Highlights
In this population-based Ontario study, decedents with multiple sclerosis were younger and had fewer comorbidities than the general decedent population, but still experienced substantial end-of-life health care needs.
Outpatient neurology involvement was common across the last 5 years of life, yet it dropped sharply near death, suggesting a transition period in which neurologic oversight may diminish just as symptom burden and care complexity rise.
Palliative care exposure was not lower overall in multiple sclerosis than in non-multiple sclerosis decedents, but rural residence was associated with less palliative care access and higher odds of dying in hospital.
Receipt of palliative care in the last 5 years of life was strongly associated with reduced odds of hospital death, supporting earlier and more deliberate integration of neuropalliative care.
Proposed Article Structure
This study is best understood through a clinically oriented structure that moves from disease burden and service gaps to methods, quantitative findings, and practice implications. The most logical framework is: clinical background and unmet need; study design and data source; major results; interpretation for neurologists, primary care clinicians, and palliative care teams; limitations and generalizability; and a concise practice-focused conclusion.
Background: Why End-of-Life Care in Multiple Sclerosis Matters
Multiple sclerosis (MS) is a chronic inflammatory and neurodegenerative disease of the central nervous system that can lead to progressive disability, cognitive impairment, bladder and bowel dysfunction, pain, spasticity, dysphagia, mood symptoms, recurrent infections, and loss of independence. While disease-modifying therapies have changed the trajectory for many patients with relapsing disease, a substantial number still accumulate severe disability over time and require complex supportive care.
The final years of life in MS often involve a high symptom burden that resembles, and in some cases exceeds, that seen in other serious chronic neurologic illnesses. Common challenges include immobility, pressure injuries, aspiration risk, urinary tract complications, caregiver strain, and decisions about hospitalization, artificial nutrition, antibiotics, and goals of care. Yet palliative care remains inconsistently integrated into routine MS care. In practice, many patients are referred late, referred only after repeated hospitalizations, or not referred at all because MS is not always recognized as a condition with substantial palliative needs.
That mismatch between need and service delivery makes this study clinically relevant. It addresses not simply whether people with MS receive palliative care, but when they receive it, how outpatient physician care evolves near death, and which factors are associated with dying in hospital versus outside hospital. These are highly actionable health-system questions.
Study Design and Methods
Kim and colleagues conducted a retrospective population-based cohort study using linked health administrative databases in Ontario, Canada. The investigators identified all adult decedents, aged 18 years or older, who died between April 2016 and March 2020. A validated algorithm was used to identify those with MS. The study then compared decedents with MS against the much larger group of decedents without MS.
The MS cohort included 1,975 decedents, while the non-MS cohort included 500,904 decedents. Administrative databases were used to capture demographics, comorbidity, outpatient physician encounters, palliative care involvement, medical interventions, and place of death. The analysis focused especially on care patterns across the last 5 years of life, with attention to the final months when care transitions often intensify.
Multivariable logistic regression models were used to identify predictors of two clinically meaningful outcomes: receipt of palliative care and hospital death. This is an appropriate approach for health services research, although interpretation should remain cautious because associations from administrative data do not establish causality.
Key Results
Who were the MS decedents?
MS decedents were, on average, younger than other decedents: mean age 68 years versus 77 years in the non-MS population. They were also more commonly female, with women comprising 66.4% of the MS group. Despite being younger, they had fewer comorbidities than non-MS decedents, which is important because it suggests that end-of-life health service use in MS is not merely a reflection of high multimorbidity seen in older populations.
Outpatient physician care changed markedly as death approached
Nearly half of MS decedents, 49.4%, had outpatient neurology visits during the last 5 years of life. This indicates that neurologists remain involved for many patients during the broader end-of-life period. However, neurology contact fell dramatically closer to death, reaching only 2.7% in the last month of life.
That pattern is clinically striking. It suggests that as patients become more disabled and medically fragile, their specialist neurologic care may recede. Some of this likely reflects appropriate transition to primary care, home care, long-term care, or palliative teams. But it may also indicate fragmentation, reduced access, transportation barriers, or the absence of structured neuropalliative pathways. For a disease in which symptom management, communication about prognosis, and anticipatory care planning are essential, the sharp late decline in neurology follow-up may represent a missed opportunity.
Palliative care use was similar overall, but timing and equity matter
Overall palliative care use was similar in MS and non-MS decedents: 58.0% versus 56.9%. This is an important finding because it counters the assumption that patients with MS uniformly receive less palliative care than patients dying of other causes.
