这是用户在 2025-8-7 23:09 为 https://app.immersivetranslate.com/pdf-pro/0ada6dc5-2663-408b-ab0a-ad3cad109bef/ 保存的双语快照页面，由 沉浸式翻译 提供双语支持。了解如何保存？

Rehabilitation after surgery for hip fracture - the impact of prompt, frequent and mobilisation-focused physiotherapy on discharge outcomes: an observational cohort study
髋部骨折手术后的康复——及时、频繁和以活动为重点的物理治疗对出院结果的影响：一项观察性队列研究

Daniel Siminiuc 1 1 ^(1){ }^{1}, Oya Gumuskaya 1 1 ^(1){ }^{1}, Rebecca Mitchell 2 2 ^(2){ }^{2}, Jack Bell 3 3 ^(3){ }^{3}, Ian D. Cameron 4 4 ^(4){ }^{4}, Jamie Hallen 5 5 ^(5){ }^{5}, Karen Birkenhead 1 , 6 1 , 6 ^(1,6){ }^{1,6}, Sarah Hurring 7 7 ^(7){ }^{7}, Brett Baxter 8 8 ^(8){ }^{8}, Jacqueline Close 5 , 9 5 , 9 ^(5,9){ }^{5,9}, Katie J. Sheehan 10 10 ^(10){ }^{10}, Antony Johansen 11 11 ^(11){ }^{11}, Mellick J. Chehade 12 12 ^(12){ }^{12}, Catherine Sherrington 13 , 14 13 , 14 ^(13,14){ }^{13,14}, Zsolt J. Balogh 15 15 ^(15){ }^{15}, Morag E. Taylor 5 5 ^(5){ }^{5} and Mitchell Sarkies 1 , 2 , 6 ∗ 1 , 2 , 6 ∗ ^(1,2,6^(**)){ }^{1,2,6{ }^{*}}
Daniel Siminiuc 1 1 ^(1){ }^{1} , Oya Gumuskaya 1 1 ^(1){ }^{1} , Rebecca Mitchell 2 2 ^(2){ }^{2} , Jack Bell 3 3 ^(3){ }^{3} , Ian D. Cameron 4 4 ^(4){ }^{4} , Jamie Hallen 5 5 ^(5){ }^{5} , Karen Birkenhead 1 , 6 1 , 6 ^(1,6){ }^{1,6} , Sarah Hurring 7 7 ^(7){ }^{7} , Brett Baxter 8 8 ^(8){ }^{8} , Jacqueline Close 5 , 9 5 , 9 ^(5,9){ }^{5,9} , Katie J. Sheehan 10 10 ^(10){ }^{10} , Antony Johansen 11 11 ^(11){ }^{11} , Mellick J. Chehade 12 12 ^(12){ }^{12} , Catherine Sherrington 13 , 14 13 , 14 ^(13,14){ }^{13,14} , Zsolt J. Balogh 15 15 ^(15){ }^{15} , Morag E. Taylor 5 5 ^(5){ }^{5} 和 Mitchell Sarkies 1 , 2 , 6 ∗ 1 , 2 , 6 ∗ ^(1,2,6^(**)){ }^{1,2,6{ }^{*}}

