Background: Emotion regulation has been established as an important concept in mental health research across a range of different diagnoses. There are numerous questionnaires used to measure emotion regulation but only one (the Perth Emotion Regulation Competency Inventory - PERCI) is based on the most recent and widely accepted model of emotion regulation (Gross' extended process model of emotion regulation). This recently developed measure has not yet been extensively used or psychometrically tested in clinical populations. However, it may be more theoretically and psychometrically sound than other measures widely used in the research literature to date.

Methods/Design: An online survey including this new measure with other relevant questionnaires will sample non-clinical and two specific clinical populations in order to explore the reliability, validity and utility of this measure.

Discussion: This will inform the ways in which emotion regulation competency is measured in future research and clinical practice.

Despite the term 'emotion' being commonly used in day-to-day language, within the research literature on emotion regulation (both within the health field and more broadly) there is great variety in the terminology and scales of measurement used. So much so that Buck describes the field as "conceptual and definitional chaos" (Buck 1990, p.330). However Gross has recently renewed his extended process model of emotion regulation and this is now a widely established model although used more frequently in cognition research than in health research (see Gross 2015 for a full description of the model)

In 1998 Gross described emotion regulation as "shaping which emotions one has, when one has them, and how one experiences or expresses those emotions" (Gross 1998). One of the key points which this model suggests is that each of the 5 steps offer potential for emotion regulation and that there is no limit as to the number or type of activities which might serve this purpose.

It is this ability to regulate emotions (i.e. a person's capacity to successfully modify the trajectory of their emotions), which has been established as an important but complex concept in health research. A number of systematic reviews have explored what is known about emotion regulation across a variety of mental health diagnoses (e.g. Sloan et al 2017, Hu et al 2014). Fernandez et al 2014 argue that it is a key transdiagnostic concept using the Research Domain Criteria framework.

As poor emotion regulation has therefore been implicated in the onset and maintenance of a range of mental health conditions, it has subsequently been the target of a number of psychological interventions. Given that affective instability and reactivity, and the use of problematic regulation strategies, are some of the core criteria for EUPD, many of the treatment approaches for this clinical group address emotion regulation. It remains unclear what the mediating or moderating role of emotion regulation is in influencing treatment outcome. A comprehensive, theoretically and psychometrically sound instrument to measure emotion regulation is therefore vital for the development or adaptation of these types of interventions and the subsequent research examining mechanisms of action.

There are numerous self-report questionnaires which are used to quantify emotion regulation. However many of these have psychometric and/or theoretical weaknesses which could potentially impact on the quality of research and the conclusions drawn from this research. Preece and colleagues (2018) describe and evaluate the existing 13 self-report measures relevant to the concept of emotion regulation. In summary they argue that none of these scales :

1. measure emotion regulation across both negative and positive emotions
2. measure across all four components of emotion regulation
3. clearly distinguish between emotion recognition/description (alexithymia) and emotion regulation.

Preece and colleagues therefore developed the PERCI on the basis of Gross' model in order to overcome these deficits. Preece and colleagues (2018) published research outlining the process of scale development and reported good indices of reliability and validity for this measure in a non-clinical population. They recommend replication of reliability and validity research in clinical populations. Factor structure was examined through the use of a series of confirmatory factor analyses. They found the factor structure to be replicable and consistent with the theoretical basis of the Gross model. In the non-clinical population (N = 1175) the PERCI total score had an excellent internal reliability (Cronbach's alpha α) of 0.95. They also reported excellent internal reliability of the four subscales of the PERCI: Negative emotion regulation (α 0.93), Positive emotion regulation (α 0.92), General facilitating hedonic goals (α 0.94) and Positive containing emotions (α 0.93). Concurrent and criterion validity analyses were also reported and confirmed the hypotheses of the researchers in that higher reporting of emotion regulation difficulties on the PERCI was associated with measures of maladaptive regulation strategies, higher levels of alexithymia, more insecure attachments and higher levels of depression, anxiety and stress.

The current study will be a cross-sectional survey using online self-report questionnaires. The study seeks to test the psychometric properties of the scale in mental health populations, as poor emotion regulation is particularly pertinent yet the psychometric properties of the scale are unknown in this population. It is possible that the factor structure and other psychometric properties of the scale are different between clinical and non-clinical populations and this could have important implications for theory, research and practice.

The clinical group will be separated in to people with low and high 'EUPD' traits to allow comparison. The diagnosis of EUPD is controversial and frequently stigmatised. Some people using mental health services will meet criteria for EUPD but not have been told this directly by clinicians, others may have been told they have EUPD informally or after insufficient assessment. This complexity and heterogeneity has led to the decision not to rely on reported diagnosis but to measure current EUPD symptomatology and allocate people to two clinical groups on the basis of clinical cut-offs.

The population accessing the NHS talking therapies service, Improving Access to Psychological Therapy (IAPT), has been found to have high rates of people meeting diagnostic criteria for EUPD (Hepgel and colleagues 2016) and even higher levels of people with traits of personality disorder. As this is a large and relatively accessible group of people this is the population the researchers will target for recruitment for the clinical group.

The researchers will initially recruit a non-clinical group from NHS staff members as they may be more representative of the general population in terms of broad demographics than a student sample.

Inclusion Criteria:

• All participants will be 18 or older. Non-clinical Group: people who score below clinical cut off on measures of anxiety and depression (scoring <10 on PHQ-9 and <8 on GAD-7) and emotionally unstable personality traits (BSL-23) and are not currently using mental health services.

Clinical Group A: people who are accessing treatment through IAPT services and score above clinical cut off on measures of anxiety and/or depression (>9 on PHQ-9 and/or >7 on GAD-7) but below clinical cut off on a measure of emotionally unstable personality traits (BSL-23) Clinical Group B: people who are accessing treatment through IAPT services and score above clinical cut off on (BSL-23) a measure of emotionally unstable personality traits (irrespective of their scores on the PHQ-9 and GAD-7).

Exclusion Criteria:

• Non-clinical Group: people who have been recruited as part of the non-clinical group but who score above clinical cut off on any of the symptom measures will be excluded rather than transferred to the clinical group as they are non-treatment seeking.

NHS staff for non-clinical group Primary care mental health service users for clinical groups