Vortragssitzung
Digitization
Talks
Market entry of digital health providers afterthe introduction of a new reimbursementpathway - Evidence from the German DiGAregulation
Sabrina Schubert, ZEW Mannheim & FAU Nürnberg
Einleitung / Introduction
Digital tools are increasingly used to complement traditional forms of health care provision. In the last five years, on average 90,000 new smartphone apps were added annually to medical and health categories in the Apple and Google app stores. For regulators in social health insurance systems, this large volume of potential solutions for their populations causes two challenges. Since most health apps are unspecific fitness and wellness apps, the first challenge is to identify those apps focusing on treatment, prevention, and management of specific clinical conditions and also showing clinical evidence for effectiveness. The second challenge is to attract app providers to this therapeutical apps market since the market entry is associated with higher regulatory and development costs. Countries globally draft strategies whether and, if so, how to include digital health applications into formal reimbursement structures. In the year 2020, Germany has been the first country world-wide to set up a structured pathway for developers of digital therapeutics to access social health insurance reimbursement - the DiGA regulation. Developers follow a structured process to demonstrate the positive care effects of their application and after successful completion of this process, the applications can then be prescribed to patients by doctors. Similar schemes are currently established in other European countries such as France and Belgium. In this paper we investigate whether this new reimbursement pathway incentivizes the development and market entry of high quality patient centered digital health solutions.
Methode / Method
We create a language-month-level app panel for all patient-centered health application in the Apple App store from January 2018 to September 2021. Our panel contains information on language availability, prices, reviews, ratings, app’s publisher and more in depth information from the App store. We exploit a feature of the regulation that a DiGA “prescription app” needs to be available in German language. In our empirical analysis, we use synthetic control and synthetic difference-in-difference models to causally identify changes in the composition of available apps and hence the effect of the “prescription apps’” reimbursement introduction on innovation in the market.
Ergebnisse / Results
Our preliminary results show increased developer activity for German digital health applications after the introduction of the scheme. However, these results are not stable over all applied methods yet reagrding statistical significance.
Zusammenfassung / Conclusion
Germany developed a new approach to reimburse patient centered digital health applications and opened the market to 73 million individuals in the German social health insurance for app providers. We find first indications that this policy increased market supply, hence more apps entered the market in terms of German language apps.
Authors
Sabrina Schubert, ZEW Mannheim & FAU Erlangen-Nürnberg
Rebecca Janßen, ZEW Mannheim, Universität Zürich
Simon Reif, ZEW Mannheim, FAU Erlangen-Nürnberg
The association of process quality and process digitization in acute somatic care – Observational study using data from hospital quality report cards and the German DigitalRadar project
Justus Vogel, Chair of Healthcare Management, University of St. Gallen
Einleitung / Introduction
Digitizing hospitals aims to improve care, reduce costs, and increase efficiency and productivity. However, the literature lacks clarity on the precise meaning of "hospital digitization," how to measure it, which aspects of quality it might affect, and how to investigate such effects empirically. We argue that only logically-linked, statistically valid data on the digital maturity of hospital processes and hospital (process) quality can show a positive digitization-quality relationship.
Methode / Method
In our study, we use detailed data from the German evaluation project DigitalRadar from 2021 describing – amongst other dimensions – process digitization in hospitals at a high level of detail. We combine these statistically sensitive data with four different risk-adjusted outcome and process quality indicators from the structured quality report cards of the external inpatient quality assurance program for four indications (osteosynthesis and hip replacement after fracture of the femur, outpatient-acquired pneumonia, inpatient-acquired bedsore/ decubitus). Our hospital sample size ranges between 664 and 1,566 hospitals, depending on the investigated indication. We run one set of stepwise, multivariate linear regressions for each indication with the respective quality indicator as dependent variable and different digitization dimensions as explanatory variables. We add control variables with each new specification.
Ergebnisse / Results
Overall, we find no significant correlation between outcome or process quality and the majority of digitization dimensions. Only digitization of documentation and diagnosis shows a consistent but weakly significant correlation (p<0.1) with the risk-adjusted ratio of inpatient cases with a new bedsore/ decubitus. We argue that this lack of statistical significance is in part due to the insufficient statistical sensitivity of the available quality indicators. Moreover, available and routinely measured quality indicators seem not to be apt to reflect digitization effects, as also other studies show.
Zusammenfassung / Conclusion
We conclude that empiric assessment of a digitization-quality relationship needs the development of more fitting and sensitive quality indicators. With existing datasets, correlations and causal relationships are not detectable with nationwide, large samples and other study designs such as small-scale pre-post intervention assessments of the introduction of specific software will remain the gold standard in digitization-quality research.
Authors
Justus Vogel, Chair of Healthcare Management, University of St. Gallen
Alexander Haering, RWI – Leibniz-Institut für Wirtschaftsforschung e.V.
