Evaluation of factors influencing the reduction of home mechanical ventilation dependency in patients planned to receive home health care

Abstract

Background

With increasing life expectancy and the growing burden of chronic diseases, healthcare systems are shifting from hospital-centered models to community- and home-based care. The use of home mechanical ventilators (HMVs) facilitates early discharge from hospital settings yet creates significant physical, psychological, and social burdens for caregivers. Identifying strategies to reduce ventilator dependency during palliative inpatient care may alleviate these challenges.

Methods

This retrospective, cross-sectional, descriptive study employed a quantitative approach with purposive sampling. The study included 75 bedridden patients who were transferred from intensive care units (ICUs) to palliative care services with a planned transition to home healthcare. A multidisciplinary rehabilitation program encompassing respiratory, physical, nutritional, and psychological interventions was systematically implemented in the palliative care unit. Data on the patients’ functional and nutritional status, as well as pressure ulcer risk scores, were collected and analyzed to evaluate their association with successful weaning from HMV.

Results

Among the study population, 77.3% (n = 58) were successfully weaned off HMV. Significant predictors of ventilator weaning included higher Early Functional Ability (EFA) scores at admission, particularly in the sensorimotor and cognitive domains, as well as greater improvement in Karnofsky Performance Scores during the palliative stay. Improvements in nutritional and pressure ulcer risk scores were not significantly associated with ventilator weaning.

Conclusions

Multidisciplinary palliative care with targeted rehabilitation interventions may effectively reduce HMV dependency in patients transitioning from intensive care. A higher initial functional status and improvements in physical performance are key determinants of successful weaning, potentially lowering the burden on caregivers in home settings.

Introduction

Significant demographic and social changes have occurred worldwide in recent years. Among the most critical of these shifts is the rapid increase in the proportion of the global population aged 65 years and older, which is expected to result in a growing demand for long-term supportive care services. According to the United Nations’ 2023 world population estimates, the global population has reached approximately 8.05 billion, of which nearly 808 million are elderly, accounting for approximately 10% of the total population [1]. This demographic trend is driving a significant transformation in healthcare systems, shifting the focus from hospital-centered care to community- and home-based care models. As life expectancy increases, the burden of chronic illnesses also increases, leading to an urgent need for new healthcare modalities. Conditions such as chronic diseases, age-related health deterioration, and respiratory failure have brought the concept of home healthcare to the forefront because many patients require medical support even after discharge from hospitals [2, 3].

As in other parts of the world, home healthcare services have expanded significantly in Turkey, profoundly impacting patient care processes and the roles of family caregivers [4]. In particular, HMVs within home healthcare facilitate hospital discharge for patients and impose substantial physical, psychological, and social burdens on caregivers [5, 6].

Patients dependent on medical devices, especially those requiring mechanical ventilation, present additional complexities in the caregiving process. Caregivers manage patients’ respiratory support and general care, operate medical equipment, handle technical malfunctions, and respond to emergencies. Lack of knowledge, limited psychosocial support, and barriers to healthcare services can exacerbate caregiving burden [6, 7].

In contrast to those used in ICUs, home-type mechanical ventilators are more compact and portable, allowing respiratory support to continue at home. However, these devices require regular monitoring, maintenance, and proper usage tailored to the patient’s condition, all of which place considerable responsibility on the caregivers. In addition to technical challenges, the use of ventilators at home may lead to increased anxiety, burnout, and psychological distress, negatively affecting the quality of life of caregivers [3, 6, 8, 9]. Notably, there is a growing body of research on the mental well-being of patients using HMVs, with device dependency often linked to depressive symptoms and emotional distress [9, 10].

The present study aimed to identify the factors that may reduce HMV dependency in bedridden patients who are transferred from intensive care units to palliative care services and are scheduled to receive follow-up care at home. By examining critical considerations during the inpatient palliative care phase, this study sought to determine the relative impact of these factors on ventilator weaning, while also evaluating the role and effectiveness of multidisciplinary interventions in this process.

