Quest for a discard survival predictive scoring system to use on board fishing vessels

Releasing tagged Atlantic cod, John Clarke Russ

The European Union Common Fisheries Policy (CFP) ban on discarding allows for animals to be discarded if “scientific evidence demonstrates high survival rates”. Estimating discard survival for fisheries has become a priority for implementation of the CFP. Limited data on discard survival and mortality is available and methods for estimation have not been standardized. Ideally, a standardized numerical scoring system can be developed and validated, based on readily observable responses and symptoms present in animals that are candidates for discarding and survival. 

RAMP is an example of a predictive scoring system for vitality, survival, and mortality, based on animal reflex actions, barotrauma symptoms, and injury that can be observed in fishing operations where real time decisions must be made about potential discarding. See post for RAMP development and validation; also Davis (2010) and Stoner (2012) for reviews of RAMP method. Other uses for RAMP are in live fisheries, aquaculture, and pollution research and monitoring.

For inspiration and alternative perspectives, examples of validated mortality predictive scoring systems can be found in human and veterinarian intensive care unit (ICU) settings, where patients present with symptoms and disease likely to result in morbidity and mortality (Rockar et al. 1994, Bouch and Thompson 2008, Timmers et al. 2011). Measurements of blood plasma and urine variables commonly made in ICU settings are not contemplated for RAMP since they are not readily made on board fishing vessels.

Below is an example of human ICU mortality prediction using the SAPS II scoring system. Note the similarity to RAMP curves for mortality prediction.

SAPS II mortality predictive scoring system, ClinCalc

Celinski and Jonas (2004) discussed scoring systems developed for the human ICU environment:

“How are scoring systems developed? All available data types and variables can potentially be used to create a scoring system. However, to make it useful, variables have to be selected to be appropriate for the predictive properties of the scoring system. The information must be unambiguous, mutually exclusive, reliable and easy to determine and collect. Ideally, the variables should be frequently recorded or measured.

The variables can be selected using clinical judgement and recognized physiological associations or by using computerized searching of data (collected from patient databases) and relating it to outcome. The variables are then assigned a weighting in relation to their importance in the predictive power of the scoring system (again by using clinical relevance or computerized databases).

Logistic regression analysis, a multivariate statistical procedure, is then used to convert a score to a predicted probability of the outcome measured (usually morbidity or mortality) against a large database of comparable patients. Lastly, the scoring system must be validated on a population of patients independent from those used to develop the scoring system.”

For discard survival prediction, groups of animals, rather than individuals, are the appropriate unit for consideration since proportion mortality is the determined outcome during index development and validation. These groups can represent various scales of resolution in fisheries of interest, i.e., single tows or traps, sets of longline, trap, or gill-net, daily catch.

Jean-Roger Le Gall (2005) discussed the appropriate use of ICU severity scoring systems:

“A good severity system provides an accurate estimate of the number of patients predicted to die among a group of similar patients; however, it does not provide a prediction of which particular patients will in fact die. Using a well-calibrated severity model, we can reasonably expect that approx. 75% of patients with a probability of mortality of 0.75 will die, but we cannot know in advance which of those patients will be among the 25% who will live. Furthermore, these 25% will not have falsified the odds but will have confirmed the validity of the probabilities. 

       The possibility that clinical decisions can be augmented by having an objective (although not always more accurate) assessment of a patient’s severity of illness is appealing. Physicians are interested in severity systems for individual patients as an adjunct to their informed but subjective opinion. Using these tools as part of the decision-making process is reasonable and prudent. Using these tools to dictate individual patient decisions is not appropriate. Decisions will and should remain the responsibility of the individual physician and should be based on a number of criteria, one of which is severity as estimated by a well calibrated scoring system.”

Stacy et al. (2013) discussed development and appropriate use of a predictive scoring system for survival in Kemp’s ridley sea turtles:

“Three mortality prediction indices (MPI) scoring systems were developed using different combinations of blood analytes, with anticipation that at least one of the three would be more accurate in predicting mortality in sea turtles within 7 days after admission. Turtles with higher scores were categorized as physiologically deranged to a degree that could result in death, and turtles that received lower scores were categorized as physiologically stable and likely to survive. Categorization of each turtle was then compared to the known outcome for that individual.

Receiver operating characteristic (ROC) analysis was used to assess the diagnostic performance of each MPI scoring system (Greiner et al., 2000; Giguere et al., 2003). The ROC analysis produces a plot that is used to estimate the area under a ROC curve, which is a summary statistic of diagnostic accuracy. A perfect test [i.e., sensitivity (SE) = 100% and specificity (SP) = 100%] will produce an area under the curve (AUC) = 1. The AUC can be used to distinguish a non-informative test (AUC = 0.5), a less accurate (0.5 < AUC ≤ 0.7), moderately accurate (0.7 < AUC ≤ 0.9), highly accurate (0.9 < AUC < 1), and perfect test (AUC = 1)."

