Patients' preferences for treatment outcomes for advanced non-small cell lung cancer: A conjoint analysis

Johns Hopkins Bloomberg School of Public Health, Baltimore, MD 21205, USA.
Lung cancer (Amsterdam, Netherlands) (Impact Factor: 3.96). 02/2012; 77(1):224-31. DOI: 10.1016/j.lungcan.2012.01.016
Source: PubMed


Treatment decisions for advanced non-small cell lung cancer (NSCLC) are complex and require trade-offs between the benefits and risks experienced by patients. We evaluated the benefits that patients judged sufficient to compensate for the risks associated with therapy for NSCLC.
Participants with a self-reported diagnosis of NSCLC (n=100) were sampled from an online panel in the United Kingdom. Eligible and consenting participants then completed a self-administered online survey about their disease and their treatment preferences were assessed. This involved respondents choosing among systematically paired profiles that spanned eight attributes: progression-free survival [PFS], symptom severity, rash, diarrhoea, fatigue, nausea and vomiting, fever and infection, and mode of treatment administration (infusion and oral). A choice model was estimated using mixed-logit regression. Estimates of importance for each attribute level and attribute were then calculated and acceptable tradeoffs among attributes were explored.
A total of 89 respondents (73% male) completed all choice tasks appropriately. Increases in PFS together with improvements in symptom severity were judged most important and increased with PFS benefit - 4 months: 5.7; 95% CI: 3.5-7.9; 5 months: 7.1; 95% CI: 4.4-9.9; and 7 months: 10.0; 95% CI: 6.1-13.9. However, improvements in PFS were viewed as most beneficial when disease symptoms were mild and as detrimental when patients had severe symptoms. Fatigue (5.0; 95% CI: 2.7-7.3) was judged to be the most important risk, followed by diarrhoea (2.8; 95% CI: 0.7-4.9), nausea and vomiting (2.1; 95% CI: 0.1-4.1), fever and infection (2.1; 95% CI: 0.2-4.1), and rash (2.0; 95% CI: 0.2-3.9). Oral administration was preferred to infusion (1.8; 95% CI: 0.0-3.6). Patients with mild and moderate symptoms traded PFS for less risks or more convenience if the severe symptoms were not experienced.
This study demonstrates the value of conjoint analysis in the study of patient preferences for cancer treatments. In this small sample of patients with NSCLC from the UK, we demonstrate that the value of improvements in PFS is conditional upon the severity of disease symptoms; and that risks are valued differently.