Methods

This was a retrospective, cross-sectional, and descriptive quantitative study. The study population consisted of patients hospitalized in the palliative care unit of Tekirdağ Dr. İsmail Fehmi Cumalıoğlu City Hospital in 2024. Interventions were recorded systematically by the healthcare team for each patient as part of routine clinical follow-up. The research was conducted in accordance with the principles outlined in the Declaration of Helsinki. Verbal consent was obtained from all participants (or legal heir).

Definition, scope and purposes of palliative care units

Palliative care in Turkey, guided by Ministry of Health regulations, now includes 437 hospital-based and home care centers. These units offer multidisciplinary management to alleviate suffering, enhance quality of life, and support families, although access remains uneven across regions. In this study, the unit is defined as providing support programs—primarily respiratory rehabilitation—for patients who have completed intensive care treatment but are unable to be discharged home.

Patients selection

The study included patients who underwent tracheostomy and received invasive mechanical ventilation for the first time during their intensive care unit stay. Patients with a prior history of ventilator use and those who had received non-invasive mechanical ventilation or HMV were excluded. After applying the exclusion criteria, 75 eligible patients were enrolled in the study. All patients were fully dependent at the time of transfer.

Transfer criteria from the ICUs to the palliative care unit

Patients who had completed intensive care treatment but were not well enough to be discharged home were transferred to our palliative unit on a HMV because they still required respiratory support.

Patients were transferred to the palliative care unit when they met the following criteria:

• Hemodynamically stable without vasoactive support,

• No acute organ failure requiring ICU-level monitoring,

• Presence of potential for weaning from HMV as assessed by the ICU team,

Assessment tools and rehabilitation interventions used in the study

Early Functional Ability (EFA) scale: This scale assesses vegetative, orofacial, sensorimotor, and cognitive domains. Although it was validated for early neurological rehabilitation, it was adopted here to evaluate functional status in patients discharged from the ICU [11].

Nutritional risk Screening-2002 (NRS-2002): evaluates nutritional risk and malnutrition [12]. The Turkish validity and reliability of the NRS have been established by Bolayır et al. [13].

Karnofsky performance scale: measures overall performance and dependency [14],

Braden pressure ulcer risk scale: assesses risk of skin breakdown and ulcer development [15].

The weaning protocol: According to the (m-BWAP) scoring system, weaning is planned for patients with 7 points and above. The m-BWAP system comprises a scoring system based on indicators of respiratory function, hemodynamic stability, and conscious examination, which determines whether the condition causing respiratory failure is treated [16].

Rehabilitation interventions: During the study, the interventions outlined in Fig. 1 were carried out by the responsible personnel at the designated frequencies. Each intervention was tailored to the patient’s clinical condition, functional status, and recommendations from the multidisciplinary team. Data were obtained retrospectively from patient records and routine documentation by the care team. The frequency of each intervention was calculated as a relative proportion of the total number of interventions applied. Interventions were then grouped according to professional responsibility. Results were presented descriptively, with the intensity of use illustrated in a frequency distribution chart.

Fig. 1.

Rehabilitation interventions in the palliative care unit

Definition of ventilator dependence

Ventilator dependence was defined as the inability to sustain spontaneous breathing for 24 consecutive hours or more without mechanical support. In this study, at the time of palliative care unit transfer, all patients required invasive or non-invasive HMV. Patients with tracheostomy were managed on invasive ventilation, while others were supported with non-invasive ventilation.

Statistical analysis

All statistical analyses were performed using SPSS and Jamovi software. Descriptive statistics were used to characterize the study population. The Mann-Whitney U test and logistic regression analysis were used to assess differences in scale scores and independent variables influencing HMV discontinuation. The effect sizes were calculated using Cohen’s d, and the results are presented in tables. The level of statistical significance was considered as p < 0.05.

