The extent to which a genomic test will be used in practice is affected by factors such as ability of the test to correctly predict response to treatment (i.e. sensitivity and specificity of the test), invasiveness of the testing procedure, test cost, and the probability and severity of side effects associated with treatment.
Using discrete choice experimentation (DCE), we elicited preferences of the public (Sample 1, N = 533 and Sample 2, N = 525) and cancer patients (Sample 3, N = 38) for different attributes of a hypothetical genomic test for guiding cancer treatment. Samples 1 and 3 considered the test/treatment in the context of an aggressive curable cancer (scenario A) while the scenario for sample 2 was based on a non-aggressive incurable cancer (scenario B).
In aggressive curable cancer (scenario A), everything else being equal, the odds ratio (OR) of choosing a test with 95% sensitivity was 1.41 (versus a test with 50% sensitivity) and willingness to pay (WTP) was $1331, on average, for this amount of improvement in test sensitivity. In this scenario, the OR of choosing a test with 95% specificity was 1.24 times that of a test with 50% specificity (WTP = $827). In non-aggressive incurable cancer (scenario B), the OR of choosing a test with 95% sensitivity was 1.65 (WTP = $1344), and the OR of choosing a test with 95% specificity was 1.50 (WTP = $1080). Reducing severity of treatment side effects from severe to mild was associated with large ORs in both scenarios (OR = 2.10 and 2.24 in scenario A and B, respectively). In contrast, patients had a very large preference for 95% sensitivity of the test (OR = 5.23).
The type and prognosis of cancer affected preferences for genomically-guided treatment. In aggressive curable cancer, individuals emphasized more on the sensitivity rather than the specificity of the test. In contrast, for a non-aggressive incurable cancer, individuals put similar emphasis on sensitivity and specificity of the test. While the public expressed strong preference toward lowering severity of side effects, improving sensitivity of the test had by far the largest influence on patients’ decision to use genomic testing.
Pharmacogenomics; Genomic medicine; Personalized medicine; Genetic testing; Discrete choice experiment; Conjoint analysis; Preference elicitation; Cancer treatment
Women with mutations in BRCA1 or BRCA2 are at high risk of developing breast cancer and, in British Columbia, Canada, are offered screening with both magnetic resonance imaging (MRI) and mammography to facilitate early detection. MRI is more sensitive than mammography but is more costly and produces more false positive results. The purpose of this study was to calculate the cost-effectiveness of MRI screening for breast cancer in BRCA1/2 mutation carriers in a Canadian setting.
We constructed a Markov model of annual MRI and mammography screening for BRCA1/2 carriers, using local data and published values. We calculated cost-effectiveness as cost per quality-adjusted life-year gained (QALY), and conducted one-way and probabilistic sensitivity analysis.
The incremental cost-effectiveness ratio (ICER) of annual mammography plus MRI screening, compared to annual mammography alone, was $50,900/QALY. After incorporating parameter uncertainty, MRI screening is expected to be a cost-effective option 86% of the time at a willingness-to-pay of $100,000/QALY, and 53% of the time at a willingness-to-pay of $50,000/QALY. The model is highly sensitive to the cost of MRI; as the cost is increased from $200 to $700 per scan, the ICER ranges from $37,100/QALY to $133,000/QALY.
The cost-effectiveness of using MRI and mammography in combination to screen for breast cancer in BRCA1/2 mutation carriers is finely balanced. The sensitivity of the results to the cost of the MRI screen itself warrants consideration: in jurisdictions with higher MRI costs, screening may not be a cost-effective use of resources, but improving the efficiency of MRI screening will also improve cost-effectiveness.
Breast cancer; BRCA; MRI; Cost-effectiveness; Screening
Resource allocation is a key challenge for healthcare decision makers. While several case studies of organizational practice exist, there have been few large-scale cross-organization comparisons.
Between January and April 2011, we conducted an on-line survey of senior decision makers within regional health authorities (and closely equivalent organizations) across all Canadian provinces and territories. We received returns from 92 individual managers, from 60 out of 89 organizations in total. The survey inquired about structures, process features, and behaviours related to organization-wide resource allocation decisions. We focus here on three main aspects: type of process, perceived fairness, and overall rating.
