Dr Umar Toseeb Hello, welcome to the Podcast series for the Association for Child and Adolescent Mental Health, or ACAMH for short. I’m Umar Toseeb, Professor of Psychology. My research focuses on special educational needs and mental health in childhood and adolescence. Today, I’ll be speaking to Dr Oonagh Coleman, author of the paper “Research Review Why Do Prospective and Retrospective Measures of Maltreatment Differ? A Narrative Review,” published in the JCPP. Oonagh, thank you so much for joining me. Dr Oonagh Coleman Hi, thank you for having me. Dr Umar Toseeb Let’s start with introduction. Can you tell us about your professional background and your research interests and your affiliations? Dr Oonagh Coleman I’m Oonagh Coleman. I’m a final year PhD student at King’s, at the Social, Genetic and Developmental Psychiatry Centre, and my PhD, which is supervised by Professor Andrea Danese and Dr Kelly Rose-Clarke, focuses broadly on risk and resilience after childhood maltreatment, with an emphasis on the, kind of, subjective experience of trauma. And this paper, this narrative review, I worked on with both of my Supervisors, Andrea and Kelly, as well as Dr Jessie Baldwin at UCL and Professor Cathy Widom at John Jay College of Criminal Justice in New York, both of whom have published extensively on this topic of agreement between maltreatment measures. And also, Professor Tim Dalgleish at Cambridge, who is a Cognitive Neuroscientist, and so, whose insights into the, kind of, memory related aspects of this research were really helpful. Dr Umar Toseeb Thank you, and before we get into the paper, let’s just start with a definition. What do you mean when you say ‘maltreatment’ in the context of this review? Dr Oonagh Coleman By ‘maltreatment’ we’re referring to any experience that happened before the age of 18, encompassing both physical, sexual and emotional abuse, as well as physical and emotional neglect, perpetrated by an adult. Dr Umar Toseeb And what are the different ways in which maltreatment is measured in research studies? Dr Oonagh Coleman Maltreatment can be assessed in various different ways, but a key, kind of, distinction is between prospective and retrospective measures of maltreatment. So, prospective measures capture experiences as they happen in childhood. Because of this, they tend to rely on external information, for example, from parents or Teachers, or a person who’s knowledgeable about the child’s experience, or in some cases they use official records. So, this could be, for example, CPS child protection reports or also court documents, that are collected during childhood. And then, in contrast to this, retrospective measures are collected later. So, this is typically either in late adolescence or in adulthood, and they use typically, self-report questionnaires, where individuals are asked to reflect back on their experiences in childhood and make a judgment of whether they experience maltreatment. So, yeah, there’s two, kind of, fundamental differences between these measures. First is in terms of the measurement source, so either other informants or self-report, and also measurement timing, so prospective measures happening during childhood, whereas retrospective measures happen later in life. And I guess historically, these two approaches have been assumed to be interchangeable with one another. They both have been thought to measure the same underlying construct, which is the measurement of maltreatment. So, therefore, they would identify the same individuals in a sample as maltreated and non-maltreated. But this, kind of, assumption was systematically tested in a meta-analysis in 2019, led by Jessie Baldwin, who’s a co-author on this paper, and they showed that actually, these two measures identify largely distinct groups of individuals. They looked at 16 studies, that had more than 25,000 individuals, and they found that over 50% of individuals identified by prospective measures of maltreatment did not have a concordant retrospective self-report of that maltreatment experience. And likewise, over 50% of individuals with retrospective self-reports of maltreatment did not have a concordant prospective measure of that maltreatment. And they also calculated the Kappa statistic, which takes into account chance agreement and this had a low value, of .19. So, essentially, the main finding from this meta-analysis is that this assumption of equivalence is actually not correct and these measures identify different groups of individuals. Dr Umar Toseeb Thank you, and then I suppose you talk about this in the introduction of the review, where depending on what measure of maltreatment you have, the relationship with psychopathology differs. Can you tell us a bit about that? Dr Oonagh Coleman So, it’s really well established that experiences of maltreatment in childhood increase risk of psychopathology in adulthood and this has been found in so many studies across a broad range of psychological disorders. But given that these two measures actually identify largely distinct groups, we were interested in seeing whether one of these measures was actually more strongly linked to psychopathology than the other, so whether one was driving the association. So, we conducted a meta-analysis last year, that was published in JAMA Psychiatry, on studies that have considered this research question, so studies that had included both prospective and retrospective measures of maltreatment in the same sample, and looked at the associations between each measure and psychopathology. And we found in this meta-analysis, that included 24 studies, that retrospective measures of maltreatment showed significantly stronger associations with psychopathology than prospective ones. So, I guess together, these findings of first, the low agreement with between measure types and their differing links to psychopathology, kind