To Örebro University

OBJECTIVES: For drugs reimbursed with limited evidence of patient benefits, confirmatory evidence of overall survival (OS) and quality of life (QoL) benefits is important. For QoL data to serve as valuable input to patients and decision-makers, it must be measured and analyzed using appropriate methods. We aimed to assess the measurement and analyses of post-reimbursement QoL data for cancer drugs introduced in Swedish healthcare with limited evidence at the time of reimbursement.

METHODS: We reviewed any published post-reimbursement trial data on QoL for cancer drugs reimbursed in Sweden between 2010 and 2020 with limited evidence of improvement in QoL and OS benefits at the time of reimbursement. We extracted information on the instruments used, frequency of measurement, extent of missing data, statistical approaches, and the use of pre-registration and study protocols.

RESULTS: Out of 22 drugs satisfying our inclusion criteria, we identified published QoL data for 12 drugs in 22 studies covering multiple cancer types. The most frequently used QoL instruments were EORTC QLQ-C30 and EQ-5D-3/5L. We identified three areas needing improvement in QoL measurement and analysis: (i) motivation for the frequency of measurements, (ii) handling of the substantial missing data problem, and (iii) inclusion and adherence to QoL analyses in clinical trial pre-registration and study protocols.

CONCLUSIONS: Our review shows that the measurements and analysis of QoL data in our sample of cancer trials covering drugs initially reimbursed without any confirmed QoL or OS evidence have significant room for improvement. The increasing use of QoL assessments must be accompanied by a stricter adherence to best-practice guidelines to provide valuable input to patients and decision-makers.

We study how much regional variation in healthcare spending is driven by place- and patient-specific factors using a random sample of 53,620 regional migrants in Sweden. We find notable differences depending on the category of care, with place-specific factors having a significantly larger impact on specialized outpatient care compared to inpatient and pharmaceutical care. The place effect is estimated to 75% of variation in specialized outpatient care, but 26% or less in variations in inpatient care, and 5% in prescription drug spending. We also find that the empirical estimator has a substantial impact on the estimates of the place-specific effect. The results based on the traditional approach in the literature with two-way fixed effects and event-study models produce much larger estimates of the place-specific effect compared to results based on recently developed heterogeneity-robust models. For total healthcare spending, the traditional two-way fixed effects model estimates a place effect of 78%, while the heterogeneity-robust estimator finds a place effect around 10%. This finding indicates that previous results in this literature, all based on traditional two-way fixed-effects regressions, should be interpreted with care.

BACKGROUND: Unilateral posterior crossbite is a common malocclusion, and early treatment is recommended to enable normal growth. There are several possibilities regarding choice of appliances used for correcting this malocclusion; however, when treatment is financed by public funds the decision needs to be based not only on the effects but also on the effect in relation to the costs. OBJECTIVES: The aim was to perform a cost-effectiveness analysis comparing quad helix (QH) and rapid maxillary expanders (RME; hyrax-type) in children in the early mixed dentition.

MATERIAL AND METHODS: Seventy-two patients were randomized to treatment with either QH or RME, at two different centres. Data were collected from the patient's medical records regarding success rate, number of visits, total treatment time, emergency visits, and so forth, together with answers from patient questionnaires concerning absence from school and use of analgesics. A cost-effectiveness analysis with both an intention-to-treat (ITT) and a per-protocol approach was performed, as well as a deterministic sensitivity analysis.

RESULTS: The success rate, one year after the completion of the expansion, was equal between groups according to the ITT approach. From a healthcare perspective, the mean cost difference between RME and QH was €32.05 in favour of QH (P = 0.583; NS). From a societal perspective, the mean cost difference was €32.61 in favour of QH (P = 0.742; NS). The total appliance cost alone was higher in the RME group €202.67 resp. €155.58 in the QH group (P = 0.001). The probability of RME having a higher cost was 71% from a healthcare perspective and 62.7% from a societal perspective. The total treatment time was 97 days longer in the QH group. In the deterministic sensitivity analysis, when using a higher valuation of the children's educational loss, the QH becomes €58 more costly than the RME. There was a statistically significant difference in chair time and visits between centres (P < 0.001).

CONCLUSION: The difference in costs between RME and QH is not statistically significant, however, there is a slightly higher probability that RME is more expensive than QH with a mean cost of an additional €32 per patient from a healthcare perspective. Different work procedures at different centres indicate that logistics around the patient's treatment is a more important aspect than appliance used to decrease the number of visits and save chair time and thereby also costs.

BACKGROUND AND OBJECTIVE: Cancer drug costs have increased considerably within healthcare systems, but many drugs lack quality-of-life (QoL) and overall survival (OS) data at the time of reimbursement approval. This study aimed to review the extent of subsequent literature documenting improvements in OS and QoL for cancer drug indications where no such evidence existed at the time of reimbursement approval.

METHODS: Drug indications with claims of added therapeutical value but a lack of evidence on OS and QoL that were reimbursed between 2010 and 2020 in Sweden were included for review. Searches were conducted in PubMed and ClinicalTrial.gov for randomized controlled trials examining OS and QoL.