About one-half of respondents indicated that their organization used a formal process for resource allocation, while the others reported that political or historical factors were predominant. Seventy percent (70%) of respondents self-reported that their resource allocation process was fair and just over one-half assessed their process as ‘good’ or ‘very good’. This paper explores these findings in greater detail and assesses them in context of the larger literature.
Data from this large-scale cross-jurisdictional survey helps to illustrate common challenges and areas of positive performance among Canada’s health system leadership teams.
Resource allocation; Priority setting; Survey research; Canada
Among women with surgically removed, high-risk HER-2/neu-positive breast cancer, trastuzumab has demonstrated significant improvements in disease-free and overall survival. The objective of this study is to evaluate the cost-effectiveness of the currently recommended 12-month adjuvant protocol of trastuzumab using a Markov modeling approach and real-world cost data. Over the long term, treatment of HER-2/neu mutation positive breast cancer with a 12-month protocol of trastuzumab in the adjuvant setting is predicted to be cost-effective in a Canadian context.
Among women with surgically removed, high-risk HER-2/neu-positive breast cancer, trastuzumab has demonstrated significant improvements in disease-free and overall survival. The objective of this study is to evaluate the cost-effectiveness of the currently recommended 12-month adjuvant protocol of trastuzumab using a Markov modeling approach and real-world cost data.
A 10-health-state Markov model tracked patients' quarterly transitions between health states in the local and advanced states of breast cancer. Clinical data were obtained from the joint analysis of the National Surgical Adjuvant Breast and Bowel Project and North Central Cancer Treatment Group, as well as from the metastatic study conducted by Norum et al. Clinical outcomes were adjusted for quality of life using utility estimates published in a systematic review. Real cost data were obtained from the British Columbia Cancer Agency and were evaluated from a payer perspective. Costs and utilities were discounted at 5% per year, respectively, for a 28-year time horizon.
In the base case analysis, treatment with a 12-month adjuvant trastuzumab regimen resulted in a gain of 1.38 quality-adjusted life years or 1.17 life years gained at a cost of $18,133 per patient. Thus, the cost per QALY gained for the base case is $13,095. Cost per LYG is $15,492.
Over the long term, treatment of HER-2/neu mutation positive breast cancer with a 12-month protocol of trastuzumab in the adjuvant setting is predicted to be cost-effective in a Canadian context.
Cost-benefit analysis; Drug costs; Breast neoplasms; Antibodies; Monoclonal
Research has shown that people from higher socioeconomic status (SES) have better hepatocellular carcinoma (HCC) survival outcomes, although no such research has been carried out in Canada. We aimed to assess if an association between SES and HCC survival existed in the Canadian context.
We conducted a population-based cohort study linking HCC cases identified in the Ontario Cancer Registry between 1990 and 2009 to administrative and hospital data. Logistic regression and chi-squared tests were used to evaluate associations between SES (income quintile) and covariates. The Kaplan-Meier method was used to estimate survival. Sequential analysis of the proportional-hazards models were used to determine the association between SES and HCC survival controlling for potential prognostic covariates. During the period 1990–2009, 5,481 cases of HCC were identified. A significant association was found between SES and curative treatment (p = 0.0003), but no association was found between SES and non-curative treatment (p = 0.064), palliative treatment (p = 0.680), or ultrasound screening (p = 0.615). The median survival for the lowest SES was 8.5 months, compared to 8.8 months for the highest SES group. The age- and sex-adjusted proportional-hazards model showed statistically significant difference in HCC survival among the SES groups, with hazard ratio 0.905 (95% confidence intervals 0.821, 0.998) when comparing highest to lowest SES group. Further adjustments indicated that potentially curative treatment was the likely explanation for the association between SES and HCC survival.
Our findings suggest that a 10% HCC survival advantage exists for the higher SES groups. This association between SES and HCC survival is most likely a reflection of lack of access to care for low SES groups, revealing inequities in the Canadian healthcare system.
Multi attribute utility (MAU) instruments are used to include the health related quality of life (HRQoL) in economic evaluations of health programs. Comparative studies suggest different MAU instruments measure related but different constructs. The objective of this paper is to describe the methods employed to achieve content validity in the descriptive system of the Assessment of Quality of Life (AQoL)-6D, MAU instrument.
