Review of 2021 Census Topics

The proposal is to collect data documenting the pain experience of all Australians. In particular, how many Australians (inclusive of children and adolescents) experience intrusive chronic/persistent pain?
There was already substantial impetus to include health topics related to smoking and long-term health conditions. The latter might be a single question listing several options from which people may select as many as apply. I believe this is a great opportunity to better capture and track Australian population trends with regard to intrusive chronic pain.
This would enable better-informed public health and clinical policy and service planning. Furthermore, mapping and longitudinally tracking the gap between rates of intrusive chronic pain amongst Australian Indigenous and non-Indigenous populations stands to inform strategies to ‘Close the Gap’. Anecdotally this gap is unconscionably high.
Chronic pain must to be considered in addition to other long-term health conditions because:
• Although it is an essential feature of some long-term conditions (eg. migraine and back pain), pain is not a necessary feature of most. Pain is common but not universal amongst people who self-identify as having other long-term conditions that have appeared in previous ABS surveys eg. diabetes, cardiovascular (heart or blood vessel) disease, respiratory (lung) disease, osteoporosis (bone density problem) or even rheumatoid or osteoarthritis arthritis (that may be well-controlled or based on X-ray changes alone). Therefore, these long-term condition labels cannot be used as surrogate measures of Australian chronic pain experience.
• Pain can be experienced when no other long-term conditions have been diagnosed or self-identified. It is widely considered a condition in its own right.
• Pain may be a modifier of disability, and need for carers, associated with long-term health conditions. It may also be a modifier of relative socio-economic advantage and disadvantage.

Pain, and chronic pain in particular, are topics of current national and global importance.
Chronic pain is estimated to affect almost one in five adult Australians (Blyth 2001, Henderson 2013). For an estimated one in twenty Australians, chronic pain is moderately or severely limiting (Blyth, 2001). There are no Australian population-based data on chronic pain in childhood but international literature and clinical load suggest the prevalence is similar to that in young adults.
Chronic pain carries high economic and human capital costs (Access Economics, 2007, http://fpm.anzca.edu.au/documents/thehighpriceofpainfinal-185.pdf). The total cost of chronic pain in Australia in 2007 was estimated at $34.3 billion, with 58% of costs borne by individuals and families/friends, 27% by government (22% federal; 5% state/territory), and 15% by employers and society.
In 2010, a National Pain Summit in Canberra, attended by a broad range of stakeholders, led to the National Pain Strategy (http://www.painaustralia.org.au/static/uploads/files/national-pain-strategy-2011-wfvjawttsanq.pdf) that articulates the mission, goals, and priority objectives agreed by the Summit. Objectives 5 and 6 address quality improvement, evaluation and research. The establishment of ePPOC, the Electronic Persistent Pain Outcomes Collaboration (https://ahsri.uow.edu.au/eppoc/index.html), was an important step, but only measures the experiences and outcomes of those currently accessing tertiary pain services.
There is a clear need for more granular, population-based data and time-trend data to inform policy and healthcare planning. Such data could inform the public health agenda (Blyth, 2015), or clarify workforce and training needs of regional general practices and Aboriginal Community Controlled Health Services. Together with ePPOC data, Census data could be used to determine whether all Australians have equitable access to tertiary services, irrespective of cultural, socio-economic and remoteness status, and develop policies to address inequities.
These data would enable a national research agenda to improve the lives of Australian children, adolescents and adults living with pain.

References
Blyth FM, March LM, Brnabic AJM, Jorm LR, Williamson M, Cousins MJ. Chronic pain in Australia: a prevalence study. Pain 2001; 89: 127-134.
Blyth FM, Van Der Windt DA, Croft PR. Chronic disabling pain: a significant public health problem. Am J Prev Med 2015; 49(1): 98-101.
Henderson JV, Harrison CM, Britt HC, Bayram CF, Miller GC. Prevalence, causes, severity, impact, and management of chronic pain in Australian general practice patients. Pain Medicine 2013; 14: 1346-1361.
 

