A study of risk analysis and prognosis models for the mortality of sepsis based on real-world data in China

Background: The study aimed to explore the factors associated with the mortality of sepsis and to develop prognosis models for predicting outcomes based on real world data in China. Methods: Data regarding sepsis patients medical records were extracted from the hospital information systems in four hospitals. The data included general information, laboratory tests, score systems, and supportive treatment for sepsis. In total, 507 medical records with complete data were available for data analysis. Multiple variable regression (MR) analysis used to explore associations, and to develop prognosis models Results: The mortality of sepsis was 0.3124 in the total sample. A univariate analysis indicated 23 variables significantly associated with the mortality of sepsis (p <0.05 for all). The MLR analysis showed independent and significant variables of age, GCS, SOFA, shock, breath rate, TBIL, CHE, BUN, LAC, OI, HCO3, IMV, and ALB (P <0.05 for all). Prognosis models have a high predictive performance (AUC = 0.885, 95% CI: 0.854 to 0.917 in model2). Conclusion: The study showed evidence of independent and significant factors associated with the mortality of sepsis, including age, GCS, SOFA, septic shock, breath rate, TBIL, CHE, BUN, LAC, OI, HCO3, IMV, and ALB. Prognosis models with a high performance were developed.

Download Full-text

The unforeseen business and medical consequences of EHR data collection for a real-world data multiple myeloma registry.

Journal of Clinical Oncology ◽

10.1200/jco.2019.37.15_suppl.e19528 ◽

2019 ◽

Vol 37 (15_suppl) ◽

pp. e19528-e19528

Author(s):

Kevin M. Keogh ◽

Andrew J. Belli ◽

Monica M. Matta ◽

Kathryn A. Tanenbaum ◽

Kaeleigh Farrish ◽

...

Keyword(s):

Multiple Myeloma ◽

Pilot Study ◽

Real World ◽

Medical Records ◽

Real World Data ◽

World Data ◽

Return Rates ◽

Research Collaborations ◽

Transfer Of Information ◽

Efficient Transfer

e19528 Background: Record retrieval on behalf of a consenting patient within the context of real-world data is not well understood. As the need for real-world data continues to expand, methods for the efficient transfer of information across consenting parties will be critical to enable research collaborations. This need was highlighted through a partnership between COTA and the Multiple Myeloma Research Foundation (MMRF). As one component of the pilot study, COTA managed the record retrieval of consenting patients before going on to abstract the data for use in a registry. Methods: The pilot study identified 23 patients that consented to the release of medical records at 54 institutions across 20 states. COTA partnered with a retrieval vendor and employed its own outreach efforts for acquisition. Outreach and retrieval techniques were similar across COTA and the vendor, including targeted calls, delivery of IRB-approved consent materials, and on-site requests. The 30-day release parameter for a covered entity under 45 CFR 164.524(b)(2) of HIPAA’s Privacy Rule was used to evaluate the observed return rates. Results: A total of 56 medical records were requested, and 48 records were retrieved. The mean (±SD) retrieval time across all sites was 33 (±58) days. We found that 54% of records were released in < = 30 days, and 32% of records > 30 days. 14% of requested records were never released, despite a median of 19 outreach attempts (range 10 to 43). Cited issues for delay or non-release consisted of 22 institutions questioning the validity of the certified electronic patient signature, 6 requiring a physical signature, and 5 requiring their own authorization forms. In no cases did the records contain structured metadata, such as LOINC, MedDRA, and RxNorm. Conclusions: This pilot showed an unpredictable variance associated with the release of records in the context of real-world data. This variance contributed to barriers and delays in broader research efforts. The lack of accompanying metadata with released records resulted in additional required data processing. Future studies should be conducted to establish best practices in the release and retrieval of medical records used to support real-world data research.

Download Full-text

PRO74 Pilot Study of a Novel Patient-facing Medical Records Collection Platform to Support Real World Data Generation in Rare Disease

Value in Health ◽

10.1016/j.jval.2021.04.1059 ◽

2021 ◽

Vol 24 ◽

pp. S211

Author(s):

E. Brimble ◽

G. Beek ◽

L. Wilson ◽

K. Muirhead ◽

S. Reichert ◽

...

Keyword(s):

Pilot Study ◽

Rare Disease ◽

Real World ◽

Medical Records ◽

Data Generation ◽

Real World Data ◽

World Data

Download Full-text

Factors associated with insufficient response to acute treatment of migraine in Japan: analysis of real-world data from the Adelphi Migraine Disease Specific Programme

BMC Neurology ◽

10.1186/s12883-020-01848-4 ◽

2020 ◽

Vol 20 (1) ◽

Author(s):

Koichi Hirata ◽

Kaname Ueda ◽

Wenyu Ye ◽

Yongin Kim ◽

Mika Komori ◽

...

Keyword(s):

Real World ◽

Acute Treatment ◽

Real World Data ◽

World Data ◽

Factors Associated ◽

Insufficient Response ◽

Disease Specific

Download Full-text

PMU68 Emulating the Main Outcomes of Randomised Controlled Studies through Statistical Analysis of Real World Data Directly Extracted from Electronic Medical Records.

