Bla g 1 and the Bedroom: Cockroach Allergen, Childhood Asthma, and Pest Control as a Respiratory Intervention

1. Introduction: the best evidence we have

Much of this journal has dealt with literatures that are thinner than practitioners assume. The mental health evidence reviewed elsewhere here rests substantially on anecdote. The rodent abundance literature is barely present in Canada. The consumer product evidence base was largely absent for fifty years.

This topic is different. The link between cockroach allergen exposure and childhood asthma morbidity is established in the New England Journal of Medicine, replicated across multiple cohorts, supported by dose-response relationships and defined exposure thresholds, and consistent across countries.

1.1 The honest counterweight

And yet the evidence that removing the cockroaches improves the asthma is weaker than the evidence that their allergen harms. Section 9 deals with that squarely, because a paper that cited only the exposure literature would be making an argument rather than reporting one.

2. The inner-city asthma study

The foundational work is Rosenstreich and colleagues in the New England Journal of Medicine, arising from the National Cooperative Inner-City Asthma Study.¹

2.1 The cohort

Children had a mean age of 6.2 years with a range of 4 to 9. The study families were poor, primarily Black or Hispanic, and had a substantial number of psychosocial problems.¹

That description matters and we quote it rather than paraphrase it, because the population in which this effect was characterised is the same population identified elsewhere in this journal as carrying the highest infestation risk and the least capacity to purchase a private solution.

2.2 Sensitisation rates

Frequency of reactivity on skin testing was 36.8 per cent for cockroach allergen, 34.9 per cent for dust-mite allergen and 22.7 per cent for cat dander.¹

2.3 Exposure rates

Elevated concentrations of the cockroach allergen Bla g 1 were found in dust samples from 50.2 per cent of the children's bedrooms.¹ Other studies found high levels, above 2 U per gram, of the related allergen Bla g 2 in dust from 37 to 85 per cent of urban homes.¹

Sensitivity to cockroach allergen is found in 23 to 60 per cent of urban residents with asthma,¹ and allergic persons with asthma experience acute episodes when exposed to cockroach allergen in bronchial provocation testing.¹

That last point is worth isolating. Bronchial provocation is a controlled clinical procedure in which the allergen is administered directly. It establishes causation in a way observational exposure data cannot.

3. Sensitised and exposed

The study's design contained a specific hypothesis worth understanding, because it is what gives the result its force.

For each allergen and measure of morbidity, the primary hypothesis assessed was that asthma morbidity would be highest among children who were both sensitive to a given allergen and exposed to high levels of that allergen in their bedrooms. This was tested by comparing that group against the other three groups in a two-sided, preplanned comparison.¹

3.1 Why the four-group design matters

Sensitised and unexposed, exposed and unsensitised, neither, and both. Only the combination should produce an allergic response, so finding elevated morbidity specifically in the both group is evidence of a mechanism rather than of a correlation.

3.2 The result

Inner-city children with asthma who were both sensitised to cockroach and exposed to high levels of cockroach allergen in their homes presented higher asthma morbidity, indicated by more hospitalisations, more medical visits and more reported symptoms.³

Exposure to Bla g 1 levels greater than 8 U per gram in children's bedrooms was strongly associated with increased hospital admission and other parameters of asthma-related morbidity.⁴

Several studies in subsequent years have confirmed the association of cockroach exposure with increased asthma morbidity in inner-city areas of the United States and in other countries.³

4. The hospitalisation finding

A later study provides the sharpest single comparison in this literature.

Participants were 86 atopic asthmatic children living in New Orleans. Sensitisation status was determined by serum specific IgE testing and vacuum dust samples were collected for allergen analysis.²

Approximately 44 per cent of children were exposed to Bla g 1 levels above 2 U per gram and 40 per cent to house dust mite levels above 2 micrograms per gram. Twenty four per cent reported at least one hospitalisation in the previous four months.²

4.1 The median comparison

The median Bla g 1 level was significantly higher in the homes of children who had been hospitalised than in those with no hospital admissions: 7.2 U per gram against 0.8 U per gram.²

A ninefold difference in median bedroom allergen between children who were hospitalised and children who were not is the kind of separation that rarely appears in environmental health data.

4.2 The odds ratio

In multivariable models, the odds of hospitalisation were significantly higher in children exposed to Bla g 1 above 2 U per gram, with an adjusted odds ratio of 4.2 and a 95 per cent confidence interval of 1.24 to 14.17.²

The confidence interval is wide, reflecting the modest sample, and we note that rather than quoting the point estimate alone. The lower bound above 1 is what establishes significance.