Download full-text


Available from: Ateesha Farah Mohamed, Jan 21, 2015
  • Source
    • "For each attribute, an effects coding system (1, 0, −1 for a three-level attribute instead of a dummy coding 1, 0, 0) was used to estimate a parameter for each attribute level [19] [20]. With this system, the omitted categories were estimated as the negative sum of the included categories [16] [21], making the mean effect of the model zero. The standard errors for each omitted category were also estimated using the variance–covariance matrix. "
    [Show abstract] [Hide abstract]
    ABSTRACT: The aim of the study was to quantify patient preferences for outcomes associated with oral antidiabetic medications (OAMs) in Sweden and Germany through a discrete-choice experiment. Adults taking OAMs who had a self-reported physician's diagnosis of type 2 diabetes mellitus (T2DM) made a series of nine choices between pairs of hypothetical profiles. Each profile had a predefined range of attributes: blood glucose control, frequency of mild-to-moderate hypoglycaemia, annual severe hypoglycaemic events, annual weight gain, pill burden and frequency of administration, and cost. Choice questions were based on an experimental design with known statistical properties. Bivariate probit analysis estimated the probabilities of choice of medication administration from patient characteristics and, conditional on that choice, preferences for treatment outcomes. The final sample consisted of 188 Swedish and 195 German patients. For both countries, weight gain was the most important attribute, followed by blood glucose control. Avoiding a 5-kg weight gain was 1.5 times more important in Sweden and 2.3 times more important in Germany than achieving moderate blood glucose control, thereby, suggesting that blood glucose control is relatively more important to Swedish than to German patients. Least important outcomes were the number of daily pills (Sweden) and frequency of mild-to-moderate hypoglycaemia (Germany). Patients in both Sweden and Germany preferred OAMs not associated with weight gain.
    Diabetes & Metabolism 07/2013; 39(5). DOI:10.1016/j.diabet.2013.06.001 · 3.27 Impact Factor
  • Source
    • "A number of researchers have highlighted the benefits of using AHP to explore user needs in healthcare [29,30], and in particular for including patient opinions in health technology assessment [31,32], choosing treatments [33], and improving patient centred healthcare [34,35]. Other methods that have attempted to elicit and quantify user needs in healthcare are conjoint analysis (CA) [36] , discrete choice experiments [37] and best-worst scaling [38]. A growing number of articles have focused on comparing AHP with these methods, and in particular with CA. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Background The rigorous elicitation of user needs is a crucial step for both medical device design and purchasing. However, user needs elicitation is often based on qualitative methods whose findings can be difficult to integrate into medical decision-making. This paper describes the application of AHP to elicit user needs for a new CT scanner for use in a public hospital. Methods AHP was used to design a hierarchy of 12 needs for a new CT scanner, grouped into 4 homogenous categories, and to prepare a paper questionnaire to investigate the relative priorities of these. The questionnaire was completed by 5 senior clinicians working in a variety of clinical specialisations and departments in the same Italian public hospital. Results Although safety and performance were considered the most important issues, user needs changed according to clinical scenario. For elective surgery, the five most important needs were: spatial resolution, processing software, radiation dose, patient monitoring, and contrast medium. For emergency, the top five most important needs were: patient monitoring, radiation dose, contrast medium control, speed run, spatial resolution. Conclusions AHP effectively supported user need elicitation, helping to develop an analytic and intelligible framework of decision-making. User needs varied according to working scenario (elective versus emergency medicine) more than clinical specialization. This method should be considered by practitioners involved in decisions about new medical technology, whether that be during device design or before deciding whether to allocate budgets for new medical devices according to clinical functions or according to hospital department.
    BMC Medical Informatics and Decision Making 01/2013; 13(1):2. DOI:10.1186/1472-6947-13-2 · 1.83 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background: While the application of conjoint analysis and discrete-choice experiments in health are now widely accepted, a healthy debate exists around competing approaches to experimental design. There remains, however, a paucity of experimental evidence comparing competing design approaches and their impact on the application of these methods in patient-centered outcomes research. Objectives: Our objectives were to directly compare the choice-model parameters and predictions of an orthogonal and a D-efficient experimental design using a randomized trial (i.e., an experiment on experiments) within an application of conjoint analysis studying patient-centered outcomes among outpatients diagnosed with schizophrenia in Germany. Methods: Outpatients diagnosed with schizophrenia were surveyed and randomized to receive choice tasks developed using either an orthogonal or a D-efficient experimental design. The choice tasks elicited judgments from the respondents as to which of two patient profiles (varying across seven outcomes and process attributes) was preferable from their own perspective. The results from the two survey designs were analyzed using the multinomial logit model, and the resulting parameter estimates and their robust standard errors were compared across the two arms of the study (i.e., the orthogonal and D-efficient designs). The predictive performances of the two resulting models were also compared by computing their percentage of survey responses classified correctly, and the potential for variation in scale between the two designs of the experiments was tested statistically and explored graphically. Results: The results of the two models were statistically identical. No difference was found using an overall chi-squared test of equality for the seven parameters (p = 0.69) or via uncorrected pairwise comparisons of the parameter estimates (p-values ranged from 0.30 to 0.98). The D-efficient design resulted in directionally smaller standard errors for six of the seven parameters, of which only two were statistically significant, and no differences were found in the observed D-efficiencies of their standard errors (p = 0.62). The D-efficient design resulted in poorer predictive performance, but this was not significant (p = 0.73); there was some evidence that the parameters of the D-efficient design were biased marginally towards the null. While no statistical difference in scale was detected between the two designs (p = 0.74), the D-efficient design had a higher relative scale (1.06). This could be observed when the parameters were explored graphically, as the D-efficient parameters were lower. Conclusions: Our results indicate that orthogonal and D-efficient experimental designs have produced results that are statistically equivalent. This said, we have identified several qualitative findings that speak to the potential differences in these results that may have been statistically identified in a larger sample. While more comparative studies focused on the statistical efficiency of competing design strategies are needed, a more pressing research problem is to document the impact the experimental design has on respondent efficiency.
    The patient 12/2012; 5(4):279-94. DOI:10.2165/11641100-000000000-00000 · 1.90 Impact Factor
Show more