5. Cross-sectional study methods
- 5.1 Aims and objectives
- 5.2 Study design
- 5.3 Study population
- 5.4 Sample size
- 5.5 Contact strategy and recruitment procedures
- 5.6 Data collection instrumentation and measures
5.1 Aims and objectives
The Australian Gulf War Veterans’ Health Study was designed to investigate whether Australian Defence Force personnel who served in the Gulf War have a higher than expected rate of several adverse health effects; and, if so, whether these effects are associated with exposures and experiences that occurred in the Gulf War.
Therefore, the specific research null hypothesis to be addressed was:
‘The health (both physical and psychological) of Australian veterans of the Gulf War does not differ significantly from similar Defence Force personnel who were not deployed to the Gulf War.’
Within this overall hypothesis are several specific research questions that the study was designed to answer:
- Do Australian Gulf War veterans have increased prevalence of psychological disorders including depression, anxiety and substance disorders; and if so, are these associated with exposures and experiences that occurred in the Gulf War?
- Do Australian Gulf War veterans have increased prevalence of symptoms, symptom clusters and medical conditions, related to several body systems; in particular psychological, respiratory, neurological; musculoskeletal and skin, and if so, are these associated with exposures and experiences that occurred in the Gulf War?
- Do Australian Gulf War veterans have increased prevalence of chronic fatigue syndrome; and if so, is this associated with exposures and experiences that occurred in the Gulf War?
- Do Australian Gulf War veterans have significantly lower lung function than expected; and if so, is this associated with exposures and experiences that occurred in the Gulf War?
- Do Australian Gulf War veterans have an increased prevalence of laboratory test results that are indicative of adverse health effects, including evidence of increased rates of markers of infection; and if so, are these associated with exposures and experiences that occurred in the Gulf War?
- Do Australian Gulf War veterans have an increased risk of having a child with a major congenital malformation or increased risk of childhood cancer, or have an increased risk of infertility, following return from the Gulf and, if so, are these associated with exposures and experiences that occurred in the Gulf War?
- Do Australian Gulf War veterans have increased rates of cancer and mortality?
5.2 Study design
The Australian Gulf War Veterans’ Health Study was designed to:
- Establish a cohort of Australian Gulf War veterans to determine all-cause mortality and cancer incidence rates using data from the National Death Index and the National Cancer Statistics Clearing House. This cohort can be followed prospectively to measure cancer and mortality experience and to monitor other health indicators.
- Undertake a cross-sectional study of this Gulf War veterans’ cohort and a frequency matched comparison group to investigate the other research questions of the study.
This chapter describes the methodology for the cross-sectional study. The mortality and cancer incidence cohort study is described in chapter 17.
5.3 Study population
5.3.1 Definition of the Gulf War veteran group
The Gulf War veteran group has been defined as all Australian Defence Force (ADF) personnel, from the Royal Australian Navy, Royal Australian Army, and the Royal Australian Air Force, who served in the Gulf War and who are listed on the Department of Veterans’ Affairs Nominal Roll for that conflict. To be included on the Nominal Roll an ADF member must have been deployed to the Gulf sometime during the period 2 August 1990 to 4 September 1991.
The Nominal Roll includes:
- ADF personnel on temporary as well as permanent postings. The majority of ADF members listed on the Nominal Roll deployed as part of Operation Ozone and Operation Damask, or with overseas forces as part of Operations Desert Shield and Desert Storm.
- Personnel who were members of the Navy, Army or Air Force Reserves.
- Personnel on board the second deployment of HMAS Darwin which reached the Gulf just after the war ended, serving from 13 June 1991 to 4 September 1991.
- Personnel who went to Kurdistan in northern Iraq as part of Operation Habitat to provide humanitarian aid from 16 May 1991 until 30 June 1991.
- Operation Blazer personnel who deployed to Iraq immediately after the war ended in support of the United Nations Special Commission (UNSCOM) to oversee the destruction of weapons of mass destruction.
- Support personnel who deployed to the Middle East to provide logistic support to the ships and aircraft. These include personnel from the Royal Australian Air Force (RAAF) Mobile Air Terminal Unit (MATU), and the Navy Logistic Support Element (LSE).
- Those personnel who were on other Defence duties in the Middle East at the time of the Gulf War and who were deployed in support of other Defence duties or United Nations (UN) peacekeeping operations. These include personnel serving with the UN Truce Supervision Organisation (UNTSO) in Beirut, South Lebanon, Israel and Syria as well as those deployed as part of the Multinational Force and Observers (MFO) in the Sinai.
- Subjects who were in non-Defence roles in the area of the Gulf, for example embassy personnel and reporters.
- Personnel deployed to UNSCOM as part of Operation Blazer after 4 September 1991.
- Personnel deployed to the Gulf and the Red Sea in support of Operations Damask IV - IX after 4 September 1991.
5.3.2 Definition of the comparison group
Comparison group subjects for this study were defined as Australian Defence Force personnel who were operational in the Royal Australian Navy, Royal Australian Army or Royal Australian Air Force at the time of the Gulf War, eligible to be deployed to the Gulf, but either not sent to the Gulf or not otherwise eligible for inclusion on the Gulf War Nominal Roll according to the criteria provided above.
To identify the eligible comparison group two lists were compiled: one list of ADF personnel posted to Maritime, Land or Air Operational Units as at 1 August 1990 and a second list of those posted to Maritime, Land or Air Operational Units as at 1 August 1991. This included personnel on either permanent postings or temporary attachments to ships, units or squadrons including members of the Navy, Army or Air Force Reserves. The two dates were selected to ensure that those personnel who were posted to operational units during the period of the Gulf War were not missed for comparison group selection. The lists were merged and then Gulf War veterans, as defined by their inclusion on the Gulf War Nominal Roll, were removed leaving a total, eligible comparison sample of 5481 Navy, 6481 Army and 14494 RAAF personnel.
From the total eligible comparison sample for each service type, subjects were randomly selected using frequency matching to the Gulf War veteran group to achieve the sample sizes described below. The criteria used for the frequency matching varied across Service type. Navy comparison group subjects were matched with Navy Gulf War veteran group subjects according to sex and 3-year age bands. The rank distribution, within the Army units which were deployed to the Gulf War, was considered not representative of the larger Army operational force, and therefore the Army comparison group subjects were matched with Army Gulf War veteran group subjects according to sex, year of birth and two service rank categories (‘Officer’ and ‘Other ranks’). Similarly the distribution of personnel in aircrew versus non-aircrew roles, within the Air Force units which were deployed to the Gulf War, was considered not representative of the larger Air Force operational force, and therefore the Air Force comparison group subjects were matched with Air Force Gulf War veteran group subjects according to sex, year of birth and the two job categories Aircrew and Non-Aircrew.
5.4 Sample size
5.4.1 Gulf War veteran group
At commencement of the Australian Gulf War Veterans’ Health Study the Gulf War Nominal Roll totalled 1873 people. Of these, 1581 were from the Navy (including 8 known deceased), 123 were from the Army (including 1 known deceased) and 169 were from the Air Force (including 4 known deceased).
The entire Nominal Roll list was included in the Gulf War veteran group. Inclusion of the entire cohort was considered both feasible by the study team and necessary to maximise the study's potential to evaluate differences between the Gulf War veteran group and the comparison group and to allow subgroup analyses within the Gulf War veteran group.