4.3 The dust mite comparison

In the same cohort, exposure to house dust mites was not associated with any measure of morbidity.²

This internal control is valuable. Both allergens were measured in the same homes by the same methods in the same children. One produced a strong association and the other produced none, which argues against the finding being an artefact of poor housing generally.

5. Independent of sensitisation

The most scientifically interesting result in this literature is one the authors flag as not fitting the expected mechanism.

The elevated odds of hospitalisation held independent of sensitisation status.² The authors conclude that exposure to cockroach allergen was strongly associated with increased hospitalisation in children with asthma, and that this effect cannot be explained entirely by IgE-mediated inflammation.²

5.1 What might account for it

The paper does not resolve the mechanism and neither will we. Plausible contributors include non-allergic airway irritation from cockroach-derived material, the correlation of heavy infestation with other exposures such as poor ventilation and damp, and the insecticide exposures discussed elsewhere in this journal, where fogger use raised household pesticide residues by a median of eighty five fold.

We flag that last possibility as our own speculation rather than a reported finding. It is testable and, so far as we can tell, untested.

6. Defined thresholds

Unusually for an environmental exposure, this literature has converged on specific numbers.

Threshold levels of cockroach exposure above which susceptible individuals would be at increased risk have been defined as 2 U per gram of dust for sensitisation and 8 U per gram for asthma symptoms, for both Bla g 1 and Bla g 2.³

6.1 Why thresholds change the conversation

A defined threshold converts a vague concern into a measurable target. It means a home can be assessed rather than described, and it means an intervention can be evaluated against a number rather than against an impression.

It also makes the §9 result legible: an intervention that reduces allergen but leaves it above threshold has not achieved anything clinically meaningful, however impressive the percentage reduction looks.

7. The potency of Bla g 2

One comparative finding establishes how unusual this allergen is.

Bla g 2 was able to induce IgE production at levels of exposure 10 to 100 fold lower, below 1 microgram per gram of dust, than cat and mite allergens.³

Cat allergen is the household exposure most people regard as the serious one. On this evidence cockroach allergen sensitises at concentrations one to two orders of magnitude below what cat requires.

7.1 The implication for low-level infestation

Exposure to low levels of Bla g 1 and Bla g 2 has been associated with wheezing among infants in the first three months of life and with increased proliferative T cell responses.³ Other reports show association of cockroach allergen exposure with persistent childhood wheezing and with severe asthma.³

This undermines a common operational assumption. A light infestation is frequently treated as a low priority because the insect count is low. If sensitisation occurs at sub-microgram concentrations, allergen burden and visible infestation are not the same variable, and a home that looks nearly clear may not be.

8. Exposure before birth

The timing evidence extends the window earlier than most practitioners would assume.

A birth cohort study of Dominican and African American pregnant women in New York City tested whether domestic cockroach allergen measured prenatally would predict cockroach sensitisation in early childhood. Prenatal exposure to cockroach allergen was associated with a greater risk of allergic sensitisation.⁵ Results showed prenatal exposure associated with greater risk of sensitisation by the age of 5 to 7 years.³

8.1 The interaction with combustion products

The risk was increased by exposure to non-volatile polycyclic aromatic hydrocarbons, with children null for the GSTM1 mutation particularly vulnerable.⁵ Sensitisation to cockroach is described there as one of the strongest identified risk factors for greater asthma morbidity in low-income urban communities.⁵

The practical reading is that allergen exposure combines with other housing exposures rather than acting alone, and that genetic susceptibility modifies who is affected. Neither of those is something a pest control programme can address, but both argue for reducing the one exposure that is addressable.

9. The intervention that did not hold

This is the section that a company selling cockroach control would prefer to omit.

The National Cooperative Inner-City Asthma Study included an environmental intervention to reduce cockroach allergen exposure in inner-city homes. The intervention did appear to lower Bla g 1 levels in the kitchen but only for a few months. Even with the decrease, levels still were well above those reported to cause respiratory symptoms among asthmatic subjects. Compliance with cleaning instructions did not increase the effectiveness of the intervention.⁶

9.1 What this does and does not mean

It does not mean abatement is futile. It means the abatement attempted in that trial was insufficient, which is a different claim and an entirely plausible one given the housing context described in §10.

The later hospitalisation study makes the gap explicit, concluding that controlled interventional trials are needed to determine whether isolated cockroach abatement improves asthma control.² That call was made in 2011 and, so far as we can establish, the definitive trial has not been run.

