00I99) Hypertensive and ischaemic heart diseases (I10I25) All other diseases of the circulatory system (I00I09, I26I99) Diseases of the digestive system (K00K93) Symptoms, signs and abnormal clinical and laboratory ndings not elsewhere classied (R00R99) Symptoms and signs involving the circulatory and respiratory systems (R00R09) General symptoms and signs (R50R69) Symptoms and signs involving the digestive system and abdomen (R10R19) Abnormal clinical and laboratory ndings (R70R94) Symptoms and signs involving the skin and subcutaneous tissue (R20R23) Symptoms and signs involving cognition, perception, emotional state and behaviour (R40R46) Symptoms and signs involving the nervous and musculoskeletal systems (R25R29) Symptoms and signs involving the urinary system (R30R39) Diseases of the musculoskeletal system and connective tissue (M00M99) Diseases of the respiratory system (J00J99) Other diseases of upper respiratory tract (J30J39) Chronic lower respiratory diseases (J40J47) All other diseases of the respiratory system (J00J29, J60J99) Diseases of the genitourinary system (N00N99) Injury, poisoning and other external causes of morbidity and mortality (S00T98, V01Y98) Mental and behavioural disorders (F00F99) Mood (affective), neurotic, stress-related and somatoform disorders (F30F48) All other mental and behavioural disorders (F00F29, F48F99) Neoplasms (C00D48) Malignant neoplasms (C00C97) Melanoma and other malignant neoplasms of skin (C43C44) All other neoplasms (D00D48) Diseases of the skin and subcutaneous tissue (L00L99) Dermatitis and eczema (L20L30) Papulosquamous disorders (L40L45) Endocrine, nutritional and metabolic diseases (E00E90) Diabetes mellitus (E10E14) All other endocrine, nutritional and metabolic diseases (E00E07, E15E90) Diseases of the nervous system (G00G99) Certain infectious and parasitic diseases (A00B99) Factors inuencing health status and contact with health services (Z00Z99) Diseases of the eye and adnexa (H00H59) All other diseases of the eye and adnexa (H15H59) Disorders of eyelid, lacrimal system, orbit and conjunctiva (H00H13) Diseases of the ear and mastoid process (H60H95) Diseases of the blood and blood-forming organs and certain disorders involving the immune system (D50D89)
*Subjects may appear in more than one category, groupings were determined to be of potential a priori interest.

n (prevalence %*) 103 (38) 65 (24) 63 (23) 87 (32) 65 (24) 28 (10) 22 (8) 9 (3) 9 (3) 4 (1) 4 (1) 3 (1) 2 (1) 62 (23) 61 26 25 22 (23) (10) (9) (8)

50 (19) 33 (12) 32 (12) 32 (12) 14 (5) 30 (11) 27 (10) 5 (2) 4 (1) 28 (10) 11 (4) 8 (3) 23 (9) 17 (6) 6 (2) 21 (8) 19 (7) 17 (6) 16 (6) 13 (5) 3 (1) 11 (4) 5 (2)