of, led us to explore why these discrepancies may exist and what they reveal about mechanisms, underlying risk or resilience to trauma-related psychopathology. And we looked and this had not been comprehensively explored or mapped before, so we decided to, kind of, put together this narrative review, integrating multidisciplinary research to systematically identify potential reasons for measurement discrepancy. Dr Umar Toseeb So, just to make sure that I’ve understood this correctly. So, if you ask people, assuming it’s the same source, if you ask people during childhood about their maltreatment and then follow them up and ask ‘em about mental health later on, there’s less of a relationship there than if you ask people in adulthood, for example, to reflect on their childhood maltreatment experiences and… Dr Oonagh Coleman Hmmm. Dr Umar Toseeb …also their current psychopathology, there is a relationship there? So, when you’re thinking back about the maltreatment experiences, that’s associated with psychopathology, but when you ask people at the time of maltreatment, or around the time of maltreatment, that’s not necessarily associated with subsequent psychopathology. Is that what you’re saying? Dr Oonagh Coleman Yeah, pretty much. I guess the, kind of, crucial difference, as well, is that you’re not necessarily asking the individuals themselves at the time, because prospective measures tend to rely on either other informant reports or, kind of, official, kind of, legal categorisation of what maltreatment is. So, you’re not getting that first-person perspective through those prospective measures. You’re getting the, kind of, external perspective of what other people would consider to be maltreatment. And then, these retrospective measures that are asking individuals themselves tend to be much more strongly associated with psychopathology. So, it’s indicating that it’s, kind of – the link with psychopathology may be more driven by individuals’, like, appraisals and memories of the maltreatment experience, more so than perhaps a, kind of, objective categorisation of what maltreatment may be. Dr Umar Toseeb And you’ve done a really good job in the paper of breaking up some of the reasons and give – providing us a structure for why there are inconsistencies in the reports of maltreatment, and you talk about measurement, memory and motivation. So, let’s go through those individually. Let’s start with ‘measurement’. Dr Oonagh Coleman And also, I would just say that while we separate these three mechanisms into these different sections, they are very much inter-related and, kind of, not mutually exclusive. But yeah, starting with ‘measurement’, there’s several methodological factors that may be contributing to low agreement between prospective and retrospective measures, which are really important to consider. The first one that we discuss is that the measures, and particularly retrospective self-report measures, don’t always show strong test-retest reliability. So, studies have found that even if you’re asking the same individuals about the maltreatment at two different timepoints, using the same methodology, their responses can change over time. And this is important because it adds error variants and likely contributes to some of this low agreement between these two different measures. But beyond this, there’s also multiple, kind of, systematic differences between these two measures that may be contributing to low agreement. So, one is the differences in definition, so the two measures may differ systematically in how they define maltreatment. So, for example, prospective measures and particularly ones that use official records, will follow a really strict criteria used to trigger either social care or Police actions, and this is really different to retrospective self-reports that are designed for research purposes that may capture a much broader range of experiences and also be more open to interpretation. Following on from that, also detection accuracy will differ between these two measures. So, again, these official prospective records tend to have high measurement specificity. So, that means that they are really accurate in confirming true cases of maltreatment because of the, kind of, rigorous process of substantiation that’s used. That same process of substantiation means that they have high false-negative rates or low measurement sensitivity, because many cases will never, kind of, come to official attention. And in contrast to this, retrospective self-reports that use these, kind of, simple questionnaires will capture a much broader range of experiences, including cases that will be missed by prospective measures. So, they’re more sensitive, but again, they’re more open to interpretation and potential sources of bias, which may lead to lower specificity. Another difference is in terms of sources of information. So, prospective measures rely, again, on these external informants, while retrospective measures depend on self-report, and this is important because caregivers and professionals in these prospective measures may not always be aware of a child’s experience. And again, differences in perspective, interpretation or personal definitions of maltreatment will exist between different informants. And we see this, actually studies have found, kind of, similar discrepancies across different informants, so comparing self-versus-parent reports for a range of different childhood measures. So, for example, parenting quality, different environmental exposures, like community violence and even psychopathology assessments. And then, the final, kind of, systematic difference that may exist between these measures, or at least the final one that we discuss, there’s probably many more, is that sometimes prospective measures will cover a different reporting period to retrospective assessments. So, there’s