Abstract  摘要

Purpose To determine the relationship between three postoperative physiotherapy activities (time to first postoperative walk, activity on the day after surgery, and physiotherapy frequency), and the outcomes of hospital length of stay (LOS) and discharge destination after hip fracture. Methods A cohort study was conducted on 437 hip fracture surgery patients aged ≥ 50 ≥ 50 >= 50\geq 50 years across 36 participating hospitals from the Australian and New Zealand Hip Fracture Registry Acute Rehabilitation Sprint Audit during June 2022. Study outcomes included hospital LOS and discharge destination. Generalised linear and logistic regressions were used respectively, adjusted for potential confounders. Results Of 437 patients, 62 % 62 % 62%62 \% were female, 56 % 56 % 56%56 \% were aged ≥ 85 ≥ 85 >= 85\geq 85 years, 23 % 23 % 23%23 \% were previously living in a residential aged care facility, 48 % 48 % 48%48 \% usually walked with a gait aid, and 38 % 38 % 38%38 \% were cognitively impaired prior to their injury. The median acute and total LOS were 8 (IQR 5-13) and 20 (IQR 8-38) days. Approximately 71% ( n = 179 / 251 n = 179 / 251 n=179//251n=179 / 251 ) of patients originally living in private residence returned home and 29 % ( n = 72 / 251 ) 29 % ( n = 72 / 251 ) 29%(n=72//251)29 \%(n=72 / 251) were discharged to a residential aged care facility. Previously mobile patients had a higher total LOS if they walked day 2-3 (10.3 days; 95% CI 3.2, 17.4) or transferred with a mechanical lifter or did not get out of bed day 1 ( 7.6 days; 95 % 95 % 95%95 \% CI 0.6 , 14.6 0.6 , 14.6 0.6,14.60.6,14.6 ) compared to those who walked day 1 postoperatively. Previously mobile patients from private residence had a reduced odds of return to private residence if they walked day 2-3 (OR 0.38; 95% CI 0.17, 0.87), day 4 + 4 + 4+4+ (OR 0.38 ; 95 % 95 % 95%95 \% CI 0.15 , 0.96 0.15 , 0.96 0.15,0.960.15,0.96 ), or if they only sat, stood or stepped on the spot day 1 (OR 0.29; 95% CI 0.13, 0.62) when compared to those who walked day 1 postoperatively. Among patients from private residence, each additional physiotherapy session per day was associated with a -2.2 ( 95 % Cl − 3.3 , − 1.0 95 % Cl − 3.3 , − 1.0 95%Cl-3.3,-1.095 \% \mathrm{Cl}-3.3,-1.0 ) day shorter acute LOS, and an increased log odds of return to private residence (OR 1.76; 95% CI 1.02, 3.02).
目的 确定三种术后物理治疗活动（首次术后行走时间、术后第一天活动情况以及物理治疗频率）与髋部骨折住院时长（LOS）和出院去向之间的关系。方法 对来自 2022 年 6 月澳大利亚和新西兰髋部骨折登记处急性康复快速审计的 36 家参与医院的 437 例髋部骨折手术患者进行队列研究。研究结局包括住院时长和出院去向。分别使用广义线性回归和逻辑回归模型，并调整潜在混杂因素。结果 在 437 例患者中， 62 % 62 % 62%62 \% 例为女性， 56 % 56 % 56%56 \% 例年龄为 ≥ 85 ≥ 85 >= 85\geq 85 岁， 23 % 23 % 23%23 \% 例此前居住在养老院， 48 % 48 % 48%48 \% 例通常使用助行器行走， 38 % 38 % 38%38 \% 例在受伤前存在认知障碍。急性期和总住院中位数为 8（IQR 5-13）和 20（IQR 8-38）天。约 71%（ n = 179 / 251 n = 179 / 251 n=179//251n=179 / 251 ）最初居住在私人住宅的患者返回家中， 29 % ( n = 72 / 251 ) 29 % ( n = 72 / 251 ) 29%(n=72//251)29 \%(n=72 / 251) 例被转至养老院。 先前能够活动的患者如果在术后第 2-3 天行走（10.3 天；95% CI 3.2, 17.4）或使用机械搬运器转移或术后第 1 天未下床活动（7.6 天； 95 % 95 % 95%95 \% CI 0.6 , 14.6 0.6 , 14.6 0.6,14.60.6,14.6 ），其总住院时间（LOS）会比术后第 1 天就行走的患者更长。来自私人住宅的患者，如果在术后第 2-3 天行走（OR 0.38；95% CI 0.17, 0.87）、第 4 + 4 + 4+4+ 天（OR 0.38； 95 % 95 % 95%95 \% CI 0.15 , 0.96 0.15 , 0.96 0.15,0.960.15,0.96 ），或仅坐、站或原地踏步（OR 0.29；95% CI 0.13, 0.62），与术后第 1 天就行走的患者相比，返回私人住宅的几率会降低。在来自私人住宅的患者中，每天每增加一次物理治疗与急性期住院时间缩短 2.2（ 95 % Cl − 3.3 , − 1.0 95 % Cl − 3.3 , − 1.0 95%Cl-3.3,-1.095 \% \mathrm{Cl}-3.3,-1.0 ）天相关，并增加了返回私人住宅的日志几率（OR 1.76；95% CI 1.02, 3.02）。

Conclusion Hip fracture patients who walked earlier, were more active day 1 postoperatively, and/or received a higher number of physiotherapy sessions were more likely to return home after a shorter LOS.
结论 早期行走、术后第 1 天更活跃以及/或接受更多物理治疗次数的髋部骨折患者，更有可能在更短的住院时间内返回家中。
Keywords Key performance indicator, Walking, Ambulation, Perioperative care, Recovery, Audit, Fracture neck of femur, Clinical quality registry, Orthogeriatric, Physiotherapy
关键词 关键绩效指标、行走、活动能力、围手术期护理、康复、审计、股骨颈骨折、临床质量登记、骨科老年医学、物理治疗