At the same time, palliative care occurred somewhat earlier in the MS population. Within the last 5 years of life, 61.9% of MS decedents who received palliative care had done so during that broader period, compared with 59.6% in non-MS decedents. The difference is modest, but directionally consistent with the concept that MS may involve a prolonged period of supportive needs before death.
Still, similar aggregate rates should not be overinterpreted as evidence of optimal care. Administrative definitions of palliative care do not fully capture quality, intensity, discipline mix, or whether services aligned with neurologic symptom burden. A single billing code and a sustained home-based palliative approach are not equivalent clinical experiences.
Rural residence emerged as a major disparity factor
Among MS decedents, rural residence was associated with increased odds of hospital death, with an odds ratio of 1.81 and a 95% confidence interval of 1.21 to 2.70. Rural residence was also associated with lower odds of receiving palliative care, with an odds ratio of 0.74 and a 95% confidence interval of 0.55 to 0.99.
This finding is highly relevant for policy and service delivery. It implies that geography remains a strong determinant of end-of-life experience, even in a publicly funded health system. Rural patients may face limited home-based palliative resources, fewer specialist clinicians, longer travel times, reduced caregiver support, and less access to timely symptom management, all of which can increase the likelihood of hospital transfer near death.
Palliative care was linked to lower odds of hospital death
Receipt of palliative care within the last 5 years of life was associated with substantially lower odds of hospital death in MS decedents, with an odds ratio of 0.47 and a 95% confidence interval of 0.37 to 0.60. This is among the most clinically meaningful findings in the study.
Although observational data cannot prove that palliative care directly prevented hospital death, the magnitude and precision of the association support a plausible and important relationship. Earlier palliative involvement may improve symptom control, facilitate advance care planning, help families navigate crises at home or in long-term care, and reduce default escalation to acute care settings.
Which patients were more likely to receive palliative care?
Higher comorbidity was associated with greater odds of receiving palliative care, with an odds ratio of 3.64 and a 95% confidence interval of 2.50 to 5.29. Increased outpatient visits to urology were also associated with a higher likelihood of palliative care, with an odds ratio of 1.47 and a 95% confidence interval of 1.20 to 1.80.
These associations are clinically intuitive. Greater comorbidity likely increases contact with the health system and heightens recognition of serious illness needs. The urology signal is particularly interesting in MS, where neurogenic bladder, catheter complications, recurrent infections, and urinary symptoms are common drivers of morbidity. Frequent urology use may function as a marker of advanced disability and multidimensional care needs that should trigger a palliative assessment.
Clinical Interpretation
The central message is not that MS decedents are entirely underserved by palliative care. Rather, it is that care remains uneven, hospital-centered for many, and vulnerable to geographic inequity. The study also suggests that neurologic care and palliative care may not be optimally synchronized in the final phase of illness.
For practicing neurologists, the decline in outpatient neurology visits near death should prompt reflection on transition planning. Patients with advanced MS often have predictable trajectories of worsening mobility, dysphagia, infections, communication difficulty, and caregiver burden. Those changes can and should trigger discussions about goals of care before recurrent crises occur. Neurologists do not need to provide all palliative care themselves, but they are often best positioned to identify when a patient would benefit from a formal palliative approach.
For primary care and palliative clinicians, the findings reinforce that MS should be considered a legitimate indication for longitudinal palliative involvement, not only terminal referral. The disease often has a prolonged advanced stage in which symptom management, psychosocial support, medication rationalization, and caregiver support are essential. Earlier integration is especially relevant because prognosis in MS can be difficult to estimate precisely, making prognosis-based referral alone inadequate.
For health systems, rural disparities deserve urgent attention. If palliative care access reduces hospital death, then improving rural home-based and community-based palliative capacity may have both human and system benefits. Telemedicine, integrated neurology-palliative consultations, and standardized referral triggers could help narrow the gap.
How This Fits With Existing Literature and Guidance
The study aligns with a growing body of literature showing that people with serious neurologic illness benefit from palliative care, yet often receive it late or inconsistently. Contemporary palliative care frameworks increasingly emphasize needs-based referral rather than diagnosis-specific or prognosis-limited models. For MS, needs-based triggers might include recurrent infections, progressive dysphagia, severe spasticity or pain, cognitive decline, frequent emergency visits, repeated urinary complications, or escalating caregiver distress.