Data is needed for small population groups:
o Research indicates that sex and age both affect the prevalence and types of pain experience. As the population ages, patterns can be tracked by repeated census.
o It is already recognised that culturally and linguistically diverse groups require tailored pain assessment and pain management programs. Granular data will help predict needs and optimal locations for services.
o Intrusive pain may be more prevalent amongst Aboriginal and Torres Strait Islander Peoples for many reasons including a higher burden of chronic conditions, injuries and trauma. Elders cite rates of one in three Indigenous Australians experiencing chronic pain. 23% of the children and adolescents seen in our Children’s Complex Pain Service are Aboriginal. Yet we cannot compare Indigenous and non-Indigenous population-based data about prevalence by age, severity, disability, education attainment and socio-economic correlates.
o Many ABS health surveys and research studies have excluded children <16 years. However, the census is an accepted means of gaining parent-proxy data regarding household children.
Data is needed for small geographic areas:
o On a national level, mapping population pain prevalence data versus pain management resources, could lead to better matching of services to needs across each jurisdiction, and predict where new services might be needed as the population grows and ages.
o Government and Local Health Districts will be able to comprehensively describe the local population with pain that they must serve, to identify service needs and gaps.
o Being able to compare ABS statistical area/LGA data with existing health service data would greatly improve service evaluation and planning at the Local Health District level.
o Designing services to meet the needs of Australians living in rural and remote areas requires small population data regarding pain, disability, and access to transport, and/or technology. For instance, designing telehealth systems to support communities may be of little benefit if the community does not access telecommunications.
The value of including pain in the Census is enhanced through combining with other characteristics collected:
o Pain by all ages and genders within the household enables a comprehensive population description, unavailable elsewhere.
o Pain by granular geographic regions is critical for policy and planning as discussed above.
o Associations between SEIFA indices and tertiary pain service access are evident in our own local service data. I have submitted a proposal to include SEIFA indices in the next version of ePPOC to enhance outcomes reports and research. The inter-relationship between social determinants of pain, and the socio-economic consequences of pain, will be an important area of future Australian research.
o Associations between other long-term conditions and pain/pain-related disability will highlight the need for pain management to be included as part of the specific chronic disease management strategy.
o Clustering of chronic pain experiences within families or households is of research interest.

Pain is a widely understood term. Because it is a ‘sensory and emotional experience’ (IASP 1994, 2012) self-report is the gold-standard. Also, because pain is an aversive experience, people generally recall if it is present over time, and approximately how long it has been a problem for them.
Although it is an internal experience, humans communicate their pain experiences verbally and non-verbally to those around them to ascertain help and/or comfort. Generally, adults within a house hold will know if a child, another adult, or an elder experiences chronic pain. Carers are valid proxies when a person cannot self-report pain. It is expected that respondents will easily know the answer to the question(s).
The exact wording of the question can be determined through consensus between ABS, clinical/policy stakeholders and researchers. The following interim questions are based on the IASP field study of the new classification of chronic pain in ICD-11 (Rief, 2010), because this is the most likely international standard for coming decades:
Chronic pain is pain that persists or recurs for longer than 3 months. Do you have chronic pain? Yes / No
If YES, does pain cause significant emotional distress (anxiety, anger/frustration or depressed mood) AND/OR functional disability (interference in daily life activities and reduced participation in social roles)? Yes / No


References
International Association for the Study of Pain (IASP) Classification of Chronic Pain. Originally published 1986, with second edition published in 1994, and updates in 2011 and 2012.
Rief W, Kaasa S, Jensen R, Perrot S, Vlaeyen JWS, Treede R, Vissers KCP. The need to revise pain diagnoses in ICD-11. Pain 2010; 149(2): 169 – 170.
 

Asking about some health topics may be considered intrusive, but most people with pain genuinely appreciate being asked about it.
The question does not ask for intrusive detail. By combining emotional and physical impact in a combined severity-qualifying sub-question, people are not required to explicitly disclose emotional impact. This is particularly important for those who require a third part to assist them in completion or submission of Census forms.
Although people with chronic pain may be stigmatized in some settings (eg. workplace, litigation) it is unlikely that the Census would be viewed as stigmatizing. Respondents are likely to be willing and able to answer accurately.
These questions or similar have been used in state-wide population surveys within Australia, the first instance being the 1997 NSW Adult Health Survey conducted by the then NSW Department of Health (Blyth, 2001), and others including the South Australian Omnibus Survey (WA Department of Health).