ROC analysis of Kemp's ridley sea turtle mortality predictive scoring system, Stacy et al. 2013

"It is clear that results of mortality prediction indices (MPI) scoring systems cannot be used indiscriminately to make euthanasia decisions, because this would result in euthanasia of some turtles with a falsely positive MPI score that would otherwise survive. As with other health scoring systems in human and veterinary medicine, the MPI scores should not prevent clinicians from providing care to an individual, and euthanasia decisions should only be made in light of numerous other clinical factors, including neurological status, vision, ability to forage, ability to swim, pain and suffering, and duration of illness. Finally, MPI scores may be useful when applied retrospectively in a stranding event for comparison of various treatment outcomes within a facility or among different facilities. Thus, the MPI could provide an objective assessment tool of treatment success and contribute to the advancement of medical care in sea turtles.”

Effect of hypoxia, injury, and facilitated recovery on reflex impairment (RAMP) in migrating sockeye salmon

Sockeye salmon and Dolly Varden, J Armstrong/UW

A study on the effects of hypoxia and injury associated with gill net fisheries and facilitated recovery in sockeye salmon migrating in the Fraser River showed important results for the use of RAMP to measure sublethal effects of stressors (Nguyen et al. 2014).

The authors state:

“Here, we examined sources of delayed fisheries-related mortality in relation to three known factors influencing postrelease behavior and mortality in fish: physiological exhaustion (stress through air exposure), physical damage (via gill net entanglement), and facilitated recovery (using Fraser boxes). We used sockeye salmon (Oncorhynchus nerka) in the lower Fraser River as a model for this research, given conservation concerns regarding a number of sockeye populations (see Cooke et al. 2012). The study was designed to simulate gill net fisheries because high levels of delayed mortality may have important implications for harvest management in exploited and non-target salmon populations. Our primary objective was to distinguish the relative consequences of physical injury and air exposure stress using an experimental approach coupled with reflex assessments (Davis 2010), physiological sampling (non-lethal blood samples; see Cooke et al. 2005), and telemetry tracking of postrelease migration success (Donaldson et al. 2008). Specifically, we used assessments of reflex impairment and blood physiology to characterize the relative impacts of our experimental treatments. Our secondary objective was to test whether Fraser recovery boxes could reduce delayed mortality and improve migration speed for captured fish exposed to varying degrees of stress and injury.”

The study used experimental stressor treatments: C - Captured-only; A - captured and Air exposed; I - captured and Injured; IA - captured and Injured + Air exposed.

The authors found:

“RAMP is intended to be a rapid, simple, and inexpensive means of assessing fish vitality (Davis 2010). It has also been validated as a predictive measure for delayed mortality in coho salmon caught in beach seine fisheries (Raby et al. 2012). RAMP scores indicated sublethal effects resulting from the A treatment but not from the I treatment. Thus, either RAMP may not capture sublethal effects from injuries, even though fish were clearly stressed (elevated plasma lactate and cortisol), or the I treatment used here was not severe enough to impair reflexes. Further research investigating a large range of physical injury might be useful in resolving this issue. Until this is done, we believe it is unwise to rely solely on a RAMP score for predicting delayed mortality of injured migrating adult sockeye salmon. Previous studies show that RAMP scores are positively correlated with intensity of capture stressors (e.g., Davis 2005, 2007; Davis and Ottmar 2006; Humborstad et al. 2009; Raby et al. 2012), but none considered the potential linkage between RAMP and physical injury. Nonetheless, wounds inflicted in fish during capture, which can be highly variable, are a major source of mortality for discards and escapees (Trumble et al. 2000; Suuronen et al. 2005). In the interim, quantitative indexes for physical injuries in fishes have been developed and used in field settings such as visual assessments (e.g., Trumble et al. 2000; Davis 2005; Baker and Schindler 2009) or use of forensic techniques (e.g., fluorescein) to detect nonmacroscopic injuries (Noga and Udomkusonsri 2002; Davis and Ottmar 2006; Colotelo et al. 2009) and might be useful to include when predicting mortality.”

Clearly, further research and validation is needed to establish relationships between RAMP, injury, vitality, and mortality.  As suggested for fish that show barotrauma symptoms, it may be appropriate to consider the inclusion of scoring for presence or absence of injury types in combination with reflex impairment. The effects of injury on reflex impairment differ among species, as shown for fish (Davis and Ottmar, 2006) and crabs (Stoner et al. 2008). Also, at lower levels of stress in some species, reflex impairment may not occur, indicating that the animals are responding to stress in an adaptive manner. Consideration and inclusion of injury in RAMP is important because of it's potential relationship with delayed onset of disease associated with tissue exposure to pathogens.