Introduction  引言

Hip fractures are a catastrophic injury for older people, responsible for substantial reductions in physical function, and high levels of morbidity and mortality [1]. Even in advanced health systems, approximately half of those with hip fractures do not regain their previous level of function and more than 10 % 10 % 10%10 \% require a change in residence to a residential aged care facility after hip fracture [ 2 − 5 ] [ 2 − 5 ] [2-5][2-5]. More than 1.66 million hip fractures are estimated to occur annually worldwide with projections that this number will rise to approximately 6 million fractures each year by 2050 [6]. Hip fractures also represent a considerable cost, estimated at over USD $ 43 , 669 $ 43 , 669 $43,669\$ 43,669 of health and social care costs per person in the 12 months following injury [7].
髋部骨折对老年人来说是一种灾难性的伤害，导致身体机能大幅下降，并造成高发病率和死亡率[1]。即使在先进的医疗体系中，大约有一半的髋部骨折患者无法恢复到之前的功能水平，并且超过 10 % 10 % 10%10 \% 的髋部骨折患者需要在骨折后 [ 2 − 5 ] [ 2 − 5 ] [2-5][2-5] 搬到一个养老院居住。据估计，全球每年发生超过 1.66 万例髋部骨折，预计到 2050 年这一数字将上升到每年大约 600 万例骨折[6]。髋部骨折也代表着相当大的成本，据估计，在受伤后的 12 个月内，每个人的医疗和社会护理成本超过 USD $ 43 , 669 $ 43 , 669 $43,669\$ 43,669 [7]。
One of the key goals of care after hip fracture surgery is to return to walking and the highest possible level of function. The first postoperative days are crucial for recovery, for example, early mobilisation within 1-2 days is recommended after surgery [8, 9] as is thought to accelerate functional recovery [10], is associated with a reduction in the likelihood of death while receiving inpatient hospital care [11, 12], and increases the likelihood of hospital discharge within 30-days postoperatively [13]. Higher frequencies of physiotherapy sessions (three times daily) in acute care have been shown to expedite functional recovery and reduce total hospital length of stay (LOS) by up to 10 days [14]. Furthermore, longer duration (greater than 2 h ) of physiotherapy in the first postoperative week is associated with discharge within 30-days [15], discharge home, survival, outdoor mobility recovery, and lower readmission rates [16].
髋部骨折手术后的康复的一个关键目标是恢复行走和尽可能高的功能水平。术后最初几天对康复至关重要，例如，手术后 1-2 天内进行早期活动被推荐[8, 9]，据认为这可以加速功能恢复[10]，与住院期间死亡风险降低相关[11, 12]，并增加术后 30 天内出院的可能性[13]。在急性期护理中，更高频率的物理治疗（每日三次）已被证明可以加速功能恢复，并将总住院时间（LOS）缩短最多 10 天[14]。此外，术后第一周内更长时间的物理治疗（超过 2 小时）与术后 30 天内出院、回家、存活、户外活动恢复以及再入院率降低相关[16]。
In Australia and the United Kingdom (UK), over 90% of hip fracture patients were offered the opportunity to mobilise by the day after surgery [17, 18], but less than 50 % 50 % 50%50 \% actually walked the day after surgery in Australia [17]. There is limited understanding of the barriers that may hinder the ability to improve rates of day 1 walking postoperatively, such as delirium and dementia, postural hypotension, postoperative anaemia, uncontrolled pain, drowsiness, and process and systems of care [19]. Furthermore, while some patients might not be able to mobilise the first day after surgery, they may succeed another day or achieve other types of activity apart from walking on day 1 (e.g. sitting or standing). It is not clear whether the benefits in hospital discharge outcomes from
在澳大利亚和英国（UK），超过 90%的髋部骨折患者术后第二天有机会活动[17, 18]，但在澳大利亚，实际上不到 50 % 50 % 50%50 \% 的患者术后第二天能行走[17]。对于可能阻碍术后第一天行走率提高的障碍因素，如谵妄和痴呆、体位性低血压、术后贫血、疼痛未控制、嗜睡以及护理流程和系统等，目前了解有限[19]。此外，虽然有些患者可能术后第一天无法活动，但他们在第二天可能成功，或除行走外还能进行其他类型的活动（例如坐着或站着）。目前尚不清楚住院出院结果中的这些益处是否来自
day 1 walking can be achieved via other types of activity, apart from walking. We sought to address these gaps in the literature by examining the impact of time to first postoperative walk, different levels of day 1 activity, and the frequency of physiotherapy on hospital discharge outcomes.
第 1 天的行走可以通过除行走以外的其他类型的活动来实现。我们试图通过检查术后首次行走的时间、第 1 天不同活动水平以及物理治疗频率对医院出院结果的影响来解决文献中的这些空白。

