Considerations for APMS 2022: Suggested content to add

Survey content

The primary purpose of the consultation was to consider the APMS 2022 questionnaire and prioritise topics and questions that should be:

• added
• amended, or
• removed.

These are considered in turn.

The consultation survey asked respondents to choose, from a list of suggested topics, their top five priorities to add to the 2022 APMS. The list of suggested topics was extensive (28 items) and generated by initial discussion with NHS Digital, DHSC, and the Academic Network around the time the 2022 survey was being commissioned (Table 3). The item order was randomised for each respondent, to adjust for the tendency for participants in surveys to select items at the top of a long list.

Eating disorders were last included in the 2007 APMS; and not included in 2014. Over half (57%) of all consultation survey respondents selected eating disorders as a priority topic to add. Of respondents who reported having lived experience of mental health issues, 72% selected eating disorders as a priority topic. In focus groups support for the prioritisation of eating disorders was expressed by all types of stakeholders: representatives from across government and third sector organisations, academics and data users, and people with lived experiences indicated that eating disorders should be included in 2022. Specifically, there was interest in collecting information on:

• a broad range of disordered eating (including, but not only, anorexia nervosa and bulimia nervosa)
• the severity of symptoms

APMS 2007 used the SCOFF questionnaire, a 5-item self-reported questionnaire that picks up on general patterns of disordered eating (Morgan et al. 1999). The measure was also used on the Health Survey for England.

One suggestion was to include the Eating Disorder Examination Questionnaire (EDE-Q). The EDE-Q is a 28-item self-reported questionnaire, designed to assess the range and severity of features associated with a diagnosis of eating disorder using four subscales (restraint, eating concern, shape concern and weight concern) and a global score. It is freely available for non-commercial research use, so it is not necessary to seek permission for use. There is a 12-item short form (EDE-QS) of the EDE-Q which was developed and validated by Gideon et al. (2018). It should be noted that body dysmorphic disorder is distinct from eating disorders and not captured by the SCOFF or EDE-Q.

In the consultation survey multiple respondents suggested using the SCOFF questionnaire, in order to compare data to previous surveys. However, there was general agreement, both in focus group discussions and guided feedback interviews, that the improved measurement that would result from switching from the SCOFF to the 12-item EDE-QS would outweigh the fact that the change in method would mean it was no longer possible to look at temporal change.

Aside from the EDE-Q, survey respondents also suggested the following alternative or additional measures:

• collection of height and weight to calculate Body Mass Index (BMI), in order to compare this to results from the eating disorder responses. It was noted that self-report height and weight is often not accurately reported by survey participants, and the bias varies by gender.
• Eating Disorder-15 (15-item): this scale is usually used in clinical settings, as a session by session measure of eating disorder cognitions and behaviours
• Eating Attitudes Test (26-item): a self-completion screening measure to help determine whether a person might have an eating disorder that needs professional attention
• Eating Disorders Diagnostic Scale (22-item): a self-report scale for people aged 13-65 years old that assesses for anorexia, bulimia, and binge-eating disorder by asking the respondent about body image, eating habits, and compensatory behaviours over the last 3-6 months.
• Body Shape Questionnaire Short-form (8-item): a self-report measure of the body shape preoccupations typical of anorexia and bulimia
• Eating Disorders Quality of Life Scale (40-item): questionnaire designed to measure 12 domains of disordered eating-related quality of life. It is typically used as a treatment outcomes measure

Other than the ascertainment of eating disorder, participants did not request other eating disorder specific questions be added, except for the suggestion to include self-reported height and weight. 

