Opinion: 'Nowhere to Die' Is Not a Medical Issue—Taiwan Needs Public Options for Peaceful Homes

When a person is no longer suitable for continued hospitalization, yet has no home where they can be cared for, where exactly should they go to complete their final journey in life? 'Wishing to pass away at home' is often seen as the warmest vision of palliative care. However, for individuals living alone, elderly couples caring for each other, those with unstable rental housing, property-less individuals, the homeless, and economically disadvantaged people, this phrase is sometimes not an option, but a luxury. They may not lack medical assessments or willingness to be cared for; what they truly lack is a place where they can rest assured, be accompanied, access professional medical care, and not be excluded due to inability to afford accommodation and living costs. As a result, patients may shuttle back and forth between acute care beds, emergency rooms, temporary shelters, and unsuitable living environments. This is not merely a medical issue, but a public policy gap woven from housing, care, poverty, and patient autonomy. Taiwan’s end-of-life care system is still missing the 'housing' piece. Taiwan is not without institutional foundations for palliative and autonomous medical care. The Hospice Palliative Care Act protects the medical wishes of terminally ill patients; the Patient Autonomy Rights Act also allows patients meeting specific clinical criteria to legally refuse life-sustaining treatments, artificial nutrition, and fluid feeding. The importance of these systems lies in gradually returning the decision-making power over 'how to receive medical care' to patients themselves. However, medical autonomy does not equate to having a stable living situation. The current payment structure for home-based palliative care focuses on medical staff visits and care services. If a terminally ill patient lacks suitable housing, has no available caregivers to take turns, or their rental environment cannot accommodate care, home-based palliative care easily stops at the edge of policy design. We cannot simultaneously encourage aging in place and dying at home, while assuming that everyone has a room, family members, sufficient financial capacity, and caregiving resources. More bluntly, if a peaceful death is only accessible to 'those who own property' or 'those capable of caregiving,' it is not a complete public policy. Anning House is not another institution—it fills the middle ground. Therefore, policy must consider the option of 'Anning House': small-scale, homelike community spaces that provide meals, daily living support, and companionship; medical, nursing, pharmaceutical, and long-term care services would be connected through qualified home medical care, home nursing, and related professional teams. It is not about covertly turning private homes into nursing homes, nor replacing safety with lower standards. Rather, it aims to build a currently non-existent path that balances safety, dignity, and a sense of daily life. Such a system design must honestly face regulatory realities. According to current nursing facility classification and setup standards, home nursing stations and nursing homes that provide residential, full-time care have different service orientations. Small spaces involving residential care may also face challenges in building use classification and fire safety requirements. Therefore, Anning House cannot simply launch with a slogan of 'regulatory relaxation.' The correct approach should be to pilot legally, review case by case, establish performance-based safety and alternative evacuation plans, and transform institutional barriers discovered during the pilot into the basis for future legal amendments and setup standards. This is a slower, yet more responsible path. Landlords’ concerns cannot be addressed by moral persuasion alone. Another barrier to implementing Anning House is housing availability. Many landlords still feel psychological pressure, leasing concerns, and uncertainty about post-occupancy repair responsibilities when someone dies naturally in their property. If the government merely calls for 'friendly landlords' without offering standard contracts, repair responsibility sharing, dispute resolution mechanisms, tax incentives, or risk-sharing frameworks, goodwill may not last. The current Residential Act already provides institutional tools for authorities to lease private housing, sublet it, manage it on behalf of owners, and utilize existing housing resources. This offers a direction for deeper development: using public housing, available social housing units, and vacant private homes as a foundation to build a transparent matching platform. For willing landlords, clear contracts, reasonable repair support, and approved risk-sharing mechanisms can reduce anxiety. The key is not to demand that landlords 'be more compassionate,' but to ensure that every willing participant knows where the boundaries of responsibility lie, what the government will bear, and how disputes will be resolved. The responsibility of public policy is precisely to transform individual goodwill into a trustworthy system. Hospitals must be able to discharge patients—only then can homes receive them. 