Aim  目的

To determine the relationship between three activities: 1) time to first postoperative walk, 2) activity on the day after surgery, and 3) physiotherapy frequency, and hospital LOS and discharge destination after hip fracture.
确定三种活动之间的关系：1）术后首次行走时间，2）术后当天活动情况，3）物理治疗频率，以及髋部骨折后的住院时间和出院目的地。

Methods  方法

Study design  研究设计

A cohort study was conducted from the ANZHFR Acute Rehabilitation Sprint Audit and is reported according to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. The ANZHFR is a clinical quality registry that collects information on patient demographics, care at presentation, and pre-, peri- and post-operative care, and mortality for 104 public and private hospitals across Australia and New Zealand that provide surgical management for people with a hip fracture [17]. In 2022, the ANZHFR conducted an acute rehabilitation sprint audit to collect additional variables over a defined period on acute rehabilitation practices for patients with hip fracture, with a specific focus on early mobilisation [20]. De-identified data from the audit were linked to routinely collected data in the ANZHFR for 437 surgically managed hip fracture patients across 36 hospitals that opted-in to the audit. In 2022, the ANZHFR included data from 16,395 individual patient records across 97 hospitals. Ethical approval was granted for the ANZHFR in each Australian state and in New Zealand for the ANZHFR and for the sprint audit (except Queensland due to state Public Health Act legislation requirements).
一项队列研究由 ANZHFR 急性康复冲刺审计开展，并根据流行病学观察性研究报告指南（STROBE）进行报告。ANZHFR 是一个临床质量登记系统，收集澳大利亚和新西兰 104 家提供髋部骨折手术管理的公立和私立医院的患者人口统计学信息、就诊时护理情况以及术前、围手术期和术后护理及死亡率信息[17]。2022 年，ANZHFR 开展了一次急性康复冲刺审计，在特定时间段内收集髋部骨折患者急性康复实践的额外变量，特别关注早期活动[20]。审计中的去识别数据与 ANZHFR 常规收集的数据链接，涉及 36 家选择参与审计的医院中的 437 名接受手术管理的髋部骨折患者。2022 年，ANZHFR 包括 97 家医院中 16,395 份个体患者记录的数据。 伦理批准已获得，适用于澳大利亚各州和新西兰的 ANZHFR，以及用于冲刺审计（除昆士兰州外，由于该州公共卫生法案立法要求）。

Data source and participants
数据来源和参与者

To be included in the ANZHFR, a patient must be aged ≥ 50 ≥ 50 >= 50\geq 50 years, with a hip fracture following a minimal mechanical trauma less than 14 days prior to presentation (including in-hospital fractures). All ANZHFR hospitals (excluding Queensland) were invited to voluntarily take part in the sprint audit. Recruitment of participating
要被纳入 ANZHFR，患者必须年满 ≥ 50 ≥ 50 >= 50\geq 50 岁，且在就诊前 14 天内因轻微机械创伤导致髋部骨折（包括医院内发生的骨折）。所有 ANZHFR 医院（除昆士兰州外）都被邀请自愿参与冲刺审计。参与者的招募
hospitals was supported through ANZHFR newsletters and direct invitations.
医院的支持通过 ANZHFR 通讯和直接邀请获得。
The protocol was developed in consultation with the Australian Physiotherapy Association, members of the ANZHFR Steering Group and Research Subcommittee, and external collaborators to ensure the data collection provided the most valuable contributions to understand existing care processes and identify opportunities for improvement. The protocol and dataset definitions were developed from a review of the UK’s 2017 physiotherapy ‘hip sprint’ audit [21], hip fracture guidelines [8], and the Australian Commission on Safety and Quality in Health Care’s National Safety and Quality Health Service Standards [9]. The audit questions were piloted at five hospitals before being added to the ANZHFR minimum data set for consecutive, eligible patients admitted from 1st to 30th of June 2022. Data were collected by healthcare providers who routinely collect ANZHFR data from hospital medical records at the participating hospital [22].
该方案是在咨询澳大利亚物理治疗协会、ANZHFR 指导小组和研究分委会成员以及外部合作者的基础上制定的，以确保数据收集能够为理解现有护理流程和识别改进机会提供最有价值的贡献。该方案和数据集定义是基于对英国 2017 年物理治疗“髋部冲刺”审计[21]、髋部骨折指南[8]以及澳大利亚医疗安全和质量委员会的国家安全和质量医疗服务标准[9]的回顾而制定的。审计问题在添加到 ANZHFR 最小数据集之前，已在五家医院进行了试点，该数据集用于收集 2022 年 6 月 1 日至 30 日期间连续入院的有资格患者的数据。数据由参与医院的医疗保健提供者收集，他们通常从医院病历中收集 ANZHFR 数据[22]。

