Written by Dr Kelly Jones, Research Fellow, Research Centre for Palliative Care, Death & Dying

Over the past five decades, palliative care has changed from helping patients at the end of life into a highly dedicated service focused on delivering supportive care to patients with life-limiting illnesses throughout the disease trajectory. (1) To date there is no measure that has been developed to standardise the evaluation of end of life as it relates to both individual and environmental literacy in a population. This blog describes the development process to create a reliable and valid tool to provide a measure of the literacy of end of life.

Scale development can appear daunting. It is a complex mix of both qualitative and statistical techniques and descriptions of scale development and their validation can be filled with what feels like a foreign language and can be resource intensive to build. Scales and indexes are important as they enable exploration of how individual modifiable risk factors may or may not relate to or predict health behaviours and use of services. They enable a standardised and systematic way to undertake evaluation of health interventions and can be used to measure changes over time. Scales are important in public health as they allow for and reflect both the frequency of any impact measured and lay judgments about the severity or effect of what is being measured.

There are very distinct phases and steps in scale development.

Phase 1: Confirmation of domains and framework. Conceptual frameworks are essential to the development of tools that are both valid and clinically or socially useful. They provide the theoretical underpinning to domains of concept which exploration. DeVellis describes scales as essentially a ‘manifestation of latent constructs; they measure behaviours, attitudes, and hypothetical scenarios we expect to exist as a result of our theoretical understanding of the world but cannot assess directly’. (2)

Phase 2: Item and response generation. Item generation can be done both deductively and inductively. Deductive approaches can be done by reviewing current scales and questionnaires which asses the domains selected and gaps for further item generation identified and inductively, whereby qualitative data exploration leads to generation of items (questions) through exploring the narratives of individuals own journeys and experiences. These can then be used to develop or refine a list of unambiguous items which ‘capture the lived experiences of the phenomenon by target population’ and are mapped against domains.

Phase 3: Scale construction. Steps in scale construction include pre-testing proposed items, administering the survey, reducing the number of items, and understanding how many factors the scale captures using cognitive testing via interviews with a population sample.

Phase 4: Confirmatory analysis. The number of items, reliability, content validity and consistency of the questionnaire items is assessed on a population sample. Scale construction relative to the domains is confirmed using exploratory factor analysis to replicate the hypothesised fit of questions to domains and inter-item correlation analysis and examination of item characteristics, scale score and test-retest reliability calculations are performed.
Many scales are used in both community and aged care end of life settings however the greater proportion of them focus on clinical conditions and assessments and not patient or provider experiences of care. New pathways for population level end of life research are emerging and scales are a consistent way to measure a range of social, psychological, and health behaviours and experiences and compare programs and services. As novel research questions emerge, new scales become necessary.

Researchers working with Professor Jennifer Tieman at RePADD are building a scale to measure end of life literacy and join others in the CNHS doing novel research using self-reported outcomes in scale development to determine better care provision and service delivery in end of life settings.

Boateng et.al provide an excellent primer on best practice for developing and validating scales for researchers and describe the ontology and methodology of such an endeavour for those not familiar and wanting to explore the ‘how to‘ of building a valid scale to understand a range of health, social, and behavioural outcomes. (3)

References

1. Al-Ansari AM, Suroor SN, AboSerea SM. et al. Development of palliative care attitude and knowledge (PCAK) questionnaire for physicians in Kuwait. BMC Palliat Care 18, 49 (2019).
2. DeVellis RF. Scale Development: Theory and Application. Los Angeles, CA: Sage Publications (2012).
3. Boateng GO et al. Best Practices for Developing and Validating Scales for Health, Social, and Behavioural Research: A Primer. Frontiers in public health, Vol. 6 149. 11 Jun. 2018.