Understanding home ventilation patients' chosen place for their final moments

Year Submitted: 2025
Published Date: 23 January 2026

Tags: Applied/Translational Research, Quantitative Research, Care Continuum, End-Of-Life Care

About this Content

Aims

Our study aims to identify patient factors that are associated with different Place of Death (POD) preference among Home Mechanical Ventilation (HMV) patients.

Background

Advance Care Planning (ACP) documents an individuals healthcare preferences, including preferred place of death (POD) (HealthHub, 2021). Hospital deaths often require more medical resources than home or hospice death (Hyun et al., 2013), hence identifying the preferred POD can pinpoint gaps in our health system and encourage death at home. One group of patients that undertake ACP are Home Mechnical Ventilation (HMV) patients.

HMV can improve symptoms in patients with motor neurone disease and prolong their survival but does not prevent disease progression (Wilson et al., 2024). One study examined the complexity and evolution of HMV patients end-of-life decision-making, highlighting the importance of support from healthcare professionals in the decision-making process (Wilson et al., 2024). Despite HMV posing a significant care and financial burden on patients and the health system (Tan et al., 2019), to our knowledge, there is little information on the patient factors that may influence the preferred POD for patients on HMV. Studies which have identified sociodemographic factors, patient experiences with illness and care, and access to resources as factors associated with POD preferences pertained mostly to patients with terminal cancer (Gomes et al., 2012; Thomas et al., 2004; Wales et al., 2018). Our study aims to identify patient factors that are associated with different POD preference among HMV patients.

Methods

A retrospective cohort study was conducted. Data from decedents of HVRSS enrolled between 2009-2018 were extracted from an existing database. Decedents enrolled in HVRSS but did not complete their Advance Care Planning (ACP) were excluded. Preferred POD information available from ACPs of included subjects was categorised into preferring home death or healthcare institution death. Subject characteristics examined were selected from previous literature (Gomes et al., 2012; Thomas et al., 2004; Wales et al., 2018), and included: Sociodemographic factors (e.g. gender, race, marital status, age of enrolment and age of death) ; experiences with illness and care (e.g. type of diagnosis, number of comorbidities and hospitalisation within 3 months prior to death); access to palliative care, medical social worker involvement, having paid caregiver, the number of paid caregiver available and the number of family living in the same house as the subject). STATA was used for the statistical analysis. Univariate statistical analysis was conducted to examine the relationships among the subject characteristics. Statistically significant characteristics were defined as having p-value less than 0.1 and are retained for multivariate analysis. Multivariate logistic regression was conducted for the retained characteristics to examine potential associations between these characteristics and preferred POD. Statistically significant characteristics were defined as having p-value less than 0.05. Patients considerations and concerns relating to their POD preferences were also documented in the ACP and informed the interpretation of the quantitative results.

Results

A total of 52 HVRSS subjects were included in this study, with 38 subjects preferring home death and 14 subjects who indicated healthcare institution/s as their preferred POD. Overall, most subjects were male (n = 33, 63.5%), aged 59.5 years (SD: 2.5 years) on average at enrolment. The enrolment age was slightly higher in those who preferred home death (60.7 years, SD: 17.2 years) as compared to those who preferred healthcare institution death 56.0 years (SD: 19.8 years). Other subject characteristics examined are shown in table 1. From the univariate analysis, only the number of hospitalisation within 3 months prior to death and access to palliative care had significant differences between subjects preferring home death and preferring healthcare institution death.

Multivariate logistic regression conducted for selected characteristics suggested that only the number of hospitalisation within 3 months prior to death was significantly associated with preferred POD (Table 1). An increase in hospitalisations was associated with 70% lower odds of preferring home death (OR: 0.30, p = 0.048, 95% CI: 0.09 0.99).

Conclusion

Our findings can help healthcare professionals and policymakers better understand how peoples experiences of illness and care, both at home and in hospital, influence end-of-life choices. Our study addresses an evidence gap for patients on HMV, which is a group of patients with very limited existing evidence on their end-of-life choices. The long study time frame improves reliability as it is less affected by short-term fluctuations, indicating that the ACP choices made truly reflect patients choices. The limitations of our study includes a small sample size, which can limit the number of factors adjusted for in the multivariate logistic regression. Our analysis is also limited by characteristic selection and may not be all encompassing. The findings are also not generalizable to other types of patients given the focus on patients on HMV. Future work can examine the barriers families and caregivers face in caring for patients on HMV, with implications for the latters POD; and how healthcare providers could ameliorate or overcome the barriers.

Lessons Learnt

We found that increased hospitalisation is associated with a lower likelihood of preferring home death. This finding could reflect greater disease burden, suggesting that current HMV treatment might not be sufficient for a comfortable death at home. Family members may not be confident enough in supporting their loved ones in their final days at home, highlighting the need for greater support for HMV patients and their caregivers.

Additional Information

Singapore Health and Biomedical Congress (SHBC) 2025. Student Awards (Open Category) Undergraduate: Bronze Award

Keywords

End-of-life care, Hospitalisation

Innovators' Details

Connect with this contributor!

Ji Shuo - ji_shuo@u.nus.edu

Project Attachment

Ji Shuo SHBC item 1 revised 23092025.pdf