334 OCCUPATIONAL MEDICINE

Table 5. Recall of symptom(s) in (i) the rst 5 years and (ii) after the rst 5 years after the visit(s) to Porton Down [ranked according to prevalence (%)] Symptoms in rst 5 years Headaches Irritability/outbursts of anger Feeling un-refreshed after sleep Fatigue Sleeping difculties Itchy or painful eyes Dry mouth Sore throat Loss of concentration Increased sensitivity to light Feeling jumpy/easily startled Dizziness Distressing dreams Unable to breathe deeply enough Tingling in ngers and arms Flatulence or burping Numbness or tingling in ngers or toes Forgetfulness Feeling short of breath at rest Persistent cough Increased sensitivity to noise Wheezing Tingling in legs and arms Feeling distant or cut off from others Rapid heartbeat Nausea Feeling disoriented Stomach cramp Faster breathing than normal Diarrhoea Chest pain Unintended weight gain .10 lbs Passing urine more often Double vision Constipation Vomiting Loss of interest in sex Shaking Feeling feverish Intolerance to alcohol Lump in throat Loss or decrease in appetite Sexual problems Unintended weight loss .10 lbs Burning sensation in the sex organs Pain on passing urine Sub-fertility Any disease in genital organs Total (%) 87 83 81 79 78 65 59 59 59 57 55 53 53 51 50 50 49 46 45 44 44 42 42 42 41 41 40 38 37 37 36 33 32 31 30 29 28 25 25 23 21 20 17 16 14 13 13 6 (32) (31) (30) (29) (29) (24) (22) (22) (22) (21) (20) (20) (20) (19) (19) (19) (18) (17) (17) (16) (16) (16) (16) (16) (15) (15) (15) (14) (14) (14) (13) (12) (12) (12) (11) (11) (10) (9) (9) (9) (8) (7) (6) (6) (5) (5) (5) (2) Symptoms after rst 5 years Fatigue Feeling un-refreshed after sleep Sleeping difculties Loss of concentration Forgetfulness Irritability/outbursts of anger Numbness or tingling in ngers or toes Passing urine more often Tingling in ngers and arms Chest pain Dry mouth Itchy or painful eyes Headaches Flatulence or burping Unable to breathe deeply enough Tingling in legs and arms Loss of interest in sex Dizziness Distressing dreams Rapid heartbeat Feeling short of breath at rest Feeling jumpy/easily startled Unintended weight gain .10 lbs Increased sensitivity to noise Increased sensitivity to light Feeling disoriented Wheezing Sore throat Sexual problems Faster breathing than normal Persistent cough Feeling distant or cut off from others Stomach cramp Diarrhoea Shaking Constipation Nausea Loss or decrease in appetite Intolerance to alcohol Double vision Feeling feverish Pain on passing urine Vomiting Unintended weight loss .10 lbs Burning sensation in the sex organs Lump in throat Sub-fertility Any disease in genital organs Total (%) 159 152 140 134 133 120 120 118 115 107 107 106 105 105 104 104 102 99 99 94 94 93 90 89 80 79 75 72 71 70 70 68 64 60 57 56 53 50 48 47 41 39 37 37 31 30 28 19 (59) (57) (52) (50) (49) (45) (45) (44) (43) (40) (40) (39) (39) (39) (39) (39) (38) (37) (37) (35) (35) (35) (33) (33) (30) (29) (28) (27) (26) (26) (26) (25) (24) (22) (21) (21) (20) (19) (18) (17) (15) (14) (14) (14) (12) (11) (10) (7)

at Porton Down but also those associated with becoming a member of a support group, plus any factors associated with responding to a survey. The factors associated with membership of any support group such as the PDVSG are likely to include concern about the potential for long-term health effects and, possibly, the presence of symptoms. Considering medical histories, the most commonly reported diagnoses and reasons for hospital admission

were diseases of the circulatory system (38% of 269). In contrast to quality of life, normative data for doctordiagnosed illnesses and admissions to hospital over the adult lifetime were not readily available. For reproductive history, however, the distribution of the number of livebirths seems to generally t the average family size expected for men born in the United Kingdom between approximately 1930 and 1950 [14]. When interpreting

S. ALLENDER ET AL.: SYMPTOMS, ILL-HEALTH AND QUALITY OF LIFE IN THE PDVSG 335

100 90 80 70 60 50 40 30 20 10 0

a) General health perception

100 90 80 70 60 50 40 30 20 10 0

b) Physical function score

35-44

45-54

55-64

65-74

75-84

35-44

45-54

55-64

65-74

75-84

100 90 80 70 60 50 40 30 20 10 0

c) Role physical
100 90 80 70 60 50 40 30 20 10 0

d) Role emotional

35-44

45-54

55-64

65-74

75-84

35-44

45-54

55-64

65-74

75-84

100 90 80 70 60 50 40 30 20 10 0

e) Social functioning
100 90 80 70 60 50 40 30 20 10 0

f) Mental health

35-44

45-54

55-64

65-74

75-84

35-44

45-54

55-64

65-74

75-84

100 90 80 70 60 50 40 30 20 10 0

g) Energy/vitality

100 90 80 70 60 50 40 30 20 10 0

h) Bodily pain

35-44

45-54

55-64
PDVSG

65-74

75-84

35-44

45-54

55-64

65-74

75-84

Health Survey for England 1996 British ONS survey 1992

Oxford Healthy Lifestyle Survey III 1997 Oxford Healthy Life Survey 1991-92

Figure 1. SF-36 scores for male PDVSG respondents and population-based estimates by age. Dimension scores presented from all surveys for men only. Age groups with less than ve excluded.