Value in Health ◽

10.1016/j.jval.2021.04.780 ◽

2021 ◽

Vol 24 ◽

pp. S157

Author(s):

D. Drake ◽

Wouwe L Van ◽

Rhali A El ◽

R. Abdi ◽

M. Kouki

Keyword(s):

Statistical Analysis ◽

Electronic Medical Records ◽

Real World ◽

Medical Records ◽

Real World Data ◽

World Data ◽

Randomised Controlled

Download Full-text

Factors associated with treatment persistence among people with Type 2 Diabetes Mellitus (T2DM) initiating basal insulin – real-world data from Germany

Diabetologie und Stoffwechsel ◽

10.1055/s-0036-1580778 ◽

2016 ◽

Vol 11 (S 01) ◽

Cited By ~ 1

Author(s):

E Moennig ◽

I Hadjiyianni ◽

T Otto ◽

D Cao ◽

M De Koven ◽

...

Keyword(s):

Diabetes Mellitus ◽

Type 2 Diabetes ◽

Type 2 Diabetes Mellitus ◽

Real World ◽

Basal Insulin ◽

Real World Data ◽

Treatment Persistence ◽

World Data ◽

Factors Associated

Download Full-text

Factors on survival for patients of EGFR mutation advanced lung adenocarcinoma with EGFR-TKIs combined with radiotherapy: Real-world data from single center.

Journal of Clinical Oncology ◽

10.1200/jco.2021.39.15_suppl.e21126 ◽

2021 ◽

Vol 39 (15_suppl) ◽

pp. e21126-e21126

Author(s):

Yuxiang Wang ◽

Wenjian Yu ◽

Jian Shi ◽

Nan Jiang ◽

Zhoufan Wang ◽

...

Keyword(s):

Lung Adenocarcinoma ◽

Real World ◽

Egfr Mutation ◽

Clinical Stage ◽

Univariate Analysis ◽

Stage Iii ◽

Total Response ◽

Real World Data ◽

World Data ◽

Egfr Tkis

e21126 Background: The aim was to retrospectively evaluate survival and prognostic factors for patients of advanced lung adenocarcinoma with EGFR mutation by a real world data. Methods: From Jan, 2015 to Dec, 2020, lung adenocarcinoma with EGFR positive mutation, advanced (clinical stage III-IV), without surgery, received EGFR-TKIs and radiotherapy (RT) were enrolled. OS and PFS were calculated. Results: 238 patients were included, 101 in males and 137 in females; the median ages was 61. The RT was performed at the time of no-PD in 71 and PD in 167 cases. The median dose of RT was 50 Gy. The rate of acute AE during RT was 36.1% in grade 1-2 and 4.8% in grade 3-4. The follow-up ended at 30, Dec, 2020. The 1-, 2-, 3-years OS,PFS1and PFS2 were 84.4%,59.7%, 38.7%, 52.0%,29.4%, 16.3%, 86.6%, 64.6%, 45.6%. mOS, mPFS1and mPFS2 were 30.3, 14.1, 33.6 months, respectively. Multivariate analysis showed the independent factors for OS were ages (median were 34.0 and 24.9 months for ≤65 and > 65) and clinical stage (median were 22.8, 34.8 and 27.4 months for stage III (33), IVA (84) and IVB (121 cases), respectively), chemotherapy (CT) (median were 27.4, 28.1, 32.5, 44.3 months for TKIs alone(120), TKIs concurrent CT (43), TKIs sequential CT (32), CT sequential TKIs (43 cases), respectively) and total response (the medians were 32.5, 27.2, 33.3 months for CR(7)+PR(107), SD(70) and PD(54 cases), respectively); Independent factors: PFS1 was the time of RT at no-PD and PD (the median were 17.8 and13.2 months), chemotherapy, (the median were 16.1, 12.1, 11.0, 13.5 for TKIs alone, TKIs concurrent CT, TKIs sequential CT, CT sequential TKIs), and total response (the medians were 18.8, 11.7, 10.9 months for CR+PR, SD and PD); PFS2 was the total response (the medians were 58.3, 21.6, 20.9 months for CR+PR, SD and PD, respectively). Univariate analysis also showed doses of RT (the median was 12.3 months for < 50 Gy and 15.5 months for ≥ 50 Gy, p = 0.018) was associated with PFS1, the first-line drugs of TKIs (the median were 39.1, 35.0, 25.6 and 38.6 months for Gefitinib (89), Icotinib (109), Erlotinib (29) and others (11 cases), respectively, p = 0.044) and chemotherapy (the medians were 45.6, 30.6, 18.7, 41.4 months for TKIs alone, TKIs concurrent CT, TKIs sequential CT, CT sequential TKIs, respectively, p = 0.007) were related with PFS2; but the type of EGFR mutation (94 in 19 exon, 118 in 21 exon and 26 cases in others) was not related with survival. Conclusions: EGFR-TKIs combined RT was tolerable and efficient for patients of advanced lung adenocarcinoma with EGFR mutation. TKIs add RT at the time of no-PD could improve PFS1. CT sequential TKIs was probably in favour of OS, even PFS. The better total response (CR+PR) was associated with longer OS, PFS1 and PFS2. But the result need to be proved furtherly.

Download Full-text