Source: APMS 2022 Consultation survey

While the APMS has always included questions on use of health services and treatment, gaps around health care access, delays and waiting lists were raised by consultation participants. Around half (49%) of all survey respondents, and 61% of respondents with lived experiences of mental health issues, chose this as a top priority to add. Government focus group participants explained that they would benefit from having data on delays to service access to understand how long people are in need without support. Suggestions included

• to ask those with a mental health diagnosis about the time (in weeks) between onset of mental health symptoms, seeking help from a health professional, receiving a diagnosis and receiving treatment
• to ask whether individuals had sought private healthcare for diagnosis or treatment of mental health conditions
• to ask whether people had been refused treatment or rejected by a service; or discharged against their wishes 
• to ask whether people had queried or contested a mental health diagnosis, particularly those diagnosed with personality disorder
• to ask whether individuals had ever been admitted as an inpatient for mental health symptoms, and the severity of symptoms before admission
• to ask if people engage in self-prescribed treatment, for example by obtaining medication through online purchases. Online self-prescribing was chosen by 12% of the online survey participants as a priority topic to add.
• barriers to seeking support for mental health symptoms: e.g. lack of awareness about treatment options and referral pathways, time/delays to securing appointments, not meeting threshold to access support, perceived efficacy of mental health treatment, mental health related stigma, anticipated or experienced discrimination, self-esteem and not feeling entitled to support 
• what additional help people would have liked
• the impact of COVID-19 on healthcare access and mode -for more information read the ‘COVID-19’ section 

Those interested in treatment inequalities also rated this topic highly. Questions about access to support could potentially help with understanding why some groups of people with symptoms were more likely to be in receipt of mental health services than other groups of people with similar symptoms. It was also suggested that coverage of use of social services could be improved.

Given the anticipated profound implications of the COVID-19 pandemic for mental health and for the determinants of mental health (such as socioeconomic context and physical health), there was widespread support for inclusion of questions on people’s experiences of it. A third (37%) of survey respondents were in favour of adding this topic. Similarly, focus group participants from all sectors were in support of a discrete module on COVID-19.

In summary, the proposed COVID-19 topics included:

• infection and severity and persistence of symptoms (long COVID)
• social distancing measures and experience of isolation and loneliness
• health service help-seeking and use and remote/digital healthcare access
• whether individuals were asked to (or chose) to shield during the pandemic and for how long
• parenting and family life during lockdowns
• bereavement during the pandemic
• experience of domestic violence during the pandemic 
• employment status during the pandemic, including whether they were classified as a key worker.

It should be noted that the survey already has questions on violence, bereavement and socioeconomic context (job loss, debt, redundancy). The Office for National Statistics (ONS) Opinions and Lifestyles survey series and ‘Coronavirus and the social impacts on Great Britain’ surveys were suggested as useful sources for harmonised questions. Similarly, the UK birth cohort surveys (National Child Development Study, British Cohort Study 70, Millennium Cohort Study) were also suggested.

While the lack of questions on social media was regarded by a range of participants, including 26% of the survey respondents, as a potential topic to add for APMS, it was also recognised that social media questions could be framed in a number of ways. One recommendation was to draw on NHS Digital’s Mental Health of Children and Young People Survey, which included questions on how survey respondents felt when using social media and whether they spent more time online than intended. Others relatedly noted that questions on screen time were almost impossible to capture meaningfully, and that it was more useful to focus on behaviours and feelings while doing different activities online.

Perinatal mental health and parenting was selected by 26% of survey respondents as a priority topic. Government and academic focus group participants explained that they need to know more about women’s mental health in relation to the perinatal period. They defined this as the time from conception through to the child reaching around age 2 years. They expressed interest in any mental health conditions experienced during this time, even if the issues started before this time. Specific suggestions included:

• the timing of the mental health symptoms developing, and whether the parent had previously had a disorder that could predispose them
• if the parent was admitted to hospital for mental health symptoms
• if the family had got support, fathers as well as mothers; what service they were using and what support parents found useful

Additionally, the following scales were suggested:

• Edinburgh Postnatal Depression Scale (10 items): this scale is intended for use at six to eight weeks after birth, so may not be appropriate for a general population survey like APMS.
• Mother Object Relations Scale – Short Form (14 items): this scale can be used with parents of infants and captures parental feeling towards their infant. Registration is required before use.
• Parental Stress Scale (18 items): a measure of the levels of stress experienced by parents. It captures both positive and negative aspects of parenting

There was recognition that without a sample boost the number of parents with recent experience of pregnancy and early years parenting or who had experienced conditions, such as perinatal psychosis, was likely to be too small for robust analysis.

