Preferences and decision making in health care |
 |
| |||
| |||||
 | |||||
ATN Centre for Metabolic FitnessKey Objectives CHERE researchers are investigating the costs and outcomes associated with an intervention aimed at changing behaviour in adolescents at greatest risk of becoming obese adults. Short-term data are available identifying a treatment effect, so cost-effectiveness methods can be used to estimate the resource use associated with the intervention, and then to generate information about the cost effectiveness of CBT for reducing overweight and obesity among adolescents. However, these results need to be extended to information about the cost-effectiveness of preventing overweight and obesity in adults to be relevant for policy makers. The significant methodological issue to be dealt with is the extrapolation of the short term results into long-term meaningful endpoints, such as reduction in morbidity or mortality. It is hoped that this study will identify a suitable method for doing this or identify the reasons why the current approaches to this problem are flawed. Funding source CHERE staff Collaborators 1 RMIT University ATN Centre for Metabolic FitnessKey objectives The prevalence of overweight and obesity is increasing in Australia. The latest results from the Obesity and Lifestyle Study (2000) suggest that approximately 60% of the adult population is overweight or obese. This is 2.5 times higher than in 1980. Similarly the prevalence of overweight and obesity in children and adolescents has risen 2-4 times since 1985. When compared with 1997/1998 data, the overall percentage of overweight or obese adults in NSW in 2004 has risen from 44% to 48.4%. Most notably the proportion of the population who are considered to be obese has risen from 12.8% on 1997/1998 to 16.9% in 2004. To date there has been limited analysis of issues associated with overweight and obesity using data from the NSW Health Survey. While many studies have examined the factors associated with being overweight or obese, few have compared individuals’ perceptions of their weight with more objective measures such as Body Mass Index (BMI). This study uses unit record data from two NSW Health Surveys to identify factors associated with the accuracy of adults’ perceived weight, and whether those factors changed over time. The sample size ranged from 35,005 in 1997/1998 to 13,025 in 2004. Descriptive methods and logistic models using STATA are used to quantify the effects of a number of demographic, socio–economic, behavioural and health–related variables on the accuracy of self–assessed weight. It was found that significant differences exists in body mass perception and accuracy by a number of characteristics, including participants’ current BMI, sex, self assessed health and SES. Males are significantly more likely than females to understate their self–assessed body mass in relation to their BMI. Similarly, respondents with lower socioeconomic status (SES) are more likely than those with high SES and respondents who are overweight or obese are more likely than those who are thin to understate their self assessed body mass. Respondents who record very good or good health status are significantly more likely to over state their body mass. Funding source CHERE staff ATN Centre for Metabolic FitnessKey Objectives Alongside the “Whyalla Community Trial – Managing Obesity and Associated Health Issues”, CHERE researchers have investigated the factors that motivate people to adopt beneficial lifestyle changes. Using Discrete Choice Experiments (DCE) this project has identified preferences for particular interventions, products or programs and the factors affecting uptake among the target population. For the sample as a whole, the emphasis at baseline indicated preference for individually designed exercise, reasonably structured diet programs and preference for high levels of support. A trade-off was apparent in exercise with greater hours taken in conjunction with flexibility and shorter hours with daily exercise schedule. The overall emphasis was one of high structure and support. At 4 months the focus shifted to a clear preference for exercising on their own with no supervision and an aversion for organised group exercise. Support was no longer significant and a strong preference emerged for the intervention diet. The general move was away from structure and supervision and towards more flexibility. Cost realities were also becoming an issue. At 12 months cost became a major factor in preferred program and there was evidence of renewed interest in having higher support. Flexible diets and individual exercise that is not overseen and weekly meal program. Preference for individual exercise over committed group exercise remains. While no firm comparison of gainers and losers can be made because of possible scale differences, there does appear to be a greater tendency on the part of losers to take an active interest in their program and to assess their own needs rather than relying exclusively on others’ directions. Funding source CHERE staff Collaborators 1 Faculty of Business, UTS Adolescents and young adults with a life threatening illness: Preferences for support servicesKey Objectives Life-threatening illnesses in young people are traumatic for patients and their families. Support services can help patients and families