The findings are also consistent with broader evidence that palliative care is associated with less acute care utilization and a lower chance of hospital death. In neurologic disease, this may be particularly important because communication impairments, cognitive changes, and complex disability can amplify the consequences of fragmented care.
Strengths of the Study
This analysis has several notable strengths. First, it is population-based, reducing referral and selection biases that often affect single-center studies. Second, the sample size is large for a disease-specific end-of-life MS analysis. Third, the use of linked administrative data allows examination of real-world physician use patterns and place of death across an entire jurisdiction. Fourth, the validated algorithm for identifying MS improves confidence in case ascertainment.
These strengths make the study valuable not only as a description of care patterns in Ontario, but also as a framework for other jurisdictions seeking to evaluate end-of-life service delivery in chronic neurologic disease.
Limitations and Cautions
As with all administrative database studies, the analysis is limited by the variables available. The data cannot fully characterize symptom severity, disability stage, caregiver capacity, patient preferences, quality of palliative encounters, or the specific reasons behind hospitalization. It also cannot determine whether death in hospital reflected an unwanted outcome or a patient-preferred setting in individual cases.
Second, palliative care identification based on administrative coding may underestimate some services and overgeneralize others. A coded palliative encounter does not reveal whether care was interdisciplinary, sustained, home-based, or aligned with best practices in neuropalliative medicine.
Third, residual confounding is likely. Patients who receive palliative care may differ systematically from those who do not in ways that are not fully measured, including social support, disease severity, institutional residence, and clinician practice style. Therefore, the association between palliative care and reduced hospital death should be interpreted as supportive, not definitive proof of effect.
Finally, generalizability may vary across health systems. Ontario’s publicly funded infrastructure, home care patterns, and specialist distribution differ from those in other countries. Even so, the broad messages about late specialist decline, rural inequity, and the value of earlier palliative integration are likely widely relevant.
Practice and Policy Implications
Several practical implications emerge from this work. First, advanced MS should prompt proactive palliative assessment. Referral should not wait for the final weeks of life. Second, neurology clinics may benefit from standardized triggers for palliative involvement, including recurrent hospitalizations, severe disability, dysphagia, or repeated urologic complications. Third, rural service models need strengthening through telepalliative care, outreach teams, and better integration across neurology, primary care, and community services. Fourth, quality metrics in MS care should move beyond relapse control and imaging to include goal-concordant care, symptom burden, caregiver support, and place-of-death outcomes.
The study also gives weight to the concept of neuropalliative care, a model that blends disease-specific neurologic expertise with palliative symptom management, communication, and advance care planning. In MS, such integration may be especially useful because disability evolves over years and often intersects with bladder dysfunction, pain, fatigue, mood symptoms, and cognitive changes that require both neurologic and palliative perspectives.
Conclusion
This population-based study shows that people with MS approaching the end of life do receive palliative care at rates similar to the broader decedent population, but this should not be mistaken for optimal care. Hospital death remains common, outpatient neurology involvement drops sharply near death, and rural patients face a double disadvantage of less palliative access and greater odds of dying in hospital. The finding that palliative care was associated with roughly halved odds of hospital death strengthens the case for earlier, structured neuropalliative integration. For clinicians and policymakers alike, the message is clear: the challenge in MS is no longer simply recognizing palliative need, but delivering timely, equitable, coordinated care before the final crisis.
Funding and Trial Registration
No ClinicalTrials.gov registration applies to this retrospective administrative database study. Funding details were not provided in the abstract and should be verified from the full Neurology article.
References
Kim W, Watt CL, Fung SG, Yoo S, Webber C, Rotstein D, Howard M, Tanuseputro P, Ramanathan U. Multiple Sclerosis and Outpatient Physician Care in the Last Years of Life: A Population-Based Study. Neurology. 2026-05-08;106(11):e218067. PMID: 42102338.
Oliver D, Borasio GD, Caraceni A, de Visser M, Grisold W, Lorenzl S, Veronese S, Voltz R. A consensus review on the development of palliative care for patients with chronic and progressive neurological disease. Eur J Neurol. 2016;23(1):30-38.
Kavalieratos D, Corbelli J, Zhang D, Dionne-Odom JN, Ernecoff NC, Hanmer J, Hoydich ZP, Ikejiani DZ, Klein-Fedyshin M, Zimmermann C, Morton SC, Arnold RM, Heller L, Schenker Y. Association Between Palliative Care and Patient and Caregiver Outcomes: A Systematic Review and Meta-analysis. JAMA. 2016;316(20):2104-2114.