'Nowhere to die' often manifests as a referral breakdown. When patients are discharged from hospitals, they need more than just a referral slip—they need secure handover of medical summaries, medication information, symptom management plans, care goals, emergency contacts, and patient wishes. If this information is interrupted between hospitals, homes, home medical care, and long-term care services, the patients and their families will bear the risk first. Therefore, Anning Houses should establish a 'green channel' with hospitals’ discharge planning services and palliative care teams: with patient consent and personal data protection, dedicated staff should complete pre-admission assessments and confirm arrangements with home physicians, home nurses, pharmacists, long-term care providers, and emergency transfers. This is not about beautifying administrative procedures, but about preventing terminally ill patients from being forced back into emergency rooms or hospital wards during their most vulnerable moments due to information gaps. Equally important is clear human resource division. The value of palliative companions and community volunteers lies in 'being present'—assisting with daily life, listening, supporting families, observing, and reporting promptly—not in replacing doctors, nurses, or pharmacists. Existing home-based palliative care already sets training requirements for different types of professionals; new systems should strengthen interdisciplinary collaboration and supervision, rather than creating a group of caregivers with unclear responsibilities forced to overstep boundaries. Discussing 'death by fasting' and 'tube-free dying' requires upholding both autonomy and professionalism. Anning Houses should not avoid societal discussions on 'death by fasting' and 'tube-free dying,' but must not simplify complex ethical issues into slogans. When patients face choices regarding artificial nutrition, fluid feeding, or life-sustaining treatments, the core is not 'whether to accelerate death,' but whether the patient, with full information, decision-making capacity, and free from undue pressure, makes a choice aligned with their values through legal and professional procedures. The Taiwan Hospice Palliative Care Association has pointed out that a peaceful death does not need to take 'fasting' as its goal; for terminally ill patients, comfort care, symptom relief, and patient wishes should be prioritized, and different situations—such as natural reduction of food intake, stopping artificial nutrition, and intentional refusal of food—must be clearly distinguished. This reminds us: any health education must simultaneously explain applicable conditions, medical and ethical evaluations, informed consent or advance directives, and the patient’s right to change their mind at any time. True respect for autonomy does not push people toward a particular choice; it ensures that even in vulnerability, they retain the rights to understand, choose, withdraw, and be properly cared for. Starting with six sites and thirty beds—making peaceful death no longer a privilege tied to assets. Anning Houses do not need to launch on a massive scale. The author suggests using the years 2027 to 2031 (Minguo 116 to 120) as a five-year pilot period, starting with a cautious scale of six sites and thirty beds across different regions. Then, annual service numbers, proportion of vulnerable users served, referral efficiency, satisfaction, safety incidents, and financial transparency should be used to evaluate progress—this is more feasible than promising large-scale expansion from the start. These goals are pilot indicators of policy design and must undergo public scrutiny and iterative adjustments, not just be numbers set for appearance’s sake. More importantly, financial principles must be clear: National Health Insurance payments should return to medical services; long-term care payments should be based on assessments and actual services provided. Costs for accommodation, living support, facility repairs, subsidies for vulnerable groups, platform operations, and training should be jointly borne by public budgets, social welfare, and transparent public-private collaboration. Public donations can make spaces warmer and companionship more diverse, but they must not become the sole prerequisite for vulnerable individuals to access basic end-of-life support. A mature society’s capacity for care should be judged not only by its ability to extend life, but also by whether, at life’s end, no one is deprived of proper care due to poverty, loneliness, or lack of housing. The significance of Anning House is not to add another institution, but to fill a truly choosable public option between 'hospital' and 'a home they cannot return to.' Only when we are willing to acknowledge that 'nowhere to die' is a social issue that must be faced, can a peaceful death cease to be a privilege for the few.