Reflex impairment can measure sublethal effects of temperature, hypoxia, and injury

Epaulette shark, Aquarium

The epaulette shark is a tropical reef shark that can live in an environment with cyclical periods of low oxygen concentration and high temperature (Wise et al. 1998). After exposure to hypoxic conditions, epaulette shark reflex actions were tested, including righting, response to touch, rhythmic gill movements, rhythmic swimming, and movement of diagonally opposed fins for locomotion. Reflex actions were not impaired by hypoxia, suggesting that animals were adapted to reef conditions and would not show maladaptive responses to hypoxia. Subsequently, this species was found to have important adaptations to hypoxia (Soderstrom et al. 1999, Hickey et al. 2012).

Animal responses to stressors can be adaptive or maladaptive. When animals are exposed to stressor types or intensities for which they are not adapted, stress responses can be too much of a good thing. Adaptive responses help the animal move away from and avoid stressors to return to normal unstressed behavior and homeostatic states. Maladaptive responses cause the animal to become more stressed and eventually can result in morbidity and mortality. 

Adaptive and maladaptive responses of crustaceans to stressors and stress, Stoner 2012

Reflex impairment is an ecologically relevant measure of vitality loss, fitness, and the spectrum of adaptive and maladaptive responses to stressors. Measurement and summation of a suite of reflex actions (as in the case of RAMP or barotrauma modified RAMP) is a powerful means for testing experimental hypotheses about the effects of stressors on animals. For aquatic animals, temperature (elevated or depressed), hypoxia, and injury are important master variables associated with stress induction (Davis 2002, Suuronen 2005, Gale et al. 2013).

The key to testing for vitality, stress, and fitness using reflex actions or barotrauma symptoms is to derive an emergent index for reflex impairment by summing presence or absence of many individual actions or symptoms. The resultant emergent impairment index (RAMP) is a powerful synthesis of the many mechanistic systems (behavior, physiology, neural, muscle, organ) that have been identified in a whole animal. RAMP expresses, with quantitative data, the resultant vitality and health state of the animal.

Some examples of relationships between reflex impairment and temperature, hypoxia, and injury are illustrated below. In these examples reflex impairment was significantly different among stressor treatments and was related to the severity of stressors.

Reflex impairment in crabs exposed to low temperature, Stoner 2009

Reflex impairment in Atlantic cod exposed to hypoxia, Humborstad et al. 2009

Reflex impairment in coho salmon exposed to hypoxia, Raby et al. 2012

Reflex impairment in sockeye salmon exposed to injury from different gear types, Donaldson et al. 2012

Reflex impairment in blue rockfish exposed to injury from barotrauma, Hannah et al. 2008

Barotrauma, RAMP, and discard survival

Yelloweye rockfish with barotrauma, OSU

Capture of fishes with physoclist gas (swim) bladders can result in impairment by barotrauma. Symptoms of barotrauma offer additional measures of impairment that can be used as predictors for survival and mortality when combined with reflex impairment to calculate RAMP (Diamond and Campbell 2009, Campbell et al. 2010). 

Calculation of RAMP modified for barotrauma is as follows. A reflex action is scored not impaired (0) when strong or easily observed, and scored impaired (1) when not present, weak, or there is a question about being present. A barotrauma symptom is scored absent (0) when not present or there is a question about presence, and present (1) when easily observed. Barotrauma and reflex impairment scores for an individual animal are then summed and divided by the total observable impairments possible to calculate proportion impairment (RAMP score). 

  

Snapper (Lutjanus sp.) and rockfish (Sebastes sp.) often show many of the barotrauma symptoms listed above in the table. Pressure from an expanded gas bladder results in many forms of swelling, distention, and eversion. Severely impaired discarded fish have difficulty leaving the water surface and returning to depth.  These fish have low survival, which can be improved with recompression devices designed to lower fish to capture depths, where healing and recovery can occur.

NOAA

Behavioral observations of rockfish with barotrauma have shown that fish trapped on the surface have a higher incidence of reflex impairment when caged at depth for recompression. Survival during healing and recovery will be dependent on return to normal behavior, e.g. avoidance of predators, return to feeding, and appropriate habitat choice.

Blue rockfish impairment after barotrauma, Hannah et al. 2008

In a tagging study, rockfish recompressed in a cage and released were shown to recover from barotrauma and resume normal behavior in the sea (Hannah and Rankin 2011).

Copper rockfish movement recovery after barotrauma, Hannah and Rankin 2011

Gadid species such as Pacific cod and Atlantic cod do not show overt symptoms of barotrauma when captured at depth. Instead their gas bladder ruptures and relieves the potential pressure of an expanded gas bladder (Nichol and Chilton 2006, Midling et al. 2012). Discards of these species can more easily descend to depth after release, with healing and recovery occurring within short periods of time.