deal with various non-medical impacts of diagnosis, disease and treatment. The aim of this study was to determine which types of support are most valued by adolescents and young adults (AYA) with cancer or blood disorders and their families. A discrete choice experiment (DCE) was performed. Separate experiments were conducted with AYA and their guardians. Types of support included in the experiment were: assistance returning to school/work; emotional support for the patient and/or family; financial support; spiritual support; and cultural support. Completed surveys were returned by 83/88 AYA and 78/79 guardians. AYA preferred emotional support for themselves (either by counsellors and/or peers), emotional support for their family, financial support and assistance returning to school/work over services relating to cultural and spiritual needs. Covariate analysis indicated female AYA were more likely than males to prefer emotional support, while males were more likely to prefer assistance returning to work/school and to have an aversion for cultural needs. Guardians preferred emotional support for their dependants and assistance returning to school/work. To a lesser extent, they valued financial and emotional support for themselves. Like AYA, they were indifferent about services relating to cultural and spiritual needs. Providing the types of support services that people prefer should maximise effectiveness. Results from this DCE can inform evidence-based health policy decision making about the types of support services provided for AYA and their families. This study was presented at AHES, Brisbane 2007. Funding source CHERE staff Body mass in AustraliaKey objectives The prevalence of obesity has risen in many developed countries over the last century. In Australia weight levels of men are rising faster than females and males in couples of working age have the highest level of obesity. There are three components to this research which all rely on data from ABS National Health Surveys: 1. We examine misperceptions of body mass, how this has changed over time and provide a profile of those who misperceive their BMI category (Ayyar, Savage and Vu). 2. We survey the literature on food misreporting and investigate the relationship between body mass and food intake in Australia using the linked 1995 National Health Survey and Nutrition Survey and (Jones, Propper, Savage). 3. We pose the question: do exercise and labour force participation of the female partner influence BMI category for working age males and females in couple income units (Gablinger, Jones, Propper, Savage)? Over time there is evidence of increasing body mass for both male and female adults with increasing misperception for the overweight. The findings show that working females have significantly lower body mass than their non-working counterparts, but that partner body mass is unrelated to the labour force behaviour of the female. Exercising is associated with lower weight in males with working partners. Surprisingly we find that higher energy intake is associated with lower BMI. This is consistent with international findings and suggests misreporting of food intake. We investigate methods for addressing misreports. In 2007 the research was presented at the iHEA World Congress in Copenhagen. Two papers are being prepared for journal submission. Funding source CHERE staff Collaborators 1. University of Bristol Validation and calibration of the SF-36 health transition question in the Household, Income and Labour Dynamics in Australia (HILDA) surveyKey Objectives Cross-sectional population surveys depend on retrospective self-report if they are to estimate changes in health status over time. An example is the health transition question (HTQ) from the SF-36 health survey that ask the respondent rate his/her health compared to one year ago. However little has been done to estimate the clinical magnitude of change in health related to responses to health transition questions. Calibrating the response categories of the HTQ against an external measure of known clinical change will help in interpreting the clinical meaning of the HTQ categories and increase its usefulness as a stand alone item in cross-sectional surveys. We are using the HILDA study to obtain some estimates of the size of prospective change in health status on the SF-36 scales for the HTQ and comparing this against the size of prospective change for respondents who have recently developed a new long-term health condition. CHERE Working paper 2007/15 has been produced. An article is being prepared for submission to the Australian and New Zealand Journal of Public Health. Funding source CHERE staff Care and outcomes of care for prostate cancer in New South WalesKey Objectives The NSW Prostate Cancer Care and Outcomes Study (PCOS) is following a group of men with prostate cancer from diagnosis for up to 5 years. CHERE is leading the economic evaluation component of the study which will investigate patient preferences, describe the use of health services and the costs of care. A Discrete Choice Experiment (DCE) has been used to elicit men’s preferences for treatment options, based on the relative tolerability of different side-effects of treatment and how these are offset by expected survival gains. Analyses have been reported in CHERE Working Paper 2006/14. Further analysis