some studies that we looked at where the prospective measure would cover experiences between zero and 12 and then, the retrospective assessments would ask, kind of, about experiences broadly throughout childhood. So, this could be anything up to 18, and this mismatch in timeframes would obviously introduce inconsistencies in classification. So, I guess these systematic differences exist, but they can’t fully explain these measurement inconsistencies, ‘cause even when studies controlled for all of these factors, they still found considerable discrepancies. Dr Umar Toseeb So, I just want to pick up on two of the points that you’ve made. The first one is around test-retest reliability. Dr Oonagh Coleman Hmmm. Dr Umar Toseeb So, I know that in lots of different measures, not necessarily maltreatment, but you could look at, like, mental health and wellbeing and whatever, and test to retest reliabilities is often an issue. Like, people report different things at different times, even if those timepoints aren’t that distant. But I… Dr Oonagh Coleman Hmmm. Dr Umar Toseeb …with maltreatment, it’s interesting because even as an adult, when you reflect and – back on childhood, I would’ve thought it would’ve been more black and white than it is with other measures, as in, like, it either happened or it didn’t happen. Why would there be low test-retest reliability when it comes to something like that, which is an experience or not an experience? Like, I think, like, with wellbeing, for example, you might have low test-retest reliability between, like, a week or two weeks because it’s a subjective interpretation of how you’re feeling and, like, you know, that might differ. Dr Oonagh Coleman Hmmm. Dr Umar Toseeb But with an experience that’s happened already, why would there be low test-retest reliability? Dr Oonagh Coleman Yeah, that’s a really good question, and I guess some of this will, kind of, be covered in terms of what we’ll go onto discuss, but there’s many reasons why memories can, kind of, shift over time. Due to processes of reappraisal, due to new life experiences, new perspectives on an experience, potentially memory biases coming into play, also, different motivations at different times. Someone may feel comfortable in one context disclosing maltreatment, but that again, may shift over time. Yeah, it’s a really interesting question and it's actually – we’re doing a lot of work on that at the moment. I’m currently working on a meta-analysis of studies that have looked at test-retest reliability of self-reports of maltreatment over time, and we’re doing a number of different moderation analyses to see factors that may moderate that association, including – I mean, a, kind of, big one is the potential for psychopathology to bias memory at one or other timepoint. But yeah, there’s multiple reasons that exist, which hopefully, we’ll cover some of them today. Dr Umar Toseeb Thank you, and then, the other one I have is the issue of source bias, and I think you’ve already talked about this, but I wonder whether the reason why there’s a relationship between me reporting on my past experiences of maltreatment and current psychopathology, is because I’m doing both of the reporting, the retrospective one, like there’s somebody else reporting. So, I would guess that you can take any two measures, whatever they might be, not maltreatment and not psychopathology, but if the same person’s reporting on both the measures, you’re more likely to have a relationship between the two than if you have two different people doing the reporting. Dr Oonagh Coleman Yeah, that’s absolutely true. There’s a common source of bias called ‘common source bias’ and the same, kind of, emotional and cognitive processes may be coming into play in terms of how an individual reports on their psychological symptoms and also how they report on experiences during their childhood. And we, in fact, found in our meta-analysis on the association between each of these measures in psychopathology, in one of the moderation analyses we ran, retrospective measures were more strongly associated with psychopathology when psychopathology was self-reported, in comparison to reported by other individuals. So, when there was that common source between self-report for retrospective measures and self-report for psychopathology, there was a stronger association. But there was still an association between retrospective measures in psychopathology when it was reported by others. So, this, kind of, source of bias, while it may explain some of the association, it can’t explain all of it. Dr Umar Toseeb And that’s a really good point to move onto ‘biases in memory'. So, where do these memory biases creep in? Dr Oonagh Coleman Beyond measurement issues, the second key mechanism that we consider that may contribute to low agreement is memory processes, and we’re focusing really on how these processes affect retrospective measures which rely on autobiographical memories of childhood events, just by their nature. We know that memories are not objective records of reality. They’re inherently subjective and they’re shaped by a number of different cognitive and emotional factors, and there are several reasons why memories either may be absent, due to processes related to memory formation or recall, or also why they may, in some way, not align with more objective categorisations or classifications of maltreatment. And these discrepancies may arise because of differences in appraisal or re-appraisal over time, as well as biases and distortions that may influence memory over time. So, in the review we categorise memory processes into two key stages and these were first, memory formation, so this is factors that influence both whether and how an experience was encoded and stored in memory. And then memory recall, so factors that shape how memories are retrieved, reconstructed