|
|
|
espa�ol - fran�ais - portugu�s |
Search
|
|
|
|
espa�ol - fran�ais - portugu�s |
|
Search
|
(Continued)
A. PANEL PROCEDURES WITH REGARD TO SCIENTIFIC EXPERTISE 6.1 The Panel recalled that paragraph 2 of Article 11 of the SPS Agreement provided that:
Noting that this dispute involved scientific or technical issues, the Panel consulted with the parties regarding the need for expert advice. The Panel noted how valuable such expert advice had been during its previous examination of this matter, and further that the evidence submitted to it included several new risk analysis reports. The Panel decided to seek scientific and technical advice as foreseen in paragraphs 1 and 2, first sentence, of Article 13 of the DSU, and pursuant to paragraph 2, first sentence, of Article 11 of the SPS Agreement . 6.2 The Panel initially considered seeking advice from two of the four experts which had advised the panel in the original dispute, as well as from a third expert with experience in the area of the application of sanitary measures. The parties were invited to comment on this suggestion by the Panel and in particular to state any compelling objections they might have with regard to any individual, or to suggest other experts. The Panel than selected three individuals taking into account the comments of the parties and the need for expertise in a number of areas. These experts were requested to serve, in their personal capacities, as individual advisers to the Panel. 6.3 The Panel, in consultation with the parties, prepared specific questions which it submitted to each expert individually. The experts were requested to provide their responses, in writing, to those questions they felt qualified to address. The parties agreed that their written submissions to the Panel, including the written versions of their oral statements, be provided to each of the selected experts. The written responses of the experts were provided to the parties, and the parties were given the opportunity to comment on these. 6.4 The experts were invited to meet with the Panel and the parties to discuss their written responses to the questions and to provide further information. A summary of the written responses provided by the experts is presented below.131 6.5 The experts selected to advise the Panel were:
B. QUESTIONS TO THE EXPERTS - COMPILED RESPONSES Question 1. Does the 1999 Import Risk Analysis on non-viable salmonids and non-salmonid marine finfish (1999 Report), in particular in so far as it relates to salmonids,
(a) evaluate the likelihood, i.e. probability, of entry, establishment or spread of the diseases of concern to Australia identified in the report?;
(b) evaluate the likelihood, i.e. probability, of the potential biological and economic consequences associated to these diseases?;
(c) evaluate the likelihood, i.e. probability, of entry, establishment or spread of the diseases of concern according to the sanitary measures which might be applied?
6.6 Dr. Br�ckner agreed that the 1999 IRA submitted by Australia was a qualitative assessment for the reasons outlined by Australia in the 1999 IRA and in their first submission (paragraphs 104 - 106). He further considered that the 1999 IRA was conducted in accordance with the OIE guidelines for import risk assessments and in accordance with the requirements of Article 5.2, 5.3 and Annex A(4) of the SPS Agreement . He observed that one of the main arguments Canada raised against the 1999 IRA was that it did not evaluate likelihood, because the IRA was qualitative and probabilities were not expressed in quantitative terms but in alleged subjective terms such as "low", "moderate", etc. A quantitative assessment was not required by the SPS Agreement . The fact that other quantitative IRAs existed (assessment by Vose on A. salmonicida and R. salmoninarum), did not put Australia under any obligation to do the same for an assessment of the same products or commodities concerned. Even if a quantitative assessment were possible, Dr. Br�ckner questioned whether a quantitative assessment for the same purpose and of the same magnitude as the 1999 IRA, would have produced a different outcome in terms of the evaluation of the likelihoods identified in the question. He noted that no evidence had been submitted to prove the contrary. 6.7 Dr. Br�ckner further noted that Canada had raised questions about the use of alleged subjective and vague indicators ("low", "moderate", etc). However, no alternative terminology had been suggested to be used in the context of a qualitative assessment or to enable an explicit indication of probability that the risk did not exist. The use of these terms should be evaluated in the context of the IRA in general and in respect of the process and methodology used to come to these conclusions. The use of these terms should also be judged as a way of expressing the outcome of a structured evaluation of several factors - i.e. to determine if risk management interventions needed to be considered or not. He noted that the conclusions in the New Zealand IRA of 1997 were also expressed in a similar qualitative manner. 6.8 Dr. Br�ckner considered that the crucial question was whether the use of these qualitative terms would make the evaluation of the likelihood, i.e. the probability, of entry, establishment or spread of the diseases of concern according to the sanitary measures which may be applied, impossible or questionable. Judged on the process and methodology applied - especially in respect of the assessment and risk management factors for the diseases concerned - he believed that the 1999 IRA succeeded in evaluating the likelihood of entry, establishment or spread of diseases, as well as the potential consequences, and this on the basis of the measures which might be applied. 6.9 With regard to the probability of entry, Dr. McVicar observed that there was international recognition, incorporated into legislation, that removal of viscera from fish carcasses reduced the risk of disease transfer. The 1999 IRA identified the two main areas of remaining risk which were of particular significance - firstly, that blood and residual blood rich organs are a major focus in the body of fish of important viral and bacterial diseases and secondly that viscera and other inedible parts of the fish body are of low value and may be disposed of either by safe legal or unsafe means. The risk associated with both of these areas was addressed and the level of viable infectious agents likely to be remaining in gutted carcasses in the parts usually removed and disposed of before human consumption was considered by Australia to warrant additional safeguards. The qualitative risk analysis undertaken was transparent in the criteria used to establish which diseases were of concern to Australia and in the identification of areas where risk management measures could be used to reduce the probability of entry of the pathogen. The analysis provided a well-reasoned argument why the measures proposed differed from these previously used in international trade in similar products 6.10 In terms of the probability of establishment or spread, Dr. McVicar noted that the IRA took into account the available published information on mechanisms of transmission of the diseases of concern and made a valid assessment of the probability of the establishment or spread of each in Australia. 6.11 With regard to evaluation of the likelihood of potential consequences, the diseases identified as of concern by Australia were all internationally recognised as serious diseases of salmonids with significant biological and economic consequences. There was no reason to consider that if established in Australia they would not have similar consequences in that country in susceptible species. Possible effects on other species, where there was no previous information available, were speculative. 6.12 In terms of evaluation of probabilities according to the measures which might be applied, the 1999 IRA recognised that even in eviscerated product which may contain some viable infection, there was a level of risk which was related to the extent containment measures could be implemented during further processing in Australia and to the proportion of the import which was rejected and subsequently disposed of. This risk was progressively reduced the more the product was processed and there was less potentially infectious material being discharged with effluent or rejected. The residual risk remaining after gutting and washing was considered by Australia to exceed their ALOP and this concern was addressed by introducing measures to limit the amount of non-consumer ready product being imported and, where further processing may occur, by controlling plants involved in this activity. 6.13 The level of infective agent in the source material at the point of origin critically influenced the level of the agent which risk reduction measures sought to manage at different steps through the chain of events leading to the final risk associated with imported salmonid product. As the levels of fish disease in both farmed and wild populations of fish were subject to substantial fluctuations, a key element in the reduction of risk from fish diseases was the regular maintenance of a good awareness of the level of the diseases of concern in the fish population providing the product. Appropriate data could only be obtained through a satisfactory system of regular inspection and monitoring, as required by Australia. Similarly, the use of inspection procedures, which were a normal part of quality assurance mechanisms in fish processing factories, to screen out clinically diseased fish would have a marked positive effect in risk reduction. 6.14 Dr. Wooldridge noted that Chapter 4 of the 1999 IRA, entitled "Risk Assessment: salmonids", was where the evaluation of the risk from each of the diseases should logically be found. The chapter contained a section, 4.2, entitled "Risk Assessments for High Priority Diseases", within which sections 4.2.1 to 4.2.15 each considered a specific disease. This was defined (section 4.1.3) as being an unrestricted risk estimate; that is, no safeguards had been considered in the assessment presented at that point. In addition, for each disease, this unrestricted risk estimate was compared with Australia's 'appropriate level of protection' (ALOP), which automatically lead to a decision on whether any risk management measures were warranted. Section 4.3 was a summary of the import risk assessment for salmonids. 6.15 With respect to the evaluation of the probability of entry, establishment of spread of diseases, for each disease considered, there was a section entitled Release Assessment, and one entitled Exposure Assessment. For each section, information was given, then an estimate of the probability of occurrence was given in qualitative terms, ranging from negligible to high (definitions: page 17). In addition, these findings were summarised in boxes at the end of each disease, along with a summary of Probability of Disease Establishment. On initial examination, it therefore appeared that the probability of entry, establishment and spread had been evaluated qualitatively. 6.16 Whether this was the case, however, in fact depended upon whether the information available had been utilised in an appropriate manner, and this in turn depended upon: the following issues: 6.17 whether or not any conclusion based upon primary information given was reasonable (relevant especially to Release Assessment and Exposure Assessment);
6.18 Looking first at the Release Assessment sections, Dr. Wooldridge indicated that the conclusions given for probability of "Release" in general, based upon the information given, were reasonable. However this would have been much easier to ascertain with certainty if the information presented, and the conclusions reached, had been separated into sections corresponding to probability of infection of fish and, given infection, probable titre levels by tissue. Presenting the information in this format would maximise transparency of a risk assessment. Currently, the arguments were presented as an amalgam of these two issues. In addition, Dr. Wooldridge expressed her opinion that some specific arguments were highly likely to lead to bias which, given the subjective nature of qualitative risk assessments, might inadvertently lead to unsafe conclusions. She gave the following example:
6.19 Did the risk conclusion follow reasonably from all the information? There appeared to be a difference in the probability of virus being present in different salmonid lifecycle stages. The overall release probability was given as low. The release probability for certain groups was given as moderate. This implied that there were groups for whom the release probability was less than low (as the implication was that low was an average for all groups). Groups where the release probability was lower than low were not mentioned in the release assessment, however, one example of such a group was (it would appear) identified in the key findings as having an "exceptionally low" titre (i.e. adults). If juveniles were being mentioned specifically in the release assessment, Dr. Wooldridge thought it would be logical to also mention this group of adults in the release assessment, in order specifically to reduce the likelihood of perception bias. Given the subjectivity involved in qualitative risk assessments, perception bias might well affect assessment and must be reduced wherever possible. This difficulty in interpretation of information to ensure logical conclusions would be much more obvious (and concurrently reduced) by differentiating, as suggested above, between probability of infection, and probable titre of pathogen given infection. 6.20 Dr. Wooldridge expressed the view that some significant information had been systematically left out of the sections entitled Exposure Assessment for each disease considered. For thirteen of the 15 diseases considered, an exposure assessment was given (for the remaining two, it was not applicable as the risk of release was considered negligible). This assessment comprised: seven diseases at Very Low; five at Low; and one at Low/extremely low. These outcomes appeared to be based only on the information given in the specific disease sections. 6.21 However, section 1.7 also dealt with Exposure Assessment in general terms, and in particular on page 34 there was a diagram of the assumed exposure pathways, with what appeared to be an indication of the proportion of total imported product likely to pass along each of those pathways, indicated by thickness of line. From this, it would appear that the probability of the product itself getting into the aquatic environment was likely to be exceptional by most pathways (five of seven). The two remaining pathways were the Domestic Sewerage system (annotated as heavily diluted, and itself not a highly probable pathway), and the Use as Bait pathway which, judging from the text (for example, section 1.2, page 5), was a route with a much higher probability of applicability to the non-salmonid group of fish which may sometimes be imported specifically as bait. One might therefore reasonably conclude that the Bait pathway for salmonids for human consumption was an exceptional route. In addition, the 1999 IRA on page 35 described the "extremely low probability of imported product following rare or exceptional pathways ...". 6.22 Dr. Wooldridge stated that in her opinion this section of information and the conclusions drawn from it did not appear to have been taken into account in the individual exposure assessments for specific diseases in salmonids. Taking this into account alongside a thorough re-examination of the disease specific information might well lead to the conclusion that for each disease the overall probability of aquatic exposure to salmonid product was exceptionally low, at the highest. 6.23 With regard to the sections entitled Probability of Disease Establishment, taking only the disease sections as they stand, each appeared to be internally consistent. However, if the additional information regarding exposure assessment detailed above was taken into account, then she believed it was highly probable that different conclusions would be reached, with the probability of establishment being lower in all cases. In summary, therefore, Dr. Wooldridge indicated that in general the probability of disease entry had been evaluated (but with reservations, including those expressed in her response to Question 2). However, she did not believe that the probability of the establishment or spread of disease had been evaluated. 6.24 In terms of the evaluation of the probability of the potential consequences, for each disease considered there was a section entitled Consequence Assessment, and an estimate of the probability of the consequences considered had been given in qualitative terms, ranging from negligible to catastrophic (definitions, page 19). In addition, these findings hade been summarised in boxes at the end of each disease. On initial examination, it therefore appeared that the probability of the consequences considered had been evaluated qualitatively. Further, Dr. Wooldridge opined that, in general, the arguments were internally consistent within each disease section, and the consequences of the disease, if it became established in Australia, had been assessed according to the definitions given (box 1.6, page 19). However whether those terms were then used reasonably to compare one disease with another, and to assess the necessity for risk management procedures was a separate issue which Dr. Wooldridge addressed in her response to Question 2. 6.25 Regarding the third part of the question, on the evaluation of the probabilities according to measures, Dr. Wooldridge observed that Chapter 5, entitled "Risk Management: Salmonids", was where the evaluation of the effect of safeguard measures for each disease should logically be found. Section 5.2 of the chapter described available safeguard measures, and sections 5.3 to 5.6 dealt with the application of these safeguard measures. Whether a particular disease in a particular stratum of the fish population required safeguards was based upon whether it met Australia's ALOP. If, in section 4, it was concluded that this ALOP was met by the unrestricted risk assessment, then no risk management measures were considered necessary. If the unrestricted risk assessment exceeded Australia's ALOP, then the implementation of risk management measures was considered warranted. 6.26 Dr. Wooldridge indicated that there were two issues here. The first concerned the derivation of the ALOP criteria, and this she addressed in her response to Question 2. The second was whether the probability of entry, establishment or spread of the disease had been evaluated according to the sanitary measures which might be applied. For each disease which did not meet the ALOP criteria, risk factors had been identified, and a list of possible risk management measures described. In addition, the particular risk factor which each measure would address was indicated. However, Dr. Wooldridge noted that she was unable to find any indication that the probability of any individual (or indeed any combination) of measures had actually been assessed specifically with regard to the likelihood of bringing the assessed risk below Australia's ALOP. Therefore, in her opinion, the probability of entry, establishment or spread had not been evaluated according to the sanitary measures which might be applied. Question 2. Do you for any other reason consider that the 1999 Report is not a proper risk assessment? If so, why? 6.27 Dr. Br�ckner recalled his view that the 1999-Report was conducted in accordance with the guidelines of the OIE for import risk analysis. Furthermore, the 1999 Report fulfilled the requirements of Article 5.1, 5.2 and 5.3 of the SPS Agreement without hindering the application of other relevant provisions of the SPS Agreement. He believed the 1999 Report could therefore be considered as a proper risk assessment. 6.28 Dr. McVicar stated that the 1999 Report fulfilled the OIE-outlined requirements of a qualitative risk assessment by identifying the hazards of concern, the possibility of their transfer to Australia, the possible consequences of transfer and the management steps which could be taken to reduce risk to an acceptable level. The underlying principle that quantitative risk analysis be developed as soon as possible by accumulating numerical data on the main risk areas was severely constrained with fish diseases in general due to the lack of adequate data in key areas. In both quantitative and qualitative risk assessment, there were inevitable difficulties and differences of opinion in deciding exactly what constituted an acceptable level of risk. Science could not provide definitive answers to this essentially social or political problem. 6.29 Dr. Wooldridge replied that the 1999 Report was set out in an appropriate manner, and contained the appropriate information, in order that it might be described as a risk analysis report containing within it a risk assessment. However, Dr. Wooldridge believed that the risk assessment was flawed and therefore may be considered not to be a proper risk assessment. She indicated that it did not use appropriate methods to properly assess the risks. 6.30 With respect to terminology, the terms used to describe the probability of an event occurring (Box 1.4, page 17) were, in themselves, acceptable qualitative terms. However, given the unavoidably subjective nature of such terms, Dr. Wooldridge did not believe that it was possible, in qualitative assessments, to easily grade probabilities with discriminations as fine as "low", "very low" and "extremely low" except as comparisons within the considerations of one specific disease, when for example a safeguard measure had been put in place which might be considered to reduce a probability from perhaps "low" to "very low". She certainly did not believe that one could with any accuracy discriminate as finely as this when comparing across diseases, unless a quantitative value range was assigned to each description. And if a quantitative value range were assigned, then in order to assign the qualitative designator correctly according to one's own definition, one MUST have performed a quantitative assessment to know in which range it fell. 6.31 Looking at the definitions given for these terms, she observed that "low" was defined as unlikely, that "very low" was defined as rare, and that "extremely low" was defined as very rare. In her view, this was merely changing one set of subjective words for another. To attempt to discriminate so finely, and to be certain that the assigned probability when applied to one disease indicated precisely the same level (or range levels) of risk as it did when applied to another disease, quantification must be undertaken. This use of these qualitative terms was therefore misleading in its implication of a level of precision which could not be achieved by qualitative methods. 6.32 With respect to the use of the risk evaluation matrix and the ALOP criteria., in the formulation of this matrix, a similar level of unachievable precision appeared to be assigned to the consequence assessment terminology. This, in conjunction with the implied, but unachievable, precision assumed in the estimation of disease establishment was combined to produce a matrix used for decisions on whether each unrestricted risk for each specified disease fell above or below Australia's ALOP. Thus the decision on whether further measures were required was based on finely categorized but highly imprecise and subjective discrimination techniques. A slight subjective shift in terminology could completely unintentionally quite easily move a specific disease from "yes" to "no" and vice versa. While a matrix such as this could reasonably be used as a guide as to which were the diseases of greatest concern, Dr. Wooldridge opined that this methodology was not appropriate for the highly "precise" use to which it was being put, and lead to unsafe conclusions. Such precision in discrimination could only be obtained by utilising quantitative methods. 6.33 In general, Dr. Wooldridge noted that her concerns over the methodology used thus far to assess the risks, and in particular the area dealing with exposure, had resulted in a lack of confidence in the final level of risk attached to the establishment of disease. This in turn made it very difficult for her to answer a number of subsequent questions based on the outcomes of the evaluation of these risks, in particular those concerning the management actions, for example Questions 4, 9, 17 and 20. When considering parts of some of these questions, she initially tried to compare different sections from the various risk analysis documents, to see if the necessity for, and level of, safeguards applied to one fish product in one situation was consistent with the estimated level of risk when compared with the safeguards and estimated levels of risk for another product in another situation. However, conclusions from this type of exercise meant assuming that the estimated levels of risks were reliable, and since in her opinion they were not, she came to no useful conclusions from this type of exercise.
Question 3. When carrying out a qualitative risk analysis, should consideration be given to volumes of imported commodities and to time periods?
