Back

A novel hospital-at-home model for patients with COVID-19 built by a team of local primary care clinics and clinical outcomes: A multi-center retrospective cohort study

Tsujimoto, Y.; Kobayashi, M.; Oku, T.; Ogawa, T.; Yamadera, S.; Tsukamoto, M.; Matsuda, N.; Nishihira, M.; Terauchi, Y.; Tanaka, T.; Kawabata, Y.; Miyamoto, Y.; Morikami, Y.; The KISA2-Tai Osaka,

2022-10-31 primary care research
10.1101/2022.10.28.22281588 medRxiv
Show abstract

BackgroundHospital-at-home (HaH) care has been proposed as an alternative to inpatient care for patients with COVID-19. Previous reports were hospital-led and involved patients triaged at the hospitals. To reduce the burden on hospitals, we constructed a novel HaH care model organised by a team of local primary care clinics. MethodsWe conducted a multi-center retrospective cohort study of the COVID-19 patients who received our HaH care from Jan 1st to Mar 31st, 2022. Patients who were not able to be triaged for the need for hospitalization by the Health Center solely responsible for the management of COVID-19 patients in Osaka City were included. The primary outcome was receiving medical care beyond the HaH care defined as a composite outcome of any medical consultation, hospitalization, or death within 30 days from the initial treatment. ResultsOf 382 eligible patients, 34 (9%) were triaged for hospitalization immediately after the initial visit. Of the remaining 348 patients followed up, 37 (11%) developed the primary outcome, while none died. Obesity, fever, and gastrointestinal symptoms at baseline were independently associated with an increased risk of needing medical care beyond the HaH care. A further 129 (37%) patients were managed online alone without home visit, and 170 (50%) required only one home visit in addition to online treatment. ConclusionsThe HaH care model with a team of primary care clinics was able to triage patients with COVID-19 who needed immediate hospitalization without involving hospitals, and treated most of the remaining patients at home.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.