5.4.2 Comparison group
The comparison group sample size was initially chosen to equal the size of the Gulf War veteran group of 1873 subjects. It was calculated that a recruitment rate of 80% per group would result in approximately 1500 Gulf War veteran and 1500 comparison group subjects participating in the study. With this number of subjects providing symptom information, the study was estimated to have at least 90% power, at a two-sided 5% significance level, to detect increases in the prevalence of symptoms in Gulf War veterans of the order of 20% to 100% (corresponding to odds ratios of 1.3 to 2.0) for symptoms with a prevalence of 2% to 30% (in reverse order) among the comparison group.
After six months of recruitment, it was determined that an 80% recruitment rate could be achieved for the Gulf War veteran group but that only a 60% recruitment rate could be achieved for the comparison group. Based on these revised recruitment forecasts, the comparison group sample size was increased by randomly sampling additional subjects so as to yield an equal number of eligible Navy comparison group respondents, and twice the number of Air Force and Army comparison group respondents, as compared with the Gulf War veteran sample. This was intended to bring the study power back in line with the original estimates and also allow a buffer in the event that the final recruitment rate was slightly lower than that forecast. The doubling of the Army and Air Force comparison group samples would also result in additional power to detect moderate to large effects when these small groups were analysed separately from the larger Navy group.
The Navy comparison group was further supplemented in response to the finding that approximately 11% of the original Navy comparison sample were ineligible to participate as they had been discharged from the RAN prior to 1 August 1990.
The final comparison group sample size totalled 3192 subjects with 2384 from the Navy, 338 from the Army and 470 from the Air Force.
5.5 Contact strategy and recruitment procedures
5.5.1 DVA-based Contact and Recruitment team
The provisions of the Federal Privacy Act (1988) required that Monash University and Health Services Australia (HSA) not be allowed access to DVA-held information pertaining to the names and addresses of the Gulf War veteran and comparison group subjects until those subjects consented to participate in the study. Thus a DVA-based Contact and Recruitment team was established. The role of the Contact and Recruitment team was to:
- Obtain and maintain current contact details for the Gulf War veteran and comparison groups.
- Undertake initial mail contact with eligible subjects.
- Follow-up subjects who did not respond to the initial mail contact by attempting phone and/or e-mail contact.
- Respond appropriately to the queries and concerns of eligible subjects so that they could make an informed decision about participating in the study.
- Record the outcome of the recruitment effort for each Gulf War veteran and comparison group subject. For example, record whether the subjects agreed to participate, declined participation, were overseas, not contactable and so on.
- Forward the personal contact details of subjects agreeing to participate to HSA.
5.5.2 Contact procedures
All eligible subjects were initially invited to participate via mailed invitation. Mailouts of invitation packages to Navy subjects commenced in September 2000, followed by mailouts to Air Force subjects commencing in February 2001 and mailouts to Army subjects commencing in May 2001. Generally, during these initial mailouts, invitation packages were mailed in batches of 300 per week and these went to subjects from both the Gulf War veteran group and the comparison group each week. Those subjects who did not respond to the initial invitation package, within two weeks of its dispatch, were sent a reminder information package.
Phone and if possible, e-mail contact was then attempted for those subjects who did not respond to the reminder invitation package within two weeks of its dispatch. Phone contact was attempted at a variety of times across the day, including on nights and at weekends, to ensure maximum contact success.
The initial invitation package contained:
- A personally addressed letter of endorsement from the Minister for Veterans' Affairs.
- A personally addressed letter of endorsement from members of the study Consultative Forum.
- A letter of invitation to participate in the study from the Monash University and HSA study team.
- The study Explanatory Statement.
- A copy of the study Consent Form.
- A page providing Freecall 1800 contact numbers for the Contact and Recruitment team and for the Monash University/HSA study team.
The complete set of these mailout materials is provided at Appendix 1.
The reminder package was identical to the initial package, with the exception that the Minister's letter was replaced with a personal letter from the Principal Investigator of the Monash University/HSA study team. This letter is also provided at Appendix 1.
If subjects did not wish to participate, they were asked to indicate this by phoning the Freecall 1800 contact number so that the Contact and Recruitment team knew that the invitation package had been received and so that reminder notices about participation would not be sent.
The initial set of address and phone contact details for study subjects, used by the Contact and Recruitment team, were those recorded as 'last known' on the Department of Defence Personnel Management System database. Where addresses or phone contact details proved to be incomplete or out of date, the Contact and Recruitment team carried out the following search procedures to identify current contact details:
1. Search the DVA Client Database.
2. Search the CD-ROM version of the Telstra White Pages.
3. Search addresses, provided by the Australian Electoral Commission, as listed on the most recent version of the Electoral Roll.
4. Search the DVA Client Database for possible relatives in the area listed for "Address on Enlistment" on the study database.
5. Check for new addresses, or addresses for any Next of Kin, on the relevant Department of Defence Personnel Management System database.
6. Forward invitation packages to ComSuper, from where packages were further forwarded on to study subjects if ComSuper held an address different to that known by DVA.
7. Request that the Health Insurance Commission (HIC) provide a more current address, if one was held, from the Medicare database.
When a new address was found, a new information package was dispatched and this was followed up with a reminder package and follow-up phone contact if necessary.
5.5.3 Methods to maximise participation
In an effort to maximise participation in the study, and in addition to the mailouts and follow-up efforts described above, the following strategies were employed:
- Promotion of the study via the mass-media including Ministerial press releases, television coverage, newspaper articles and advertisements in Defence-related publications,
- Informed promotion of the study by the Consultative Forum to the members of the organisations they represented,
- Presentations to the members of interested groups and associations,
- Liaison with Defence Force personnel to facilitate the location of serving Defence Force members. This included the logging of Ship locations and expected return dates to facilitate contact of deployed Navy subjects,
- Organisation of medical teams to conduct medical assessments at locations remote to the standard HSA offices. For example, medical assessments were conducted by assessment teams sent to Cairns in Queensland and to the Western Australian HMAS Stirling Naval base,
- Reimbursement of any loss of income incurred as a result of participation,
- Reimbursement of travel and accommodation costs incurred as a result of participation,
- Provision of vouchers for meals and refreshments during the course of attendance at the medical assessment,
- Agreement by the Department of Defence that serving Defence Force members be able to attend as part of their duties,
- Availability of flexible appointment times, including some late afternoon and Saturday appointments.
5.5.4 Participation options
Full participation in the study involved completing a written postal questionnaire and attending a medical assessment at a Health Services Australia (HSA) Clinic. The questionnaire was estimated to take up to two hours to complete and the medical assessment was estimated to take up to five hours with the inclusion of a break. Travel time to and from the HSA clinics varied considerably depending on a subject’s residential address. In some cases, where the travel time was lengthy, participants were accommodated overnight.
Where subjects were willing to participate in the study, but were unable to undertake the travel or offer the time necessary for full-participation, they were offered the option of becoming a 'postal questionnaire only’ participant. These participants completed the postal questionnaire but did not attend for the medical assessment. The postal questionnaire was forwarded to participants in the mail along with a reply-paid envelope for its return.
5.5.5 Telephone-questionnaire option
Subjects who were contacted but who declined to complete either the medical assessment or the postal questionnaire, were offered the option of completing a brief telephone-administered questionnaire. The telephone questionnaire (Appendix 2) contained the Short-Form-12 Health Survey (SF-12) (see section 5.6.1.3.1) and some demographics and lifestyle questions including country of birth, highest education level, occupational status and smoking history.
The telephone questionnaire was administered by the DVA contact and recruitment team, and also by HSA administrative staff.