336 OCCUPATIONAL MEDICINE

the nding that 10% of the group reported consulting a doctor for difculties conceiving, it is interesting to note that in a large survey of male UK nuclear industry workers a similar proportion of men .40 years of age (11%) reported that they had contacted a medical practitioner regarding difculties conceiving [15]. Considering fatigue symptoms, while we have not identied appropriate normative data for older populations, a postal survey was carried out of all patients aged between 18 and 45 years in three London and three rural or semi-rural general practitioner practices in the United Kingdom; of 6532 men surveyed 31% met the denition for a case of fatigue, compared to 65% in the present study [16]. It is possible, however, that the higher prevalence in the respondent members of the PDVSG reects their older age. For other symptoms, it is possible to compare the present ndings with those from veterans of the rst Gulf War who, although younger, are also military veterans. The ve most prevalent symptoms reported as occurring in the rst 5 years after tests at Porton Down were similar to those reported by British Gulf War veterans [6,17]. For symptoms reported as experienced since the rst 5 years after visiting Porton Down, three of the top ve, and six of the top ten, symptoms are similar to those reported by the Gulf War veterans. The significance of this similarity is unclear because there is considerable overlap in symptom patterns reported following wars and other potentially harmful experiences [18]. The older age of many of the PDVSG membership and the long interval since visiting Porton Down does not rule out post-traumatic stress disorder, which can occur in later life [19]. Our study should also be placed in the context of other information on the morbidity of veterans of chemical warfare research programmes assembled from the United States. The Institute of Medicine in the United States reported in 1993 the ndings of a committee convened to survey the literature relevant to health effects of mustard gas and lewisite exposure [2]. As part of its work, this committee held a public hearing and responses were received from 257 veterans of the American chemical warfare research programme. A wide variety of diagnoses were reported but the committee commented that it was difcult to interpret this information. In the United Kingdom, detailed information on the experience and health of several individual Porton Down veterans was presented during a parliamentary debate in 2005 and is available in Hansard [20]. In contrast to these other sources of information, the present study has made comparisons with published normative data and has shown that the PDVSG members who responded to the survey had worse current quality of life than the general population. It might have been expected that veterans whose health was affected by their experience at Porton Down would have been particularly likely to join this support group.

While this survey has not identied any striking pattern of specic morbidities among the respondents, it cannot rule out any specic types of ill-health as potentially associated with experience at Porton Down, which may have policy implications. Regarding the implications of this survey for the future analysis of the ongoing cohort study of mortality and cancer incidence, the variety of diagnoses reported by the PDVSG is wide and these results have not allowed a focus on specic types of ill-health. Among the several conclusions and recommendations made in 1993 by the US Institute of Medicine committee was that . . . morbidity and mortality studies should be accomplished . . . comparing . . . cohorts to appropriate control groups in order to resolve some of the remaining questions about the health risks associated with exposure to these agents [2]. In the United Kingdom, an important step has been taken with the start of such a cohort study. This includes 20 000 veterans who attended Porton Down between 1939 and 1989 together with 20 000 military controls. As well as studying cause-specic mortality, this study will collect information on other significant conditions contributing to death, and will also study cancer registration. The latter is important given that the chemicals used as chemical warfare agents include several known carcinogens.

Key points This study is an important preliminary step in the UK programme of research on veterans of the human testing programme at Porton Down. It provides descriptive information about the current health of members of the PDVSG which points to impaired quality of life among the respondent men without identifying a clear explanation. The main part of the research programme will follow up this nding by means of a cohort study of the cause-specic mortality and cancer incidence of 20 000 veterans who attended Porton Down between 1939 and 1989 together with 20 000 military controls.