9.2 The honest position for an operator

We can say that exposure above defined thresholds is associated with hospitalisation. We cannot say that hiring a pest control company will reduce a child's asthma admissions, because that study does not exist. Any company telling a parent otherwise is going beyond the evidence.

10. Why abatement is hard in the relevant housing

The §9 failure is comprehensible in light of everything else in this journal.

The building, not the unit. German cockroach populations in attached housing are connected between units, so a single-unit intervention is recolonised, as argued in the housing analysis published elsewhere here.

Allergen persists after the insects. Bla g 1 and Bla g 2 are proteins in dust, frass and shed cuticle. Killing the population does not remove the accumulated allergen, which requires physical cleaning of reservoirs that include soft furnishings and cracks.

Cleaning compliance did not help. That finding⁶ is consistent with a reservoir that ordinary household cleaning cannot reach.

Behavioural resistance may block baits. Where glucose aversion is present, as described elsewhere in this journal, the bait-led approach that is otherwise correct for cockroaches in occupied housing may simply fail.

Consumer foggers make it worse. They do not control cockroaches and they raise household pesticide residue substantially, in homes that already contain a child with asthma.

10.1 What an adequate trial would need

Building-wide rather than unit-scoped treatment, bait matrix selected against local aversion status, concurrent allergen reservoir removal rather than surface cleaning alone, elimination of consumer fogger use in the building, and follow-up measured against the 2 and 8 U per gram thresholds³ rather than against percentage reduction.

11. What this makes cockroach control

Collecting the evidence produces a reframing that the industry has not adopted.

A cockroach infestation in a household containing an asthmatic child is an exposure to an agent that, above defined thresholds, is associated with a fourfold adjusted odds of hospitalisation,² that sensitises at concentrations one to two orders of magnitude below cat allergen,³ and whose prenatal presence predicts sensitisation years later.⁵

That is not a comfort or an aesthetic problem. Whether the sector describes it as a health intervention or not, it is one.

11.1 The obligation this implies

For a landlord, it means the priority of a cockroach report should be informed by who lives in the unit. For an operator, it means asking is reasonable and cheap. For a public housing provider, it means cockroach prevalence is a paediatric health indicator that happens to be recorded as a maintenance statistic.

None of this requires overstating the abatement evidence of §9. The exposure case stands on its own and is sufficient to change prioritisation.

12. The Manitoba position

Three points specific to this province.

The species is here and thermally protected. As set out elsewhere in this journal, Blattella germanica survives here entirely inside heated buildings, so infestations persist through winter with no seasonal reduction in allergen production.

The housing type matches. The evidence base comes from older multi-unit urban rental housing with connected units, which describes a substantial share of Winnipeg's rental stock.

No local data exist. We have found no Manitoba or Canadian measurement of domestic Bla g 1 or Bla g 2 levels, and no linkage of cockroach prevalence to paediatric asthma admissions here. Given that dust sampling is routine and the assay is established, this is among the more tractable of the gaps this journal has identified.

13. Limitations and open questions

The abatement gap is the main one. We have stated it in §9 rather than buried it. Exposure evidence is strong; intervention evidence is weak, and the call for controlled trials remains open.²⁶

Sample sizes vary considerably. The hospitalisation odds ratio rests on 86 children with a confidence interval from 1.24 to 14.17.² The point estimate should not be treated as precise.

The cohorts are American and urban. Findings come from inner-city populations in the United States,¹²⁵ though the association has been confirmed in other countries.³

Confounding is difficult to exclude fully. Heavy cockroach infestation correlates with other housing exposures. The dust mite null result in the same homes² is the strongest available argument against pure confounding, but it is not conclusive.

Our pesticide speculation in §5.1 is ours. No cited source links insecticide residue to the non-sensitised association, and we have labelled it as speculation.

We sell cockroach control. The commercial incentive runs toward overstating the benefit of what we sell, which is precisely why §9 and §9.2 are in the paper.

14. Conclusion

Elevated Bla g 1 was present in 50.2 per cent of inner-city children's bedrooms, 36.8 per cent of those children were skin test reactive to cockroach, and those both sensitised and exposed showed more hospitalisations, more medical visits and more symptoms.¹³ Median bedroom allergen was 7.2 U per gram in hospitalised children against 0.8 in those not hospitalised, with an adjusted odds ratio of 4.2 above the 2 U per gram threshold, while dust mite exposure in the same homes predicted nothing.² Bla g 2 sensitises at concentrations 10 to 100 fold below cat and mite allergens,³ and prenatal exposure predicts sensitisation years later.⁵

This is the strongest health evidence attached to any structural pest, and it should change how the sector prioritises cockroach reports in family housing.