5.5.6 Administration of appointments for medical assessments
When subjects agreed to fully participate in the study, by agreeing to undertake the medical assessment and postal questionnaire, their contact details were forwarded via weekly electronic dispatch to a designated recipient at the HSA Head Office in Canberra. HSA then undertook responsibility for arranging the completion of the postal questionnaires and the medical assessments according to the following procedure:
1. Upon import of the electronically transferred data, subject's postcodes were used to allocate individuals to that HSA office which appeared to be the closest to their home address. Study participants could be allocated to any one of ten HSA offices, these being located at:
- Canberra ACT
- Parramatta NSW
- Newcastle NSW
- Wollongong NSW
- Surry Hills NSW
- Melbourne VIC
- Adelaide SA
- Perth WA
- Brisbane QLD
- Darwin NT
Tasmanian study participants were allocated to the Melbourne HSA office and flown to that office for their medical assessment. HSA also undertook some assessments in Cairns, Queensland and on the ADF Naval base HMAS Stirling in Western Australia.
2. Designated and trained administrative officers at each HSA office would then contact the study participant in order to arrange a suitable appointment time for the medical assessment.
3. The subject could be reallocated to a different HSA office if requested.
4. Once an appointment had been made, the administrative officer would mail an Appointment Confirmation package to the subject (Appendix 3). This included:
- A letter with the appropriate appointment time and location;
- An Instructions to Participants document requesting that the postal questionnaire be completed and brought to the medical assessment, and outlining the medical assessment requirements relating to the medical assessment, eg refraining from smoking one hour prior to the appointment;
- Documentation required for claiming compensation for travel costs from DVA;
- The study postal questionnaire.
5. Participants were then recontacted by phone, 24 - 48 hours prior to their appointment, to remind them of their appointment details and to answer any final questions about the requirements for the medical assessment or postal questionnaire.
6. Upon arrival at the HSA office for their medical assessment appointment, participants were seen first by a nurse, secondly by a psychologist and thirdly by a medical officer, as described in section 5.6.2.
5.6 Data collection instrumentation and measures
Full participation in the study involved two stages of data collection, with participants undertaking:
- completion of a postal self-administered questionnaire, and
- a medical assessment conducted by a trained HSA medical team, each comprising a nurse, clinical psychologist and medical doctor.
Where possible previously validated data collection instruments were used in the study, and the availability of normative comparisons was considered in the selection of instruments. Some instruments and the design of new questions were based on questionnaires used in overseas Gulf War studies, to allow comparisons where appropriate. For example the symptom questionnaire was based on the instrument used by the King’s College Gulf War Illness Research Unit in London (with their permission and acknowledgment).
There was some intentional overlap in the assessment of important health outcomes. For example, symptoms of posttraumatic stress disorder and alcohol use were assessed in both the postal questionnaire and by psychological interview. Overlap of health assessments in this way enhanced the quality of data collection for health outcomes of particular interest and facilitated checks of data validity and reliability.
Equipment was standardised across HSA offices and calibrated as required. Standardised procedures were developed for the performance of the medical assessments (section 5.6.8) and all HSA staff were trained in these procedures (section 5.6.9).
5.6.1 Postal self-administered questionnaire
All participants were asked to complete a self-administered questionnaire (Appendix 4) which was sent to them in the post. The postal questionnaire was used to assess self-reported health outcomes as well as self-reported exposures and experiences. The postal questionnaire sections, and accompanying question numbers and section labels or headings, included:
- A definition of a Gulf War veteran for the purposes of the study and a self-report of the participant’s perception of their Gulf War status (Question 1).
- Demographic information (A. Personal Details. A1-12).
- Assessment of self-reported exposures through:
- History of military service postings (B. Military Postings. B1-4),
- Gulf War deployment history (C. Gulf War Deployment. C1-13),
- Active (war or peacekeeping) deployment history (D. Deployments. D1a-9),
- Stressful military service experiences (E. Military Service Exposures. E1),
- Civilian occupational history (F. Civilian Occupational History. F1-4).
- Assessment of health outcomes (G. Health. G1-47) through:
- The Short-Form-12 Health Survey (SF-12) (G1-7),
- The 12 item General Health Questionnaire (GHQ-12) (G8-19),
- A 63-item symptom questionnaire (G20. q1-63),
- A 17-item neurological symptom questionnaire (G20. q64-80),
- A 61-item questionnaire relating to diagnosed or treated medical conditions (G21-22),
- General medical history (G23-27),
- Smoking and tobacco consumption (G28-30),
- The AUDIT questionnaire for alcohol intake and its effects (G31-40),
- The Posttraumatic Stress Disorder Checklist (PCL-S) (G41a-41b),
- Self-report of reproductive outcomes and any children with congenital malformations or cancer (G42-46),
- Self-report of infertility (G47)
- Open questions allowing the reporting of additional and important military experiences, exposures or health concerns (H. Final Questions. H1-3).
- Confirmation of current contact details (I. Contact Details).
- Opportunity to nominate a medical practitioner to receive a copy of their medical report (J. Nomination of a Medical Practitioner).
The questionnaire took up to two hours to complete and was undertaken by participants in their own time at home prior to attending for the medical assessment.
5.6.1.1 Demographic information
Participants were asked to provide demographic information in relation to gender, date and country of birth, Aboriginal or Torres Strait Islander origin, language spoken at home, current marital status, any change in marital status since August 1991, highest education qualification completed, current occupational status, any period of unemployment greater than 3 months since August 1991 and current main source of income (A. Personal Details).
5.6.1.2 Exposure assessment
Exposure assessment included Gulf War specific exposures, as well as potential confounding exposures experienced during non-Gulf War military life and civilian occupational life that may also have influenced health outcomes.
Gulf War veterans were asked about medications and immunisations that were administered during the Gulf War deployment.
Participants were asked about chemical and environmental (C & E) exposures, which may have occurred during:
- the Gulf War deployment only, eg SMOIL;
- non-Gulf War active deployments, eg exposure to heat or cold;
- military activities when not on deployment, eg exposure to exhaust fumes; or
- civilian occupational activities, eg exposure to pesticides.
Recall of occupational exposure to specific substances can be unreliable,[252] therefore subjects were asked for details of their military and civilian job histories, such as their job titles.
Participants were asked about psychological stressors specific to military activity (eg coming under enemy attack), that may have occurred during:
- the Gulf War deployment only;
- non-Gulf War military activities, including active deployments other than the Gulf War and other military activities, including training exercises, when not on deployment.
Table 5.1 lists the chemical, environmental, medical and psychological exposures that were investigated by self-report, the corresponding question number of the postal questionnaire and the sources of the questions.
|
Exposure
|
Question number
|
Source
|
|---|---|---|
| Chemical Warfare | D3, D4, C6 | Monash, [22] |
| Respirator Use | D4 | Monash, [22] |
| NBC suit use | D5 | Monash, [22] |
| Pesticide exposed | D2 q7, 8, 19-23. B. F | Monash, [16, 22] |
| Pesticide user | D2 q7, 8 | Monash, [16, 22] |
| Insectrepellent | D6 | Monash, [16, 28] |
| Depleted Uranium | D2 q1, 2. C5, C6, E1 q2 | Monash, [22, 28] |
| SMOIL/D | D2 q10. C4, F | [16, 22, 28] |
| Dust | D2 q10. | Monash |
| Infectious agents | D2 q4-7, 17, 18. E1 q6-8. B. | Monash, [16, 22] |
| Possibly contaminated food | D2. q4, 5. E1 q6 | [22] |
| Possibly contaminated water | D2. q6-9. E1 q6, 7 | Monash, [16, 22] |
| Insects | D2 q17, 18 | [16] |
| Exhaust | D2 q15, B. F | Monash, [16, 22] |
| Fuel | D2 q7, 9, 12. E1 q6, 7. B. F. | Monash |
| Solvents | D2 q13. B. F | Monash, [22] |
| Solvent/skin | D2 q8, 9, 14. E1 q7 | [16, 22] |
| Heat or Cold | E1 q5 | Monash, [16] |
| CARC paint | D2. q3 | [16, 28] |
| Sun screen | D2 q16 | Monash |
| Immunisations | C7-9, 13 | [21, 61, 157] |
| Medications | C10-12, 13 | [21] |
| Post deployment experiences | D9 | Study focus group of ADF veterans and personnel, Consultative Forum, Monash, DVA |
| Psychological stressors | E1 q1-44 | Monash, study focus group of ADF veterans and
personnel, [21, 22, 253-255] |
Exposure to passive smoking, ionising and non-ionising radiation were not addressed in this study.
