Measuring the total impact of a health condition
When I think of the ‘cost’ of having a disability, I have to say I usually think about the effect of the disorder alone on the ability to do valued activities – I haven’t usually thought about the impact of the treatment itself on how people live their lives. I came across this paper by Gerald Devins while searching for ways to measure the effectiveness of interventions on the lives of people attending the Pain Management Centre in which I work. (I had been looking for something that could measure ‘interference’ of pain on activity and found a couple of good measures such as the Interference subscale from the Westhaven Yale Multidimensional Pain Inventory and the Interference bank of statements from the PROMIS bank.)
Anyway, Devins introduced me to the concept of ‘illness intrusiveness’ as an underlying construct that identifies both the illness-related limitations AND the treatment-related impact on the ability to engage in activities that are important - and these activities include those that are not directly related to the disease impact itself such as normally rewarding opportunities (like travelling overseas being limited by sitting tolerance, for example). In other words, this model looks at the effect of illness on the disruption of lifestyles, activity and interests – and considers that illness intrusiveness is a ‘fundamental determinant of the psychosocial impact of chronic disabling and life-threatening disease.’ (p 592).
There are a couple of reasons this model struck a chord in me – the first was the reference to valued activities. Illness intrusiveness is thought to interfere with psychological wellbeing by reducing the availability of the usual positive effects of engaging in valued activities (ie it does take more effort to get involved in enjoyable activities when you have pain – even though the activities are fun!) and it also reduces personal control by reducing the ways to obtain positive outcomes and/or reducing the ability to avoid negative ones. In other words choices are restricted.
It also appeals to me because it starts to include not only the disease effects, but also the impact of treatment on how well people feel. I don’t think the effects of treatment are always appreciated – such as the time needed for appointments; the side effects; the social effects of feeling like a ‘patient’; the ups and downs of hope and disappointment as treatments are tried and often fail (especially in chronic pain) – even the stigma of being seen by a psychologist or needing to take antidepressant medication (although I don’t see it as a stigma!). All of these things really affect the sense of how much a health problem ‘gets in the way of’ life.
Devin’s work, along with others, has explored various factors that interact to influence both illness intrusiveness and subjective wellbeing.
From the diagram you can see how this model integrates contextual factors and social/psychological factors (yay!) that can either exacerbate or mitigate the effects of a disease/disorder on the individual. It’s wonderful to see how these factors have been included because it brings home yet again how the biopsychosocial model MUST include the social as well as the psychological and biophysical. *steps off hobby horse quickly*
As well as introducing me to this construct, Devins’ paper also discusses the pretty extensive support for it across a range of disorders and populations. From my understanding, it seems that, as we’d expect
- the more severe the disease, the more intrusive it is
- side effects and treatment schedules are associated with more intrusion
- reducing the disease by various means is associated with reduced intrusion
- the intrusiveness of a disease is associated with a sense of less personal control
- illness intrusiveness also correlates with other, more commonly measured outcomes such as mood, quality of life and so on – and it’s not equivalent to the impact of the disease or treatment alone
- about eight factors that influence psychosocial impact of a disorder don’t themselves directly influence illness intrusiveness – for example, stigma, self concept, age, stage in the family life cycle, role importance, culture and the context of events in whcih disease and treatment occur (need to explore that one more to understand it)
Devins has developed a 13-item scale to measure, through self-report, the intrusiveness of a disorder. It’s been used in 36 different disease groups and with thousands of indviduals (although each disease group varies in number from 42 to 2300). The scale itself has been found to form three subscales – relationships & personal development, intimacy and instrumental.
The paper I’ve referred to discussed psychometric properties of this scale, and it seems to have very good internal reliability and test-retest reliability (.80-.90 alpha coefficient for internal consistency) and test-retest coefficients of .80 and .90. over 9 months, 18 months and one three month test for end stage renal disease. Construct validity, always a tricky one to measure, has been established via converging findings in all the various factors across various conditions including both medical and mental health. The paper also discusses other measures of validity such as discriminant validity, criterion-related validity and so on. It’s always an ongoing process to determine the validity of an abstract construct, so this work will continue.
In terms of the factor structure (ie the three subscales) this has been demonstrated across eight disorders including one for psychiatric outpatients with anxiety disorders, and this also seems to remain over different languages.
So, here we have a measure that seems to measure something quite different from other questionnaires; it’s reliable and validity is promising; it provides consistent subscales – but it’s not easily interpreted at present. Hence one of my reasons for discussing it - Devins readily acknowledges that this measure requires some really sound normative data with large groups or smaller ‘local’ norms. There isn’t a specific standard cut-off for ‘mild’, ‘moderate’, or ‘severe’ intrusion and this will be needed before it is useful clinically, although it has been used to compare the effectiveness of different treatments, for example. Devins also acknowledges that as our cultures have widely divergent ideals about what a ‘good life’ looks like, testing this measure in different cultures and subcultures will be important – for example, in a rural farming area of New Zealand, the intrusiveness of a chronic pain problem could be more intrusive than the same problem in an office worker in an city centre in Australia.
A couple of points I want to make here:
- It’s exciting to come across a new concept to describe how people cope with illhealth – and particularly one that considers more than the individual in isolation
- This assessment measure has already had a good deal of research carried out on it, across many of the health disorders that affect quality of life, and its been translated into different languages
- The psychometric properties are pretty good – it seems reliable, and validity is being established, it also seems to be sensitive to change and measure something quite unique from other measures
- It still needs more work to give us usable information relevant to individuals in New Zealand – it shows some of the complexity involved in developing new measures, and when considering social and contextual factors
- BUT it might be worth considering as part of an outcome measure for pain management programmes and when informing policy makers of the real impact of alternative treatments
BTW the paper actually provides the items of the scale. They include health, diet, work, active recreation (sports), passive recreation (reading, music), financial situation, realtionship with your spouse, sex life, family relations, other social relations, self-expression/self-improvement, religious expression and community/civic involvement.
Devins, G. (2010). Using the Illness Intrusiveness Ratings Scale to understand health-related quality of life in chronic disease Journal of Psychosomatic Research, 68 (6), 591-602 DOI: 10.1016/j.jpsychores.2009.05.006