References
Blyth FM, March LM, Brnabic AJM, Jorm LR, Williamson M, Cousins MJ. Chronic pain in Australia: a prevalence study. Pain 2001; 89: 127-134.
WA Department of Health and the SA Department of Human Services for the ABS. Results of Field Testing Three Modules of Core Health Questions for Inclusion in a Nation-Wide Preventable Chronic Disease and Behavioural Risk Factor Survey Module Manual. (undated) https://health.adelaide.edu.au/pros/docs/reports/report_health_survey.pdf

As outlined in Assessment Criteria 3, I propose a simple opening question with dichotomous answer that will be easy to code and process. It will not significantly add to the time or cost burdens associated with the Census.
If respondents answer ‘No’ they could be directed to skip to the next topic.
If respondents answer ‘Yes’ they would complete one severity-qualifying sub-question. Although it is tempting to recommend an ordinal scale here, the dichotomous response type is simple, will identify the proportion with intrusive chronic pain, and therefore satisfies the main purpose.
Alternatively, the ordinal scale for pain interference used in the NHS 2014-15 could be considered – refer to http://www.abs.gov.au/ausstats/abs@.nsf/Lookup/by%20Subject/4363.0~2014-15~Main%20Features~Bodily%20pain~33.

The topic will remain relevant in the future. The number of Australians with chronic pain is predicted to increase over coming decades as our population ages and as it expands numerically (Access Economics, 2007 http://fpm.anzca.edu.au/documents/thehighpriceofpainfinal-185.pdf). As the burden of disease due to chronic non-communicable diseases continues to grow, the contribution of chronic pain to long-term condition disability and costs will become increasingly important.
To date, prevalence and cost predictions have been derived from rates of chronic pain in regional/state population samples being extrapolated to the known Australian population at risk. Policy and service planning requires whole population and time series data to ensure public health and service planning remains responsive to real-world and regional changes.

In Australia, identifiable and re-identifiable health information collected for clinical and administrative purposes is protected by privacy legislation. Any information submitted to disease or device registries requires ethics approval and informed consent.
We have examined ABS surveys and other data sources including:
1) National Health Survey (NHS) 2014-15:
o Module 7: Mental Well-Being includes questions about the frequency and severity of bodily or physical pain in the last 4 weeks are under (PAIN_Q01 and PAIN_Q02) http://www.abs.gov.au/ausstats/abs@.nsf/Lookup/by%20Subject/4363.0~2014-15~Main%20Features~Bodily%20pain~33
o Module 8: Disability also includes questions related to pain, in particular, 2. Chronic or recurring pain [that has lasted, or is likely to last, for six months or more] (first appears at DIS_Q02). This leads into sub-questions about the primacy of the various conditions, their effects on everyday activities, and their association with needing care. Also in Module 8, but listed separately, is 18. Any condition that restricts physical activity or physical work (e.g. back problems, migraines).
o Despite having a large sample of some 19,000 people across urban, rural and remote areas it only private residences were sampled and within them, only one adult plus one child (if appropriate) were included to allow comprehensive data collection. Many items including those addressing pain were only reported for those >15 years of age.
o Data are not searchable at granular geographic levels.
2) Australian Aboriginal and Torres Strait Islander Health Survey (AATSIHS) 2012-13:
o Long-term health conditions examined included asthma, ear diseases and hearing problems, diabetes, heart and circulatory diseases, and kidney disease.
o Chronic pain was not included per se.
o Although also a large population-based sample, only one adult and one child (as appropriate) were selected and there was differential data collection from those who lived in remote areas compared with others.
3) Electronic Persistent Pain Outcomes Collaboration 2012+:
o Only persons referred to a tertiary chronic pain service are eligible to participate.
o This data source is valuable but only for the small subset of people with chronic pain who are selected, due to severity, complexity, advantage, advocacy, or proximity, to attend a pain service.
o It is not a substitute for Census level data.

Letters of support for this proposal may.