The AQoL program introduced the use of psychometric methods in the construction of health related MAU instruments. To develop the AQoL-6D we selected 112 items from previous research, focus groups and expert judgment and administered them to 316 members of the public and 302 hospital patients. The search for content validity across a broad spectrum of health states required both formative and reflective modelling. We employed Exploratory Factor Analysis and Structural Equation Modelling (SEM) to meet these dual requirements.
Results and Discussion
The resulting instrument employs 20 items in a multi-tier descriptive system. Latent dimension variables achieve sensitive descriptions of 6 dimensions which, in turn, combine to form a single latent QoL variable. Diagnostic statistics from the SEM analysis are exceptionally good and confirm the hypothesised structure of the model.
The AQoL-6D descriptive system has good psychometric properties. They imply that the instrument has achieved construct validity and provides a sensitive description of HRQoL. This means that it may be used with confidence for measuring health related quality of life and that it is a suitable basis for modelling utilities for inclusion in the economic evaluation of health programs.
Quality of Life; MAU instrument; HRQoL; Economic evaluation
To examine variation in patients' access to a set of cancer drugs through publicly funded provincial drug programs.
Data Sources/Study Design:
We surveyed provincial drug program managers about their highest-expenditure intravenous and oral cancer drugs. We then investigated whether the same cancer drugs account for the highest expenditures across the provincial programs. We also compared the rates at which these drugs are accessed through these programs.
While there is moderate consistency in the selection of cancer drugs that account for the highest provincial expenditures, considerable differences were found in the rates at which some drugs are accessed across provincial programs.
The study demonstrates the existence of interprovincial variation in publicly funded access to cancer drugs even after these drugs have been approved for public coverage.
To evaluate the validity of cancer-specific and generic preference-based instruments to discriminate across different measures of cancer severities.
Patients with breast (n = 66), colorectal (n = 57), and lung (n = 61) cancer completed the EORTC QLQ-C30 and the FACT-G, as well as three generic instruments: the EQ-5D, the SF-6D, and the HUI2/3. Disease severity was quantified using cancer stage, Eastern Cooperative Oncology Group Performance Status (ECOG-PS) score, and self-reported health status. Comparative analyses confirmed the multi-dimensional conceptualization of the instruments in terms of construct and convergent validity.
In general, the instruments were able to discriminate across severity measures. The instruments demonstrated moderate to strong correlation with each other (r = 0.37-0.73). Not all of the measures could discriminate between different groups of disease severity: the EQ-5D and SF-6D were less discriminative than the HUI2/3 and the cancer-specific instruments.
The cancer-specific and generic preference-based instruments demonstrated to be valid in discriminating across levels of ECOG-PS scores and self-reported health states. However, the usefulness of the generic instruments may be limited if they are not able to detect small changes in health status within cancer patients. This raises concerns regarding the appropriateness of these instruments when comparing different cancer treatments within an economic evaluation framework.
Quality of life; cancer-specific instruments; generic instruments; external validity: responsiveness; disease severity; utilities
Moral distress — the physical and emotional response to feeling prevented from carrying out ethically proper action — can have serious consequences for health professionals and healthcare organizations. We investigated perceived moral distress qualitatively with managers in two BC health authorities.
Respondents described conditions under which they experienced distress: when they set priorities within highly resource-constrained environments, when they observed inequities between budget allocations and management responsibilities, and when organizational priorities did not align with their personal values. When coping proved insufficient, managers would respond by leaving positions, organizations or the healthcare field altogether.
Respondents asked for leadership development and the creation of spaces in which moral distress could be openly discussed. However, formal training in priority setting did not appear to be helpful on its own. Rather, it increased managers' awareness of the ethical dimensions of resource allocation without (in this instance) entrenching supports that would help them resolve these concerns.
In the HPV FOCAL trial, we will establish the efficacy of hr-HPV DNA testing as a stand-alone screening test followed by liquid based cytology (LBC) triage of hr-HPV-positive women compared to LBC followed by hr-HPV triage with ≥ CIN3 as the outcome.