will be done in 2008 to prepare papers for publications. Funding source CHERE staff Collaborators 1. School of Population Health and Health Services Research, USyd Choice experiments for complex choices: the case of contraceptivesKey Objectives The range and complexity of contraceptive choices introduced over the past 5 years pose a significant challenge for GPs to provide information and recommendations to women, in the limited consultation time available. No detailed data are available about the factors which will influence a woman’s choice of method or the way GPs will deal with these issues. This research will quantify the trade-offs that women make in assessing different contraceptive alternatives, provides information about how they will choose under different circumstances, and seeks to predict uptake of new products. These data are necessary to inform GPs in providing appropriate advice and recommendations to women. During 2007, the pilot study and the main survey for the women’s choice experiment was conducted. 528 women participated in the experiment, and analysis of the results is underway. Two experiments for the GP survey are currently in the design phase. Funding source CHERE staff Collaborators 1 School of Economics, UNSW Decisions about cervical screening: What influences women and providers?Key Objectives Discrete choice experiment surveys were used to collect stated preference data from women in the target population and GPs. The findings showed women were more likely to have a Pap test when the GP was female, she/he was their regular GP, if they were due or overdue for a screening test and when the GP recommended they have a Pap test. They were discouraged by cost and any decrease in the accuracy of the test, but were not influenced by the recommended screening interval, or whether the GP received an incentive payment. In making a recommendation that a woman have a Pap test, GPs took into account whether she was due or overdue for a test and any cost to the patient. They were less likely to recommend a screening test if the reason for consultation was a serious health problem or if the woman was aged 70 years or older. This ongoing research allows us to predict the uptake of new technologies and provide greater understanding of how variables such as price, test accuracy, and screening interval affect participation in the screening program. A CHERE Working Paper has been produced and the research was published in Health Economics in 2006. Further papers are currently under review. During 2007 a follow-up survey was conducted. This study was conducted after the introduction of the human papilloma virus vaccine in Australia in April 2008. Analysis of these data and comparison with the previous study is currently underway. Funding source CHERE staff Collaborators 1 School of Economics, UNSW Evaluation of asthma medicationsKey Objectives Asthma is a significant health problem in Australia affecting 16% of children and 12% of adults. Poor patient compliance is a major challenge for physicians. Understanding what influences patients’ preferences for medication is important. Discrete choice experiments (DCE) were used to elicit stated preferences for products and programs on a range of topics including asthma management. This study was embedded within a multi-centre, cross-over, randomised controlled trial of three preventive asthma medications. Four DCE surveys were completed: one at entry to the trial, and one at the completion of each of the three treatment phases. This design allowed for testing of the stability of preferences over time, and the response of patients as their experience of medication and its outcome changed. All the DCE surveys have been analysed. In 2007, results from the first DCE were published in Respirology and results from the second DCE were published in Health Economics. The findings show patients preferred medications which enabled them to participate in usual daily and sporting activities and gave them minimal symptoms experience. In terms of medication side-effects tremors, palpitations, nervousness and headache were considered worse than oral thrush, or occasional hoarseness of speech. Preliminary results also show that the preferences were stable over the DCE surveys; in another word, patients were making similar trade-offs across the same attributes and levels. The analysis looking at stability of preferences and effect of medication experience on preferences will be ongoing in 2008. Funding source CHERE staff Collaborators 1. School of Economics, UNSW Incorporating the contribution of informal carers into the economic evaluation of community palliative careKey Objectives The provision of care at home for people with a terminal illness necessitates a substantial care input from family and friends (informal care). This study aims to investigate the carers’ preferences for support with providing this care and uses a discrete choice experiment to identify: • the support services carers prefer, and • whether carers prefer to receive support services or financial assistance? The analysis of the carers’ preferences for different types of support services found that while all carers valued nursing services, preferences for other services varied over the palliative process. Domestic help, transport and coordination of treatment and information sharing