The study, known by the acronym MATHRU, which stands for Multifaceted, Comprehensive, Appropriate Care Package for People with Advanced and Terminal Illnesses Needing Intensive Care Support in India, was published as a study protocol in BMC Health Services Research and is funded by the Indian Council of Medical Research under its Centre for Advanced Research programme. Led by Naveen Salins and Arun Ghoshal of the Department of Palliative Medicine and Supportive Care at Kasturba Medical College, Manipal Academy of Higher Education, the collaboration brings together intensivists, palliative care physicians, ethicists, health economists, legal policy experts, and public health researchers from institutions spanning the country, from Safdarjang Hospital in New Delhi to Christian Medical College in Vellore and Tata Memorial Hospital in Mumbai. It is prospectively registered with the Clinical Trials Registry-India, a detail that matters in a field where implementation studies often proceed without formal registration.

At the heart of the programme lies the intervention itself, the Multifaceted, Comprehensive, Appropriate Care Package, or MCACP. Rather than a single protocol, it is a structured bundle of seven interlocking domains: prognostication and recognition of patients approaching the end of life; interdisciplinary review and consensus among the clinical team; structured communication with patients and families; shared decision-making; comprehensive anticipatory care planning; care delivery; and care evaluation and learning. The logic is that dying well in an ICU is not the product of one heroic conversation between doctor and family but of a chain of coordinated processes, each of which can fail independently. A clinician may accurately recognise that a patient is dying, but if the team cannot reach consensus, or if the family meeting is handled poorly, or if no anticipatory plan is documented, the system defaults to futile escalation.

What distinguishes MATHRU from many previous attempts to export end-of-life care models to low- and middle-income settings is its methodological scaffolding. The study is explicitly grounded in the Medical Research Council framework for developing and evaluating complex interventions, the Consolidated Framework for Implementation Research, Normalization Process Theory, and the Implementation Research Logic Model. These are not decorative citations. Each framework does specific work: CFIR helps the team systematically characterise the barriers and facilitators present in each ICU’s context, from leadership culture to resource constraints; Normalization Process Theory examines whether new practices become embedded in routine work or wither once the research team leaves; and the logic model links intervention components to mechanisms and outcomes in a testable chain. The reporting will follow the StaRI standards for implementation studies, and the intervention will be described using the TIDieR template, making it reproducible by others.

The study unfolds in five sequential phases over five years. Phase 1 is devoted to developing and specifying the intervention through evidence synthesis, a two-step Delphi process to build expert consensus, the Nominal Group Technique, human-centred co-design, and usability testing. This is where the patient and public involvement component becomes critical: in a country where family members typically serve as primary caregivers and medical decisions are often made collectively, a care package designed without families at the table would fail on contact with reality. Phase 2 pilots the package in one open ICU and one closed ICU, testing whether the intervention survives contact with two very different models of intensive care organisation. Only then do the large-scale evaluation phases begin.

Phases 3 and 4 constitute the evaluative core of the programme. The package will be implemented across eight ICUs, deliberately split between four with established palliative care access and four without, a design choice that acknowledges the reality that Indian critical care exists on a spectrum of palliative support rather than at a single baseline. Phase 3 uses a mixed-methods process evaluation to understand how, for whom, and under what conditions the package works, while Phase 4 employs a quasi-experimental pre-post effectiveness design. The planned sample is substantial: 150 patient episodes per ICU in each of the pre- and post-implementation periods, yielding thousands of episodes across the network. The investigators have been candid about the limitations of this non-randomised design, acknowledging its susceptibility to secular trends, regression to the mean, and residual confounding, and planning to interpret findings through covariate-adjusted analyses alongside rich contextual data on fidelity, reach, and implementation mechanisms.