6.34 Dr. Br�ckner responded that the historical events preceding the introduction of an SPS measure (i.e. whether diseases have been recorded or have been introduced in the absence, during the time-period, of a new proposed measure), could be of some value in making a qualitative judgement of the likelihood of a disease being introduced with or without the implementation of a new measure. However, the relative weight allocated to available historical information in respect of volumes imported over a certain time period should be applied with caution. The fact of the non-occurrence of diseases as result of unrestricted imports of non-viable salmonids prior to 1975, was for example weighted relatively heavy in the 1995 Draft Report of Australia; the same was not done in the 1999 IRA. This could be attributed to the fact that the 1999 IRA was conducted as a more structured way of assessment based on scientific facts to reach a decision on the feasibility of applying risk management procedures. 6.35 In the 1997 New Zealand IRA, where the authors explain the relative value of qualitative assessment, it is also rightfully stated that "� the risk will not vary from tonne to tonne or from year to year as result of an earlier result i.e. no disease introduction during importation of 1,000 tonnes of product does not increase the probability of introduction with the next 1,000 tonnes imported". 6.36 Dr. McVicar observed that qualitative risk assessment relied heavily on previous experiences, and a lack of previous episodes lead to an increasing perception of low risk. However, the consequence of an incident happening specifically as a result of an activity would immediately and completely change the perceived risk. To use a practical example from personal experience, as ISA had not occurred in salmon farming in Scotland during 20+ years of existence, with trade controls in place, the risk from this disease was considered to be low for over 10 years after its discovery in Norway. This was despite the proximity of the two countries. However, the appearance of the disease in eastern Canada in 1996-97, with no apparent transfer links, indicated a much higher risk of a similar outbreak in Scotland. Similarly, the first outbreak of a fish disease in Australia associated with an imported product would elevate what may have been considered a low risk to high risk status. 6.37 Irrespective of the disease level in the source of the commodity, each import episode would carry the same level of risk of the disease agent being present, with potentially the same consequences arising from its establishment. As the risk was repeated on each occasion as if the previous risk had not occurred, the frequency and time span that the commodity had been imported was not relevant. This assumed that the risk level did not change. However, within an individual import, the number of fish would have a bearing on the risk level, particularly if a disease is present at low levels. This was illustrated by the practice during disease surveillance when a probability table was used to determine confidence level of the presence or absence of disease. For example, a sample of 150 fish from a population of 100,000 or over would give a 95 per cent confidence that at least one infected fish would be detected if the disease prevalence level was equal to or greater than 2 per cent. For a 95 per cent confidence of detecting a 5 per cent level of disease from the same population, 60 fish needed to be sampled. Thus in a commodity with a disease level of 10 per cent, at least one in 60 fish would carry the disease, and for 2 per cent, one in 150. 6.38 The concept of a minimum infective dose was frequently used in the fish disease field. This suggested that if there were large single challenges or an accumulation of infective agents to reach these high "critical" levels of challenge, frequent and/or large volumes of import could be significant in initiating establishment of disease in an available susceptible population. However, this was a concept which was poorly understood, even in experimental conditions, and Dr. McVicar was not aware of any fish disease case where this had been properly addressed or quantified for field situations. 6.39 He also be noted that the absence of disease incidents associated with frequent large volume commodity imports did not indicate that this material posed no risk. The absence of previously associated disease problems only indicated that the import had a low risk in relation to the specific materials and conditions where that importation had previously occurred (locality, access to susceptible fish etc), not that the practice was safe under all conditions. Conclusions made from the absence of recorded incidents were only valid if appropriate monitoring had been in place to detect any problems which might have occurred. Dr. McVicar also drew attention to his response to Question 12. 6.40 Dr. Wooldridge indicated that if a given quantity (a specified unit) of product imported carried with it a certain assessed risk, then more of that product carried a larger total risk. This fact should be borne in mind whatever the type of risk assessment, as a guide to eventual decision making. In addition, if information on volumes of import was available, it made practical and methodological sense to collect it along with all other relevant information. 6.41 If the import risk per unit had been assessed as negligible (using the definition 'Chance of event so small it can be ignored in practical terms'; box 1.4, page 17), then this approximated to zero (it was not actually zero, of course), and any number of multiples of it might reasonably be considered also to approximate to zero. Therefore it could reasonably be argued that there was no necessity to consider time periods or volumes. Conversely, if the import risk per unit had been assessed as high, or probably even moderate, one would probably not contemplate the import of the product as it stands, therefore consideration of time periods or volumes was probably irrelevant at that point. The problem of multiples of a given quantity was only likely to occur when the risk was assessed as somewhere between moderate and negligible (with or without safeguards) per unit, when the amount per year might theoretically alter the risk from a level which a specific country was prepared to accept, to a level which that country was not prepared to accept. 6.42 However, a qualitative risk assessment was not generally undertaken with the kind of precision necessary to assess the risk in terms of units in any situation other than that of negligible risk, and if a per-unit risk estimate was required it would generally be necessary to undertake a quantitative risk assessment. To summarize, consideration to volumes and time periods should be considered in a qualitative assessment, but its significance was situation-dependent.
Question 4. Was any new scientific evidence referred to in the 1999 Report other than that used in the 1995 Draft Report? If so, was it of such importance that it warranted different quarantine measures than those proposed in the 1995 report?
6.43 Dr. Br�ckner responded that the approach and presentation of the scientific information in the 1999 Report differed substantially from that in the 1995 Draft Report. The information on some of the diseases that were presented in the Draft 1995 Report was essentially the same. However, the scope had been expanded to also incorporate non-salmonid marine finfish. The evaluation of scientific data was also structured to clearly distinguish between the assessment factors and the rational relationship thereof with the risk management factors. It could be reasonably accepted that the way in which the scientific facts were presented differently merited a re-evaluation of the quarantine measures to assess their justification in terms of Article 2.2 of the SPS Agreement. The different quarantine measures proposed were scientifically justifiable in view of the above. 6.44 Dr. McVicar indicated that much new information was included in the 1999 Report, reflecting the fact that research on fish disease was highly active and new data were continually becoming available. He had previously advised (to New Zealand) that Import Risk Assessment on fish diseases should be a continuously dynamic process and it was therefore appropriate that Australia had made good use of an extensive volume of relevant new scientific information which had become available during the last few years. For example, in 1995 ISA was not a recognised disease in Canadian or Scottish waters and this was properly addressed in the 1999 IRA. 6.45 Dr. Wooldridge noted that the date of some references was after 1995, therefore additional scientific information was referred to, but it had not been demonstrated to her that it was of such importance as to warrant different quarantine measures. However, in her opinion, it could not in any event be so demonstrated until the flaws which she had identified in the baseline assessment of the risk were addressed.
Question 5. Is the criticism in Canada's first submission, paras. 49-68, that "Australia's evaluation of likelihood is highly subjective" (paragraph 52) justified in particular when it comes to applying the terms "low", "moderate", � to event probabilities (critique in paras. 49-61); assigning relative probabilities to different disease agents (critique in paras. 62-66); and consequences of disease establishment (critique in paragraph 67)? Please assess the specific inconsistencies raised by Canada, but with a focus on the report in general.
6.46 Dr. Br�ckner observed that Canada's critique related in essence to the relative value attached to conclusions and probability expressions used in qualitative versus quantitative risk assessments. Although specific examples related to specific diseases were quoted to illustrate the alleged unacceptability of the terms "low", "moderate", their use must also be judged in relation to the report as a whole and the fact that the Report must not be assessed as a quantitative risk assessment. The specific examples quoted for A. salmonicida , IPNV, ISAV VERV and Vibrio anguillarum were used to illustrate the same argument. Dr. Br�ckner agreed that one of the dangers of a qualitative risk assessment was the possibility that the manner in which an outcome of an assessment was semantically expressed could be subject to a difference of opinion depending on the perception and opinion of the scientific evaluator. He drew attention also to his response to Question 1. 6.47 Dr. McVicar replied that, because of the lack of good available data, both the New Zealand Risk Analysis on Aeromonas salmonicida and the Vose Report had to make many general assumptions on biological aspects of disease and on aspects of transmission. Some of these in critical areas were highly subjective and were open to dispute. When in-depth studies had been undertaken on fish diseases, it had been demonstrated that there were a high number of interactive determinant factors which could influence transmission of infection and contribute to the subsequent variations in the level and effects of fish disease. For the diseases considered to be of potential significance to Australia, there was in general insufficient information available to conduct such detailed analyses, and it was appropriate that Australia chose a qualitative approach. 6.48 In any risk analysis (whether quantitative or qualitative) there was likely to be dispute between what was an acceptable level of risk and what was unacceptable, with the proponents of relaxation wishing higher levels than those desiring a more precautionary approach. Science sat uncomfortably with the relative subjectivity of the two stances. 6.49 With respect to Aeromonas salmonicida , the main risk of import of this agent was not from furuncules (which as indicated should be removed by inspection) but from the occurrence of infection in tissues in covertly infected fish. For this disease, it was frequent that fish were acutely infected and even died without showing clinical signs of the disease and heavily infected fish harvested from these populations were likely to be missed during inspection. Such fish could be expected to harbour high levels of bacteria in their blood and tissues which would remain after evisceration and washing. There was therefore justification in the assessment level of risk of entry as "moderate", taking into account the level of risk of establishment through potential pathways as agreed by both countries. 6.50 With respect to the relative probabilities assigned to IPNV and ISA, it would appear that there was a range of risk levels possible within one category and that several different factors could contribute to the assigning of a disease to any one category. For example, it was known that ISA more easily transmitted horizontally than IPNV, balancing out its more restricted known host range. It should be noted that there were reports in the scientific literature of ISA also infecting rainbow trout (Onchorhynchus mykiss) and sea trout (Salmo trutta). Recently there has also been a press notice from the Scottish Executive that ISA virus had also been found in eel Anguilla anguilla. IPNV also had greater stability than ISAV in the environment. 6.51 VERV could be treated differently in Australia from other important viral diseases (listed by OIE) because this virus already occurred in Australia, different strains had different host ranges and the virus was usually associated with juvenile fish which would not be imported. Dr. McVicar did not agree with the statement made in the First Submission by Canada (paragraph 65) that, A. salmonicida and R salmoninarum were rarely reported in older fish. This had not been the case in his personal experience so he did consider this stated discrepancy to be valid. 6.52 The belief by Canada that Australia overstated the potential damage which could occur to its disease and chemical residue free image as a consequence of A. salmonicida introduction, had some substance. Unless fish were farmed in quarantine-like conditions, it was inevitable that local diseases occurred, some of which required treatment. Australia was, and would continue to be, no exception. 6.53 Dr. Wooldridge indicated that she had replied to this question in her answer to Questions 1 and 2. In her view, Canada's criticisms as described in this question were justified in any event on methodological grounds. Furthermore, she did not understand why there had been no attempt to undertake a quantitative assessment (Canada's first submission, paragraph 49) as, although it was not specifically required by the SPS Agreement, it would have simplified and clarified the issues.