Results

A total of 75 patients who were transferred from the ICUs to the palliative care unit and were dependent on HMV were included in the study. The mean age of the population was 60.73 ± 16.62 years (range: 18–92 years), and 68% (n = 51) were male. The mean length of stay in the palliative care unit was 29.25 ± 20.82 days (range: 7–98). There is no statistically significant difference between the groups in terms of age, gender, comorbidity, and duration of stay in the palliative unit (p > 0.05 in all. During inpatient follow-up, 77.33% of the patients (n = 58) were successfully weaned from HMV, while 22.67% (n = 17) were discharged with continued ventilator dependency and referred to home care services. Comorbid conditions were present in the majority of patients, with only five individuals having no additional diagnoses. Among these, neurological diseases (such as stroke) were the most common, followed by cardiovascular conditions (including previous myocardial infarction and coronary artery disease) and respiratory disorders (Table 1).

Table 1.

Demographic characteristics of patients

Variable	Category	Weaned (n = 58) %*	Non-weaned (n = 17) %*	Total (n = 75) %*	p
Age (years)	Mean ± SD	59.1 ± 17.9	66.3 ± 10.4	60.73 ± 16.7	0.856
Gender	Male	42 (72.4%)	9(52.9%)	51(68.0%)	0.318
	Female	16 (27.6%)	8(47.0%)	24(32.0%)
Length of stay in the palliative care unit(days)		33.2 ± 22.0	15.59 ± 4.8	29.25 ± 20.82	0.122
Comorbidity	Yes	53 (91.3%)	17(100.0%)	70(93.3%)	0.673
	No	5 (8.6%)	0	5(6.6%)
Primary disease groups	Neurological	39 (67.2%)	7(41.1%)	46(61.3%)
	Cardiovascular	7(12.0%)	4(23.5%)	11(14.6%)
	Respiratory	7(12.0%)	3(17.6%)	10(17.3%)
	No comorbidity	5(8.6%)	-	5(6.6%)

*Column percentage(s)

Upon admission to the palliative care unit from the ICUs, all patients were assessed using the Early Functional Ability (EFA) scale. The Mann-Whitney U test revealed statistically significant differences (p < 0.001) across all EFA criteria between patients who were successfully weaned from HMV (n = 58) and those who remained ventilator-dependent (n = 17). Effect size analyses, using Cohen’s d, demonstrated moderate to large effects, with all EFA subscale values exceeding 0.5. Significant differences were observed between the two groups in the total EFA scores across the four subdomains of vegetative, orofacial, sensorimotor, and cognitive functions. The mean total EFA score for the non-weaned group was 30.29, compared to 50.10 for the weaned group. The corresponding Cohen’s d value is 0.74, indicating a moderate effect size. This suggests that the initial functional status, as assessed by the EFA scale, was a meaningful predictor of ventilator weaning success (Table 2). All subdomains of the EFA scale effectively distinguished patients who were successfully weaned from those who remained dependent on the HMV. At the time of transfer from the ICUs to palliative care service, the vegetative and orofacial subdomain scores demonstrated moderate effect sizes for ventilator weaning (Cohen’s d = 0.55). In contrast, sensorimotor and cognitive scores exhibited large effect sizes (Cohen’s d = 0.71 and 0.68, respectively), further emphasizing the predictive value of initial functional capacity in determining weaning outcomes (Table 2).

Table 2.

The effect of total EFA and subdomain scores on weaning from home mechanical ventilation

EFA Measure	Weaning Status	Mean ± SD	Min–Max	Shapiro-Wilk (p)	Mann-Whitney U (p)	Cohen’s d
EFA Total Score	Not Weaned (0)	30.29 ± 8.81	20–54	0.002	< 0.001*	0.74
	Weaned (1)	50.10 ± 16.00	20–82
Vegetative Function	Not Weaned (0)	5.76 ± 2.02	4–9	0.002	< 0.001*	0.55
	Weaned (1)	9.86 ± 4.53	4–18
Orofacial Function	Not Weaned (0)	7.41 ± 3.12	4–15	0.005	< 0.001*	0.55
	Weaned (1)	11.14 ± 4.36	4–20
Sensorimotor Function	Not Weaned (0)	8.06 ± 1.89	6–12	0.021	< 0.001*	0.71
	Weaned (1)	13.95 ± 5.05	6–30
Cognitive Function	Not Weaned (0)	9.12 ± 5.01	5–23	0.027	< 0.001*	0.68
	Weaned (1)	15.14 ± 4.61	5–25