It should not, however, be used to claim more than it supports. The one major environmental intervention trial lowered kitchen allergen only for a few months, to levels still above those reported to cause symptoms, and resident cleaning compliance did not help.⁶ The call for controlled trials of isolated cockroach abatement was made in 2011 and appears still to be open.²

So the accurate statement is this. Cockroach allergen exposure above defined thresholds is associated with paediatric asthma hospitalisation, and that association is strong, replicated and mechanistically supported. Whether the pest control industry can reduce that exposure enough to matter clinically is a question nobody has properly answered, and answering it would require an approach considerably more thorough than the unit-scoped treatment the market currently buys.

References

1. Rosenstreich, D.L., Eggleston, P.A., Kattan, M. et al. (1997). The Role of Cockroach Allergy and Exposure to Cockroach Allergen in Causing Morbidity among Inner-City Children with Asthma. New England Journal of Medicine, 336, 1356–1362. Source for the cohort description and mean age of 6.2 years, skin test reactivity of 36.8 per cent to cockroach, 34.9 per cent to dust mite and 22.7 per cent to cat dander, elevated Bla g 1 in 50.2 per cent of bedrooms, Bla g 2 above 2 U per gram in 37 to 85 per cent of urban homes, cockroach sensitivity in 23 to 60 per cent of urban residents with asthma, acute episodes on bronchial provocation, and the preplanned four-group comparison design. https://www.nejm.org/doi/full/10.1056/NEJM199705083361904
2. Cockroach exposure independent of sensitization status and association with hospitalizations for asthma in inner-city children. Annals of Allergy, Asthma and Immunology. Source for the 86-child New Orleans cohort, exposure rates of approximately 44 per cent above 2 U per gram Bla g 1 and 40 per cent above 2 micrograms per gram house dust mite, 24 per cent reporting hospitalisation in the previous four months, median Bla g 1 of 7.2 against 0.8 U per gram, the adjusted odds ratio of 4.2 with 95 per cent confidence interval 1.24 to 14.17 independent of sensitisation status, the null result for house dust mite, the conclusion that the effect cannot be explained entirely by IgE-mediated inflammation, and the call for controlled interventional trials. https://www.annallergy.org/article/S1081-1206(10)00957-9/abstract
3. Review of cockroach allergens and asthma. PubMed Central. Source for the NCICAS findings of higher morbidity among children both sensitised and exposed, confirmation of the association in the United States and other countries, wheezing among infants in the first three months of life at low Bla g 1 and Bla g 2 levels with increased proliferative T cell responses, Bla g 2 inducing IgE at 10 to 100 fold lower exposure than cat and mite allergens, the defined thresholds of 2 U per gram for sensitisation and 8 U per gram for symptoms, and the association of exposure with persistent childhood wheezing and severe asthma. https://pmc.ncbi.nlm.nih.gov/articles/PMC3909726
4. Cockroach allergens and asthma. Journal of Allergy and Clinical Immunology. Source for the finding that exposure to Bla g 1 levels greater than 8 U per gram in children's bedrooms was strongly associated with increased hospital admission and other parameters of asthma-related morbidity. https://www.jacionline.org/article/S0091-6749(01)30848-5/fulltext
5. Early-life cockroach allergen and polycyclic aromatic hydrocarbon exposures predict cockroach sensitization among inner-city children. Journal of Allergy and Clinical Immunology. Source for the New York City birth cohort of Dominican and African American pregnant women, the association of prenatal cockroach allergen exposure with greater risk of allergic sensitisation, the augmenting effect of non-volatile polycyclic aromatic hydrocarbons with GSTM1-null children particularly vulnerable, and the description of cockroach sensitisation as one of the strongest identified risk factors for asthma morbidity in low-income urban communities. https://www.jacionline.org/article/S0091-6749(12)02646-2/fulltext
6. Results of the National Cooperative Inner-City Asthma Study environmental intervention to reduce cockroach allergen exposure in inner-city homes. Source for the finding that the intervention lowered kitchen Bla g 1 levels only for a few months, that levels remained well above those reported to cause respiratory symptoms, and that compliance with cleaning instructions did not increase effectiveness. https://www.sciencedirect.com/science/article/abs/pii/S009167499970477X