Premium
Risk of cancer in patients with medically diagnosed hay fever or allergic rhinitis
Author(s) -
Hemminki Kari,
Försti Asta,
Fallah Mahdi,
Sundquist Jan,
Sundquist Kristina,
Ji Jianguang
Publication year - 2014
Publication title -
international journal of cancer
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.475
H-Index - 234
eISSN - 1097-0215
pISSN - 0020-7136
DOI - 10.1002/ijc.28873
Subject(s) - medicine , hay fever , cancer , incidence (geometry) , population , prostate cancer , breast cancer , cancer registry , testicular cancer , endocrine system , esophageal cancer , cohort , genitourinary system , case control study , gastroenterology , asthma , hormone , physics , environmental health , optics
Data on allergic conditions as risk or protective factors for cancer are controversial probably because most studies have used self‐reported data on mixed groups of allergies in a case–control setting. We define cancer risks in medically diagnosed hay fever/allergic rhinitis patients in a nationwide cohort study. A total of 138,723 hay fever/allergic rhinitis patients were identified from three Swedish health care databases and standardized incidence ratios (SIRs) were calculated for subsequent cancers identified from the Swedish Cancer Registry. Overall cancer risk was not changed (SIR 1.03). For individual cancers, the highest SIR was observed for nasal cancer (SIR 2.63), followed by testicular (1.46) and endocrine tumors (1.42), and kidney (1.31), prostate (1.18) and breast (1.11) cancers. The results were consistent in the three sources of data and all SIRs were above unity, albeit mainly not statistically significant. The SIRs for nervous system tumors were above unity and of borderline significance. SIRs were decreased for esophageal (0.50), liver (0.62) and lung (0.78) cancers, and the three sources of data agreed in the direction of the effect. The increased risks for testicular, renal, prostate and endocrine cancers may be explained by immunological mechanisms. Excess risk for these cancer accounts for a significant population attributable fraction. Nervous system cancers showed a borderline increase and none of the histological types were significantly decreased, providing strong evidence against the published case–control studies, which have reported protective effects. The reasons for the reduced risks for esophageal, liver and lung cancer remain to be explained.