Meta-analysis topic ideas in haematology

Haematology covers three different literatures: non-malignant blood disorders such as anaemia, sickle cell disease and haemophilia; thrombosis and anticoagulation; and haematological malignancy. The first has small trials and many rare-disease studies, the second has large trials that have been pooled many times, and the third has fast-moving drug approvals with short follow-up.

Because many conditions are rare, a review in haematology is often limited by sample size and by differences in definitions, not by a lack of questions. Choosing a question that can be answered with the studies that exist is what makes a topic workable.

Find a haematology topic that has not been done

SynthGap searches the published literature, checks each candidate against existing systematic reviews, and proposes topics with the primary studies behind them. Start on the Free plan: no card needed, and every run shows every topic it finds.

Where gaps tend to hide in haematology

Rare disorders

Sickle cell disease, thalassaemia, haemophilia and rare anaemias have small trials and many observational studies. Syntheses of long-term outcomes, quality of life and access to care are scarce, and pooling methods for rare events matter.

Anticoagulation in special situations

Cancer-associated thrombosis, pregnancy, obesity, kidney disease and thrombosis at unusual sites are under-represented in the large trials, so pooled analyses of subgroups or of smaller trials are a source of topics.

Supportive care in malignancy

Transfusion thresholds, infection prophylaxis, iron therapy and the management of neutropenia are studied in many trials that are rarely compared directly.

Newer targeted therapies

Response and survival with newer agents come from single-arm and early-phase studies. Indirect comparisons and pooled safety reviews depend on being clear about the era and the line of therapy.

Transfusion and patient blood management

Restrictive versus liberal thresholds, iron before surgery and tranexamic acid are studied across surgical populations, and a review may be best defined by procedure.

Outcomes and effect measures that pool well in haematology

Overall survival and progression-free survival: hazard ratios, using the generic inverse variance method. Individual patient data reconstructed from Kaplan-Meier curves can add information but should be flagged as a method with its own error.
Response and remission rates in malignancy: risk ratios for randomized comparisons, and proportions with an appropriate transformation for single-arm studies, noting that between-study differences are usually large.
Venous thromboembolism recurrence and major bleeding: risk ratios over a fixed follow-up window, with the definition of major bleeding (ISTH or another) stated.
Haemoglobin and ferritin: mean differences, noting the units and the time of measurement relative to treatment.
Sickle cell pain crises or haemophilia bleeding rates: rate ratios, using annualised rates and the number of patient-years where available.

Common traps in haematology reviews

Combining single-arm studies and randomized trials in one estimate of treatment effect.
Using median overall survival as an effect measure: it cannot be pooled directly, and hazard ratios should be used.
Treating early-phase studies of newer agents as if they were comparative evidence.
Counting a paediatric and an adult leukaemia protocol as the same population.
Ignoring that definitions of response and remission differ between versions of the disease-specific criteria.

Check before you commit

Before committing months to a topic, check that nobody has beaten you to it. Search PubMed with its systematic review filter, the Cochrane Library (the Cochrane Haematology group), PROSPERO for registered reviews, and Epistemonikos. In malignancy, check the date of the last search carefully, because new agents change the evidence within a year or two. Step-by-step guide.

Starting points to try in SynthGap

These are areas to explore, not claims that they are open. Some may already be well covered, and finding that out quickly is the point.

Iron supplementation before elective surgery and transfusion rates
Hydroxyurea and outcomes in children with sickle cell disease in lower-income settings
Direct oral anticoagulants for cancer-associated thrombosis in people with low body weight or kidney impairment

Questions

How do I check whether a haematology meta-analysis already exists?

Search PubMed with the systematic review filter, the Cochrane Library, PROSPERO and Epistemonikos. In malignancy, check how recent the last review is, because new agents change the field quickly.

Can I do a meta-analysis of a rare disease?

Yes, as long as you design for it: the question should be about outcomes or proportions you can pool, you should expect high heterogeneity, and you should say clearly that confidence in the estimate is low if the studies are small or non-randomized.

What if there are only single-arm studies?

A pooled proportion is possible, for example the proportion who respond, but it is not a comparison of treatments. Describe it as a summary of reported outcomes, not as an effect.

Other specialties

This page is general methodological guidance, not medical advice, and it does not list specific unreviewed topics. SynthGap's suggestions are AI-generated from the published literature, so verify them before registering a protocol.