Government participants made a request for body dysmorphic disorder (BDD) to be included in APMS. The Body Dysmorphic Questionnaire (BDDQ) was suggested as a possible new measure. The BDDQ is a brief (up to 7 items) self-report screening measure for BDD.

Covered in the suggested content to amend section.

Problem gambling was last included in the 2007 APMS, and not included in 2014. Government and third sector focus groups participants explained that the availability of community-based data that spans both gambling behaviours and mental health is limited. They also noted that information about the association between gambling, substance use and debt in the context of mental ill-health is limited but needed for policy and service development. Government participants have an increasing focus on gambling, and so robust data is needed to understand the level of harm to inform treatment pathways and estimate demand for services.

APMS previously used a set of questions that operationalise the ten criteria listed in the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) to assess problem gambling (10-items). The Problem Gambling Severity Index (PGSI) was suggested as an alternative scale. It combines negative consequences from gambling with experience of psychological symptoms. This measure is therefore thought to better identify a range of types of gambling related harm. However, criticisms of the PGSI are that it may be less good at identifying problem gambling. Compared with the DSM-IV, the PGSI produces more conservative estimates of problem gambling.

Switching from DSM to PGSI will also mean that the APMS series will be unable to examine change in prevalence between the 2007 and 2022 surveys. The PGSI has 9-items and was developed by Ferris and Wynne (2001) for use among the general population rather than in a clinical context. It has been used on the Health Survey for England, the Scottish Health Survey and the Welsh Problem Gambling survey

Other alternative scale suggestions included

• Short Gambling Harm Screen (10-items): this scale screens for the presence and degree of harm caused by gambling
• DSM-V (10-items): The updated version of the DSM-IV used in 2007 APMS

See the COVID-19 section.

Self-esteem was highlighted by 20% of the consultation survey respondents. It was not raised as a topic in the open text responses, nor did it feature in the focus groups.

Those interested in addressing the drivers of the mortality gap faced by people with experience of mental health conditions were interested in collecting data on physical activity, as well as sedentary behaviour, diet, smoking and other such factors that contribute to mortality risk (also see P18). Physical activity was a priority topic for 17% of consultation survey respondents. It was acknowledged in discussion with academics that to measure physical activity well, a lengthy module would be required and that given the pressure on space in the survey this was unlikely. However, it was noted that for physical activity and diet in particular proxy questions might exist, for example asking about consumption of fizzy drinks as a proxy for poor diet.

Gaming disorder is emerging as a newly recognised mental health condition, sometimes considered alongside online problem gambling. It was prioritised by 15% of consultation respondents. It was raised as a ‘good to have’ in interviews with government representatives and some academics. No particular measures were recommended, although contacts were suggested that could be followed up.

Personality and positive psychology were not raised much in the focus groups nor in the open text responses from consultation survey respondents.

The idea of the gig economy and the importance of secure work for mental health was raised in several groups, including by DWP, and prioritised by 13% of the survey respondents. Family Resources Survey was recommended as a potential source for harmonised questions on this topic

Access to green – as well as blue – space was raised by some academics and prioritised by 13% of consultation survey respondents. They suggested drawing on the People and Nature Survey, for possible items.

There was interest in collecting information about social prescribing among government focus group participants (and 12% of survey respondents). However, there was concern that survey respondents may not know if they have been offered or taken up this treatment pathway. Specific questions on social prescribing were not suggested.

See the health care: access, delays and waiting lists section.

See the physical activity section.  

See the eating disorders section.

Although not prioritised by many of the consultation survey respondents (5%), capturing the changing nature of nicotine product use was considered important by some policy and academic representatives in focus groups and events. Given the survey already includes some questions on this topic, adapting the existing questions may be sufficient to address data user needs. Some participants flagged concern about the routing of existing questions, wanting these items to be asked of all participants and not just those trying to cut back on smoking.

There was a suggestion to include questions about period health more generally. No particular measure was suggested for this. Other low prevalence disorders not previously covered in the survey series did not get widespread support for addition on the survey.