TL;DRIt is now nearly a decade and a half since the Marmot Review's housing chapter set out what was then the state of the evidence on housing as a determinant of health, and the intervening years have added substantially to…
It is now nearly a decade and a half since the Marmot Review's housing chapter set out what was then the state of the evidence on housing as a determinant of health, and the intervening years have added substantially to the literature. The cohort studies have matured, the trial portfolio has expanded, the cost-effectiveness analyses have been refined, and the policy interventions of the period have themselves produced data on real-world outcomes that the earlier evidence base could only model. This article is a consolidated 2026 summary of where the evidence stands across the principal intervention categories.

The summary is honest about uncertainty. The healthy-homes literature has matured but not reached the point where it produces simple causal claims of the kind that infectious-disease epidemiology typically delivers, and the policy debate is too often conducted as if the evidence were either sufficient to justify any intervention or insufficient to justify any. The reality, as is usual in environmental epidemiology, sits between those poles, and the practical work is in distinguishing what is established from what is contested and what is still to be settled.

Cardiovascular outcomes: established, with magnitude debated

The relationship between cold indoor temperatures and cardiovascular outcomes is established with sufficient robustness to support clinical action. The physiological mechanisms are characterised, the cohort epidemiology is consistent, and the intervention literature shows measurable improvements in blood pressure and other intermediate outcomes following thermal upgrade. The magnitude of the benefit on hard outcomes — myocardial infarction, stroke, cardiovascular mortality — is harder to pin down, with credible estimates spanning a meaningful range. The uncertainty does not undermine the qualitative claim; it does mean that population-level cost-effectiveness analyses should report ranges rather than point estimates and that policy debates should engage with the range rather than select the convenient end of it.

Respiratory outcomes: established for damp, less clear for thermal alone

The damp-and-respiratory relationship, particularly in paediatric asthma, is one of the strongest findings in the literature. The intervention literature shows that addressing visible mould and chronic damp produces measurable reductions in symptom frequency and healthcare utilisation. The thermal-only interventions — heating-system upgrades or insulation work that does not address an existing damp problem — produce smaller respiratory benefits, consistent with the mechanism that respiratory effects are mediated principally by humidity and allergen exposure rather than by temperature directly. The implication for intervention design is that respiratory outcomes call for ventilation-and-damp work as a primary target rather than thermal upgrade alone.

Mental health: emerging, with operational implications

The mental-health literature has expanded considerably and now supports a credible claim that fuel poverty contributes to anxiety, depression, and chronic stress in affected households, and that intervention produces measurable benefit in the affected population. The mechanisms are partially characterised — chronic stress, sleep disruption, social isolation — and the dose-response relationships are reasonably consistent. The implication for intervention design is that mental-health considerations should be factored into pathway design rather than treated as an incidental co-benefit, and the integration with formal mental-health services is now operationally warranted in a way it would not have been ten years ago.

Falls and musculoskeletal: emerging co-benefits

The falls-prevention literature has begun to incorporate the home-environment dimension in a more systematic way, and the evidence supports a meaningful co-benefit from retrofit interventions in older households. The magnitude of this co-benefit is not yet well-pinned-down, and trials sized to detect it as a primary outcome are scarce. Within those limits, the qualitative claim is supported and the operational implication — designing whole-house assessment to include falls-prevention components — is reasonable.

Indoor air quality: established for outdoor pollution, more contingent indoors

The outdoor air-quality literature is robust and has informed UK air-quality policy for decades. The indoor air-quality literature is less mature but has expanded considerably, and the principal indoor sources — combustion-based cooking, solid-fuel heating, household smoking — are now sufficiently well-characterised to support source-control intervention as a defined target. The benefit of air-purification interventions is smaller than commercial discourse suggests, and the policy emphasis should be on source control rather than on filtration. Within the indoor environment, particulate matter is the most studied pollutant category; the volatile organic compound and biological-aerosol literature is more contingent and merits cautious interpretation.

Cognitive and developmental outcomes: contested

The literature on housing-quality effects on cognitive and neurodevelopmental outcomes has expanded but remains harder to interpret than the cardiovascular or respiratory literature. The associations are smaller in magnitude and more vulnerable to confounding, particularly by socioeconomic factors that themselves have well-established effects on the outcomes of interest. The qualitative claim — that childhood housing conditions affect cognitive trajectories — is supported, but the quantitative claim about the magnitude of the effect after appropriate confounding adjustment is contested. The policy implication is that this category of outcome should not bear the weight of intervention cost-effectiveness claims unless the analysis is unusually robust.