HPV-FOCAL is a randomized, controlled, three-armed study over a four year period conducted in British Columbia. It will recruit 33,000 women aged 25-65 through the province's population based cervical cancer screening program. Control arm: LBC at entry and two years, and combined LBC and hr-HPV at four years among those with initial negative results and hr-HPV triage of ASCUS cases; Two Year Safety Check arm: hr-HPV at entry and LBC at two years in those with initial negative results with LBC triage of hr-HPV positives; Four Year Intervention Arm: hr-HPV at entry and combined hr-HPV and LBC at four years among those with initial negative results with LBC triage of hr-HPV positive cases
To date, 6150 participants have a completed sample and epidemiologic questionnaire. Of the 2019 women enrolled in the control arm, 1908 (94.5%) were cytology negative. Women aged 25-29 had the highest rates of HSIL (1.4%). In the safety arm 92.2% of women were hr-HPV negative, with the highest rate of hr-HPV positivity found in 25-29 year old women (23.5%). Similar results were obtained in the intervention arm HPV FOCAL is the first randomized trial in North America to examine hr-HPV testing as the primary screen for cervical cancer within a population-based cervical cancer screening program.
International Standard Randomised Controlled Trial Number Register, ISRCTN79347302
To date there has been relatively little published about how research priorities are set, and even less about methods by which decision-makers can be engaged in defining a relevant and appropriate research agenda. We report on a recent effort in British Columbia to have researchers and decision-makers jointly establish an agenda for future research into questions of resource allocation.
The researchers enlisted decision-maker partners from each of British Columbia's six health authorities. Three forums were held, at which researchers and decision-makers from various levels in the health authorities considered possible research areas related to three key focus areas: (1) generation and use of decision criteria and measurement of 'benefit' against such criteria; (2) identification of so-called 'disinvestment' opportunities; and (3) evaluation of the effectiveness of priority setting procedures. Detailed notes were taken from each forum and synthesized into a set of qualitative themes.
Forum participants suggested that future research into healthcare priority setting would benefit from studies that were longitudinal, comparative, and/or interdisciplinary. As well, participants identified two broad theme areas in which specific research projects were deemed desirable. First, future research might usefully consider how formal priority setting and resource allocation projects are situated within a larger organizational and political context. Second, additional research efforts should be devoted to better understanding and improving the actual implementation of priority setting frameworks, particularly with respect to issues of change management and the resolution of impediments to action on recommendations for resource allocation.
We were able to validate the importance of initial areas posed to the group and observed emergence of additional concerns and directions of critical importance to these decision-makers at this time. It is likely that the results are broadly applicable to other healthcare contexts. The implementation of this research agenda in British Columbia will depend upon the ability of the researchers and decision-makers to develop particular projects that fit within the constraints of existing funding opportunities. The process of engagement itself had benefits in terms of connecting decision-makers with their peers and sparking increased interest in the use and refinement of priority setting frameworks.
A key function of cancer genetics services is to provide risk information. Yet, to date there has been little consistency in the way that breast cancer risk perception has been measured. The aims of the study were to measure estimates of (i) population risk (ii) absolute risk (iii) comparative risk of developing breast cancer for Ashkenazi Jewish women, and to determine predictors of breast cancer risk perception. Of 152 women, 107 (70%) completed all questions. The mean (s.d.) estimate for population risk, absolute risk and comparative risk were 22.7% (15.9), 31.8% (20.6) and 1.9-fold (1.9), respectively. Most women over-estimated population risk. Women at population risk generally over-estimated the population risk and their own absolute risk, yet understood they are at the same risk as the population. Those with a family history understood they are at increased risk, but underestimated the extent to which their familial risk is increased. Anxiety, high estimation of population risk and lesser family history predicted over-estimation of absolute risk, while high estimation of population risk and a strong family history predicted under-estimation of comparative risk.
Ashkenazi; breast cancer; perceived risk; genetic counselling; BRCA1; BRCA2
To describe the methods and innovations used in constructing the VisQoL, a vision-related utility instrument for the health economic evaluation of eye care and rehabilitation programs.