Question 6. Is Canada's critique justified that the 1999 Report does not in any substantial way evaluate the relative effectiveness of the risk reduction measures in reducing the overall disease risk linked to imports of Canadian salmon (first submission, paras. 69-79)? Please assess the specific inconsistencies raised by Canada, but with a focus on the report in general.
6.54 Dr. Br�ckner indicated that the risk reduction measures (risk management measures) as stated by Australia were a set of measures in accordance with the ALOP set by Australia in terms of their sovereign right to do so. This was done to accommodate a combination of measures to allow one set of requirements for the imports of salmon. The alternative would be to have several sets of requirements to be applied on a case-by-case basis depending on the disease presence or absence in the exporting country concerned. His opinion was that the measures developed did not place an undue trade restriction on the commodity concerned, with the exception of the requirements for "consumer ready" and "not consumer ready" products. His views on were contained in the response to Question 7 below in respect of the requirement for skin-on products. 6.55 Dr. Br�ckner agreed with the statement in paragraph 72 of the Canadian submission that the addition of one or two additional risk mitigation measures relative to the risk posed by a specific disease (i.e. presence of disease only in juveniles and non-spawning adults), might have a different outcome in terms of the overall risk reduction measures in respect of a specific disease. It would, however, restrict the purpose and scientific value of the 1999 IRA if the risk assessment of specific diseases were only to focus on certain age groups within a species (e.g. only adults) and ignore the risk posed by age groups not considered. In the import conditions outlined in AQPM 1999/51, the eight primary risk reduction measures reflected the outcome of a total evaluation. This approach incorporated the common and individual risks of the diseases concerned and not only of one particular disease (i.e. requirement 3 in respect of the exclusion of juveniles and spawners). The end result was the setting of requirements higher than the acceptable international standard (evisceration) only when scientifically justified by the 1999 IRA. This method of setting a combination of import requirements common to several diseases was not uncommon practice in respect of other animal food commodities. 6.56 Dr. McVicar replied that there were few cases of disease outbreaks occurring as a consequence of movement of processed fish for human consumption, that of whirling disease (Myxobolus cerebralis) being the most quoted example of a fish disease which had probably been spread with frozen fish and fish products. Australia identified practical options available for risk reduction associated with salmonid products. There was but limited relevant, quantified data available of the decrease in the level of pathogen present in the commodity after preparation to a consumer-ready state, supporting the logic that removal of inedible or low value parts would reduce (but not eliminate) the risk of this material coming into contact with waters containing susceptible fish. On this basis Australia made a judgement on their likely effectiveness in reducing this risk which was both transparent and logical. 6.57 Regarding the inconsistencies raised by Canada, Dr. McVicar recalled his comments on the occurrence of furunculosis and BKD in market-sized salmon (response to Question 5) and indicated that it was commonly reported that IHN-infected broodfish represented an important source of infection to the next generation. For all infectious diseases of salmonids, there was a close relationship between the severity of a disease and stress in the general sense and it was likely that any disease persisting in a population through from juvenile to adult to sexually mature fish could erupt under adverse conditions for the host. It was too simplistic to take the generalisation that diseases could be most prevalent in juveniles or sexually mature fish as indicating these were the only areas of risk, although for many diseases they did represent the highest period of risk. 6.58 Dr. Wooldridge drew attention to her reply to Question 1, part (c). In her opinion Canada's critique was justified.
Question 7. What risk is avoided by the removal of skin from Canadian salmon? And what risk is avoided by the requirement that skin-on product weigh less than 450 grams? Would the risk related to imports of Canadian salmon be any greater without these requirements? If so, would the higher risk be such that it exceeds Australia's acceptable level of risk, namely "a high or very conservative level of protection aimed at reducing risk to very low levels, while not based on a zero-risk approach" (Australia's first submission, paragraph 147)?
6.59 Dr. Br�ckner stated that no rational scientific justification could be found in the 1999 IRA for the specific requirements for "consumer ready" products (i.e. skin-on for less than 450 grams and skin-off for more than 450 grams). The only reference made in respect of this requirement (Australia submission paragraphs 66-67) was not a convincing scientific opinion. Mention was made of the preferences of the trade for specific products (Exhibit H). Trade preferences should however, not blur scientific judgement on risk. 6.60 It was also unclear why products greater than 450g should be processed at pre-release facilities within Australia. No reason was given why this could not be done in the country of origin in AQIS-approved facilities. Both the requirements for "consumer-ready" and products other than "consumer-ready" could be interpreted as trade restrictive measures in terms of Articles 2.2, 2.3 (second sentence) and 5.6 of the SPS Agreement. 6.61 Dr. McVicar replied that, with respect to skin, two questions should be considered. The first was whether fish skin contained infection at levels sufficient to provide a risk of transmission of disease agents. Recent information on the occurrence of disease agents such as Aeromonas salmonicida and ISA indicated that infection levels were high on the surface of skin and gills in live fish and that for ISA virus, blood, mucus and body fluids adhering to surfaces were important carriers of infection (e.g. on contaminated equipment) capable of transmitting this disease. Washing of carcasses was a requirement to decrease surface levels of infection in product, and this would undoubtedly remove much of the mucus with associated infection. However, the extent to which this reduction was achieved under normal factory conditions had not been quantified. As salmonid skin was not a blood rich organ and its actual tissues were not recognised as a significant site of infection of the diseases of concern to Australia, it was unlikely that salmonid skin or washed skin surfaces were important areas of infection risk in gutted carcasses. 6.62 The second question to consider was what was the risk of skin containing viable infection coming into contact with susceptible fish? Skin was a low value waste component which might be discarded in uncontrolled ways with the risk that any associated infectious agents present would be transferred to open environment. The removal of skin on non-consumer ready products before entry into Australia would undoubtedly remove this particular risk. 6.63 With regard to skin-on product less than 450 grams, Dr. McVicar noted that product less than 450 grams may be considered to be portion size and in a form which was acceptable for direct cooking without further processing. The risk associated with discarded low value parts of product was therefore again reduced by requiring imports of salmon to be in a consumer ready form. 6.64 In his view, based on current knowledge on the diseases of concern to Australia, the removal of skin from Canadian salmon was unlikely to make a significant contribution to risk reduction. 6.65 Dr. Wooldridge indicated that she was not competent to answer whether any particular pathogenic agent was likely to be highly localised to the skin, nor whether any such agents (if they existed) were likely to be found in Canadian salmon. However, if agent was not localised to the skin, then in her opinion removal of the skin would not affect the risk in any meaningful way. If it was, then removal of the skin prior to entry to Australia, provided that the skin did not also enter Australia, would reduce the risk of pathogen entering Australia. If, however, the skin was removed in Australia, then the total risk would be the same as if it were not removed, unless additional safeguards on skin disposal were put into place at the same time which reduced the aquatic exposure risk from skin. This would only be necessary if the total risk was unacceptable. 6.66 If the pathogen was skin-localised, then that pathogen was presumably also in the skin of skin-on products weighing less than 450 grams, which may also be derived from the same sources. Looking simply at the risk-releasing capability of the product therefore, the requirement to ensure a weight lower than 450 grams would not affect the risk. 6.67 However, Australia's argument (1999 Report, 5.2.2, page 199) appeared to be based on exposure pathways, and assumptions regarding human behaviour patterns, namely that consumers having purchased human food grade salmon products were more likely to use them as fish food or bait if they had skin attached and weighed more than 450 grams. Given the availability of (presumably) cheaper products for these purposes, it seemed unlikely that this would be a common occurrence, however a psychologist or home economics expert might be more appropriate to estimate this probability. There would probably be a differential effect depending upon the price difference between the products. Either way, it seemed unlikely that this use would constitute a large proportion of the total imported fish volume sold for human consumption. 6.68 The important point from a methodological point of view was that release and exposure pathways had already been considered (on a total imports basis) within the risk assessment part of the report. Exposure had been assessed as (at the maximum) low (and Dr. Wooldridge had already explained why she believed they should probably be even lower). Given what appeared to be the probable amounts (and their probable release potential) which would be disposed of by the specific exposure pathway postulated above, in her opinion it seemed highly unlikely that the total pathogen concentration in a particular area would vary significantly from the baseline assessment. An appropriate quantitative assessment would greatly help to clarify this issue. Question 8. Please give your views on the assertion by Norway (in its third party submission, paragraph 21) that "[i]t is difficult to see how a requirement that fish and fillets be 'not more than 450 grams' and 'pan size' has a relevant bearing on the risk, no rational explanation has been given by Australia in 1999-IRA". See also EC third party submission, paragraph 11.
6.69 Dr. Br�ckner replied that he supported in full the opinions expressed by both Norway and the EC, as outlined in his respnse to Question 7. 6.70 Dr. McVicar responded that the cut-off point of 450 grams did not reflect any known significant difference in the infection pattern of salmonids. As he had indicated in response to Question 7, a possible reason was that 450 grams was considered to be a maximum individual portion size, above which further processing, with the associated risks from disposal of effluent and unwanted waste, was likely to increase. 6.71 Dr. Wooldridge also recalled her answer to Question 7. There might be a highly unlikely theoretical potential justification but its necessity in practice remained totally unproven.
Question 9. Please give your views on Australia's assertion that "[t]he disease-based risk evaluation process destroyed once and for all assumptions that risk consistent with an ALOP could only be managed by applying the same measures to all products. The IRA's demonstrated that a comparison of risks between different products on the basis of measures applying to diseases in common is totally unscientific �" (paragraph 12 of its first submission) and that "[g]eneralisations about the relative effectiveness of controls on the internal movement of fish and fish products as part of risk management, as well as of the economic consequences are alarmingly unscientific" (paragraph 71). Please do the same in respect of the three points raised in Australia's submission at paragraph 124.