Cost-effectiveness: favourable across the established categories

The economic literature on healthy-homes intervention has matured to the point where most categories of fabric-based intervention clear the standard cost-effectiveness thresholds applied in UK health technology assessment, with reasonable confidence across plausible assumption ranges. The interventions that perform best in this calculation are whole-house upgrades targeted at the lowest-EPC properties occupied by households with one or more clinically vulnerable members. Single-measure interventions and broader population coverage produce smaller per-pound benefits and are correspondingly less robustly justified.

The remaining methodological challenges in the cost-effectiveness work concern the appropriate discount rate for benefits accruing across long time horizons, the inclusion or exclusion of co-benefits across health domains, and the treatment of distributional effects. Different defensible methodological choices produce meaningfully different headline cost-effectiveness ratios, and the published literature uses an inconsistent enough set of choices that headline comparisons between studies should be made carefully.

What is still genuinely unknown

Several questions remain genuinely open and deserve restraint in policy and clinical conversation. The first is the durability of intervention benefits over decade-plus timescales — most evaluations have followed cohorts for two to five years, and the question of whether the early benefit is sustained or attenuates with time is poorly characterised. The second is the interaction of healthy-homes interventions with broader social-determinant interventions, and whether the effect of the home is amplified or attenuated by simultaneous changes in income, employment, or community connection. The third is the effect of interventions delivered at population scale, where general-equilibrium effects on housing markets, construction-sector capacity, and household behavioural responses may produce outcomes that differ from those observed in controlled trial settings.

How clinical practice has been responding

The clinical workforce has not stood still while the evidence has been accumulating. The integration of home-environment factors into routine primary-care review has progressed unevenly across the UK, with notable variation between regions, between integrated care systems, and between individual practices. The published surveys of practice in this area suggest that a meaningful minority of GP practices now incorporate at least some home-environment screening into routine asthma and cardiovascular reviews, and that this minority has grown materially over the last several years. The remainder either do not screen routinely or screen in a less structured way that produces lower yields.

The factors that distinguish the practices that have integrated effectively from those that have not are reasonably consistent: dedicated time within the review for the conversation, named referral routes that have been verified to function, and feedback loops that allow the clinical team to see what happens after referral. Practices with these elements report sustained engagement and growing referral volumes; practices without them report initial enthusiasm followed by decline as the operational frictions accumulate.

The secondary-care response has been more limited but is also visible. Respiratory and cardiology services in several integrated care systems have begun to incorporate home-environment factors into discharge planning and into the structured care pathways for patients with established disease. The scale of this incorporation remains modest relative to the scope of the underlying patient population, but the trajectory is in the direction the evidence supports, and the operational templates that have emerged from the earliest adopters are being adapted by services elsewhere.

What the next five years should produce

For commissioners, evaluators, and clinical leaders considering where the field should go from here, several priorities present themselves. The first is the closure of the durability gap in the evaluation literature. Long-term follow-up of intervention cohorts is operationally demanding but is the only way to settle the question of whether the early benefit is sustained or attenuated, and the answer to that question matters considerably for the lifetime cost-effectiveness calculation that underpins commissioning.

The second is the consolidation of the operational templates. The early adopters have produced reasonably consistent patterns of what works in pathway design, in workforce integration, and in the cross-organisational interface. Codifying these patterns into accessible operational guidance would shorten the learning curve for the next wave of services and would produce a more rapid scaling of the evidence-based pathway than ad-hoc adoption produces.

The third is the development of the data infrastructure that will support routine outcome evaluation at scale. The studies that produced the existing evidence were in many cases stand-alone research projects with their own data collection. Routine commissioning of the intervention pathway requires routine outcome data, and that requires linked datasets connecting housing intervention records with health outcomes over years. The technical work to produce these linkages is not trivial but is well within reach, and the value of the resulting evidence base would be substantial.

The 2026 evidence base supports active healthy-homes intervention as a defined health policy lever, with the strongest evidence in the cardiovascular and respiratory categories, robust evidence in the mental-health and indoor-air-quality categories, and emerging evidence in the falls-prevention and developmental categories. The cost-effectiveness calculation is favourable across the established categories under most reasonable assumption sets. Several questions remain genuinely open and should temper but not defer policy action. The honest summary is that the evidence is sufficient to justify the intervention pathway as it now exists, sufficient to justify expansion in a well-targeted form, and not yet sufficient to justify either uncritical scale-up or the position that the case is unsettled. The work of the next several years is operational delivery, evaluation refinement, and the gradual closure of the remaining evidential gaps.

For clinicians: signpost patients to evidence-led referral pathways →