Some of the exposure questions were modified to make them more relevant to Australian veterans and to maintain consistency with the style of the rest of the questionnaire. Alterations to the questions were also made on the basis of consultation with a focus group of veterans of a number of deployments held in Melbourne in June 2000, with the Scientific Advisory Committee and Consultative Forum for the study.
back to top5.6.1.2.1 History of military service postings
Participants were asked to specify their service type, the year they joined the ADF, year of departure (unless still serving) and rank achieved on Jan 1991. They were also asked to list any non-reserve postings held for 3 months or more and to report whether they regularly worked with or handled pesticides, solvents, fuels or engine exhaust during such postings.
The military service postings history was used to:
- Subgroup by type of service
- Subgroup by rank
- Identify specific chemical and environmental exposures so that, if necessary, the exposure could be adjusted for in any investigation of a possible association between a health outcome and similar exposures that occurred in the Gulf War.
5.6.1.2.2 Gulf War deployment history and specific exposures
This section of the questionnaire assessed exposures specific to the Gulf War deployment including services with specific Defence Force groups, service in specific locations, exposure to smoke, oil and dust from burning oil wells and exposure to medications and immunisations.
General Gulf War deployment details that were requested included deployment dates, the identity of ship, unit or squadron with which the veteran deployed, the rank held at the time of deployment, primary duties, job title or trade and any service with Operation Habitat.
The general deployment details were used to subgroup members of the Gulf War veteran study group by whether their deployment was completed before the commencement of the air war on 17th January 1991.
Proximity to the ground war and to significant events such as the Camp Doha fire and the demolition of weapons at Khamisiyah was assessed. Subjects were asked to identify the position and timing of any land based activities using maps which were based on those developed for a UK study.[28]
Exposure to smoke and oil and dust (SMOIL) was assessed through self-report of number of days of exposure and number of hours, outside or on upper ship decks, on each of these days.
Exposure to medications and immunisations (referred to as vaccinations in the postal questionnaire) specific to the Gulf War deployment were assessed using questions based on those used by Kings College and Manchester University research groups.[21, 157]
Participants were asked:
- whether they had their World Health Organisation (WHO) yellow vaccination booklet to refer to (C7),
- the number of immunisations they received (C8),
- the timing of these in relation to their deployment and transit to the Gulf (C8),
- the time period over which they received them (C8),
- the individual immunisations that they think they received (C9) (including routine and biological warfare vaccines),
- self-reported significant reactions to immunisations or medications that they received, and an indication of the severity of any such reactions (C13),
- whether they took:
- anti-nerve agent pills (C10)
- malaria tablets (C11)
- anti-biological warfare tablets (C12), and if so, for how long and in what quantities (C10-12).
The difficulty in recalling immunisations received ten years ago was acknowledged, and a number of measures were used to increase or assess the accuracy of recall. These were:
- participants were asked to refer to their WHO yellow vaccination booklets, if available, when completing this section of the postal questionnaire and to bring these with them to their medical assessment so they could be photocopied,
- efforts were made to access policy documents relating to the immunisation schedules for each ship to try and validate participants’ immunisation experience where uncertainty existed
Efforts were made to verify self-reported information on the use, and duration of use, of pyridostigmine bromide, antimalarials and anti-biological warfare medications against Australian Defence Force policy documents. However there was likely to be considerable variation in individual compliance with self-administered medication such as pyridostigmine bromide and the study analysis relied primarily on the self-reported exposure data.
A deployment immunisation and medication history was not sought for veterans other than those who were deployed to the Gulf War. A comparison of the background exposure to immunisations and medications for the Gulf War veterans and comparison group was considered in a manner similar to other exposures; that is, through a comparison of other active deployments that were reported and through their duration of military service.
5.6.1.2.3 Active (war or peacekeeping) deployment history and deployment exposures
All participants were asked to indicate whether they had taken part in any active deployments and to identify specific chemical or environmental exposures during those deployments. Gulf War veterans were additionally asked to describe the exposures that took place during the Gulf War.
The participants were provided with a list of active deployments, supplied by the ADF (D1b). The list included war and non-warlike operations, UN peacekeeping and peacemaking operations, service protected evacuations, humanitarian aid operations, humanitarian assistance and was restricted to operations with more than 10 persons that were not security sensitive. Participants were asked to nominate the active deployments they had been on, the year first deployed, the duration of the deployment, and whether they were ordered to serve or volunteered (D1a, b q1-26). Space was provided for participants to nominate additional deployments. Goodwill deployments and training exercises were excluded.
Participants were then asked about chemical and environmental exposures (D2-7) and post deployment experiences (D9) in relation to these active deployments. Participants were asked to indicate whether they were exposed (yes/no) and the extent of exposure (rarely/sometimes/never) to a variety of chemical and environmental exposures. The chemical and environmental exposures of interest are summarised in Table 5.1.
The participants were asked to assess the extent of some exposures during the Gulf War and also during their entire military history (Section E) eg experiences of heat and cold. The participants were asked to assess other exposures during the Gulf War and during other active deployments (Section D) eg to locally sourced/non-military food. In addition, participants were asked to identify the pesticides that they used or were exposed to, the chemical warfare agents that they believed that they had been exposed to, the frequency that they used chemical protective clothing and respirators, and the reasons why they used this protective equipment.
A 17-item Post Deployment Experiences questionnaire (D9. q1-17) was developed to assess the experiences of veterans upon return from active deployments. The focus group, which was run for the purpose of constructing the Military Service Experience questionnaire (section 5.6.1.2.4) provided input about a variety of post deployment experiences which potentially contributed to a veteran’s overall experience of a deployment. Themes explored in the Post Deployment Experiences questionnaire included whether veterans felt recognised and acknowledged by the ADF, Australian Government and Australian people for their deployment activities, whether they felt adequately debriefed, proud of their activities, improved as a leader, more knowledgeable about world issues and more respectful of other veterans.