David Kuklinski, Chair of Healthcare Management, University of St. Gallen
Alexander Geissler, Chair of Healthcare Management, University of St. Gallen
The potential of a PROM-based DHI in knee and hip replacement to reduce post-surgery cost and health service utilization – Secondary analyses of PROMoting Quality
David Kuklinski, University of St. Gallen
Einleitung / Introduction
Western healthcare systems increasingly suffer from shortages of physicians and care workers, and simultaneously experience a rise in demand for healthcare service due to an aging population. These challenges will not only prevail but increase in the upcoming years. To ensure the security of high-quality care provision and control exploding healthcare costs, it is essential to increase productivity of healthcare workers and to reduce unnecessary and avoidable healthcare services. One potential lever is to use digital technologies, such as digital health interventions, to complement and support medical services. In this study, we investigate whether a digital PROM monitoring and alert system for hip and knee replacement patients can be used to reduce health expenditures and decrease the utilization of inpatient and outpatient health services, one-year post-surgery.
Methode / Method
We use data from the multicenter randomized controlled trial PROMoting Quality, which included 3697 hip and 3110 knee replacement patients between the years 2019 and 2020 of nine German hospitals. Analyses were run on a subset of 546 hip and 492 knee replacement patients for which we received longitudinal billing data from 24 public health insurances. Patients were equally divided into two groups: one receiving a PROM-based monitoring and alerts intervention, the other standard care. All patients completed digital PROM assessments at 12 months after surgery. The intervention group was additionally assessed at 1, 3, and 6 months post-surgery for potential critical recovery detection. We compare differences in one-year post-surgery expenditures and physician visits between intervention and control group using Wilcoxon-Mann Whitney rank sum test, student’s t-tests, and a mixed-effect regression model.
Ergebnisse / Results
The digital PROM monitoring and alert system was able to significantly reduce one-year post-surgery health expenditures by 9.4% (376 €) for hip replacement patients and by 7.1% (375 €) for knee replacement patients. Furthermore, for hip replacement, we found a decrease of physician visits for the intervention group. These differences in cost and visits, mostly originated from ambulatory care (~1.5 fewer visits), in particular from GP visits, fewer prescriptions and remedies. For knee replacement, main determinants of the cost differences were fewer prescriptions and fewer aids. The mixed-effects model supports our findings.
Zusammenfassung / Conclusion
Digital health interventions such as our intervention cannot only improve quality of care but can also save scarce healthcare resources. It can be hypothesized that through the “being taken care of” effect, hip and knee replacement patients need less affirmation that their recovery is on track and thus, spare GP visits and other expenditures.
Authors
David Kuklinski, University of St. Gallen
Lukas Schöner, Technical University Berlin
Alexander Geissler, University of St. Gallen
Justus Vogel, University of St. Gallen
Theories, models and instruments of predicting user acceptance of eHealth interventions: a systematic review of validation studies
Juliana Schmidt, Universität Bielefeld, Fakultät für Gesundheitswissenschaften, Arbeitsgruppe für Gesundheitsökonomie und Gesundheitsmanagement
Einleitung / Introduction
Despite the increasing availability of digital and technological interventions in healthcare, few show sustained use and acceptance by patients or providers. Although the literature provides various instruments that aim to measure and explain technology acceptance by patients and providers, there is no systematic overview or comparison of their measurement properties. There is therefore no consensus on which instruments are best suited for the measurement of acceptance. Consequently, the aim of this work is to identify validation studies of theories, models or instruments of user acceptance of technology interventions and to evaluate their assessment of measurement properties.
Methode / Method
A systematic literature review was conducted in the MEDLINE, Web of Science, CINAHL and PsycINFO databases in January 2023. Studies were included if they examined digital interventions with therapeutic uses (I.e. not merely the digitization of processes), ≥ 50 % of the study population were prior users of the eHealth intervention in question, ≥ 2 measurement properties were assessed, and they were published in English or German in a peer-reviewed journal.
Ergebnisse / Results
Of the 4,233 potentially relevant titles (duplicates excluded), 36 were included. The instruments evaluated acceptance of interventions ranging from mHealth to AI from either a healthcare professional, patient or relative perspective. A plurality (n=14) of studies used instruments based on the Unified Theory of Technology Acceptance (UTAUT/UTAUT2) or a modified UTAUT. A further 9 studies used instruments based on the Technology Acceptance Model (TAM) or a modified TAM. Modifications of TAM and UTAUT used a number of additional constructs, with no clear consensus emerging. The remaining 13 studies used various other instruments. Although it is considered to be the most important measurement property, 14 studies did not provide information on content validity. Twelve studies assumed content validity based on translation or prior studies, although adjustments or combinations were made. Content validity evaluated by adequate methods was determined in 10 studies. Other measurement properties were extracted.
Zusammenfassung / Conclusion
A preliminary finding of this review is the variety of instruments of technology acceptance in healthcare being tested in validation studies. Studies show a concentration of instruments based on (context-related modifications of) UTAUT and TAM. Furthermore, few studies report the content validity of their instruments. Therefore, the overall validity of these instruments can be questioned. This systematic literature review is a first step towards reaching a consensus concerning the most appropriate instrument for measuring eHealth user acceptance.