The VisQoL disaggregates vision into six items. Utilities were estimated for item worst responses (the worst level for each item, with all other items at their best level) and VisQoL all-worst responses (all items at their worst level) using the time trade-off procedure. Time trade-off questions require people to imagine living a fixed number of years with a particular health condition and then indicate how many of those years of life they would be willing to trade to have perfect health. Where respondents indicated a health state was “worse than death” negative utilities were estimated. Time trade-off questions minimized the “focusing effect,” which occurs if respondents discount the fact that all other aspects of health are at their best when answering questions, by using pictorial and verbal aids.
Item utilities were combined using a multiplicative model, and VisQoL model utilities placed on a scale where 0.00 and 1.00 represent full health and death, respectively. The VisQoL allows utilities to be calculated for a wide range of vision-related conditions.
The 6-item VisQoL has excellent psychometric properties and is specifically designed to be sensitive to vision-related quality of life. It is the first instrument to permit the rapid estimation of utility values for use in economic evaluations of vision-related programs.
Vision-related quality of life; utility instrument; cost-utility analysis; economic evaluation; time trade-off
Doctors and managers have to make tough decisions about what services to provide from their budgets. Economic approaches can help, but they also need to take into account the practical and ethical challenges faced by healthcare professionals
Recent increases in the bulk-billing rate have been taken as an indication that the Federal government's Strengthening Medicare initiative, and particularly the bulk-billing incentives, are 'working'. Given the enduring geographic differences in the supply of general practitioners (GPs) it is timely to reconsider the impact that this increase in the provision of 'free care' will have on access to Medicare-funded GP services in rural and urban areas of Australia. Utilisation has been modelled as two different stochastic processes: the decision to consult and the frequency of consultation.
In the decision to consult model the supply of FFS GPs is a more important predictor of utilisation than the bulk-billing rate. Paradoxically the modelling predicts that ceteris paribus increases in either GP supply or the bulk-billing rate appear to have perverse effects in some areas by decreasing utilisation. In the frequency of consultation model, GP density is not a predictor and increasing the bulk-billing rate will unambiguously increase the frequency of consultation across all areas. In both models, the positive impacts associated with changes in supply and cost are constrained outside the inner metropolitan area by reduced geographic accessibility to Medicare-funded GP services. The modelling also shows that people are more likely to consult a GP in areas of high socioeconomic disadvantage, although socioeconomic status is not a predictor of frequency of consultation.
Bulk-billing rates and the supply of FFS GPs are important features of the Australian health care system that are, potentially, amenable to policy manipulation. The implications of this research are that government policies designed to achieve similarity in these characteristics across geographic areas will not result in equity of access because they fail to address problems caused by geographic inaccessibility in rural and remote areas. Attempting to increase bulk-billing rates in some of these areas may, in fact, reduce access to FFS GP services.
In order to profile the health services research community in Australia and New Zealand and describe its capacity, a web-based survey was administered to members of the Health Services Research Association of Australia and New Zealand (HSRAANZ) and delegates of the HSRAANZ's Third Health Services Research and Policy Conference.
Responses were received from 191 individuals (68%). The responses of the 165 (86%) who conducted or managed health services research indicated that the health services research community in Australia and New Zealand is characterised by highly qualified professionals who have come to health services research via a range of academic and professional routes (including clinical backgrounds), the majority of whom are women aged between 35 and 54 who have mid- to senior- level appointments. They are primarily employed in universities and, to a lesser extent, government departments and health services. Although most are employed in full time positions, many are only able to devote part of their time to health services research, often juggling this with other professional roles. They rely heavily on external funding, as only half have core funding from their employing institution and around one third have employment contracts of one year or less. Many view issues around building the capacity of the health services research community and addressing funding deficits as crucial if health services research is to be translated into policy and practice. Despite the difficulties they face, most are positive about the support and advice available from peers in their work settings, and many are actively contributing to knowledge through academic and other written outputs.
If health services research is to achieve its potential in Australia and New Zealand, policy-makers and funders must take the concerns of the health services research community seriously, foster its development, and contribute to maximising its capacity through a sustainable approach to funding. There is a clear need for a strategic approach, where the health services research community collaborates with competitive granting bodies and government departments to define and fund a research agenda that balances priority-driven and investigator-driven research and which provides support for training and career development.
Health services research; capacity; resources; funding; outputs; Australia; New Zealand