5.6.1.2.4 Military Service Experience questionnaire
Psychological stressors were assessed using the Military Service Experience (MSE) questionnaire (E1. q1-44). The MSE questionnaire was a 44-item questionnaire designed by the Monash study team for this study, to measure the occurrence and severity of stressful military service experiences relevant to ADF personnel. Existing research instruments designed to measure military or combat-related stressors, such as the Laufer Combat Scale[253] and the Combat Exposure Scale (CES),[254] were considered not fully relevant to the Australian study population. These instruments focus largely on combat exposure in Army infantry who have experienced direct military attack. The Australian Gulf War veteran population, in contrast, were primarily Navy personnel whose experiences involved threat of military attack but few actual encounters. Similarly, many of the recent active deployments and peacekeeping operations, undertaken by Australian military forces, have involved few direct combat encounters and few casualties. Nonetheless, these deployments often involved high risk of military attack including threat of nuclear, biological and chemical warfare, extreme environmental conditions and, often, social isolation and poor organisational conditions. Therefore the MSE questionnaire was designed to explore the severity of stress associated specifically with Australian military experiences. Both war-like deployments and training or other routine military activities were included in the circumstances under which these experiences may have occurred, as it has been during training exercises and routine deployments that the Australian military have encountered several of their most severe accidents in recent times. These include the Black Hawk helicopter accident in June 1996, in which 18 ADF personnel died and a further 12 were injured,[256] and the accidental fire on HMAS Westralia in May 1998 which resulted in a further four deaths.[257]
In developing the MSE questionnaire, existing questionnaires measuring combat severity were searched for items deemed relevant to the Australian military experience. Instruments included in the search were the Laufer Combat Scale, modified for studies of United States military personnel involved in Operation Desert Storm,[253] the CES[254] and the War Zone Stress Exposure Scale (WZ-SES).[255] Also included in the search were the exposure questionnaire items used by the Kings College research team in their Gulf War veteran research.[21] Many of these items were subsequently used in the Health Study of Canadian Forces Personnel Involved in the 1991 Conflict in the Gulf.[22] In addition to items drawn from previously used instruments, several new items were constructed, based on information about known Australian military experiences including accidents and anecdotal evidence provided by veterans of the Gulf War and other deployments.
A total pool of 76 possible items was collated covering experiences such as military attack, threat of chemical warfare, lack of preparation, fear of death, responsibility for the lives of others, physical discomfort, risk of disease, accidents causing death or injury, social isolation and poor unit cohesion.
The 76 items were subsequently circulated to a focus group of ten male and two female Australian former, or current, ADF personnel. This group included veterans of the Gulf War, veterans of other deployments and some personnel who had not been deployed. The focus group participants were invited to:
1. List any additional key experiences which were missing from the list and which they considered to be important stressors relevant to Australian military activity.
2. Strike out any listed experiences which they considered NOT important or relevant to Australian military activity.
3. Indicate, in order of severity, the ten listed items which they considered the most important stressors relevant to Australian military activity.
4. Indicate the ten listed items which they considered the least important stressors relevant to Australian military activity.
Focus group participants were then invited to meet as a group with Monash study team members and representatives of DVA and the Australian Centre for Posttraumatic Mental Health. Discussions centred on stressful military activities that were most relevant to members of the group. Some new issues, such as experiences of sexual harassment, were tabled during the focus group discussions and some events considered irrelevant or trivial, such as ‘falling overboard’, were removed from the lists.
Subsequent to the focus group discussions the Monash study team pared the original 76-item pool to a final 44 items, which comprised the MSE questionnaire and which were included in the postal questionnaire.
back to top5.6.1.2.5 Civilian occupational history
Participants were asked to state whether they had ever been members of the Country Fire Authority and to list civilian jobs held for more than 6 months including duration, job title, industry sector and company/employer for each job. They were also asked for each job to assess their exposure to pesticides, fuels, engine exhaust, solvents, infectious diseases and trauma. If they used pesticides in any of these jobs, they were asked to list them and describe what they used them for. We asked about exposure to specific chemical and environmental hazards so that if necessary, the exposure could be adjusted for in any investigation of a possible association between a health outcome and similar exposures that occurred in the Gulf War.
5.6.1.3 Health outcomes assessment
Health outcomes assessment included assessment of participant’s physical and psychological health through data collected in both the postal questionnaire and the HSA medical assessment. The following sections describe the health outcome measures that were included in the postal questionnaire.
5.6.1.3.1 Short-Form-12 Health Survey (SF-12)
The Short-Form-12 Health Survey (SF-12) (G1-7) is a self-administered generic measure of health status.[258] It was developed to be a brief, yet valid, alternative to the longer SF-36[259] for use in large surveys of general and specific populations as well as large longitudinal studies of health outcomes. It has become one of the most widely used instruments for monitoring the health of populations,[258] and was included in the Australian National Survey of Mental Health and Wellbeing of Adults.[260]
The instrument’s twelve questions explore eight concepts commonly represented in health surveys, namely physical functioning, role limitations due to physical problems, bodily pain, general health, energy and fatigue, social functioning, role limitations due to emotional problems, mental health and change in health. Responses are differentially weighted and combined to produce a ‘Physical Component Summary’ (PCS-12) score between 0 and 100 as an indicator of physical health, and a Mental Component Summary’ (MCS-12) score between 0 and 100 as an indicator of mental health
Both the PCS-12 and the MCS-12 use the same 12 items but different weights. These weights were constructed by the developers of the SF-12 to produce PCS-12 and MCS-12 scores that would have a mean of 50 and a standard deviation of 10, if applied to the US general population.[258] US population norms are used to allow results to be compared with other studies. The higher the score, the better the physical or mental health status. The items in the scale refer to the four weeks prior to the completion of the questionnaire. Responses to all 12 items are required for summary scores to be derived.
Whilst not quite displaying validity as high as the SF-36, the SF-12 is believed to represent a very satisfactory trade-off between reduced questionnaire length and reduced precision for large study groups.[258] Test-retest reliability for the PCS-12 scale has been found to be 0.89 in a US general population survey[261] and 0.86 in a UK general population survey.[262] Reliability coefficients of 0.76 and 0.77 were found for the MCS-12 scale in the US and UK studies respectively. Observed PCS and MCS scores, using the SF-12, can be reliably compared with those observed in studies using the SF-36, with both instruments shown to reach similar statistical conclusions about group differences.[263]
5.6.1.3.2 Twelve item General Health Questionnaire (GHQ-12)
The GHQ-12 (G8-19) is a twelve item, abbreviated version of the 60-item General Health Questionnaire.[218] It is a brief, self-administered screening instrument designed to detect current, diagnosable psychiatric disorders in general population surveys, primary medical care settings or among general medical outpatients. The instrument covers four identifiable elements of distress; those being depression, anxiety, social impairment and hypochondriasis. The instrument is not intended to distinguish among psychiatric disorders or to be used in making a diagnosis of an actual disorder. Emphasis is on changes in condition, not on absolute level of a problem, so items compare the present state to the person’s normal situation.
Studies have found that the scale shows consistently high reliability and validity measures.[218, 264] Split half reliability on the 12-item version has been reported at 0.83[218] with Cronbach’s coefficient alphas[265, 266] ranging from 0.82 to 0.90.[267] Sensitivity and specificity ranges are reported between 74.2% and 95.0%.[218]
The GHQ-12 was used by the King’s College Gulf War Illness Research Unit in their study of 4248 Gulf War veterans.[21] This instrument was also used in the Australian National Survey of Mental Health and Wellbeing of Adults.[260]
There are three commonly employed methods of scoring the GHQ-12.[268, 269] The standard method[218] involves coding the four possible responses to each question as 0 – 0 – 1 – 1 and then summing the binary scores, giving a total score ranging from a minimum of 0 to a maximum of 12. Another method of scoring, generally known as the Likert method, makes use of a four point scale, coded 0 – 1 – 2 – 3. When these values are summed a total score ranging from 0 to 36 is obtained.[218] Another scoring method, known as the Chronic method,[270] proceeds in the same manner as the standard method except that responses to negative items such as ‘Have you recently lost much sleep over worry?’ are coded 0 – 1 – 1 – 1. The latter approach assumes that the response ‘no more than usual’, which would otherwise be coded as 0 (good health) using the standard method, is indicative of the presence of a chronic problem, and so would be coded as 1. For all three scoring methods, higher scores indicate an increased likelihood of psychological ill-health. A variety of thresholds or cut-off scores for determining caseness, or possible psychiatric condition, have also been employed with the GHQ-12.[264, 268, 271]
The scoring method for the GHQ-12 in this study, and determination of the most appropriate GHQ-12 caseness score, is described in chapter 11.
