An examination of machine learning to map non‐preference based patient reported outcome measures to health state utility values

Aghdaee, Mona; Parkinson, Bonny; Sinha, Kompal; Gu, Yuanyuan; Sharma, Rajan; Olin, Emma; Cutler, Henry

doi:10.1002/hec.4503

Cited by 6 publications

(8 citation statements)

References 113 publications

(108 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It could thus be more rigorous to administer a relevant questionnaire, like PROMIS Global Health-10 questionnaire. 32 Despite excluding other medical conditions that may impact QoL and decision regret before the investigation, the participants might have undetected diseases, which would influence the conclusion. This is our study's weakness.…”

Section: Discussionmentioning

confidence: 99%

“…Although the main objective of this study is to utilise the ThyPRO‐39 and DRS scales to evaluate patients with thyroid‐related diseases’ quality of life and decision regret, it should be noted that other factors such as social environment and psychological health will potentially affect the outcomes. It could thus be more rigorous to administer a relevant questionnaire, like PROMIS Global Health‐10 questionnaire 32 . Despite excluding other medical conditions that may impact QoL and decision regret before the investigation, the participants might have undetected diseases, which would influence the conclusion.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Quality of life and decision regret in patients with late‐hypothyroidism after radioiodine treatment for Graves' disease

Li,

Yang,

et al. 2023

Clinical Endocrinology

View full text Add to dashboard Cite

ObjectivePatients with Graves' disease often engage in shared decision‐making to select an individualised treatment regimen from multiple options. Radioactive iodine (RAI) is one of the treatment choices for their condition, aims to improve quality of life and well‐being. Likewise, dissatisfaction with treatment outcomes can result in decision regret. We employed validated questionnaires to assess the prospective quality of life, decision regret and relative factors involved in decision‐making of patients with late hypothyroidism after RAI therapy.MethodsA questionnaire survey was conducted among patients in hypothyroidism status for more than 1 year after RAI therapy. Disease‐specific and generic QoL were assessed using the short form of thyroid‐related patient‐reported outcome (ThyPRO‐39) questionnaire. Patient satisfaction regarding their decision to undergo RAI was assessed using the Decision Regret Scale (DRS) and patients were asked about the importance of relative factors in decision‐making.ResultsOf 254 patients who responded to the survey, the mean age of patients was 45.3 years (range: 18–78 years) and the median time from RAI therapy to survey was 4 years (range: 1–30 years). Patients' median and mean DRS score were 34.4 and 38.8 (range: 0–100), respectively. A total of 100 (39.4%) patients express absent‐to‐mild regret (score: 0–25), 154 (60.6%) patients express moderate‐to‐severe regret (score: >25). The mean score of the absent‐to‐mild regret group were significantly higher than those of the moderate‐to‐severe regret group on most ThyPRO‐39 scales. A statistically significant positive correlation was observed between DRS score and most ThyPRO‐39 scale score. There was a significant positive association between higher DRS score and longer time intervals after RAI treatment, a brief duration of hyperthyroidism, and the significance of long‐time outpatient follow‐up. More decision regret was negatively associated Iodine‐free diet, ineffectiveness of ATD, fear of surgery.ConclusionImpairment of quality of life was positively correlated with decision regret in patients with late‐hypothyroidism after radioiodine therapy. Patients with insufficient information support before decision‐making are more likely to have higher decision regret after treatment. Our findings suggest that health providers should fully communicate with patients and provide information support in multiple dimensions during the shared‐decision‐making process.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Quality of life and decision regret in patients with late‐hypothyroidism after radioiodine treatment for Graves' disease

Li,

Yang,

et al. 2023

Clinical Endocrinology

View full text Add to dashboard Cite

show abstract

“…Employing machine learning techniques for mapping is still uncommon [ 29 ]. However, some mapping studies have taken this path in recent years [ 41 , 54 , 55 ]. Because the core strength of machine learning is prediction, mapping studies appear to be a suitable application field for it.…”

Section: Discussionmentioning

confidence: 99%

“…For model selection and estimation, only the model subset was used. We selected one final model for each covariate scenario based on tenfold cross-validation, which has become increasingly popular for mapping studies in recent years (e.g., [ 37 – 41 ]). We thus randomly split the sample into 10 folds, i.e., equally sized parts of the data set.…”