The outcome measures reveal what the researchers believe genuinely matters at the end of life. Alongside implementation outcomes such as feasibility, acceptability, fidelity, reach, workflow compatibility, and normalisation, the study will track patient, family, and system outcomes including the ICU Quality of Dying and Death instrument, Family Satisfaction in the ICU scores, care process indicators such as rates of do-not-attempt-resuscitation orders and documented anticipatory care plans, resource utilisation, and costs. The inclusion of economic evaluation, led by health economists including Shankar Prinja of the Postgraduate Institute of Medical Education and Research in Chandigarh, is significant for a country where out-of-pocket health spending pushes millions of families into poverty each year. Care that reduces futile interventions may be not only more humane but also more affordable, and demonstrating that could be the key to policy uptake.

The ethical architecture of the study reflects the sensitivity of the subject matter. Ethics approval has been obtained from the Institutional Ethics Committee of Kasturba Medical College and Kasturba Hospital, Manipal, and from the Institutional Ethics Committee of St John’s National Academy of Health Sciences in Bengaluru, with additional site approvals required before each site activates. The study adheres to the Declaration of Helsinki, and the investigator team includes Roli Mathur of the ICMR Bioethics Unit, signalling that ethical oversight is built into the study’s design rather than bolted on. The collaboration also includes Dhvani Mehta of the Vidhi Centre for Legal Policy, an acknowledgement that end-of-life care in India operates within a contested legal landscape, where questions about withdrawal of life-sustaining treatment have been shaped by landmark Supreme Court rulings and where clinicians remain uncertain about the boundaries of lawful practice.

Phase 5 looks beyond the study itself toward national scale. The final phase is designed to develop a national technical resource hub and a minimum viable intervention set, a distilled version of the package that resource-constrained ICUs can adopt without the full apparatus of the research programme. This tiered approach recognises a hard truth of implementation science: interventions that succeed in well-funded academic centres frequently collapse when transplanted into settings with fewer staff, less training infrastructure, and heavier patient loads. By specifying in advance what the irreducible core of appropriate end-of-life care looks like, the investigators are attempting to build a bridge from evidence to policy that survives the end of the grant.

At the time of the protocol’s publication, Phases 1 and 2 were ongoing while the large evaluation phases had not yet commenced, so no effectiveness results exist yet. But the study’s significance is already apparent. India’s ICU population includes enormous numbers of patients with advanced illness, and the cultural, linguistic, and religious diversity of the country makes any one-size-fits-all protocol doomed to failure. MATHRU’s bet is that a defined but adaptable package, co-designed with the people it affects and evaluated with implementation science rigour, can be normalised across heterogeneous ICUs. If it succeeds, the payoff will be measured not in survival statistics but in something harder to quantify: the quality of the final days of thousands of patients, and the peace of mind of the families who love them. In a field where the default has too often been silence and escalation, simply building the machinery for honest, coordinated, compassionate care may prove to be the most consequential intervention of all.

Subject of Research: Implementation of a comprehensive palliative and end-of-life care package for terminally ill patients in Indian intensive care units

Article Title: Implementing a multifaceted, comprehensive, appropriate care package for people with advanced and terminal illnesses needing intensive care support in India: protocol for a multiphase mixed-methods implementation study

Article References: Ghoshal, A., Mathur, A., Choudhuri, A. H., Mehta, D., Sundararaj, J. J., Vallath, N., Ramanjulu, R., Mani, R. K., Mathur, R., Gursahani, R., Mishra, S., Rao, S. R., Prinja, S., Myatra, S. N., Iyer, S., Singh, S., Simha, S., Varghese, C., Shivashankar, R., & Salins, N. (2026). Implementing a multifaceted, comprehensive, appropriate care package for people with advanced and terminal illnesses needing intensive care support in India: protocol for a multiphase mixed-methods implementation study. BMC Health Services Research. https://doi.org/10.1186/s12913-026-15832-0

Image Credits: AI Generated

DOI: 10.1186/s12913-026-15832-0

Keywords: palliative care, intensive care unit, end-of-life care, implementation science, India, shared decision-making, advanced illness, mixed methods, MRC framework, health services research, ICMR, quality of dying