5.6.1.3.3 Symptom questionnaire
The 63-item self-report symptom questionnaire included respiratory, cardiovascular, musculoskeletal, dermatological, gastrointestinal, genitourinary, neurological, neuropsychological or cognitive, and psychological symptoms (G20. q1-63). It was based on the symptom questionnaire developed and used by the King’s College Gulf War Illness Research Unit,[21] which was based on the Hopkins Symptom Checklist,[272] and employed the same severity scale for symptoms reported to have occurred in the last month. It also included some symptoms used in other overseas symptom prevalence surveys. Similar symptom questionnaires and symptoms have been used in a number of overseas postal surveys investigating the health of their country’s Gulf War veterans.[16, 20-22, 73, 157, 158, 162] Such symptom questionnaires have also formed the basis for subsequent factor analyses by these research groups.[73, 154, 157, 158, 160] The neuropsychological symptoms in the symptom questionnaire were similar to those used to evaluate and compare neuropsychological function between the study groups in previous Gulf War studies.[16, 21, 28, 31, 73, 154, 157, 158, 160, 163]
In addition to enabling internal comparisons of self-reported symptoms within the study groups, the use of the symptom questionnaire allows comparisons to be made with the results of overseas studies. The symptom questionnaire also formed the basis for the factor analyses of symptoms in chapter 18. Participants were asked about the occurrence of symptoms in the past month, and, if symptoms were experienced, the severity of those symptoms according to whether they were ‘mild’, ‘moderate’ or ‘severe’.
5.6.1.3.4 Neuropathic symptom questionnaire
The 17-item neuropathic symptom questionnaire asked about neuropathic symptoms indicative of peripheral neuropathy experienced in the past month (G20. q64-80). The instrument was developed in consultation with a neurologist specifically for this study, as a suitable pre-existing instrument could not be identified. The questions were based on those included in other studies of neurological function[273, 274] and related to four parameters of peripheral neurological dysfunction:
- muscle weakness
- sensory disturbance
- autonomic function, and
- severity of neurological dysfunction.
5.6.1.3.5 Doctor diagnosed or treated medical conditions
This 61-item medical conditions questionnaire asked about medical problems or conditions that had been diagnosed or treated by a doctor (G21.q1-61). Subjects were also given the opportunity to list additional medical conditions that had been diagnosed or treated by a doctor (G22). The term ‘medical doctor’ was used to qualify the person who diagnosed or treated the problem or condition in order to standardise the reference point and context for that diagnosis or treatment. This questionnaire was based on the medical conditions questionnaire developed and used by the King’s College Gulf War Illness Research Unit[21] as well as including several conditions considered relevant to Australian veterans. As was the case with the symptom questionnaire (section 5.6.1.3.3), similar medical conditions questionnaires have been used in a number of overseas postal surveys investigating the health of Gulf War veterans.[16, 20-22, 73, 157, 158, 162] In addition to enabling internal comparisons of self-reported medical conditions within the study groups, the use of the medical conditions questionnaire allows comparisons to be made with the results of overseas studies.
Subjects who answered ‘Yes’ to a medical condition were asked to identify the year in which the condition was first diagnosed, and whether the medical condition had been treated by a doctor in the past year. These two sub-questions served to locate the diagnosis temporally in relation to the Gulf War and to give an indication of the current status of that condition.
5.6.1.3.6 General medical history
Hospitalisation
The questions relating to hospitalisation (G. 23) were based on those used in a study of US Gulf War veterans.[20] Participants were asked whether they had been hospitalised overnight or longer because of illness or injury during the past 12 months, and if so, what was the duration of, and reason for, the hospitalisation.
Functional impairment due to illness or injury
The questions relating to functional impairment due to illness or injury (G. 24) were also based on those used in a study of US Gulf War veterans.[20] Participants were asked whether they had stayed in bed or at home all or part of any day because they did not feel well or as a result of illness or injury in the past two weeks.
Current use of medication
Participants were asked about current use of medicine including tablets, creams, inhalers or other drugs (G. 26). Medications were coded using an electronic version of the Monthly Index of Medical Specialties (MIMS).
Family history
Participants were asked whether they had a family history of asthma, stroke when less than 65 years of age, a heart attack when less than 65 years of age, diabetes and cancer. Responses were coded according to the ICD-9 classification system.[275]
5.6.1.3.7 Cigarette smoking and tobacco use
Exposure to cigarette and tobacco smoke was assessed with a brief set of questions (G28-G30) designed to determine:
1. Status of current, former or never/occasional smoker
2. Age started (for current and former smokers)
3. Age stopped (for former smokers)
4. Pack years of cigarette consumption based on an estimated average number of cigarettes smoked per day, grams of tobacco smoked per day (not including from cigarettes or cigars) and cigars smoked per week, and duration of smoking in years between the age started and the age stopped as given by the participant.
5.6.1.3.8 Alcohol consumption
The Alcohol Use Disorders Identification Test (AUDIT) Core questionnaire was used to quantify current alcohol use and detect alcohol disorders (G31-40). This scale was developed by WHO-affiliated investigators for the identification of currently active, hazardous and harmful alcohol consumption.[276] The 10-item AUDIT core questionnaire measures alcohol consumption, dependence symptoms and personal and social harm reflective of drinking. Items refer to the previous year and are scored according to their frequency of occurrence rather than their presence or absence. Studies have shown the internal reliability of the AUDIT to be 0.86.[277]
In addition to the inclusion of the AUDIT in the study postal questionnaire, alcohol disorders were also investigated as part of the psychologist’s assessment (see section 5.6.2.2)
back to top5.6.1.3.9 Posttraumatic Stress Disorder Checklist – S (PCL-S)
The Posttraumatic Stress Disorder Checklist (PCL) is an easily administered self-report rating scale for assessing the 17 DSM-IV symptoms of PTSD. First developed in 1993[278] the PCL was normed in part on a Gulf War veteran sample. Diagnostic utility of the PCL was determined by using the PCL scores to predict PTSD diagnosis derived from the Structured Clinical Interview for DSM-III-R (SCID). It has excellent test-retest reliability, over a two to three day period, and internal consistency is very high for each of three groups of items corresponding to the DSM-IV symptom clusters as well as for the full 17-item scale.[278, 279] The PCL correlates strongly with other measures of PTSD such as the Mississippi Scale, the PK scale of the MMP1-2 and the Impact of Events Scale, and correlates moderately with the Combat Exposure Scale.[278-280] Data are available on Australian Vietnam veterans,[281] US peacekeeping forces,[282, 283] US Gulf War veterans[176, 177] and from many civilian studies.[279, 284, 285]
The PCL-S used in this study (G41a-G41b, q1-17) is one of three versions of the PCL available. The PCL-M is a military version with questions that refer to “a stressful military experience”. The PCL-C is a general civilian version that is not linked to a specific event. Its questions refer to “a stressful experience from the past”. The PCL-S is a non-military version that can be referenced to any specific traumatic event. The PCL-S allows the respondent to nominate the criterion event and subsequent questions refer to “the stressful experience”.
