Vortragssitzung

Economic Evaluations

Vorträge

Who benefits from technological progress? Heterogeneities in health outcomes of heart attack treatments
Matthias Westphal, TU Dortmund

Einleitung / Introduction

We estimate the effect of an invasive heart attack treatment option – that is, percutaneous transluminal coronary angioplasty (PTCA) - compared to conservative treatment on mortality using administrative data of the universe of all hospital cases in Germany from 2005 to 2007. Often, instrumental variables estimation (IV) is used to estimate the causal effect of a medical intervention on outcomes like health or costs. This technique account for the problem that – if individuals are not randomized into the treatment – recipients of the treatment often differ from those who do not receive it, based on unobservable characteristics. IV estimation may provide a high internal validity of estimated effects, given that the employed instruments are valid. Yet, a growing literature criticizes a rather limited external validity. If the treatment effect differs by individuals – for instance, some strongly benefit, others do not or are even worse off – IV estimates identify a local average treatment effect (LATE), an average effect of a certain subgroup of individuals which differs from the subgroup of the treated. This might induce two problems: First, policy makers are usually interested in average treatment effects of the entire population (ATE) or of the treated (ATT), numbers that might differ from the local average treatment effect. Thus, IV is not necessarily informative about the effect policy makers are interested in. Second, effect heterogeneities play an important role, particularly in the evaluation of medical interventions. Positive average effects might not be sufficient if there is also a large fraction of individuals that is actually harmed by the treatment.

Methode / Method

We estimate the marginal treatment effects of PTCA treatment on mortality. Marginal treatment effects, as suggested by Heckman and Vytlacil (2007) deal with the two problems of LATE. They are more fundamental parameters that allow to derive all conventional treatment parameters like the ATE, the average treatment effect on the treated (ATT) and others. Moreover, they deliver the distribution of treatment effects over the entire population, allowing to observe whether each individual benefits (if to a different degree) or some do not, or are even worse off. This makes much more comprehensive assessments of treatment impacts possible than is currently state of the art in the international literature.

Ergebnisse / Results

We use regional variations in access to invasive heart attack procedures as an instrument. The „differential distance“ to the next hospital that offers PTCA turns out to be a strong instrument. Our results suggest a 4.5 percentage point mortality reduction for patients who have access to PTCA compared to patients receiving only conservative treatment.


AutorInnen
Ansgar Wübker, RWI Essen
Hendrik Schmitz, Universität Paderborn
Matthias Westphal, TU Dortmund
Cost-utility analysis of biologic therapy vs. early surgery in ulcerative colitis
Isa Maria Steiner, Hamburg Center for Health Economics, Universität Hamburg

Einleitung / Introduction

Over the past 20 years, biological medicines have amplified the treatment options for ulcerative colitis patients. This enables patients to postpone or completely avoid surgery after the failure of conventional medical therapy. However, previous studies comparing the cost-effectiveness of biologic therapies with early surgery come to ambiguous results and should be re-evaluated. Our objective was thus to assess whether biologic therapies are cost-effective compared to early surgery in ulcerative colitis.

Methode / Method

We conducted a cost-utility-analysis for adult ulcerative colitis patients from the perspective of the German statutory health insurance. The analysis was based on a Markov model consisting of two main parts. While one part of the Markov model simulates disease progression under biologic therapy (states: remission, mild, moderate/severe), the other part simulates disease progression after colectomy (states: surgery, post-surgery remission with pouch, post-surgery complications with pouch, post-surgery with permanent ostomy). Our hypothetical starting cohort consists of 1000 patients with moderately or severely active ulcerative colitis that are 30 years old and biologic-naïve. When treated with early surgery, patients enter the model in the surgery state and never transition to any of the biologic therapy states. When treated with biologics, patients enter the model in the moderate/severe state and may transition to surgery in the following cycles. Costs and quality adjusted life years (QALYs) were used as outcomes. Transition probabilities were partly taken from the literature and a German IBD register (CEDUR). Quality of life for each state was also obtained from the literature, whereas cost data was acquired from a large German sickness fund.

Ergebnisse / Results

Preliminary results for the base case indicate that early surgery could be superior to biologic treatment in terms of quality of life and costs. However, the status of final parametrization has not been reached yet. Also, full deterministic and probabilistic sensitivity analysis are yet to be conducted.

Zusammenfassung / Conclusion

Despite their high costs, biologic therapies are nowadays a widely used treatment option in ulcerative colitis. Our first results suggest that for some patients an early surgery may be a suitable alternative. This is, however, still subject to the final parametrization.


AutorInnen
Isa Maria Steiner, Hamburg Center for Health Economics, Universität Hamburg
Stefanie Howaldt, ImmunoRegister gemeinnützige UG
Tom Stargardt, Hamburg Center for Health Economics, Universität Hamburg
Evaluation of a medical emergency call software
Marc Diederichs, Johannes Gutenberg Universität Mainz

Einleitung / Introduction

While much is known about emergency health care provision in hospitals, research about the time before patients enter hospitals is scarce. It is crucial that emergency medical dispatch acts fast and accurate in the allocation of medical vehicles and personnel. Using the introduction of a structured and standardized emergency call (SSEC) software for supporting dispatchers, this research evaluates potential benefits of using SSEC software. Specifically, we analyze if the use of SSEC software affects: (i) emergency call and emergency vehicle response times, (ii) the accuracy with which emergency vehicles and staff are allocated, (iii) the ICD-10 code specific mortality rate on the district level.