Study exposures  研究暴露

The study exposure variables were: 1) first postoperative walk; 2) type of first day activity; and 3) number of physiotherapy sessions per day. First walk was defined as the first postoperative day the patient was able to walk or step transfer categorised as day 1 , day 2 or 3 , or day 4 + 4 + 4+4+. First day activity was defined as the type of activity achieved day 1 postoperatively, categorised as walk or step transfer, stood next to the bed or stepped/marched on the spot or sat on the edge of the bed, or transferred with mechanical lifter or no activity achieved. Physiotherapy sessions were defined as the total number of physiotherapist or allied health assistant sessions provided per day for up to seven days during the acute ward admission period. Walking was defined according to the ANZHFR mobilisation definition, as “the patient managed to stand and step transfer out of bed onto a chair/commode or walk. This does not include only sitting over the edge of the bed or standing up from the bed without stepping/walking” [23].
研究暴露变量包括：1）术后首次行走；2）首次日活动类型；3）每日物理治疗次数。首次行走定义为患者术后能够行走或进行步态转移的第一天，分为第 1 天、第 2 天或第 3 天，或第 0 天。首次日活动定义为术后第 1 天实现的活动类型，包括行走或步态转移、床边站立或原地踏步/行进或床沿坐位，或使用机械辅助器转移或未实现任何活动。物理治疗次数定义为在急性病房住院期间，每天提供的物理治疗师或辅助医疗人员治疗次数，持续最多七天。行走根据 ANZHFR 活动定义，定义为“患者能够站立并从床到椅子上/便器上行走。这不包括仅坐在床沿或从床上站起而不进行步态/行走”。[23]

Study outcomes  研究结局

The study outcomes included LOS and hospital discharge destination. Reasons for being unable to mobilise day 1 postoperatively were also explored. Hospital LOS was defined as acute (number of inpatient bed days on the acute ward) or total (entire inpatient hospital stay, including acute and subacute care). Discharge destination was defined as discharge to private residence or residential aged care facility/other discharge destination for people who were previously from a private residence. The most common reasons for being unable to mobilise were reported from the patient medical record. Patients whose usual place of residence was “residential aged care” or
研究结果表明包括住院时间和出院目的地。还探讨了术后第一天无法活动的原因。医院住院时间定义为急性（住院期间急性病房的床日数）或总（整个住院医院停留时间，包括急性期和亚急性期护理）。出院目的地定义为从私人住宅出院到私人住宅或养老院/其他出院目的地。无法活动最常见的原因报告来自患者病历。居住地通常为“养老院”或
“other”, those previously non-ambulant prior to their hospital admission and those experiencing in-hospital death were excluded from the discharge destination analysis. Patients who were previously non-ambulant prior to their hospital admission and those experiencing in-hospital death were excluded from the LOS analyses.
“其他”，那些在入院前无法行走以及在住院期间死亡的患者被排除在出院目的地分析之外。那些在入院前无法行走以及在住院期间死亡的患者被排除在住院时长（LOS）分析之外。