In this study a slight modification was made to the PCL-S stem question which originally read “Please consider the event that you found most stressful or upsetting…” The question was changed to read “Please consider the event or group of events, military or non-military, in your life that you found most stressful or upsetting…” The modification was implemented on the basis that:
a) in some cases, a stressful event with the potential to precipitate PTSD could be a group of related events, rather than a single discrete event. For example, there could be several incidents encountered during a combat deployment which combine to precipitate a PTSD reaction.
b) as the primary focus of the study is military related, participants may feel that they are obliged to nominate a stressful military event, or group of events, and fail to consider a stressful civilian event.
5.6.1.3.10 Reproductive health outcomes
The following aspects of reproductive health outcomes were investigated:
- Self-report of fertility difficulties (G47).
A short series of questions relating to self-report of infertility, based on those used in a recent fertility study[286] was included in this study (G47).
- Self-report of pregnancies resulting in miscarriages, stillbirths or a termination of pregnancy (G42-43).
Participants were asked how many times they had ever been pregnant or fathered a pregnancy, and if any of these pregnancies resulted in a miscarriage, a still birth or a termination of pregnancy.
- Health outcomes in live born children (G45a, b).
For live born children, questionnaire respondents were asked to provide information such as date of birth, birth weight and term of gestation and to identify any birth defect or chromosomal abnormality, cancer or other serious health problem. If a child had died respondents were asked to provide the cause and date of death.
- Validation of reports of children with cancer
The study consent form included consent for reports of children’s cancers to be matched against the records of the national cancer registry. To facilitate this matching process, questionnaire respondents were asked to provide identifying information such as their children’s full names and dates of birth.
5.6.1.4 Nomination of a medical practitioner to receive a copy of their medical report
The participants were invited to nominate a medical practitioner for the purposes of sending out a copy of their medical report as prepared by the HSA doctor. Signed consent was required for the release of this information. Participants who did not wish to nominate a medical practitioner were not required to do so.
back to top5.6.2 Medical assessment
All participants were invited to attend for a medical assessment at one of the HSA medical centres. Assessments were conducted at one of HSA’s medical centres or by a mobile team dispatched to a more remote area or to a naval base where numbers justified this. Participants were assessed by a medical team comprising a nurse, a clinical psychologist and a medical doctor, all specifically trained for the purposes of the study. Investigations were undertaken using standardised equipment and standardised procedures, in accordance with the protocols and procedures developed by the Monash University study team. Equipment was appropriately calibrated as necessary. Each medical assessment took up to five hours, including breaks for refreshments. The data was recorded at the time of the assessment in a standard Data Collection Booklet that was developed for the study (Appendix 5). A number of measures were instituted to ensure that assessors remained blinded to the Gulf War status of the participant wherever possible. These included instructions to the assessors to refrain from discussing the participant’s military service histories. In addition, the sequence of the medical assessment components was carefully tailored to minimise the possibility of any disclosure by the participants of their Gulf War status.
In the following sections describing the medical assessment components, the page numbers in parentheses refers to the page number of the Medical Examination Data Collection Booklet where that component of the assessment was recorded.
5.6.2.1 Nurse-administered data collection
A Registered Nurse conducted the first part of the assessment process. This involved:
- Greeting the study participants and checking their identity; introducing themselves, explaining what would happen in the assessment process and answering questions,
- Obtaining and witnessing informed consent (detachable page),
- Obtaining emergency contact details (reverse of detachable page),
- Standardised measurements of height, weight, hip and waist circumference (p1),
- Standardised measurement of diastolic and systolic blood pressure according to the standard protocol recommended by the British Hypertension Society (p1),
- Standardised assessment of corrected visual acuity using the Snellen Eye Chart (p1),
- Assessment of symptoms of tiredness or fatigue for the purpose of selecting cases and controls for the collection of blood for lymphocyte subsets (p2),
- Collection of blood samples (p2),
- Completion of the interviewer-administered Respiratory Health Questionnaire with the participant (p3-7),
- Performance of skin prick tests to measure atopy (p8),
- Standardised measurement of respiratory function using the Spirocard spirometer according to the American Thoracic Society (ATS) protocol (p8,10),
- Urinalysis for blood, protein, glucose and nitrites (p9),
- Photocopying the WHO yellow vaccination booklets provided by Gulf War veterans (p9,12), and
- Checking returned postal questionnaires for completeness (done last in their examination to help maintain blindness to the participant’s Gulf War status),
- Documenting whether they remained blinded to Gulf War status (p9).
The nurses were also trained to take on a case management and quality assurance function, to ensure each participant was managed efficiently through the medical assessment process. The quality assurance function included checking all documentation to ensure it was complete and consistent before transporting it to the Monash University study team.
5.6.2.1.1 Informed Consent
Participants were asked to sign an Informed Consent Statement giving their consent to participate in the study. Participants could participate in all aspects of the study or could indicate on the Informed Consent Statement those aspects of the study in which they did not consent to participate. The purpose of the study and their role in the study was explained in the Explanatory Statement that accompanied the initial invitation to participate. The Explanatory Statement explained the nature and content of the research study, the voluntary nature of participation, what participation in the study involved, possible risks and inconveniences, issues related to the storage of blood, confidentiality and privacy, use of data and data management and participation in possible future investigations. Contact details for the study team, Scientific Advisory Committee and overseeing Ethics Committees were provided so that participants could ask questions or express complaints or concerns about the conduct of the study. Trained HSA nurses obtained and witnessed the informed consent at the commencement of the medical assessment. Consent was also sought, through an Informed Consent Statement sent to them for their signature and return, from participants who completed the postal questionnaire only.
5.6.2.1.2 Standardised measurement of height, weight, hip and waist circumference
- Height was measured in centimetres, to one decimal place using a stadiometer, as the maximum distance from the floor to the vertex of the head with shoes removed (p1).
- Weight was measured in kilograms, to one decimal place, in light clothing and without shoes, using the Tanita Body Composition Analyzers TBF-410 that were dedicated to the study (p1).
- Waist and hip circumferences (p1) were measured using a tape measure according to guidelines described by the National Health and Medical Research Council (NHMRC).[287] Waist circumference was measured in centimetres, to one decimal place, at the smallest circumference below the rib cage and above the umbilicus taken at the end of normal expiration. Hip circumference was measured, in centimetres to one decimal place, at the largest circumference at the posterior section of the buttocks.
Body mass index
Body mass index (BMI) is a simple index of weight-to-height ratio that is commonly used to classify underweight, overweight and obesity in adults. It is defined as the weight in kilograms divided by the square of the height in metres (kg/m2). It is considered to provide the most useful, although crude, population-level measure of obesity and the risks associated with it. The most recent WHO categories[288] differ slightly to those traditionally recommended for use in the study of Australian populations by the NHMRC.[287] The 1995 National Nutrition Survey adopted the World Health Organization’s recommendations for BMI categories, but included split categories of the normal or acceptable weight range to enable comparison with the NHMRC categories.[289] BMI for study participants was calculated using their height and weight measurements. Participants were then classified according to the categories used by the National Nutrition Survey as listed in Table 5.2.
|
Classification
|
BMI kg/m2
|
Risk of comorbidities
|
|---|---|---|
| Underweight | <18.50 | Low (but risk of other clinical problems increased) |
| Normal range | 18.50-<20.00 20-<25.00 |
Average |
| Overweight: | >=25.00 |
|
| Pre-obese | 25.00-<30.00 | Increased |
| Obese class 1 | 30.00-<35.00 | Moderate |
| Obese class 2 | 35.00-<40.00 | Severe |
| Obese class 3 | >=40.00 | Very severe |
Waist circumference and waist-to-hip ratio
Waist circumference and waist-to-hip ratio (WHR) are indicators of the amount of fat located predominantly in the abdominal region. WHR was calculated by dividing each participant’s waist measurement by their hip measurement. Resulting ratio scores were rounded to one decimal point. Cut-off points that may define increased risk of cardiovascular disease and all-cause mortality range from >1.0 in men and >0.85 in women[288] to >0.9 in men and >0.8 in women.[287, 289-291] This study has used the cut-points of >0.9 in men aged 19 years and over and 0.8 for women aged 19 years and over as those generally adopted in Australia[287] to define those at health risk from having a central fat distribution.