RESULTS: Of the 22 included drug indications, seven were found to have at least one trial with conclusive evidence of improvements in OS or QoL after a mean follow-up of 6.6 years. The remaining 15 drug indications either lacked subsequent randomized controlled trial data on OS or QoL (n = 6) or showed no statistically significant improvements (n = 9). Only one drug demonstrated evidence of improvement in both OS and QoL for its indication.

CONCLUSIONS: A considerable share of reimbursed cancer drug indications continue to lack evidence of improvement in both OS and QoL. With limited healthcare resources and an increasing cancer burden, third-party payers have strong incentives to require additional post-reimbursement data to confirm any improvements in OS and QoL.

BACKGROUND: Individuals of lower socioeconomic status generally have higher healthcare expenditures than individuals of higher socioeconomic status. However, little is known about how expenditures are distributed across socioeconomic groups over a lifetime, once accounting for differences in life expectancy. This study describes how lifetime healthcare expenditures are distributed across age, sex and socioeconomic groups in Sweden while adjusting for differences in life expectancy.

METHODS: Healthcare utilization from 2016 were linked to demographic and socioeconomic data for a random sample of individuals aged 20 and above in the four largest Swedish regions (n = 440 659). Mortality data were used to estimate income- and sex-specific survival rates. Expected lifetime healthcare expenditures were estimated by combining survival rates with mean healthcare expenditures over age, by sex, and income quintile.

RESULTS: We find that expected lifetime healthcare expenditures are highest among the first (lowest) income quintile despite their evident lower life expectancy. Expected lifetime expenditures were 17.9% (16.8%) higher in the first income quintile compared to the fifth (highest) quintile for women (men). Individuals in the first income quintile had higher expected lifetime expenditures for all care categories except for primary care.

CONCLUSION: We conclude that despite a lower life expectancy, the quintile of the lowest socioeconomic status still had higher lifetime healthcare expenditures.

Little is known about how patients dynamically respond to a forthcoming reduction in health care out-of-pocket prices. Using a kinked Donut Regression Discontinuity design with kinks entering and exiting the donut, we evaluate a Swedish cost-sharing policy, where primary care out-of-pocket prices were eliminated at age 85. We find evidence of forward-looking moral hazard with older adults delaying primary care visits up to four months before the out-of-pocket elimination and shifting these visits until shortly after. These health care delays are driven by non-urgent care: non-physician visits, planned visits and follow up visits. We find no evidence of severe negative health effects in the short-term as a result of the delay. Contrary to our finding of forward-looking behavior with respect to out-of-pocket prices, we do not find evidence of typical moral hazard, as we do not find a persistent increase in primary health care use after the copayment elimination.

BACKGROUND: This study aimed to describe the clinical and cost-effectiveness evidence supporting reimbursement decisions of new cancer drugs and analyze the influence of trial characteristics and the cost per quality-adjusted life years (QALYs) on the likelihood of reimbursement in Sweden.

PATIENTS AND METHODS: Data were extracted from all appraisal dossiers for new cancer drugs seeking reimbursement in Sweden and claiming added therapeutical value between the years 2010 and 2020. The data were analyzed using descriptive statistics, and logistic regression models were also used with the cost per QALY, study design, comparator, and evidence on final outcomes in the clinical trials as predictors of reimbursement.

RESULTS: All 60 included appraisals were based on trial evidence that assessed at least one final outcome (overall survival [OS] or quality of life [QoL]), although rarely as a primary outcome. Of the appraisals with a final decision (n = 58), 79% were approved for reimbursement. Among the reimbursed drugs, only half had trial evidence demonstrating improved OS or QoL. Only one drug had trial evidence supporting improvements in both OS and QoL. The average cost per QALY for reimbursed cancer drugs was estimated to be 748 560 SEK (€73 583). A higher cost per QALY was found to decrease the likelihood of reimbursement by 9.4% for every 100 000 SEK (€9830) higher cost per QALY (P = 0.03). For cost-effectiveness models without direct evidence of improvements in final outcomes, a larger QALY gain was observed compared with those with evidence mainly relying on intermediate and surrogate outcomes.

CONCLUSIONS: There are substantial uncertainties in the clinical and cost-effectiveness evidence underlying reimbursement decisions of new cancer drugs. Decision makers should be cautious of the limited evidence on patient-centered outcomes and the implications of allocating resources to expensive treatments with uncertain value for money.

There is substantial variation in drug spending across regions in Sweden, which can be justified if caused by differences in health need, but an indication of inefficiencies if primarily caused by differences in place-specific supply-side factors. This paper aims to estimate the relative effect of individual demand-side factors and place-specific supply-side factors as drivers of geographical variation in drug spending in Sweden. We use individual-level register data on purchases of prescription drugs matched with demographic and socioeconomic data of a random sample of about 900,000 individuals over 2007-2016. The primary empirical approach is a two-way fixed effect model and an event study where we identify demand- and supply-side effects based on how regional and local migrants change drug spending when moving across regional and municipal borders. As an alternative approach in robustness checks, we also use a decomposition analysis. The results show that the place-specific supply-side effect accounts for only about 5%-10% of variation in drug spending and remaining variation is due to individual demand-side effects. These results imply that health policies to reduce regional variation in drug spending would have limited impact if targeted at place-specific characteristics.