6.72 Dr. Br�ckner indicated that Articles 3.3, 5.3, 5.4, 5.6 of the SPS Agreement had a special bearing on the assertions of Australia. The general sentiment of all these relevant Articles in the Agreement was to not use an ALOP as a trade restrictive measure, consistency and the need to apply it only as far as protection of animal, human and plant life or health was concerned. Article 5.5 referred to ALOP in different situations, which could also be interpreted as "for different products". The question at stake was if an ALOP could be achieved better or in the same way if the measures were not generalised and applied differently to different products. Australia's assertion in paragraph 12 apparently warned against the generalisation of measures and formed the core of their argument for applying a disease-based risk assessment and management process (paragraph 10). The scientific argument was that disease manifested differently in different species and in respect of products of such species. It was asserted that this difference should be taken into account when determining risk management measures. In none of the Articles mentioned in the Agreement, was it required that there should be "across the board" conformity of measures to meet an ALOP. The process that was followed in the 1999 IRA also supported the view of Australia, although it could be reasoned that there were both advantages and disadvantages to this approach - especially if a measure was evaluated in terms of possible restrictions on trade that such a measure might impose. The approach of Australia appeared not to be inconsistent with the Agreement and could thus not be opposed. 6.73 The assertions of Australia in paragraph 71 were supported in the sense that they apparently did not oppose the need to bring national standards in conformity with international standards, but they added some perspective on the rationale for internal control under specific circumstances. They based their argument on the total assessment of risk, i.e. the risk posed by imports relative to the national situation and controls. It could be accepted that if an inland disease occurred in a localised manner (i.e. not endemic) but still posed a disease and economic risk, equal or non-discriminatory risk mitigation measures would apply. However, the mere fact that a disease was present within the national territory without taking the feasibility of risk mitigation measures into account in relation to the epidemiology of the disease, could be regarded as unscientific. 6.74 With respect to paragraph 124, first bullet, the 1999 IRA supported this assumption The general perception would be that fish for bait and live fish would pose a higher risk, however it must be evaluated against the release and consequence factors for the relevant diseases and in respect of the product concerned, and interpreted on scientific grounds and not from a subjective generalised perspective (i.e. "would always be higher"). 6.75 In terms of paragraph 124, second bullet, Dr. Br�ckner noted that the setting of an ALOP was not explicitly required in terms of the SPS Agreement and it could be said that the ALOP could be deducted from the measures applied. However, it did not imply that the ALOP was determined by the measure. In the normal practice of setting the ALOP (which was the prerogative of the Member), the ALOP was first determined (i.e. by risk assessment) and then followed by the measures necessary to meet the ALOP. 6.76 Paragraph 124, third bullet, raised the question of "same measure" for "one disease in common", which also related to Question 9. For the same reasons, Dr. Br�ckner concurred with this assertion. 6.77 Dr. McVicar replied that different diseases had different distributions, levels and locations of infection, and different survival capabilities and it was fully appropriate that each should be considered separately. For example, the risk management measures which could be applied for Gyrodactylus salaris (e.g. 2+ days without access to a live susceptible host species) were completely different from those appropriate for ISA virus (e.g. low pH) and completely different again from IPNV (e.g. high pH). It was true that the same methods could be effective in addressing risk for several different diseases, but the assumption could not be made that a limited suite of measures would be useful for all the diseases of concern. Dr. McVicar therefore supported the approach taken by Australia to evaluate the risks, consequences and appropriate measures to be taken on a disease-by-disease basis. Considering the three points raised in Australia's submission at paragraph 124, he recalled that the comparative risk from the 'one disease in common' between different types of product was addressed in his response to Question 10. The ALOP determined the measures required for each disease and each type of product, and, as indicated already in his response to this question, one measure might effectively achieve the ALOP for more than one disease of concern, but not necessarily for all.
6.78 Dr. Wooldridge responded that as she believed that the risk evaluation
process described by Australia in the 1999 report was seriously flawed, she not
believe one could base any further generalisations upon its results. In her
opinion it could not be said to have destroyed (or upheld) any assumptions
regarding risk management, or to have demonstrated anything regarding risks
between different products (paragraph 12 of the first submission). Using the
same rationale, in her opinion, the results of the 1999 risk assessments had not
confirmed the scientific validity or invalidity of the assumptions described in
paragraph 124.
Question 10. In your view, would the Australian measures now imposed on
imports of non-viable salmonids and those that will be imposed on other
non-viable marine finfish (in particular herring for use as bait) and live
ornamental fish, result in a similar level of protection, namely "a high or very
conservative level of protection aimed at reducing risk to very low levels,
while not based on a zero-risk approach" (Australia's first submission,
paragraph 147)? If not, is there any scientific justification for such
differentiation?
6.79 Dr. Br�ckner expressed his view that the measures imposed were in
accordance with the ALOP set by Australia and would result in a similar level of
protection.
6.80 Dr. McVicar noted that it had been shown on numerous occasions that live
fish presented the highest risk of diseases being moved between areas. Neither
ungutted fish nor gutted carcasses had been conclusively linked to extensive
transfer of infection between areas, although it was recognised that viscera
(with blood rich organs and low disposal value) posed a sufficiently high risk
for international agreement that removal was necessary to achieve sufficient
reduction in the level of associated risk. As pointed out by the 1999 Report,
all infection was not removed with viscera, because of remaining blood in
carcasses and the location of infection in other parts of the body. The
difference in risk between gutted and ungutted fish was therefore a matter of
degree and, in the absence of quantitative studies on the extent of reduction of
infectious agent present, a value judgement.
6.81 To consider whether or not there was scientific justification for a
differentiation between the measures being imposed on non-viable salmonids,
those on non-viable marine fish and live ornamental fish, Dr. McVicar indicated
that three areas should be considered:
(a) Some caution was required in the unreserved use of published host-disease
lists. Numerous reports of disease occurrence in fish species were the result of
experimental challenge or from samples taken from or in close association with
infected populations of the normal host in unnatural conditions, e.g. farms.
Because of the possible occurrence of infective carrier status, it was logical
to exercise strict controls on ornamental fish which had been in contact with
diseases of other fish, particularly in the abnormal conditions of farms.
However, under natural conditions, the process of infection might be more
difficult or even impossible in some cases due to the existence of a range of
biological or physical barriers. This finding was reflected in international
fish disease control regulations which did not usually recognise host species as
susceptible to a particular disease agent when the challenge had been abnormal
(e.g. experimentally) and natural occurrence of the infection had not been
demonstrated. Thus, for example, many of the "host" listings of IHN might not be
considered "valid".
(b) The existence of strains of the same infective agent of fish which showed
marked differences in pathogenicity and infectiveness (and therefore risk) was
well known. This might occur within one host species but was relatively common
when infections of the "same" species were found in several types of fish. When
it was known that these differences were due to inherent characteristics of the
infectious agent, not the host or environment, but the different "strains",
these could not be separated by currently approved diagnostic methods (probably
due to the inadequacy of the methods) and this caused a difficulty in how
legislative controls could logically function. To address this problem, a large
research programme was currently in progress in Europe in an attempt to refine
diagnostic methods of VHS/marine rhabdoviruses from different host species.
Similarly, it was apparent that some atypical strains of Aeromonas salmonicida
did not cause significant disease in salmonids when they came from non-salmonids.
As a generalisation, it could be concluded that infections which caused disease
in one species of fish would present the highest level of risk to stocks of the
same, or closely related species, in the importing area and less of a risk to
other species. These considerations were evident in some fish disease control
regulations (e.g. within the EU) where controlled trade in live ornamental fish
was permitted between zones under different status levels for controlled
salmonid diseases, while trade in salmonids was more tightly controlled.
(c) The risk from dispersion of identified diseases of concern from imported
fish or product and availability of susceptible fish were also relevant. A
substantial level of infectious agent ("minimum infectious dose") was required
before many infections could become established in a new individual fish or fish
population. Dr. McVicar recalled the uncertainties about the concept of minimum
infective dose for fish diseases expressed in his response to Question 3. It was
self-evident that the extent of dilution by the aquatic environment at the point
of release could be of critical importance. Thus four types of aquatic
environments, in order of increasing dilution capability (or decreasing risk),
were fish farm ponds, rivers, lakes and the open sea.
6.82 Dr. McVicar further noted that Australia intended to continue the import of
live ornamental fish and Canada raised the question essentially whether the
import of live ornamental fish, some capable of carrying certain of the salmonid
diseases of concern, posed a risk which negated the effects of the controls
being placed on salmonid products (first submission by Canada, paragraphs
92-95). Diseases of concern which were common between ornamental fish and
salmonids were IPNV and Aeromonas salmonicida. For the former, the measures
required for salmonid products were relatively undemanding in the level of risk
they accepted. For the latter, there was a recognition that non-salmonid fish in
fresh water were more likely to be infected with atypical strains than with the
typical strain of A salmonicida and that these as a whole had no significant
effect on the natural environment.
6.83 With regard to non-viable marine finfish, the same consideration of
differences in the biological "strain" of the one infective agent common between
different host species could be made. There was increasing evidence that there
was a group of marine rhabdoviruses (all identified as VHS by current diagnostic
methods) occurring in marine fish species which had a much lower infectivity and
pathogenicity than the "classic" VHS as found in fresh water rainbow trout
farms. Thus current diagnostic test methods could not differentiate between
strains which had major biological differences and might be grouping markedly
different infectious agents under a common name. Further research was ongoing in
this area in an attempt to resolve this difficulty.
6.84 The finding of IHN virus in Pacific herring was cited by Canada as an
uncontrolled risk associated with the import of herring into Australia for bait,
but it was worth noting that the records of infection were from experimental
challenge and from fish caught in the same general locality as infected farmed
salmon populations.
6.85 It was Dr. McVicar's view that there was scientific justification for a
different set of measures for different products being imported into Australia
because of variations in the nature of known infections which might be present
in the source material and because of variations in the actual and probable risk
of release of these into the environment in Australia. There was precedent in
the regulations in other areas. Host-disease lists found in the scientific
literature should not be used without careful evaluation as evidence of
susceptibility of a fish species to an infection (and therefore risk in the
product from naturally occurring infections). Dr. McVicar concluded that a very
low level of risk could be achieved by the measures Australia would now impose
on non-viable marine finfish and live ornamental fish.