Waist circumferences greater than 94 cm in males and greater than 80 cm in females indicate increased risk of obesity related complications, and waist circumferences greater than 102 cm in males and greater than 88 cm in females indication substantially increased risk of obesity related complications.[288, 292]
back to top5.6.2.1.3 Blood pressure
Blood pressure was measured twice using a mercury sphygmomanometer (p1) according to the British Hypertension Society protocol (http://w3.abdn.ac.uk/BHS/booklet/proced.htm).[293] Two readings were separated by at least two minutes and averaged to derive a single systolic and diastolic blood pressure reading. Participants were then categorised according to the guidelines provided by the Joint National Committee on Prevention, Detection, Evaluation and Treatment of High Blood Pressure[294] and the Guidelines Subcommittee of the World Health Organization-International Society of Hypertension,[295] as presented in Table 5.3.
Where a participant’s systolic and diastolic blood pressure fell into two different categories, the higher category was applied to classify the individual’s blood pressure status.[294, 295]
|
Category
|
Systolic (mmHg) | Diastolic (mmHg) | |
|---|---|---|---|
| Normal |
<130
|
and
|
<85
|
| Optimal |
<120
|
and
|
<80
|
| High-normal |
130-139
|
or
|
85-89
|
| Hypertension |
>=140
|
or
|
>=90
|
| Grade 1 hypertension (mild) |
140-159
|
or
|
90-99
|
| Grade 2 hypertension (moderate) |
160-179
|
or
|
100-109
|
| Grade 3 hypertension (severe) |
>=180
|
or
|
>=110
|
| Isolated systolic hypertension |
>=140
|
and
|
<90
|
It should be noted that this classification system[294, 295] is based on the average of two or more blood pressure readings at each of 2 or more visits in adults who are not taking any medication and who have no acute illness. These conditions have not been met in this study where it was impractical to measure blood pressure on two different visits, so the classification system has been used as a framework for comparing the study groups rather than for diagnosing hypertension.
5.6.2.1.4 Visual acuity
Corrected visual acuity was assessed for both eyes using the Snellen Chart at a distance of 6 metres. Visual acuity was recorded (p1) as a fraction, with the test distance recorded as the numerator and the last line of letters in which all letters were read correctly was recorded as the denominator, in a standardised manner.[296]
Subjects who were unable to correctly read the 6/6 line using their refractive correction, eg glasses or contact lenses if worn were categorised as having visual impairment. Visual impairment despite correction with glasses or contact lenses suggests that the defect may not be fully refracted by the glasses or contact lenses or due to an explanation other than a refractive error.[297] Visual impairment was considered as part of the neurological assessment of participants.
5.6.2.1.5 Symptoms of tiredness or fatigue
An abbreviated assessment of any symptoms of extreme tiredness or fatigue was undertaken by the nurses for the purpose of selecting ‘cases’ and ‘controls’ for the collection of blood for lymphocyte subsets testing (p2). This abbreviated assessment was based on the first two doctor’s questions relating to these symptoms in Section 2 of the doctor’s examination (p35). The procedure for the full assessment of symptoms of extreme tiredness or fatigue is described in Section 5.6.2.3.2.
5.6.2.1.6 Spirometry
Spirometry was undertaken using QRS SpiroCard spirometers with Office Medic software (QRS Diagnostic, LLC; Plymouth, Minnesota, USA) according to the standardised procedures recommended by the American Thoracic Society (ATS),[298] and Thoracic Society of Australia & New Zealand.[299]
The SpiroCard spirometer was chosen for use in this study on the basis of its compliance to ATS standards,[300] its portability and simplicity of use, and its measures for infection control. This is a flow-sensing spirometer that uses a pre-calibrated, non-sterile disposable pneumotach. On each day of use, spirometers were calibrated using a 3-litre syringe (or a 1-litre syringe where a 3-litre syringe was not available). Predicted ventilatory function values were drawn from criteria published by Knudson[301] or ECCS (European Community Coal and Steel) criteria if the participant’s parameters were outside the Knudson criteria. Correction for Body Temperature and Pressure, Saturated (BTPS) was automated. Race correction factors of 15% for Asian and Black, African participants were applied.
Participants were instructed to avoid, if possible, the use of anti-histamine based medications in the four days prior to their HSA appointment, and to avoid alcohol, caffeine and asthma medications on the day of their appointment. If any of these contraindicated asthma medications were consumed in the prescribed period, this was recorded and the spirometry was still performed. If a participant had smoked a cigarette within the previous hour, spirometry was performed if possible, at a point later in the examination.
Nurses were trained to instruct and supervise participant’s performance of spirometry to ATS criteria[298] as summarised below:
- participants were seated for the testing;
- the use of a nose clip was recommended;
- the forced expiratory manoeuvre was performed with maximum effort immediately following a maximum inspiration;
- participants were required to perform a minimum of three technically
acceptable blows with acceptability criteria defined as:
- satisfactory start-of-test,
- minimum exhalation Forced Vital Capacity (FVC) time of 6 seconds, and
- end-of-test criteria;
- participants manoeuvres were to meet the ATS criteria for reproducibility
with reproducibility criteria defined as:
- the largest and the second largest FVC must not vary by more than 0.2L, and
- the largest and the second largest Forced Expiratory Volume in 1 second (FEV1) must not vary by more than 0.2L;
- participants were to perform up to, but no more than 8 blows (unless the test was terminated on clinical grounds) until three of these were judged to be technically acceptable and the reproducibility criteria were met.
The HSA doctors were also asked to review the spirometry results in providing feedback to the participant and for quality control purposes in monitoring the performance of spirometry.
In addition to spirometry training for HSA doctors and nurses at the commencement of the study, feedback on the performance of spirometry and additional training was provided in April 2001. Subsequent monitoring and individual feedback was provided over the post-training period.
5.6.2.1.7 Skin prick testing for common allergens
Prior to testing, participants’ use of antihistamines in the previous four days was recorded (p8). Skin prick testing was used to assess participant’s mast cell reaction to four common aero-allergens: cat dander, mixed grasses, house dust mite and mould.
Allergen solutions (Bayer Australia Ltd to March 2001 then Richard Thompson Pty Ltd both of Sydney, New South Wales) used were:
- House dust mite (Standardised Mite DP Dermataphagoides pteronyssinus Strength 30 000 AU/ml)
- Grass mix (Grass mix #7, 1:20 w/v)
- Cat dander (Standardised Cat Pelt Acetone Precipitated Strength 10 000 BAU/ml)
- Mould mix (Alternaria tenuis, Aspergillus fumigatus, Hormodendrum cladosporioides, Penicillium notatum 1:10 w/v)