Section: Methodsmentioning

confidence: 99%

Mapping from SIBDQ to EQ-5D-5L for patients with inflammatory bowel disease

Steiner¹,

Bokemeyer²,

Stargardt³

2023

Eur J Health Econ

View full text Add to dashboard Cite

Objective Clinical studies commonly use disease-specific measures to assess patients’ health-related quality of life. However, economic evaluation often requires preference-based utility index scores to calculate cost per quality-adjusted life-year (QALY). When utility index scores are not directly available, mappings are useful. To our knowledge, no mapping exists for the Short Inflammatory Bowel Disease Questionnaire (SIBDQ). Our aim was to develop a mapping from SIBDQ to the EQ-5D-5L index score with German weights for inflammatory bowel disease (IBD) patients. Methods We used 3856 observations of 1055 IBD patients who participated in a randomised controlled trial in Germany on the effect of introducing regular appointments with an IBD nurse specialist in addition to standard care with biologics. We considered five data availability scenarios. For each scenario, we estimated different regression and machine learning models: linear mixed-effects regression, mixed-effects Tobit regression, an adjusted limited dependent variable mixture model and a mixed-effects regression forest. We selected the final models with tenfold cross-validation based on a model subset and validated these with observations in a validation subset. Results For the first four data availability scenarios, we selected mixed-effects Tobit regressions as final models. For the fifth scenario, mixed-effects regression forest performed best. Our findings suggest that the demographic variables age and gender do not improve the mapping, while including SIBDQ subscales, IBD disease type, BMI and smoking status leads to better predictions. Conclusion We developed an algorithm mapping SIBDQ values to EQ-5D-5L index scores for different sets of covariates in IBD patients. It is implemented in the following web application: https://www.bwl.uni-hamburg.de/hcm/forschung/mapping.html.

show abstract

“…They found that the machine learning technique performed similarly to the traditional econometric methods in three out of the four countries of their sample. Another study by Aghdaee et al [ 22 ] also found that machine learning (e.g., Lasso regression) performed marginally better if not combined with other traditional econometric methods ( n = 2015). Despite these previous findings, machine learning still has the advantage of determining the nature of the relationships without researchers trying to guess the possible combinations between them or imposing their bias on the results by selecting their preferred functional form.…”

Section: Introductionmentioning

confidence: 98%

Mapping the 12-item World Health Organization disability assessment schedule 2.0 (WHODAS 2.0) onto the assessment of quality of life (AQoL)-4D utilities

Ma,

Chen,

Badji

et al. 2023

Qual Life Res

View full text Add to dashboard Cite

Purpose The World Health Organization Disability Assessment Schedule 2.0 (WHODAS 2.0) is a widely used disability-specific outcome measure. This study develops mapping algorithms to estimate Assessment of Quality of Life (AQoL)-4D utilities based on the WHODAS 2.0 responses to facilitate economic evaluation. Methods The study sample comprises people with disability or long-term conditions (n = 3376) from the 2007 Australian National Survey of Mental Health and Wellbeing. Traditional regression techniques (i.e., Ordinary Least Square regression, Robust MM regression, Generalised Linear Model and Betamix Regression) and machine learning techniques (i.e., Lasso regression, Boosted regression, Supported vector regression) were used. Five-fold internal cross-validation was performed. Model performance was assessed using a series of goodness-of-fit measures. Results The robust MM estimator produced the preferred mapping algorithm for the overall sample with the smallest mean absolute error in cross-validation (MAE = 0.1325). Different methods performed differently for different disability subgroups, with the subgroup with profound or severe restrictions having the highest MAE across all methods and models. Conclusion The developed mapping algorithm enables cost-utility analyses of interventions for people with disability where the WHODAS 2.0 has been collected. Mapping algorithms developed from different methods should be considered in sensitivity analyses in economic evaluations.

show abstract

An examination of machine learning to map non‐preference based patient reported outcome measures to health state utility values

Cited by 6 publications

References 113 publications

Quality of life and decision regret in patients with late‐hypothyroidism after radioiodine treatment for Graves' disease

Quality of life and decision regret in patients with late‐hypothyroidism after radioiodine treatment for Graves' disease

Mapping from SIBDQ to EQ-5D-5L for patients with inflammatory bowel disease

Mapping the 12-item World Health Organization disability assessment schedule 2.0 (WHODAS 2.0) onto the assessment of quality of life (AQoL)-4D utilities

Contact Info

Product

Resources

About