Table 3.
Characteristics of Case-Control Studies Excluded From a Meta-Analysis of the Association Between Atopy and Leukemia, 1952–March 2009
| First Author, Country, and Year (Reference No.) | Study Design | Reason for Exclusion | Results |
| Manning, United States, 1957 (56) | Case-control | Specialty clinic-based control group | Suggestion of increased risk of childhood leukemia associated with allergies (i.e., asthma, eczema, hay fever, hives, other) (OR = 2.44, 95% CI: 0.97, 6.14).a |
| Stewart, United Kingdom, 1958 (57) | Case-control | Results presented for childhood cancer cases overall, not for leukemia cases. | No evidence for an association between allergic conditions and childhood malignancy (OR = 1.12, 95% CI: 0.65, 1.94).a |
| Ager, United States, 1965 (58) | Case-control | Very low prevalence of atopic disease reported among cases and controls under age 5 years. This may have been due to misclassification, since exposure was measured first via maternal interview, followed up with medical record verification only for those with a positive maternal report. | No evidence for an association between death from childhood leukemia and asthma, eczema, or hives (for asthma, OR = 0.66, 95% CI: 0.11, 4.03; for eczema, OR = 0.99, 95% CI: 0.14, 7.16; for hives, OR = 0.99, 95% CI: 0.14, 7.16).a |
| Smith, United States, 1973 (55) | Case-control | Study overlapped with Bross (20). Cell counts from this paper were used to calculate overall age-adjusted OR. | Results included in meta-analysis |
| Natarajan, United States, 1973 (59) | Case-control | Study overlapped with Bross (20) | The association between allergic disease (i.e., asthma, hives, eczema) and childhood leukemia was greater among those with maternal exposure to preconception radiation (OR = 4.6, P = 0.0001) than among those with no preconception radiation (OR = 1.9, P = 0.03). |
| Viadana, United States, 1974 (60) | Case-control | Results presented for leukemia cases aged ≥15 years. Unable to locate author to request data for 15- to 18-year-olds. | Not applicable |
| Bross, United States, 1974 (61) | Case-control | Study overlapped with Bross (20) | Association between allergic disease (i.e., asthma, hives, eczema) or bacterial disease (i.e., pneumonia, dysentery, rheumatic fever) and childhood leukemia was greater among those with maternal preconceptional, intrauterine, or postnatal radiation exposure (OR = 4.1, P = 0.0001) than among those with no prior radiation history (OR = 1.6, P = 0.23). |
| Gibson, United States, 1976 (62) | Case-control | Results presented for leukemia cases aged ≥15 years. Unable to locate author to request data for 15- to 18-year-olds. | Not applicable |
| Magnani, Italy, 1990 (63) | Case-control | Hospital-based control group | Inverse association observed between allergic diseases and childhood ALL after adjustment for SES (OR = 0.4, 95% CI: 0.2, 0.8). |
| Zheng, China, 1993 (15) | Case-control | Results presented for leukemia cases aged ≥15 years. Per personal communication with first author, unable to obtain data for 15- to 18-year-olds. | Not applicable |
| Buckley, United States, 1994 (64) | Case-control | Results presented for family history of allergy, not for personal history of allergy. | Family history of allergies (i.e., asthma, hay fever, hives, food or drug allergy) in siblings, parents, and/or grandparents was associated with a modestly increased risk of childhood ALL after adjustment for birth year, race, income, geographic region, and family size (OR = 1.3, P < 0.05). |
| Petridou, Greece, 1997 (65) | Case-control | Hospital-based control group and exposure of hospitalization for allergic disease. | Nonsignificant inverse association observed between hospitalization for allergic disease and childhood leukemia after adjustment for gender, age, location of residence, and other biomedical variables of interest (OR = 0.36, 95% CI: 0.09, 1.43). |
| Kaatsch, Germany, 1998 (66) | 2 case-control studies | Study overlapped with Schüz (22) | Nonsignificant inverse association observed between allergy and childhood leukemia after adjustment for age, sex, place of residence, and SES (OR = 0.90, 95% CI: 0.63, 1.29). |
| Schüz, Germany, 1999 (67) | 2 case-control studies | Study overlapped with Schüz (22) | Inverse association observed between allergy and childhood leukemia after adjustment for gender, date of birth, district of residence, and SES (OR = 0.6, 95% CI: 0.5, 0.8). |
Abbreviations: ALL, acute lymphoblastic leukemia; CI, confidence interval; OR, odds ratio; SES, socioeconomic status.
a
Crude odds ratio was calculated from the data provided.