6.86 Dr. Wooldridge stated that the definition of level of protection as a "high
or very conservative level of protection aimed at reducing risk to very low
levels" suffered from the same problems with regard to subjectivity inherent in
the words "high" and "very low" which she had referred to earlier. Australia had
stated that it required a "high or very conservative level of protection aimed
at reducing risk to very low levels", and if it was prepared to accept
non-viable salmonids, herring-bait and live finfish under certain conditions,
then the associated risks must, by definition, be reduced to at least what
Australia would define as "very low levels". The question, in her view, was
whether the restrictions put on non-viable salmonids actually took the risks to
an (unnecessarily) lower level that this.
6.87 This was not a simple question to answer without benefit of a risk
assessment laid out specifically to address this issue. There might be genuine
differences in the overall risk of disease establishment associated with
different fish species (requiring different safeguards) even when exposure
pathways were the same. Suppose that fish species F1 had an extremely low
prevalence of disease X, but exposure pathways which gave a high probability of
exposure (for example fish intended as bait). Overall the risk of establishment
of X would be assessed as extremely low (or lower) due to the extremely low
prevalence. It might be that no (or minimal) safeguards were considered
necessary to ensure risk was below acceptable levels.
6.88 Now suppose fish species F2 had a higher level of prevalence of disease X
than did fish species F1, and exposure pathways which gave a lower probability
of exposure than for F1 (for example fish intended for human consumption). Was
the overall risk of establishment higher or lower, or the same? This of course
depended upon exactly how much higher the prevalence was, and how much lower the
probability of exposure. It was possible that overall the risk of establishment
was higher despite the lower probability of exposure. In this case, more
stringent safeguards might be required for the fish intended for human
consumption, with regard to disease X in order to give the same level of
protection as that for the other fish product.
6.89 Although an intuitive argument, it could not therefore be assumed that a
less probable exposure pathway would automatically always lead to a lower level
of safeguard required for a specific disease. However, if the same disease was
present at the same prevalence level in two species of fish (for example a salmonid and a non-salmonid), and the probability of exposure for one was lower
than the other, then clearly that with the lower probability of exposure would
give a lower total risk (for the same quantity of product, and all else being
equal).
6.90 Some information could be obtained by looking at the risk assessment for
the same diseases in two groups of fish (for example, salmonids and
non-salmonids) and comparing risk outcomes with management required. However
since she believed that the assessments themselves were flawed, without a
re-assessment first she saw no point in this comparison. In Dr. Wooldridge's
opinion one other way to clarify this complex issue would be to attempt to
quantify the risks either for the same disease in different fish categories,
and/or for the disease considered qualitatively to have the highest risk for
each category of import, in order to ascertain whether there were differences of
order in the risks and likely safeguard levels required, compared with
safeguards required.
Question 11. Is it possible to verify in an objective manner on the basis of the
1999 Report and/or other evidence before the Panel whether the difference
referred to in question 10 exists and, if so, whether it is justified?
6.91 Dr. Br�ckner replied that the 1999 IRA supported the rationale of the
measure to achieve the same level of protection.
6.92 Dr. McVicar drew attention to his comments included in his response to
Question 10.
6.93 Dr. Wooldridge also replied that her answer to Question 10 answered this
question.
Question 12. Section 8.1 of the 1999 Report (section 8.1.2 in the Draft 1999
Report) states, at paragraph 345, that fish used for purposes such as fish feed
and bait is obviously more likely to introduce disease agents (if present in the
fish) into the aquatic environment than product imported for human consumption.
The same section notes, at paras. 346-47, that the use of thousands of tonnes of
imported pilchards, blue mackerel and herring as lobster bait over several
decades has not caused any detectable adverse effect on fish health or the
aquatic environment. The conclusion cited is that the risk of such imports
introducing an exotic disease that is capable of producing a large-scale fish
kill is either very low or does not exist at all.
At paragraph 115 of its first submission, Canada states, with reference to the
foregoing, that if the absence of disease transmission involving mere thousands
of tonnes of product from a few species in a small area is relevant to suggest
very low to non-existent risk, then it stands to reason, (according to Canada),
that the absence of disease transmission from billions of tonnes of dead,
eviscerated fish of all species moving all around the world is even stronger
evidence that the risk from such product is negligibly small. Please comment on
Canada's assertion.
6.94 Dr. Br�ckner observed that this question also related to Question 3 above
(consideration of volumes and time periods). The question could also imply
perceived discrimination between non-viable salmonids relative to non-viable
non-salmonids. The acceptance of the opinion expressed by Canada could also
imply acceptance of the historical facts as a reliable and scientifically
justifiable qualitative observation to defend a lower risk management measure.
It was, however, in contradiction to the earlier insistence of Canada (paragraph
49) for a quantitative approach. Acceptance of the assertion in paragraph 115
would nullify the need for a science-based risk assessment. It did not say that
the evaluator should not take cognizance of the historical facts, but these
should be tested and evaluated as was done in the 1999 IRA. In AQPM 1999/51,
Australia did not impose any restrictions on non-salmonid marine finfish, but
the restrictions imposed were in accordance with the outcome of the conclusions
made in the 1999 IRA.
6.95 Dr. McVicar indicated that the absence of clinical disease associated with
large amounts of fish imported as fish feed or bait might be taken as evidence
that no significant disease problems occurred associated with this product.
However, some caution should be used in directly accepting this conclusion at
its face value. The imported fish were being used as feed in highly specific
circumstances (tuna cages and as bait) mainly in the open sea. The lack of
associated infectious disease problems was relevant to these circumstances, but
not necessarily to others such as in and around salmon farming. The almost
exclusive use of processed diets in salmon farming internationally partly
reflected the risk to salmon farms associated with fresh or frozen whole fish
diets. Australia was taking similar risks with their tuna farms.
6.96 Dr. McVicar further noted that fish kills were difficult to observe in the
open sea. For example, there had been a major herring kill (estimated as 30 per
cent + of the standing stock) due to Ichthyophonus in the North Sea and Kattagat
in the early 1990s, but dead fish were only evident in the latter, relatively
confined, area. It was probable that many major fish kills in the sea had gone
unnoticed.
6.97 Dr. Wooldridge replied that in her opinion, Canada's assertion was a
logical deduction, and in the absence of acceptable evidence to the contrary it
was a statement which she would be prepared to accept.
Question 13. In paragraph 28-(a)(ii) of Australia's first submission, it notes
that imports are permitted subject, inter alia, to certification that "the fish
must be derived from a population for which there is a documented system of
health monitoring and surveillance administered by the competent authority." To
what extent can this be done for wild, ocean-caught salmon? If this requirement
is considered necessary with respect to risks associated with Canadian salmon,
should similar requirements not be necessary for non-salmonids?
6.98 Dr. Br�ckner indicated that he was not an expert on these management
systems and did not comment.
6.99 Dr. McVicar replied that, ass indicated in his response to Question 1, a
knowledge of the level of infection in the source population of the product was
beneficial in determining the extent that risk reduction measures needed to be
applied during the import process. This could be easily achieved in farmed
populations (salmonids and non-salmonids) and would be of most value there as
infected sick fish could have a prolonged survival in the absence of predation,
and a proportion passed through processing lines without being detected.
However, as sick fish did not usually survive long in the wild (largely because
of predation pressures), the level of serious disease in wild fish populations
was typically low, and few fish which were heavily infected with important
diseases were therefore likely to be caught An exception occurred during
epidemics when large numbers of heavily infected fish could appear. Current data
available indicated that such events were relatively rare, were typically of
short duration and were highly visible, particularly during any processing. With
wild fish populations, an accurate health monitoring programme was difficult to
implement, as extensive and complex monitoring studies were required to
determine the range of diseases occurring, or even the incidence of any specific
disease present. He noted that the level of health monitoring and surveillance
on wild fish populations was not specified by Australia.
6.100 The risk of disease being released from imported farmed non-salmonid
marine finfish or non-salmonid wild marine finfish into a marine environment at
concentrations where local susceptible stocks would be placed at risk was lower
than with salmonids (farmed and wild) because of the high dilution factor of the
marine environment in comparison to inland water with actual or potential
susceptible fish species. The relative levels of diseases risk between different
host species for the 'same' disease, as indicated in his response to Question
10, was also relevant to this discussion.
6.101 The comparison of the level of risk between wild and farmed fish of the
same species and between different species from the farmed and wild environments
raised many difficult problems. Fish diseases, when they occurred, often tended
to be at higher levels in farmed populations but the identification of which
diseases were there and at what level could be easily determined. Most existing
fish disease data had been derived from captive fish populations. Lower levels
of significant disease were normally found in wild populations so there was a
lower chance of dangerous levels of pathogen being imported with product.
However, the prevalence of a disease in a natural population did not directly
indicate its incidence. It was difficult to determine the range and level of
infections in wild populations, particularly in the extensive marine
environment. It was a matter of judgement at what level the inspection was
required to meet the ALOP on these different types of product.
6.102 Dr. Wooldridge replied that she was not competent to answer the first part
of this question. Given that for non-salmonids, the exposure pathways might
(generally, due to alternative use patterns) appear to present greater risks,
then (again, generally) it might be reasonably assumed that one would wish to be
at least (if not more) confident that the prevalence was lower for any given
pathogen from any given source. Given this, one would expect similar
requirements (at least) to be necessary for non-salmonids as for salmonids (but
she recalled her Question 10, that especially for a specific disease this might
not necessarily be the case).
Question 14. Australia observes in is first submission, paragraph 38, with
respect to live ornamental finfish that "Diseases may be localised, in many
instances at premises level, and disease status may alter rapidly". This is
presumably one of the reasons why the new measures require, inter alia,
certification of the health status of the premises [of export]. Is it not also
the case that farmed salmon diseases may be localised, in many instances at
premises level, and that disease status may alter rapidly? Should not
certification of the health status of farmed salmon premises be comparably
important, and effective, for risk management?