Acknowledgements
The researchers are grateful to Ken Earl, Eric Gow and members of the PDVSG for helping to formulate the questions, piloting the questionnaire, mailing out the survey materials and participating in the survey. Without their active support, this survey would not have been possible. We acknowledge invaluable discussions with Valerie Beral at the start of this project. Simon Wessely provided information on surveys of Gulf War veterans and Crispin Jenkinson on quality of life

S. ALLENDER ET AL.: SYMPTOMS, ILL-HEALTH AND QUALITY OF LIFE IN THE PDVSG 337

measurement. Charlie Foster provided advice on the management of calls to the survey telephone helpline. Mandy Roberts provided clerical support. We are also grateful for general advice from the Gulf Veterans Illness Unit of the Ministry of Defence (now the Veterans Policy Unit). The study was funded by a grant from the UK Ministry of Defence administered by the MRC (G0200288). In commissioning this study, it was pre-specied that a mailed survey of the PDVSG should be carried out and that the report should be submitted to a peer-reviewed journal. The sponsors had no role in the collection, analysis or interpretation of data or in the writing of the report. Steven Allender was supported by a research grant from the Higher Education Funding Council for England (GMP 229).

Conicts of interest
None declared.

References
1. Karalliedde L, Wheeler H, Maclehose R, Murray V. Possible immediate and long-term health effects following exposure to chemical warfare agents. Public Health 2000;114:238248. 2. Pechura CM, Rall DP, eds. Veterans at Risk: The Health Effects of Mustard Gas and Lewisite. Washington, DC: National Academy Press, 1993. 3. Lee HA, Gabriel R, Bale AJ, Welch D. Clinical ndings in 111 ex-Porton Down volunteers. J R Army Med Corps 2004;150:1419. 4. Walters SJ, Munro JF, Brazier JE. Using the SF-36 with older adults: a cross-sectional community-based survey. Age Ageing 2001;30:337343. 5. Ware JE, Kosinski M, Dewey JE. How to Score Version 2 of the SF-36 Health Survey. Lincoln, RI: Quality Metric Corporation, 2000. 6. Unwin C, Blatchley N, Coker W et al. Health of UK servicemen who served in the Persian Gulf War. Lancet 1999;353:169178. 7. Chalder T, Berelowitz G, Pawlikowska T et al. The development of a fatigue scale. J Psychosom Res 1993;37: 147153.

8. Deale A, Husain K, Chalder T, Wesseley S. Long-term outcome of cognitive behaviour therapy versus relaxation therapy for chronic fatigue syndrome: a 5-year follow-up study. Am J Psychiatry 2001;158:20382042. 9. Maconochie N, Doyle P, Davies G et al. The study of reproductive outcome and the health of offspring of UK veterans of the Gulf war: methods and description of the study population. BMC Public Health 2003;3:4. 10. Medical Research Council. Questionnaire on Respiratory Symptoms. London: MRC, 1986. 11. World Health Organization. International Statistical Classication of Diseases and Related Health Problems (10th Revision). Geneva: World Health Organization, 1992. 12. Bowling A, Bond M, Jenkinson C, Lampin DL. Short form 36 (SF-36) health survey questionnaire: which normative data should be used? Comparisons between the norms provided by the Omnibus Survey in Britain, The Health Survey for England and the Oxford Healthy Life Survey. J Public Health Med 1999;21:255270. 13. Jenkinson C, Stewart-Brown S, Petersen S, Paice C. Assessment of the SF-36 version 2 in the United Kingdom. J Epidemiol Community Health 1999;53:4650. 14. Ofce for National Statistics. Birth Statistics FM1 No 32. London: Ofce for National Statistics, 2003. 15. Doyle P, Roman E, Maconochie N, Davies G, Smith PG, Beral V. Primary infertility in nuclear industry employees: a report from the Nuclear Industry Family Study. Occup Environ Med 2001;58:535539. 16. Pawlikowska T, Chalder T, Hirsch SR, Wallace P, Wright DJM, Wessely SC. Population based study of fatigue and psychological distress. Br Med J 1994;3008: 763766. 17. Cherry N, Creed F, Silman A et al. Health and exposures of United Kingdom Gulf War veterans. Part 1: the pattern and extent of ill health. Occup Environ Med 2001;58: 291298. 18. Hotopf M, Wessely S. Can epidemiology clear the fog of war? Lessons from the 199091 Gulf War. Int J Epidemiol 2005;34:791800. 19. Busuttil W. Presentations and management of post traumatic stress disorder and the elderly: a need for investigation. Int J Geriatr Psychiatry 2004;19:429439. 20. http://www.publications.parliament.uk/pa/cm200405/ cmhansrd/cm050222/halltext/50222h03.htm#50222h03_ head0