Methode / Method

We have dispatch center specific data on emergency call- and vehicle response times, from 8 German dispatch centers that introduced the dispatch software between the years 2015 and 2020, at our disposal. We use this data to answer subquestion (i) by applying a difference in differences estimation strategy. Dispatch center specific time trends reduce our identification’s reliance on the parallel trends assumption. The same data is used to answer subquestion (ii). By comparing how often an additional emergency doctor is alarmed only after an ambulance has arrived at the emergency with and without the software, we can make judgements about changes in the efficiency of resource allocation. This analysis is done by normalizing the different dates of the software introduction to zero and subsequently using the negative binomial regression model. Subquestion (iii) is answered using 2014-2017 cause of death data from regional statistical offices. We analyze whether deaths from causes that are particularly sensitive to emergency care are reduced. This specifically includes cardiac arrest, as SSEC software provides telephonic support for reanimation by lay-persons.

Ergebnisse / Results

Our preliminary results suggest (i), that the overall calltime and the time until the emergency vehicles arrive at the emergency increases by about 10 seconds per call. While this may suggest a worse performance, this does come with a slightly increased dispatch accuracy (ii). Further analyses will evaluate how the distribution of call and response times are affected. The coefficient of the mortality rate (iii) is negative but statistically insignificant. The coefficients of the cause -specific subsample analyses are insignificant as well. Our analyses thus show mixed results, and no clear overall performance improvement from an introduction of SSEC software.


AutorInnen
Marc Diederichs, Johannes Gutenberg Universität Mainz
Reyn van Ewijk, Johannes Gutenberg Universität Mainz
Better care for less money: cost-effectiveness of integrated care in multi-episode patients with severe psychotic disorders.
Christian Brettschneider, Universitätsklinikum Hamburg-Eppendorf, Institut für Gesundheitsökonomie und Versorgungsforschung

Einleitung / Introduction

Severe mental illness (SMI) is defined by considerable and persistent impaired functioning due to mental disorders. People with psychotic disorders fulfilling SMI criteria present clinical and social challenges (e.g. comorbid mental and chronic somatic disorders, social disability, service disengagement, medication non-adherence). Assertive Community Treatment (ACT) is one approach to address these issues. In this study, we compare cost-effectiveness of integrated care with therapeutic assertive community treatment (IC-TACT) versus standard care (SC) in multiple episode psychosis.

Methode / Method

The data for the present analyses were derived from the ACCESS I and II studies. 12-month IC-TACT in patients with schizophrenia-spectrum and bipolar I disorders were compared with a historical control group. Differences were adjusted by entropy balancing. The primary outcome was cost-effectiveness based on mental health care costs from a payers’ perspective and quality-adjusted life years (QALYs) as a measure of health effects during 12-month follow-up. The incremental cost-effectiveness ratio and cost-effectiveness acceptability curves based on the net-benefit approach were calculated.

Ergebnisse / Results

At baseline, patients in IC-TACT (n=212) had significantly higher illness severity and lower functioning than SC (n=56). Differences were successfully adjusted. Over 12 months, IC-TACT had significantly lower days in inpatient (10.2±20.5 vs. 28.2±44.8; p<0.05) and day-clinic care (2.6±16.7 vs. 16.4±33.7; p=.004) and correspondingly lower costs (€-55,084). Within outpatient care, IC-TACT displayed a higher number of treatment contacts (116.3±45.3 vs. 15.6±6.3) and higher related costs (€+1,417). Both resulted in lower total costs in IC-TACT (€10,741±8,269 vs. €18,215±17,456; adjusted mean difference=€-13,248±2,975, p<0.001). Adjusted incremental QALYs were significantly higher for IC-TACT versus SC (+0.10±0.37, p=0.05). Hence IC-TACT was dominant. The probability of cost-effectiveness of IC-TACT was constantly higher than 99%.

Zusammenfassung / Conclusion

IC-TACT was cost-effective compared with SC. The use of prima facies ‘costly’ TACT teams is highly recommended to improve outcomes and save total cost for patients with severe psychotic disorders.


AutorInnen
Anne Karow, Universitätsklinikum Hamburg-Eppendorf
Christian Brettschneider, Universitätsklinikum Hamburg-Eppendorf
Hans-Helmut König, Universitätsklinikum Hamburg-Eppendorf
Christoph U. Correll, Charité Berlin
Daniel Schöttle, Universitätsklinikum Hamburg-Eppendorf
Daniel Lüdecke, Universitätsklinikum Hamburg-Eppendorf
Anja Rohenkohl, Universitätsklinikum Hamburg-Eppendorf
Friederike Ruppelt, Universitätsklinikum Hamburg-Eppendorf
Vivien Kraft, Universitätsklinikum Hamburg-Eppendorf
Jürgen Gallinat, Universitätsklinikum Hamburg-Eppendorf
Martin Lambert, Universitätsklinikum Hamburg-Eppendorf