or, kind of, distorted over time. Starting with memory formation, so this involves, kind of, two key processes, encoding and consolidation, where experiences are, essentially, stored in long-term memory and become resistant to loss, and this process is selective. So, not everything we experience is committed to long-term memory. Retaining all information and memory would completely exceed our cognitive capacity and it wouldn’t be adaptive. So, yeah, I guess that’s the first important point to consider. But when considering why some individuals may not retrospectively report maltreatment, one key factor is, yeah, whether they form these memories in the first place. So, we know, for example, that age at the time of the experience may have an impact. Adults rarely recall experiences before age three, due to a phenomenon known as infantile amnesia, and this may result from either the immaturity of the infant brain, meaning that it’s unable to effectively store memories, or from the way that early experiences are encoded, so that could render them inaccessible for later recall. And some studies have found that older age at the time of maltreatment is linked to more detailed or, kind of, stable recollections of these experiences. But it’s really challenging to isolate the effective age, because maltreatment often occurs over a, kind of, time period rather than at a single moment. So, it’s difficult to, kind of, determine how these age effects may influence recall. And then, the second key factor we discussed in terms of memory formation was sleep, which is really crucial for memory consolidation. And we know that sleep disturbances are really common in maltreated children, so this could impact whether these experiences are actually stored in long-term memory. Beyond these factors that may affect memory formation, whether and how an event is remembered depends on how it was experienced and appraised at the time that it occurred. So, if an event is highly distressing or traumatic, it will be more strongly encoded and increase the likelihood of later recall and retrospective reporting. And the way that individuals initially appraise and interpret their experiences are influenced by a number of different individual differences in cognitive and emotional processing. For example, if a child doesn’t initially appraise an event as harmful, for example due to their developmental stage or social normalisation, which could be deliberately manipulated by the perpetrator, they may fail to encode an experience as maltreatment, making retrospective recall less likely. Research has also found that individual differences in terms of sensitivity to negative experiences exist. So, for example, certain personality traits, like high neuroticism or current psychopathology, may make individuals more likely to perceive experiences as distressing or threatening, again increasing likelihood of later recall. Some studies have found that autistic individuals may perceive a broader range of experiences as traumatic due to heightened sensitivity to, kind of, negative environmental influences, and also we discuss in the review the impact of social environment. So, it’s been found that supportive social environments can buffer against, kind of, negative im – appraisals and reduce the likelihood of an individual experiencing an event as distressing or traumatic. And then, in contrast to this, loneliness is associated with increased perceptions of environmental threat. And we also know that environment can impact appraisal in other ways, for example environments characterised by, kind of, violence or chaos or neglect, may normalise these experiences, making them less likely to be appraised as harmful or unacceptable or as maltreatment. Dr Umar Toseeb So, if we just summarise the section on memory. So, what you’re saying is there are potentially two overarching sources of bias. So, you’ve got the bias at the time of forming the memory and then… Dr Oonagh Coleman Hmmm. Dr Umar Toseeb …the bias at the time of recalling the memory. Do we have any indication of where lots of the bias is coming from? Is it more dominant in one area or the other, or is it just not possible to tell because they’re so linked? Dr Oonagh Coleman Yeah, I guess it’s very difficult to, kind of, untangle the effective recall bias over time, but biases can, kind of, come into play both at formation and at retrieval. And also, kind of, in similar ways, they can work both at appraisal and reappraisal. It is possible for memory formation to happen in the absence of experience, so leading to, kind of, distorted or false memories. These aren’t common. Most people can distinguish between real and imagined events, but examples of these, kind of, distorted memories could be because of source monitoring errors. This is at formation, where individuals, kind of, misattribute memories to thoughts, dreams or suggestions. So, they think that these experience actually happened when they didn’t. And also due to suggestive questioning, so for example, repeated or misleading questions, but these aren’t common, but we do know that certain conditions can, kind of, increase their likelihood. So, for example, they are more common in younger children. Their – younger children are more vulnerable to suggestive questioning and source monitoring errors. Individuals with dissociative sende – tendencies and also source monitoring errors are more likely in individuals with psychotic symptoms or schizophrenia. But more common than false memories being created in the absence of experience is the introduction of biases during retrieval. So, that the way that something is remembered doesn’t necessarily align with how it was originally experienced, which I’ll go onto explain in a bit more detail. But in general, it’s extremely difficult to disentangle the different sources of bias that can shape recall. Dr