Among the 75 patients NRS score at admission to the palliative care unit was 5.2. Following the treatment interventions, the discharge NRS score decreased to 3.47, indicating an average reduction in nutritional risk of 33.27%. Table 3 presents the nutritional improvement rates between patients who were successfully weaned from HMV [1] and those who were not (0). At the initial admission, the mean Karnofsky Performance Scale score for the study population was 20.4. Upon discharge, this value increased to 33.6, representing an approximate 13% improvement. Table 2 compares the performance scores of weaned and non-weaned patients. Regarding the pressure ulcer risk, the average Braden score at admission was 11.95. By discharge, this score had improved to 13.66, shifting from the high-risk to the moderate-risk category, with an overall improvement of 9.40%. Table 3 details the differences in Braden scores between ventilator-weaned and non-weaned patients.

Table 3.

NRS, Karnofsky, and Braden scores at admission and discharge by ventilator weaning status

Scale	Time Point	Weaning Status	Mean ± SD	Min	Max
NRS	Admission	Not Weaned (0)	5.06 ± 0.43	4	6
		Weaned (1)	5.24 ± 0.76	4	7
	Discharge	Not Weaned (0)	3.59 ± 0.87	2	5
		Weaned (1)	3.28 ± 0.81	2	5
Karnofsky	Admission	Not Weaned (0)	11.76 ± 3.93	10	20
		Weaned (1)	22.93 ± 7.49	10	40
	Discharge	Not Weaned (0)	19.41 ± 10.88	0	40
		Weaned (1)	37.76 ± 14.99	0	80
Braden	Admission	Not Weaned (0)	11.12 ± 2.03	8	15
		Weaned (1)	12.19 ± 2.00	9	18
	Discharge	Not Weaned (0)	12.76 ± 2.84	8	17
		Weaned (1)	13.93 ± 2.63	9	18

The effectiveness of improvements in nutritional risk (NRS), functional performance (Karnofsky), and pressure ulcer risk (Braden) during palliative care was evaluated based on the patient’s ability to be weaned from home mechanical ventilation. Among these factors, only improvement in the Karnofsky performance status was significantly associated with successful ventilator weaning. The effect size for this association, as measured by Cohen’s d, was 0.63, indicating a strong effect (Table 4).

Table 4.

Improvement percentages and effect sizes for NRS, Karnofsky, and Braden scores about HMV weaning status

Scale	Weaning Status	Improvement (%) Mean ± SD	Shapiro-Wilk (p)	Mann-Whitney U (p)	Cohen’s d
NRS	Not Weaned (0)	28.43 ± 19.15	0.107	0.090	0.24
	Weaned (1)	36.55 ± 16.57
Karnofsky	Not Weaned (0)	7.65 ± 10.91	0.007	0.031	0.63
	Weaned (1)	14.83 ± 11.43
Braden	Not Weaned (0)	9.15 ± 9.61	0.017	0.907	0.014
	Weaned (1)	9.47 ± 10.07

Logistic regression analysis conducted on the full dataset revealed that the overall model was statistically significant, with all independent variables contributing significantly (χ²(39) = 80.283, p < 0.001). This indicates that the logistic regression model provides a statistically significant improvement over the null model. According to the model summary, the − 2 Log Likelihood value was 0.000, suggesting an extremely close fit. The Cox and Snell R² values were 0.657 and 0.952, respectively, indicating excellent explanatory power. However, the analysis identified a perfect fit condition, which raises concerns regarding potential overfitting and limited generalizability. The Hosmer-Lemeshow goodness-of-fit test yielded non-significant results (χ² [2] = 0.000, p = 1.000), further supporting the model fit and confirming the absence of overfitting. A revised logistic regression model was developed using only the variables showing clinical relevance and variability: changes in the EFA total and subdomain scores, Karnofsky improvement percentage, NRS score, and Braden score. This revised model showed a Nagelkerke R² of 0.563, indicating that approximately 56% of the variance in ventilator weaning status could be explained at an acceptable level for clinical predictive modeling.

