Diet and risk of multiple myeloma in Connecticut women

H. Dean Hosgood, Dalsu Baris, Shelia Hoar Zahm, Tongzhang Zheng, Amanda J. Cross

Research output: Contribution to journalArticlepeer-review

40 Scopus citations

Abstract

Multiple myeloma accounts for an estimated 19,900 incident cancer cases per year in the United States. A population-based case-control study, consisting of 179 incident cases and 691 controls, was conducted to examine the impact of diet on multiple myeloma risk. Diet was assessed using a food frequency questionnaire and odds ratios, 95% confidence intervals, and P-trends were calculated across quartiles of consumption. After controlling for potential confounders, we observed inverse associations for cooked tomatoes (P-trend = 0.002), cruciferous vegetables (P-trend = 0.01), fresh fish (P-trend < 0.001), alcohol (P-trend < 0.001), and vitamin A (P-trend < 0.001) with multiple myeloma risk. In contrast, consumption of cream soups (P-trend = 0.01), jello (P-trend = 0.01), ice cream (P-trend = 0.01), and pudding (P-trend < 0.001) were positively associated with multiple myeloma. Furthermore, there was a suggestion that carbohydrate intake may be positively associated, whereas vitamin D and calcium intake may be inversely associated, with multiple myeloma risk. Despite very limited data on dietary factors in relation to multiple myeloma, the findings from this study concur with previously published studies, suggesting an inverse association for consumption of fish, cruciferous vegetables and green vegetables, and a positive association for some dairy products.

Original languageEnglish (US)
Pages (from-to)1065-1076
Number of pages12
JournalCancer Causes and Control
Volume18
Issue number10
DOIs
StatePublished - Dec 2007
Externally publishedYes

Keywords

  • Case-control
  • Diet
  • Multiple myeloma
  • Risk factors

ASJC Scopus subject areas

  • Oncology
  • Cancer Research

Fingerprint

Dive into the research topics of 'Diet and risk of multiple myeloma in Connecticut women'. Together they form a unique fingerprint.

Cite this