Potential confounders  潜在混杂因素

Potential confounders were entered into the analysis as covariates measured according to the ANZHFR [23], identified as those considered clinically relevant and where previous research has indicated an association with the study exposures or study outcomes. These included age ( 50 − 84 y 50 − 84 y 50-84 y50-84 y vs 85 y + 85 y + 85 y+85 y+ ), usual place of residence (private residence vs residential aged care facility/other), pre-admission walking ability (with or without an aid), pre-admission cognitive state (impaired or not impaired), ASA grade ( 1 / 2 , 3 1 / 2 , 3 1//2,31 / 2,3, or 4 ), time to surgery ( ≤ 48 h ≤ 48 h <= 48h\leq 48 \mathrm{~h} vs > 48 h > 48 h > 48h>48 \mathrm{~h} ), type of fracture (intra vs extracapsular), and type of anaesthesia (general, spinal/ regional, or general and spinal/regional) [19, 24]. Some variable categories were collapsed due to low patient numbers (e.g. ASA grade 1/2) and the age cutoff was used in recognition of potentially poorer outcomes for older adults aged 85 years and older [25].
潜在混杂因素作为协变量纳入分析，依据 ANZHFR [23]进行测量，这些混杂因素被确定为在临床上有意义，并且既往研究已表明其与研究暴露或研究结果存在关联。这些因素包括年龄（ 50 − 84 y 50 − 84 y 50-84 y50-84 y vs 85 y + 85 y + 85 y+85 y+ ）、通常居住地（私人住宅 vs 养老院/其他）、入院前行走能力（有或无辅助设备）、入院前认知状态（受损或未受损）、ASA 分级（ 1 / 2 , 3 1 / 2 , 3 1//2,31 / 2,3 或 4）、手术时间（ ≤ 48 h ≤ 48 h <= 48h\leq 48 \mathrm{~h} vs > 48 h > 48 h > 48h>48 \mathrm{~h} ）、骨折类型（关节内 vs 关节外）以及麻醉类型（全身麻醉、硬膜外/区域麻醉或全身麻醉和硬膜外/区域麻醉）[19, 24]。由于患者数量较少，某些变量类别被合并（例如 ASA 分级 1/2），并且对于 85 岁及以上的老年人可能存在较差的结果，因此使用了年龄截止点[25]。

Data analysis  数据分析

Patients with complete data for exposures, outcomes, and potential confounders were included in the analyses. Patient demographics were described using frequencies and percentages. Outcomes were summarised descriptively using frequencies and percentages for discharge destination and median and interquartile range (IQR) for LOS. Chi-square tests of independence were used to compare patient demographics for older adults with a hip fracture by day 1 mobilisation rates postoperatively. Generalised linear regression was used to calculate coefficients and 95% confidence intervals (CIs) for the association between each exposure and acute ward and total hospital LOS. Logistic regression was used to calculate odds ratios (OR) and 95% CIs for the association between each exposure and discharge destination. Both crude models (including exposure and outcome only) and adjusted models (include exposure, outcome, and all covariates) were applied to each exposure and analysed outcome. Models were checked for multicollinearity and specification and statistical significance was set at p ≤ 0.05 p ≤ 0.05 p <= 0.05\mathrm{p} \leq 0.05. All statistical analyses were undertaken using STATA (StataCorp. (2023). Stata Statistical Software: Release 18. College Station, TX: StataCorp LLC).
纳入了所有暴露因素、结果和潜在混杂因素数据完整的患者进行分析。患者人口统计学特征使用频率和百分比进行描述。出院去向的结果使用频率和百分比进行描述性总结，而住院时长（LOS）则使用中位数和四分位间距（IQR）进行总结。使用卡方检验比较了髋部骨折老年患者术后第 1 天活动率与患者人口统计学特征的关联性。采用广义线性回归计算每个暴露因素与急性病房和总住院时长之间的关联系数和 95%置信区间（CI）。使用逻辑回归计算每个暴露因素与出院去向之间的比值比（OR）和 95% CI。对每个暴露因素均应用了原始模型（仅包括暴露因素和结果）和调整模型（包括暴露因素、结果和所有协变量），并分析了结果。检查了模型的多重共线性，统计学显著性水平设定为 p ≤ 0.05 p ≤ 0.05 p <= 0.05\mathrm{p} \leq 0.05 。所有统计分析均使用 STATA（StataCorp. (2023).）进行。 Stata 统计软件：版本 18。德克萨斯州 College Station：StataCorp LLC）。
The number of physiotherapy sessions and discharge destination analysis model was not linear, so a logarithmic transformation was applied to the physiotherapy
物理治疗次数和出院目的地分析模型并非线性，因此对物理治疗
  1. *Correspondence:  *通讯：
    Mitchell Sarkies
    mitchell.sarkies@sydney.edu.au
    Full list of author information is available at the end of the article
    全文作者信息详见文末