According to the regression analysis, total EFA score, vegetative function, orofacial function, sensorimotor function, cognitive function, and percentage improvement in Karnofsky performance were significant predictors of ventilator weaning. These predictors substantially contributed to model accuracy and played a crucial role in explaining the outcome (Table 5).

Table 5.

Logistic regression analysis of predictors for successful weaning from HMV

Predictor Variable	Wald χ²	df	p-value
Total EFA Score (20–100)	18.434	1	< 0.001
Vegetative Function (4–20)	11.364	1	< 0.001
Orofacial Function (4–20)	9.630	1	0.002
Sensorimotor Function (7–35)	17.415	1	< 0.001
Cognitive Function (5–25)	17.089	1	< 0.001
Karnofsky Improvement Percentage (%)	5.068	1	0.024
NRS Improvement Percentage (%)	2.907	1	0.088
Braden Improvement Percentage (%)	0.014	1	0.905

Discussion

This study presents the outcomes of a successful multidisciplinary intervention program designed to facilitate liberation from mechanical ventilation in patients who represent a significant economic burden to national healthcare systems. A key finding was the discussion of how certain assessment scales, when reinterpreted, can provide more practical value in routine clinical practice. The findings demonstrated that the EFA total score, reflecting the patient’s overall status upon transfer from the ICU, including the vegetative, orofacial, sensorimotor, and cognitive domains, was a significant predictor of successful weaning from HMV. In particular, a higher sensorimotor capacity and mental function are associated with an increased likelihood of successful weaning from the ventilator. Among the variables examined, improvement in Karnofsky performance status during palliative care had a direct and substantial effect on ventilator weaning. This finding highlights the critical role of physical rehabilitation in regaining independence from mechanical respiratory support.

The increasing occupancy rates in ICUs and challenges in managing patients with prolonged mechanical ventilation needs have prompted healthcare providers to adopt home ventilator strategies. This shift enables earlier hospital discharge and transition to home care but also introduces significant burdens for caregivers who must manage respiratory support, either invasively or non-invasively, in the home setting. Palm et al. emphasized the growing economic burden of HMV and recommended that future research focus on the impact of HMV on patient survival and healthcare costs [17].

The use of HMV not only imposes significant caregiving and financial burdens but also increases the risk of medical complications. In a study by Yanaz et al., involving 416 pediatric patients receiving invasive or noninvasive ventilation at home, approximately 50% were managed with HMV, and 60% required hospital readmission within the past year [18]. Ütebey stated that among those discharged patients with tracheostomy, 46.8% had an unplanned hospital readmission within 30 days of discharge, underscoring the vulnerability of this patient group [19]. Another major challenge faced by patients discharged with HMV is the lack of structured and accessible home healthcare services. In a study conducted by Song et al. most patients lack access to organized home healthcare, leading to considerable distress for family caregivers [20].

There is a growing impetus to conduct studies that define multidisciplinary interventions and indicators for successful ventilator weaning, as such efforts are crucial to reducing the economic and psychosocial burden of dependency and to guiding evidence-based clinical practice. In the multidisciplinary rehabilitation process that begins with palliative ICU transfer, therapeutic interventions should be prioritized using appropriate functional assessment tools. Boltzmann et al. emphasized that functional status is significantly associated with prognosis in patients with severe brain injury, and recommended the use of the EFA scale in clinical decision-making [21]. Similarly, in our unit, EFA is routinely used at admission to evaluate the general status and prognosis of patients with HMV dependency. This study demonstrated that higher initial EFA scores, particularly in the sensorimotor and cognitive domains, are significantly associated with successful weaning from ventilator support. Several studies have investigated the association between functional status and successful weaning. Schünemann et al. reported that functional improvement is critical for reducing ventilator dependency in palliative and home care settings [22]. Other studies have shown that patients with higher functional capacity are more likely to achieve successful weaning, even when using alternative scales such as the Functional Independence Measure (FIM) [23]. It should be advised to prioritize early functional assessments to identify candidates for active weaning programs.