6.103 Dr. Br�ckner recalled that in paragraph 71 of Australia's first submission
it was stated that: "In respect of hatchery-sourced fish (including recreational
trout) the most effective means of risk management will usually be
hatchery-based". It was not stated if the risk management referred to included
participation in a national disease surveillance scheme for farmed salmon
diseases. However, in Appendix 6 of the 1999 IRA details were given of the
surveillance and monitoring of fish health in Australia. Dr. Br�ckner agreed
that an official health certification of farmed salmon premises should be
carried out to support the inland disease risk management.
6.104 Dr. McVicar that it was appropriate that certification of fish from
aquaculture establishments (whether salmonid or non-salmonid) providing high
risk exports be subject to certification that either they were located in a zone
free of the diseases of concern or that the farm was equivalent to an approved
farm in a non-approved zone (relevant for live fish or eggs). Where the risk was
considered to be lower (e.g. for non-viable product), it was appropriate that
the farm populations were subject to certification. The same situation pertained
to both salmonid and non-salmonid farms and there would be no scientific reason
to have such a requirement for one and not the other. In the case of live
ornamental fish, however, there was an increased risk if stocks had been
collected from several farms or wild areas prior to export and there would be
logic that this was specifically controlled by adequate certification of the
farm of export.
6.105 Dr. Wooldridge indicated that she was not competent to answer this
question.
Question 15. Please give your views on the assertions by Norway in its third
party submission, paras. 24-25 (e.g. "That there should be a need for different
measures against other finfish imported i.a. for human consumption and
containing the same diseases, is � far from clear").
6.106 Dr. Br�ckner responded that the measures proposed in AQPM 1999/51 for
non-viable salmonids and non-salmonid marine fish, respectively, did differ in
respect of monitoring required, age restriction and requirements for the nature
of the product once imported (consumer-ready, further processing in approved
plants post-arrival). It was unclear why there were differences for human
consumption in respect of consumer-ready and the need for post arrival
processing of salmonids if greater than 450 grams same questions were raised in
respect of the scientific justification for skin-on/skin-off relative to the
weight in terms of the risk created/not created. Dr. Br�ckner agreed that the
first four requirements for other finfish did establish acceptable risk
mitigation measures supported by the release and consequence assessment
conclusions in the 1999 IRA. However, he failed to find convincing evidence for
the reasons to pose further restrictions on size and processing on salmonids but
not on other finfish.
6.107 Dr. McVicar replied that the views of Norway regarding the more burdensome
measures to be imposed on salmon imports would only be valid if there were
homogeneity in the distribution of the different diseases of concern and
pathogenic strains (e.g. strains of A salmonicida) throughout the different
source populations of fish species from which product was being imported. This
was not the case. Taking Aeromonas salmonicida as an example, the typical strain
causing classic furunculosis was commonly a serious disease in salmonids
throughout many parts of the world (excluding Australia) whereas several
atypical strains were widespread and caused ulcer disease in a wide variety of
other fish species. These atypical strains only caused serious disease in
salmonids in relatively few areas (eg Iceland, Japan). It was his interpretation
that the proposed Australian measures were specifically targeted against
salmonids because of the risk from specifically salmonid diseases such as
typical strains of A salmonicida, and that infections occurring in non-salmonids
represented a lower level of risk to this species. To this extent the
differences in the measures being applied to different products could be
justified. He also drew attention to his response to Question 10.
6.108 Dr. Wooldridge agreed that the need for different (or any specific)
measures was far from clear.
Question 16. Is the difference in measures applied to New Zealand and Canadian
salmon justified on scientific grounds, in particular the disease status of New
Zealand?
6.109 Dr. Br�ckner stated that, considering the disease status of New Zealand as
outlined in the 1997-IRA of New Zealand, he would not contest a change in the
status quo. However, it could reasonably be asked why Australia did not apply
the same measures for import as New Zealand for non-viable salmonids, since the
New Zealand measures are considerably less restrictive than those put in force
in Australia.
6.110 Dr. McVicar indicated that the import measures applied by New Zealand were
essentially similar to those of Australia. Because of their geographical
proximity, similar history regarding the occurrence of non-indigenous salmonids
and similar disease profiles (except for Whirling Disease), both areas might be
exposed to similar risks from imports of salmonid products. A degree of
similarity in risk reduction measures between the two countries would be
expected. From a scientific point of view, there was recognition of the concept
of more relaxed trading between zones of comparable fish health status than when
trade wass from zones of lower fish health status to zones of higher health
status. This was the basis for the EU Directive 91/67/EEC. On the same
scientific principle, it was therefore not illogical to have a similar
relationship between Australia and New Zealand.
6.111 Dr. Wooldridge noted that it was more appropriate for a fish expert to
answer this question.
Question 17. In view of the similar disease agents that may be present in
salmonids and non-salmonids, please comment on the validity of the distinction
made by Australia between salmonids, which may not be imported or released from
quarantine unless processed to "consumer-ready" form, and non-salmonids, which
are not required to be processed to consumer-ready form.
6.112 Dr. Br�ckner observed that this question related to a similar question
posed by Norway (Question 15). He failed to see the reasons and scientific
justification for a distinction at this level as outlined in his response to
Question 15.
6.113 Dr. McVicar also recalled his response to Question 15.
6.114 Dr. Wooldridge indicated that her answers to Questions 7 and 10 in part
answered this question. In summary, if the same disease in two different species
was present at the same prevalence, and if the exposure pathways were the same
and of the same probability, then there would be no justification for
differences. This identical situation was unlikely to arise, and the overall
risk depended upon both. There may therefore be justification for treating two
different products in different ways. However, the justification for this
particular safeguard for salmonids was, in her opinion, unproven and unlikely so
to be.
Question 18. Please give your views on paras. 15-24 of Canada's submission of 30
September 1999, in particular on the question of whether Australia's measures on
imports of salmon and those on imports of whole, uneviscerated pilchards,
including for use as bait, result in two (substantially) different levels of
sanitary protection in these two areas and, if so, whether there is any
scientific justification for such differentiation?
6.115 Dr. Br�ckner noted that the issues surrounding Pilchard herpes virus were
discussed over three paragraphs in section 6.2.1 of the 1999 IRA and not further
considered in the IRA. Canada had submitted scientific literature to verify
their concern. In the absence of any further substantial scientific evidence, it
was not possible to make a judgement on this issue. However, if the virus was
considered to be endemic, then the assumption made by Australia not to institute
risk management practices in respect of the disease or other "unknown diseases"
that might be introduced through imports was valid and justified.
6.116 Dr. McVicar responded that because Australia considered that the herpes
virus associated with the disease outbreak in Australian pilchards was endemic,
it was not included as a disease of concern and accordingly there was no desire
to introduce measures to restrict this disease in imports. The issue of VHS was
addressed in his response to Question 10. When there was a lack of evidence in
the scientific literature of infectivity and pathogenicity associated with a
disease agent, the introduction of control measures which might affect trade on
a precautionary basis would be difficult to justify.
6.117 Dr. Wooldridge believed that Canada's submission as presented did indicate
a substantial difference in levels of sanitary protection for the two products
under consideration, for which scientific justification was not immediately
apparent.
Question 19. Is it possible to verify in an objective manner on the basis of the
1999 Report and/or other evidence before the Panel whether the difference
referred to in question 18 exists and, if so, whether it is justified?
6.118 Dr. Br�ckner replied that the assumptions of Australia could possibly be
objectively justified (or proven to be incorrect) by further literature research
and/or by means of simulation disease modeling in respect of the impact of
disease introduction through the importation of pilchards.
6.119 Dr. McVicar recalled his response to Question 18.
6.120 Dr. Wooldridge observed that Canada argued that Australia was prepared to
accept whole uneviscerated pilchard imports with substantially less safeguards
than for eviscerated salmonid imports (paragraphs 14 and 15, Canada's submission
of 30 September.). The evidence that the safeguards required for such pilchards
was less than that for such salmonids was available in the final 1999 report.
Canada argued that it believed that the actual risk of disease establishment
from such pilchard imports was higher than that from eviscerated salmonids. It
presented its own evidence of the probability of disease incursion and
establishment for pilchards using as an example a disease believed by scientists
to be due to pilchard herpesvirus (various references, many from Australia's own
documents).
6.121 Section 6.2. of the Australian 1999 report was a hazard identification
exercise for "Diseases/Disease agents of non-salmonid marine finfish". Section
6.2.1 identified Viruses. In this section pilchard herpesvirus wa s identified
(pages 256-7) as a hazard. It was described as "associated with extensive
mortality in pilchard" and "reported only in Australia and New Zealand", and
"this agent is not further considered in this IRA". The paper supplied in the
Canadian submission of 30 September (Whittington, reference in footnote 88, page
3) gave evidence that the consequences of this disease in Australia had been
serious; therefore the probability was that they would again be serious in a
further outbreak. Dr. Wooldridge did not think this was in contention.
6.122 The remaining issue from an import risk assessment view was therefore what
was the probability of the disease having been imported along with fish imports.
She had not found anywhere where this probability was considered in the 1999
report (and indeed it was stated not to be, in the quotation above). The fact
that it had only been reported in Australia and New Zealand did not of itself
rule out the possibility that it had been imported (though it was a relevant
fact in estimating this probability). Whittington et al (see footnote 88), on
page 14, looked at the epidemiology of this disease in Australia and New
Zealand, and considered the probability that the infection could have come from
an external source, examples given being commercial shipping discharges and
imported pilchard baitfish. In Dr. Wooldridge's opinion, the authors of this
paper considered this external source as a higher than negligible probability,
for a number of described epidemiological reasons.
6.123 Australia's own draft document, 1999 Draft Report (section 8.1.2), also
referred to by Canada (paragraph 16, 30 September), discussed the probability of
this virus being exotic to Australia, on pages three and four of this section
(8.1.2). Of note was the submission by the Chair of the CCEAD Joint Pilchard
Scientific Working Group, which stated that:
"The Working Group is currently coordinating a national research programme �
with one of the objectives to determine whether the virus is endemic or exotic
and, if exotic, the source of the virus. Results to date do not support any
definitive conclusions �"
6.124 This doubt was, in her opinion, very relevant to the assessment of the
probability of risk from this particular disease due to the import of pilchards.