Umar Toseeb What about recall? So, we’ve talked a lot about memory formation, but what about recall? Dr Oonagh Coleman Yeah, so memory recall is the process of actually, kind of, retrieving these stored memories and predictors of recall can help explain why some individuals may have memories that are more accessible or readily recalled, while others don’t. And there’s several really well established factors, for example time. We know that longer time periods make experiences more prone to forgetting. Rehearsal, so the more that a memory is rehearsed and recalled, the stronger or more enduring it becomes over time, and mood congruency is really important. So, studies show that people tend to recall memories that match their current mood. So, depressed individuals tend to recall more negative memories than non-depressed individuals, which, in turn, strengthens these memories over time. And similar effects have been found in the opposite direction, for mentally healthy individuals and positive memories. But I guess beyond these, kind of, general influences on memory recall, there’s also specific influences on distressing or potentially traumatic memories. So, in some cases, the, like, higher emotional impact of these memories can lead to stronger memory encoding and greater likelihood of recall, but in other individuals, it can trigger avoidance strategies. Examples of these could be dissociation or thought suppression or also, over general memory, that may mean that for some individuals, these memories are less accessible or deliberately avoided. And then, beyond this, as I, kind of, touched on earlier, memory reappraisal is really important to consider. Memory isn’t this passive retrieval process, but it’s, kind of, active and reconstructive and memories can be updated over time through the processes of reconsolidation and retrieval. So, past events can be reinterpreted with new context. Events once perceived as normal may be later recognised as abusive. This could be due to cultural or societal shifts, due to therapy or just exposure to new perspectives. In terms of these biases that may also come into play at recall or reappraisal, psychopathology may bias memory recall. So, current negative mood may lead to negative reappraisals of past experiences, or also, mental health could lead to positively biased memories over time. So, yeah, these, kind of, memory biases could work in either direction, either leading to more negative or more positive recall in relation to how events were originally experienced. Dr Umar Toseeb Thank you, and let’s move onto ‘motivation’. What can you tell us about the influence of motivation in the inconsistencies in the reports of maltreatment? Dr Oonagh Coleman Beyond memory processes, we considered a number of different motivational factors that could influence reporting and contribute to low agreement between measures. And importantly, these motivational factors could influence both retrospective reports and also prospective reports that use informant reports, such as parents. So, we considered internal barriers, for example, feelings of shame and guilt that may develop after a maltreatment experience, that could shape an individual’s concept of themselves and lead to, kind of, either delays in disclosure or complete concealment of these experiences due to these negative self-conscious emotions. And then, as well as this, we considered interpersonal dynamics that could affect disclosure of maltreatment. So, in part, prospective measures that use caregiver reports will rely on these caregivers being aware of the experiences and in some cases, the child may deliberately keep maltreatment a secret, either due to these negative self-conscious emotions or due to blackmail, threats or fear of consequences. As well as this, caregivers may be aware or may be the perpetrators of maltreatment and therefore, may be unwilling to disclose maltreatment due to potential negative consequences. Particularly in – with prospective measures, there’s a threat of mandated reporting if there’s an ongoing, kind of, risk to the child, which could result in the child’s removal, legal repercussions or perpetrator imprisonment, all of which may motivate a caregiver to, kind of, remain silent about that experience. And then, finally, for both prospective and retrospective reports, factors like rapport with an interviewer if they’re using these, kind of, interviewer formats, privacy concerns or social desirability bias may lead to under-reporting, both for self-reports and informant reports. Dr Umar Toseeb And I think it might be helpful just to provide a, like, a lay summary of those three things. So, if you were going to tell somebody who’s not an expert on any of this a one sentence summary of measurement and another one of memory and another one of motivation, what would it be in terms of the inconsistencies? Dr Oonagh Coleman So, in terms of ‘measurement’, this would refer to methodological issues that may reduce agreement between prospective and retrospective measures, so for example, systematic differences between measures. In terms of ‘memory’, this is processes that affect whether an experience is remembered, whether it’s accessible for recall that – at the time of retrospective reporting and how something is remembered, so shaped by appraisals, reappraisals and various sources of bias. And then, in terms of 'motivation', I would say these are factors that affect an individual’s willingness to disclose maltreatment, which may contribute to low agreement between measures. Dr Umar Toseeb Thank you, and what are the implications of your work? And I think that for me, when I was reading this, I was thinking well, I can see definitely why this would impact on my work as a Researcher. Dr Oonagh Coleman Hmmm. Dr Umar Toseeb Is the intended audience for you when you were writing this? Were you