Nutritional status is another critical concern in patients with long-term HMV use [8, 9, 24]. In a study involving 744 ventilator-dependent patients, improvements in nutritional parameters are correlated with enhanced lung function [25]. Early recognition of nutritional risks and the implementation of targeted nutritional strategies have been shown to improve outcomes [26, 27]. Based on these findings, we hypothesized that improved NRS-2002 scores would contribute to ventilator liberation. However, contrary to existing literature, our findings do not demonstrate a direct effect. We believe that continued monitoring of the NRS parameter after ventilator liberation would yield more meaningful results.

Pressure ulcers are a common complication in HMV-dependent patients due to prolonged immobility and restrictions imposed by the ventilator equipment. In our unit, Braden scores were assessed upon admission and monitored on a weekly basis. Pressure ulcers not only increase mortality and prolong hospitalization but also place a substantial burden on both patients and caregivers and escalate healthcare costs. A Braden score ≤ 16 is directly associated with pressure ulcer development [28, 29]. This tool is widely used in similar patient groups for risk stratification and care planning [30]. In our study, Braden scores ranged from 9 to 10, categorizing patients as being in the moderate-risk category. Although the pressure ulcer risk status was monitored, it did not significantly influence the weaning outcomes. This is likely because the Braden Scale is a risk prediction tool rather than a measure of recovery. In our setting, proactive interventions may have maintained stable scores, thereby limiting observable improvements. We suggest that alternative outcome-based wound assessment tools may be helpful in evaluating the effects of pressure ulcer healing on ventilator weaning.

In our daily practice, the Karnofsky score is used more frequently. However, as demonstrated in our study, improvement in the Karnofsky score represents a stronger indicator of well-being. Therefore, we recommend that patients’ scores be assessed at initial admission and subsequently at intervals determined by multidisciplinary teams. In this way, more realistic projections can be made regarding the future for both patients and their relatives. Therefore, we recommend that upon admission to palliative care, patients with HMV dependency be enrolled in a structured rehabilitation program under the supervision of a physiotherapist, aiming to improve Karnofsky performance indicators as part of the weaning strategy.

This study has several limitations. It is based on one-year retrospective data from a single center with a relatively small sample size, which may limit generalizability. Multicenter follow-up studies that incorporate data from diverse patient groups are necessary. On the other hand, frailty assessment was not performed, although frailty is known to be a significant determinant of outcomes in critically ill and chronically ventilated patients. The lack of a validity study of the EFA scale in this patient group also represents an important limitation of our study.

Conclusion

HMV poses a significant healthcare burden for patients and their family caregivers. This burden is particularly pronounced in patients transferred from ICUs to palliative care services with ongoing ventilator dependency. Therefore, it is essential to implement a multidisciplinary approach and structured rehabilitation program during the palliative care stay before hospital discharge to facilitate weaning from HMV.

In our study, a structured rehabilitation program was implemented by a multidisciplinary team. Not only were scales administered, but the data were also processed to derive more useful parameters for predicting patients’ prognosis. Therefore, in line with the principle that “There are no diseases, only patients”, we recommend that clinics develop follow-up parameters tailored to the profiles of the patients they admit.

Author’ contributions

All authors have made substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data, and have approved the submitted version. Conceptualization A.C.B. G.K.G.Ö; Methodology, A.C.B., G.K.G.Ö.; Formal Analysis G.K; Investigation, A.C.B., G.K.G.Ö. Data Curation, G.Ö, G.K. Writing – Original Draft Preparation, A.C.B., G.K.G.Ö; Writing – Review & Editing A.C.B., G.K.G.Ö.

Funding

None.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

The ethics committee (Süleyman Demirel University ethic committee, No 94/33) granted ethical approval. The research was conducted in accordance with the Declaration of Helsinki. Written informed and verbal consent was obtained from all participants.

Consent for publication

and authors’ contributions: All authors have made substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data, and have approved the submitted version. Conceptualization A.C.B. G.K.G.Ö; Methodology, A.C.B., G.K.G.Ö.; Formal Analysis G.K; Investigation, A.C.B., G.K.G.Ö. Data Curation, G.Ö, G.K. Writing – Original Draft Preparation, A.C.B., G.K.G.Ö; Writing – Review & Editing A.C.B., G.K.G.Ö.

Competing interests

The authors declare no competing interests.