Pacific cod movement recovery after barotrauma, Nichol and Chilton 2006

Pearl perch is another example of a physoclist species that does not exhibit overt symptoms of barotrauma during capture (Campbell et al. 2014). As angled fish approach the surface, gas is released through rupture.  For this species, hook location and type were found to be important predictors of discard survival.

Pearl perch, Dave Harasti

Assumptions for use of RAMP

Loggerhead sea turtle escaping trawl, NOAA

Here is a list of key assumptions for the use of RAMP. The list is probably not exhaustive and can be added to as new perspectives and research warrant. These assumptions have been experimentally tested and validated to various degrees by peer-reviewed published research. Further validation is useful and helps to better define possible error terms in RAMP curves. Healthy, control animals are assumed to have a full complement of reflex actions present. See choices for reflex action testing. 

Vitality is inversely related to reflex impairment. Animal vitality is an abstract concept for which we have strong intuitive notions related to observing absence of injury and presence of behavior, including activity and responsiveness. Reflex actions are fixed response patterns to stimuli that clearly reflect internal state without confounding factors. By using reflex actions to quantitatively measure vitality, the confounding effects of volitional behavior and motivation that are often more related to external conditions can be eliminated. Also animals may not be injured, yet show reflex impairment and reduced vitality associated with other factors (e.g., temperature, exhaustion, hypoxia, and xenobiotics).

Reflex impairment is directly related to stressor types and intensities.  Stressors have been shown to induce reflex impairment, interpreted as symptoms of stress. Therefore reflex impairment is a useful measure of stress. Reflex impairment integrates the effects of stress in whole animal responses that are ecologically meaningful for vitality and fitness outcomes. An impaired animal can have morbidity or decreased predator avoidance, feeding, sheltering, migrating, and reproducing.

Reflex impairment occurs immediately after exposure to stressors. Time course studies for several species have shown immediate impairment after exposure to stressors.  Animals with lower levels of stress can then recover full reflex actions hours to days after exposure to stressors. Reflex actions are sensitive measures of sublethal acute and chronic stress as well as predictors of delayed mortality.    

RAMP curve is different for each species and related to stressor sensitivity. Each species has reflex responses that are evolved for habitat types in which they occur. Differences in reflex types and responsiveness among species are apparent in body types, predator avoidance, habitat choices, and feeding strategies. Some species are easily injured and reflex impaired, while others resist injury or are relatively insensitive to environmental insults (e.g., temperature, hypoxia, and hydrostatic pressure).

RAMP curve used for a species is experimentally derived by inclusion of appropriate types of stressors and animal sizes, ages, and sex. RAMP curves must be derived from reflex impairment observed in animals experimentally exposed to combinations of stressors present in systems of interest. Also animals representing size, age, and sex of interest should be included in impairment experiments. The experiments should result in animals with reflex impairment that ranges from 0 to 100%, with accompanying mortality. The curve must include the complete range of impairment and mortality to avoid extrapolation beyond available data. 

RAMP curve is stable for a species and comprehensive experimentally tested conditions. The stable RAMP curve, with defined conditions of reflex types and testing, can be used among widely different situations for measuring animal vitality, survival, and delayed mortality. Exceptions have been noted for larvae or juveniles with ontogenetically delayed development of reflex actions and spawning anadromous adults which show altered sensitivity to stressors.

Reflex actions in RAMP are given equal weighting rather than weighted differently. Reflex impairment used in a RAMP curve is the result of summing several reflex actions. This approach views the whole animal as the important entity of vitality and fitness. Different reflex actions may be affected by different stressor types. In stressor systems of interest, there are relatively unlimited sets of stressor combinations. Therefore, no a priori expectations of importance for specific reflex actions are made and all measured reflex actions can be equally important. However, the order of reflex action impairment relative to stressor intensity can give valuable information about species sensitivity and associated life history characteristics. 

Observers are assumed to objectively score presence or absence of reflex action in a replicable manner. This assumption is satisfied by using the “rule of doubt”. If any doubt exists about the presence of a reflex action, the action is scored as absent. If the reflex action is present without a doubt, it is scored as present. Further controlled comparisons of reflex scoring among observers is warranted to better define possible observer error terms.

RAMP mortality and survival predictions are dependent on the accuracy of captive holding, tagging, and biotelemetry experiments. To calculate RAMP curves, animals are observed for delayed mortality after initial exposure to experimental stressors. Mortality observed with captive holding is simply related to initial stress, assuming that holding conditions are not stressful.  Mortality observed with tagging or biotelemetry includes sources other than initial stress (e.g., additional stressors, predation, disease, and food limitation)(Thorsteinsson 2002).

Sedna, mother of all sea creatures, K. Sagiatok