Cohort studyindependent cross-checkabstract checked I-CAN-110 standard review
What cancer inpatients died of in an Indian hospital
Why do cancer patients die? A decade of experience on trends, profile, causes, and timing of deaths in India.
Overview
Doshi A et al.. Public Health Action 2026 · PMID 42830814 · DOI 10.5588/pha.26.0040
Most deaths were due to cancer progression (57%), organ failure (21%) or sepsis (15%).
Does the study support the specific claim? No · Methodological quality: low
Plain-language summary
Study explained simply
- What did the researchers want to find out?
Describe causes of death.
- How was the study done?
Hospital record review.
- How many people, animals or samples were studied?
474 deaths.
- What was actually found?
Cancer progression dominant.
- How large was the observed effect?
57% immediate; 83% underlying.
- What do the results mean in everyday life?
People with cancer mostly die from the cancer and its complications.
- What are the study's limitations?
Single centre; no cachexia category.
- What does it say about Elawady's specific claim?
Contradicts the claim.
- Do the results apply to Germany and Europe?
Low-moderate.
- What can explicitly not be concluded from this study?
Role of cachexia.
Scientific in-depth review
1. Full reference
- Authors
- Doshi A et al.
- Journal
- Public Health Action
- Year
- 2026
- Identifier
- PMID 42830814 DOI 10.5588/pha.26.0040
- Verification
- verified – PubMed efetch 2026-10-26 + Crossref
- Retracted
- no
- Corrections
- none
- Extent checked
- abstract only
- Conflicts of interest and funding
No conflict of interest declared; SORT IT/TDR-WHO training initiative
2. Research question
Trends, causes and timing of inpatient deaths among cancer patients
3. Study design and population
- Study design
- Cohort study – Retrospective hospital cohort (electronic medical records)
- Population
Inpatients of a medical oncology department, South India
- Country
India
- Period
2016-2025
- Sample size
474 inpatient deaths
- Inclusion and exclusion criteria
Inpatient deaths in medical oncology
4. Methods and endpoints
- Exposure / intervention
not applicable
- Comparator
not applicable
- Endpoints
Immediate and underlying causes of death
- Statistical methods
Descriptive; adjusted relative risks
5. Actual results
- Results
Immediate causes: cancer progression 57%, organ failure 21%, sepsis 15%; leading underlying cause cancer progression (83%). 73% had metastases.
- Absolute risks
Proportions as reported
- Authors' conclusion
Cancer progression predominates as the cause of inpatient mortality.
6. Strengths
- Physician notes rather than death certificates
- 10-year period
7. Weaknesses
- Single centre
- Cachexia not recorded as a separate category
- Inpatients only
8. Risk of bias
- Orienting tool
- ROBINS-E
- Type of assessment
- simplified assessment (abstract only)
- Full formal assessment
- no
- Overall risk of bias
- some concerns
- Rationale
Selection (inpatients), classification by treating physicians.
9. Assessment of the statistical interpretation
Descriptive proportions (recomputed counts: about 270, 100 and 71 deaths).
10. Relevance to the specific claim
Direct data on causes of death in cancer patients: the underlying cause is cancer in most cases.
11. Independent contrary findings
Does not separate cachexia, so it cannot fully test the claim.
12. Transferability to Germany and Europe
Low-moderate (Indian tertiary centre).
13. Overall methodological judgement
Limited but direct evidence against 'they do not die of cancer'.
| Multidimensional assessment | |
|---|---|
| Authenticity of the source | verified |
| Methodological quality | low |
| Risk of bias | some concerns |
| Precision of results | moderate |
| Directness | direct |
| Reproducibility | unknown |
| Consistency with the overall evidence | consistent |
| Regional transferability | low |
| Adequacy of the interpretation (by the archive) | not applicable |
14. Open questions
Multi-centre data including cachexia.
15. Linked fact checks
- Cancer patients do not die of cancer itself but of wasting and cortisone (hormonal disturbance). – Studies contradicting or limiting the claim