Abstract People with congenital or acquired immune deficiency have rates of non-Hodgkin lymphoma (NHL) that are raised 50 fold or more above population rates, but it is unknown whether risk of NHL is associated with other forms of immune dysregulation. We performed a population based case-control study of risk factors for NHL in adults aged 20–74 years in New South Wales and the Australian Capital Territory, Australia. We investigated the association of NHL risk with atopy, which is associated with a Th2 dominant immune response. In addition, we investigated the association of NHL with birth order and childhood crowding, which are known to predict atopy. Cases with NHL were selected from a cancer register (n=704), and controls (n=694) were randomly selected from state electoral rolls and frequency-matched to cases by age, sex and area. Cases with clinically apparent immune deficiency were excluded. Birth order, childhood crowding and history of atopic conditions (hayfever, asthma, eczema, and specific allergies) were assessed by questionnaire and by interview. Logistic regression models of NHL risk included the three matching variables as covariates. Being an only child or first born child of a larger family was strongly inversely associated with risk of NHL. Compared to a fourth or later born child, the odds ratios (OR) for development of NHL were 0.52 (95% CI 0.32–0.84) for an only child, 0.55 (95% CI 0.40–0.75) for a first-born child, and 0.70 (95% CI 0.51–0.96) and 0.81 (95 % CI 0.57–1.14) respectively for second and third born children (p trend < 0.0001). Indicators of crowding in later childhood, such as ever sharing a bed or bedroom, and number of years of sharing, were not associated with NHL risk. Diagnosis of atopic conditions was also inversely associated with NHL risk. Self-reported histories of hayfever, asthma, eczema and food allergies were each associated with reduced NHL risk; this reduction was significant for hayfever (OR 0.65, 95% CI 0.52–0.82) and food allergies (OR 0.29, 95% CI 0.20–0.42), but not significant for asthma (OR 0.88, 95% CI 0.67–1.17) and eczema (OR 0.79, 95% CI 0.57–1.10). In summary, being an only or other first born child and reporting a history of atopic conditions were associated with reduced NHL risk. Taken together, these data suggest a hypothesis that late exposure to infections in early life and atopic conditions, both of which are associated with a predominant Th2 response, are associated with a reduced risk of NHL.