were important at the earlier phase, while help with personal care and respite became the priority as the care recipient’s condition deteriorated. Analysis of the carers’ preferences regarding financial assistance is on-going. The results so far have been presented at three conferences and published in a CHERE working paper. Funding source CHERE staff Collaborators 1. School of Economics, UNSW The Trade-Off Between Equity and Efficiency: A Discrete Choice Experiment to Elicit Population OpinionKey Objectives Standard economic evaluation considers outcomes to be of equal value irrespective of who they accrue to. However, it is plausible that, under certain circumstances, society may decide that an outcome in one group (such as the more disadvantaged) may be relatively of greater importance. While people such as Alan Williams have highlighted the possibility of weighting outcomes in economic evaluation according to societal preferences using equity weights, this has not been undertaken on a large scale. Discrete choice experiments may be a useful way forward towards doing this. They allow investigation of complicated preferences, estimating both the effect of changing individual characteristics, but also the way that characteristics interact in the decision-making process. Funding from the Faculty of Business has provided the opportunity to undertake a pilot investigating the issue. The survey will be developed through use of small convenient samples, and then piloted using an online representative sample of the Australian population. This will consider whether baseline differences in life expectancy, social group, gender or smoking status affect the valuation of outcomes. The advantage of using an online panel is that a larger sample can be reached than through more labour-intensive methods, allowing more complicated designs. Funding source CHERE staff Multi-Attribute Utility/Discrete Choice Experiments (MAUDcE)Key Objectives The primary aims of this research are: • To develop discrete choice experiment (DCE) methods to model and measure community trade-offs for health states (‘utility scores’) for use in calculation of quality adjusted life years (QALYs) in economic evaluation; and • To provide utility weights that can be used in economic evaluation in the Australian context, and that can be compared with utility weights from other countries During 2007 the major focus of the research has been on development and implementation of the discrete choice experiment approach for application to valuation of health states. The discrete choice experiment component of the project is now fully designed and has been piloted. A key component of the pilot project was to undertake a sub-study investigating the impact of alternative experimental designs. The pilot project was conducted on a sample of 300 people. Three alternative approaches to selection of health states for inclusion in the main study were tested, and the preferred approach was identified. This sub-study forms the basis of a separate methodological paper which is currently being prepared for submission to a peer-reviewed journal. Significant progress was made towards the time tradeoff and standard gamble components of the project. For comparability in method of administration, these will be implemented via a computer- based questionnaire. The development of the computer based version of the TTO and SG has been a major activity during 2007. The project produced a working paper and three conference presentations. 1. Norman R, Cronin P, Viney R, King M et al. Issues in the quantification of health-related quality of life: The EQ-5D. Health Services Research Association of Australia and New Zealand (HSRAANZ), Auckland, 2-5 December 2007. Funding source CHERE staff Collaborators 1 University of Sheffield UK Understanding the determinants of participation: An analysis of breast cancer screening in New South Wales Key Objective Many jurisdictions have used public funding of health care to reduce or remove price at the point of delivery of services. Using the 2002 and 2004 NSW Health Survey, CHERE researchers estimated multinomial logit models on the probability of three screening behaviours; never had a mammogram, had last mammogram within the last two year and had last mammogram more than 2 years ago. The models examined the relative importance of socio-economic and geographic factors as well as the level of education and being born overseas in predicting screening behaviours for women in the target aged group 50 – 69. The findings showed that women in lower socio-economic groups were more likely to have never screened or be overdue for a screening. Place of residence and being in the younger age cohort (women aged 50 to 55) plays a significant role in predicting the likelihood of a woman to have never screened. More educated women are more likely to be overdue for screening. The research indicates that despite the existence of a ‘free’ and well established program, social disparities remain. This indicates the need for further recruitment and maintenance strategies that focus on women (i) residing in certain geographic locations, (ii) entering the target age group, (iii) on low incomes and (iv) born overseas. Funding source CHERE staff Collaborator 1. Department of Clinical Epidemiology and Biostatistics, McMaster University, Ontario, Canada |
|
