[Utilization of routine data from different statutory health insurance funds: a practical report of experience from the TIM-HF2 study about data acquisition, validation, and processing]. / Nutzung von Routinedaten unterschiedlicher gesetzlicher Krankenkassen: Ein Erfahrungsbericht aus der TIM-HF2-Studie zur Datenerschließung, -validierung und -aufbereitung.

BACKGROUND: The randomized controlled clinical trial "TIM-HF2" investigated the benefit of telemonitoring in chronic heart failure. The health economic evaluation of this intervention was based on routine data from statutory health insurance (SHI) funds. Since participants were recruited independently of their SHI affiliation, there was a large number of potential data-providing SHI funds. This resulted in both organizational and methodological challenges, from participation of the data providers to data preparation. METHOD: The procedures are described from study planning and data acquisition to data review and processing in the TIM-HF2 trial. Based on the identification of potential problems for data completeness and data quality, possible solutions have been derived. RESULTS: In total, participants were insured with 49 different SHI funds, which provided routine data for a total of 1450 participants. About half of all initial data deliveries were correct. The most common problems in data preparation occurred in the machine readability of the data. Success factors for a high level of data completeness were close communication with the SHI funds and a high level of time and personnel commitment to intensive data checking and preparation. DISCUSSION: Based on the experience of the TIM-HF2 trial, a high heterogeneity has been detected in data management and transmission of routine data. Universally applicable data descriptions are desired to improve data access, quality, and usability for research purposes.

Cost-effectiveness of noninvasive telemedical interventional management in patients with heart failure: health economic analysis of the TIM-HF2 trial.

BACKGROUND: Noninvasive remote patient management (RPM) in patients with heart failure (HF) has been shown to reduce the days lost due to unplanned cardiovascular hospital admissions and all-cause mortality in the Telemedical Interventional Management in Heart Failure II trial (TIM-HF2). The health economic implications of these findings are the focus of the present analyses from the payer perspective. METHODS AND RESULTS: A total of 1538 participants of the TIM-HF2 randomized controlled trial were assigned to the RPM and Usual Care group. Health claims data were available for 1450 patients (n = 715 RPM group, n = 735 Usual Care group), which represents 94.3% of the original TIM-HF2 patient population, were linked to primary data from the study documentation and evaluated in terms of the health care cost, total cost (accounting for intervention costs), costs per day alive and out of hospital (DAOH), and cost per quality-adjusted life year (QALY). The average health care costs per patient year amounted to € 14,412 (95% CI 13,284-15,539) in the RPM group and € 17,537 (95% CI 16,179-18,894) in the UC group. RPM led to cost savings of € 3125 per patient year (p = 0.001). After including the intervention costs, a cost saving of € 1758 per patient year remained (p = 0.048). CONCLUSION: The additional noninvasive telemedical interventional management in patients with HF was cost-effective compared to standard care alone, since such intervention was associated with overall cost savings and superior clinical effectiveness.

Mortality and morbidity 1 year after stopping a remote patient management intervention: extended follow-up results from the telemedical interventional management in patients with heart failure II (TIM-HF2) randomised trial.

BACKGROUND: The Telemedical Interventional Management in Heart Failure II (TIM-HF2) trial showed that, compared with usual care, a structured remote patient management (RPM) intervention done over 12-months reduced the percentage of days lost due to unplanned cardiovascular hospitalisations and all-cause death. The aim of the study was to evaluate whether this clinical benefit seen for the RPM group during the initial 12 month follow-up of the TIM-HF2 trial would be sustained 1 year after stopping the RPM intervention. METHODS: TIM-HF2 was a prospective, randomised, multicentre trial done in 43 hospitals, 60 cardiology practices, and 87 general practitioners in Germany. Patients with heart failure, New York Heart Association functional class II or III, and who had been hospitalised for heart failure within 12 months before randomisation were randomly assigned to either the RPM intervention or usual care. At the final study visit (main trial), the RPM intervention was stopped and the 1 year extended follow-up period started, which lasted 1 year. The primary outcome was percentage of days lost due to unplanned cardiovascular hospitalisations and all-cause mortality. Analyses were done using the intention-to-treat principle. This trial is registered with ClinicalTrials.gov, number NCT01878630. FINDINGS: Between Aug 13, 2013, and May 12, 2017, 1538 patients were enrolled (765 to the remote patient management group and 773 to the usual care group) in the main trial. 671 of 765 patients in the remote patient management group and 673 of 773 in the usual care group completed the main trial and started the extended follow-up period up to 1 year later. In the extended follow-up period, the percentage of days lost due to unplanned cardiovascular hospital admissions and all-cause mortality did not differ significantly between groups weighted mean 5·95% [95% CI 4·59-7·31] in the RPM group vs 6·64% [95% CI 5·19-8·08] in the usual care group [rate ratio 0·79; 95% CI 0·78-1·21]). However, when data from the main trial and the extended follow-up period were combined, the percentage of days lost due to unplanned cardiovascular hospitalisation or all-cause death was significantly less in patients allocated to the RPM group (382 [50%] of 765; weighted mean 9·28%; 95% CI 7·76-10·81) than in the UC group (398 [51%] of 773; 11·78%; 95% CI 10·08-13·49; ratio of weighted average 0·79; 95% CI 0·62-1·00; p=0·0486). INTERPRETATION: The positive effect of our RPM intervention on morbidity and mortality over the course of the main trial was no longer observed 1 year after stopping the RPM intervention. However, because the TIM-HF2 trial was not powered to show significance during the extended follow-up period, our results are exploratory and require further research. FUNDING: German Federal Ministry of Education and Research.