thinking this is for Researchers, or are there people in clinical practice or people who work with children and young people who might also benefit from this? Dr Oonagh Coleman Well, we hope that the implications are for both research and then also beyond that, in terms of clinical interventions. We separated the implications into four key areas. The first was in terms of the understanding these different constructs that these two measurements actually capture. So this is a, kind of, research implication. This review and the, kind of, work previously on this topic really indicates that these measures don’t capture the same underlying construct and so, care must be taken to, kind of, select the most appropriate measurement for a specific research question. So, prospective measures are reflecting this third person perspective, shaped by informants’ knowledge, their definitions of maltreatment, their willingness to disclose, or these, kind of, legal standards. And then, in contrast, these retrospective measures are capturing a first person perspective, shaped by appraisals and memory processes. So, yeah, that’s the first, kind of, implication in terms of measurement construct. The second implication is in terms of improving measures and particularly retrospective measures of maltreatment. Again, this is a research implication. Current retrospective measures really primarily focus on a, kind of, basic disclosure of childhood maltreatment and really, this, kind of, review suggests that a much richer, kind of, more nuanced understanding of the subjective experience and recall of maltreatment could be developed. So, in a similar way to, kind of, understanding traumatic memories in the context of PTSD, future research, we, kind of, suggest, could explore how these memories of maltreatment are organised, their intrusiveness, their centrality and stability and the role of, kind of, maladaptive trauma-related cognitions. And importantly, we highlight that the aim of this is not to, kind of, increase agreement with prospective measures, but more to develop a, kind of, more comprehensive understanding of this subjective experience and recall of maltreatment. And then, the third implication is in terms of our understanding of resilience and vulnerability to trauma-related psychopathology. So, we’ve identified in this review a number of characteristics that may influence whether someone retrospectively recalls or appraises an experience as maltreatment. And we know that this recall is associated with stronger associations with psychopathology than prospective reports. So, understanding and directly testing factors that predict measurement disagreement and the presence or absence of recall and how these factors may relate or map onto vulnerability or resilience factors to trauma-related psychopathology could provide insights into, kind of, modifiable mechanisms underlying mental health risk after maltreatment. And then, I guess leading on from that, is the fourth implication, which is around treatment. We know that individuals with maltreatment histories tend to have poorer treatment outcomes across a broad range of mental health disorders in comparison to non-exposed individuals. And if this retrospective recall of maltreatment is a risk factor for a broad range of psychopathology, then perhaps targeting these memory processes and the subjective experiences in therapy may help to prevent or treat maltreatment-related mental health issues. So, we draw the link to evidence-based trauma-focused treatments that are used in treating PTSD, for example trauma-focused cognitive behavioural therapy, which is – already addresses these unhelpful or maladaptive trauma-related appraisals and cognitions and memory processes. These trauma-focused treatments are only typically used when PTSD is the primary concern. So, we suggest that, kind of, future research could explore the possibility of extending and adapting these treatments for other psychiatric disorders linked to maltreatment, beyond PTSD. Dr Umar Toseeb Thank you, and you mentioned that this is part of your PhD, so what else should we look out for and that might be coming up on this topic from you? Dr Oonagh Coleman I’ve been working on both qualitative and quantitative studies looking at factors predicting measurement inconsistency, using data from the E-Risk Longitudinal cohort. The qualitative study has just been published in child abuse and neglect and a quantitative study is currently in progress, but they’re both from different angles, looking at predictors of measurement disagreement. And then, as well as that, what I mentioned earlier was this meta-analysis on test-retest reliability of self-reports of maltreatment and, kind of, moderators of the stability over time, which again, is currently in progress. Dr Umar Toseeb And finally, what’s your take home message for our listeners? Dr Oonagh Coleman Well, first, these measures are not equivalent and retrospective measures have historically been criticised due to their subjective nature and their susceptibility to these various sources of bias, in comparison to prospective measures. But I think rather than dismissing these measures on these grounds, actually understanding more about what they capture around memory and the subjective experience may hold key insights into their stronger links with psychopathology. And by focusing on how individuals experience and remember maltreatment, we may increase our ability to reduce its long-term psychological impact. Dr Umar Toseeb Thank you and I thank you for taking the time to speak to us. For more details, please visit the ACAMH website, www.acamh.org, and Twitter @ACAMH. ACAMH is spelt A-C-A-M-H, and don’t forget to follow us on your preferred streaming platform, let us know if you enjoy the podcast, with a rating or review, and do share with your friends and colleagues.