The fact that this uncertainty had not been considered in the final 1999 report
was a significant flaw in the risk assessment for non-salmonid marine finfish,
particularly given the consequences of an outbreak. It was also very odd
methodologically to have ignored this uncertainty in the final report.
6.125 This evidence clearly did not support the case for justifying more
stringent safeguards for eviscerated salmonids than for whole non-salmonid
marine finfish. In itself it said almost nothing about the safeguard measures
appropriate specifically to eviscerated salmonids. By itself, it would tend to
support a case for stringent safeguards on imported pilchards although it must,
of course, be considered in context with other evidence.
Question 20. Would any series of measures more limited than the current
Australian set of measures on imports of Canadian salmon achieve Australia's
acceptable level of risk, namely "a high or very conservative level of
protection aimed at reducing risk to very low levels, while not based on a
zero-risk approach" (Australia's first submission, paragraph 147)? Please be
specific.
6.126 Dr. Br�ckner indicated that he did not see the rationale behind this
question. One could list several other additional procedures such as compulsory
laboratory screening of consignments before export certification for the
diseases of concern, but this would just be a shopping list without answering
the question.
6.127 Dr. McVicar replied that. in general, it appeared that Australia had
identified the minimum risk reduction measures which could be implemented to
safeguard local stocks from the identified diseases of concern. There would not
appear to be a major disease risk associated with salmonid skin which would
substantially alter the level of risk by its removal, but balanced against this
was the greatly increased risk that this inedible, low value material might be
disposed of in an unsafe manner prior to cooking.
6.128 Dr. Wooldridge indicated that this would depend upon how effective each
measure was in reducing the particular risk factor(s) which it addresses, and
that she was not competent to assess this.
Question 21. Is it possible to verify in an objective manner on the basis of the
1999 Report and/or other evidence before the Panel whether any of the
alternative policy options, or any alternative set of measures, would achieve
Australia's appropriate level of protection?
6.129 Dr. Br�ckner observed that this could be possible but would imply another
risk assessment as risk management measures/policy options were the result of an
outcome of a scientifically based risk assessment. It would imply the proposal
of less stringent risk mitigation measures and an evaluation thereof in terms of
consequence. He was of the opinion that the existing 1999 IRA has followed this
approach in general to establish a rational relationship between the proposed
measures and the ALOP.
6.130 Dr. McVicar was of the view that Australia had objectively analysed the
available fish disease data to identify the diseases of concern and had
identified the most likely and practical means of reducing the risk of their
introduction with product with appropriate deference to limiting restrictions on
trade. The lack of quantified data on the extent of reduction in the levels of
infective agent present or of restriction of access of important infections to
local fish populations precluded objective assessment of the likely success of
the proposed measures.
6.131 Dr. Wooldridge noted that she was not competent to assess this.
Question 22. What is the level of risk to Australia's commercially or
recreationally important fish populations resulting from diseases such as EHNV,
GUD, VERV and EUS (referred to in paras. 136-144 of Canada's first submission),
which are endemic to some parts of Australia but exotic to others, given that no
restrictions on the domestic movement of dead finfish are imposed? Is this risk
similar or higher than that of imports of Canadian salmon under current
Australian requirements? If so, is there any scientific or technical
justification for not imposing restrictions on the domestic movement of dead
finfish, including uneviscerated fish?
6.132 Dr. Br�ckner indicated that this question could be better answered by a
fish disease expert.
6.133 Dr. McVicar responded that most fish diseases showed variations in their
occurrence both at the local and wider geographic levels. Even where diseases
were considered widespread within a country/area, control through legislative
restriction of movement of live fish or parts of fish could be beneficial where
there was either a discontinuous distribution or there were certain populations
which were particularly vulnerable (for example farmed stocks). This was an
underlying principle of several fish disease control regulations (e.g. UK
Diseases of Fish Acts, the List III classification of certain diseases in EU
Directive 91/67/EEC). Where significant risk had been demonstrated, national or
international controls could be imposed. Without detailed knowledge of the
vulnerability of the fish populations in different parts of Australia to the
diseases listed in the question, it was not possible to predict the risks to
these and benefits which might be derived from the introduction of national
controls on fish or fish products. Although there was an increasing trend
towards use of the precautionary approach in the deployment of restrictions, it
was his firm view that regulatory controls on fish diseases should be subjected
to both risk and cost benefit analyses to provide a logical scientific basis.
Also, because of variations between diseases in such factors as infection
levels, pathogen survival, use of product etc, it was not possible to conclude
that measures used to reduce risk for one disease would have a similar benefit
for other diseases. Each disease considered to be potentially significant
required individual consideration of the measures required to reduce risk of
spread. As the diseases listed in paragraph 138 of the first submission of
Canada were not those being controlled in imports of Canadian salmon into
Australia, direct comparison would not be appropriate.
6.134 Dr. Wooldridge indicated that she was not competent to answer most of this
question. To fully answer the question as to the level of risk of internal
movement of dead finfish, one would have to undertake a risk assessment
specifically to examine the risks of internal dead finfish movements. Canada
suggested that these risks would prove significant. To compare the results to
see whether this risk was higher or lower than the risk of imports into
Australia from any given country (e.g. Canada), would then be possible. If
similar fish species and products from different sources were entering a given
area of Australia, and if the exposure pathways once the product had entered
that area were the same (which would be highly likely if the products were the
same and intended for the same uses), then the overall risk comparison would
depend upon the prevalence of the disease(s) of interest within the fish being
moved in, and the quantity of such movements from each source.
6.135 Therefore, if fish products with a high prevalence of disease X were being
moved into area B from area A, and the same fish products with an equal
prevalence of the same disease X were being moved (in the same quantities, and
all else being equal) into area B from a different country, then the risks to
area B would be the same. If the level of X in the different country was
actually lower than in area A, then the risks to B would be lower, and vice
versa. Given an identified hazard and an assessed risk to area B, whether these
theoretical considerations had any practical application in risk mitigation
depended (in part) upon whether imports to a given country could legally be
allowed only into certain regions or zones of that country, as well as whether
there were internal movement safeguards. However, if there was an assessed high
risk of internal disease transmission to area B by fish product movement and
this was not addressed then, Dr. Wooldridge believed that it would seem
superfluous to attempt to stop the same disease(s) by restricting foreign
imports to the same area.
Question 23. Can A. salmonicida (typical and atypical) be detected through
visual examination? If not, what SPS measure can be adopted to imports of live
ornamental finfish known to host these agents, and to whole herring for bait
also known to host them?
6.136 Dr. Br�ckner indicated that this question could be better answered by a
fish disease expert.
6.137 Dr. McVicar responded that most species of fish appeared to be susceptible
to typical furunculosis, but the level of susceptibility, and hence associated
pathology which might be detected, was variable. Covert typical furunculosis (ie
clinically unapparent infections) due to Aeromonas salmonicida subsp.
salmonicida had long been recognised as a problem causing difficulties when
controlling this disease in salmonids. Covertly infected fish were capable of
acting as carriers and initiating infection in other fish. In cases of acute
furunculosis, high mortalities might occur in affected populations without
showing external signs of infection. In both of these cases, it was unlikely
that visual examination during inspection of carcasses would detect all infected
fish.
6.138 A typical Aeromonas salmonicida occured in a large number of host species
in fresh and sea water and the associated pathology which had been recorded was
variable. However, the most common clinical sign of atypical A salmonicida was
skin ulceration which would be detectable through visual examination. Screening
of live ornamental fish for such lesions at import and the use of quarantine
restrictions with health surveillance after import (i.e. the SPS measures
indicated by Australia for live ornamental finfish) would undoubtedly reduce the
risk of heavily infected fish being released, but would not necessarily detect
and remove covertly infected fish. It thus appeared that Australia was prepared
to accept a higher level of risk for atypical strains of A salmonicida than for
typical. Similarly, although freezing of herring would undoubtedly considerably
reduce the level of viable disease agents of concern present in imported bait,
it could be expected that a level of infectivity, and therefore risk still
remained in these. Balanced against this was the likely variability in "strains"
of agent between different host species and the likely dilution factor occurring
in the area of use after import (also referred to in his response to Question
10).
6.139 Dr. Wooldridge noted that she was not competent to answer this question.
Question 24. Does OIE listing/absence of listing of different diseases and/or
their OIE categorization reflect the outcome of risk evaluation and risk
management comparable to that of Article 5.1?
6.140 Dr. Br�ckner replied that it was accepted that the OIE listing was not
complete as indicated in the 1999 IRA and was therefore evaluated on a
continuous basis by the Fish Diseases Commission of the OIE on the
recommendations of Member Countries. Should the 1999 IRA only have focussed on
those diseases listed in the OIE Code, it would have been incomplete. Article
5.1 of the SPS Agreement also referred to risk assessment techniques developed
by international organisations. The 1999 IRA used these as a guideline, in
conformity with Article 5.1.
6.141 Dr. McVicar noted that the OIE listing of diseases was a reflection of
decades of experience of the pathogenicity and consequences of these infections
in several countries, the absence of suitable control measures, an awareness of
their restricted distribution and how amenable they might be to containment and
control by legislation. As there were no published records, it was apparent that
in no case had a formal risk assessment been carried out although, loosely, the
international awareness of and agreement on their effects could be classed as a
form of qualitative risk assessment. It was no coincidence that most of the
OIE-listed diseases were also controlled diseases in other national (e.g. United
Kingdom, United States, Ireland, Canada) or international (e.g. EU) fish disease
control regulations. OIE annually took representation on fish disease issues
from national participants, assessed the regulations and lists of diseases and
if considered appropriate, might add diseases to the lists or remove others.
6.142 Dr. Wooldridge indicated that she was not competent to answer this
question.
131 A transcript of
the meeting with the experts is attached as Annex 1 of this document.
To continue with
Return to Table of Contents |
|
