Author, year / Exposurea / Myeloid / Lymphoid
N / RR / 95% CI / N / RR / 95% CI
Pinkerton et al, 20041 / Total cohort / 15 / 1.44 / 0.80-2.37 / 3 / 0.60 / 0.12-1.75
Beane-Freeman et al, 20092 / Peak > 4 ppm / 19 / 1.78 / 0.87-3.64 / 14 / 1.15 / 0.54-2.47
Stroup et al, 19863 / All (1925-79) / 5 / --b / -- / 1 / --b / --
Hauptmann et al, 20094 / Peak > 9.3 ppm / 12 / 13.0c / 1.4-116.9 / 15 / 0.6c / 0.2-1.3
Hayes et al, 19905d / Total cohort / 24 / 1.57 / 1.01-2.34 / 7 / 0.74 / 0.29-1.53
Walrath & Fraumeni, 19836d / Total cohort / 6 / 1.46 / 0.53-3.18 / 4 / 1.54 / 0.41-3.94
Walrath & Fraumeni, 19847d / Total cohort / 6 / 1.50 / 0.55-3.26 / 0 / 0.0e / --
Meta-analysis Results 1.55 1.18-2.03 0.95 0.60-1.51
Abbreviations: CI, confidence interval; N, number of exposed cases; RR, relative risk
a For some studies, the exposure category is not the same as in Table 1 in the main paper since some studies did not provide data for both myeloid and lymphoid leukemia for the exposure category in Table 1.
b Expected numbers were not provided.
c Data on lymphatic leukemia were not provided. "Lymphoid" includes all lymphohematopoietic malignancies of lymphoid origin. Uses subjects with no embalmings as the reference group since the RR with < 500 embalmings as the reference group is not provided for both.
d These three studies involve the same cohort of embalmers and funeral directors as Hauptmann et al 2009. They are presented since they provide RRs specifically for lymphatic leukemia.
e 2.2 cases expected.
Explanation of Supplemental Table 1. This table provides RRs for all studies that provided separate data for both myeloid and lymphatic leukemia. In every study where RRs are provided except for one (Walrath and Fraumeni, 1983), the RR is greater for myeloid leukemia than for lymphatic leukemia. A meta-analysis of these data was done using the same statistical methods described in the main paper and excluding the studies of Stroup et al 1986 (no RRs provided) and Hauptmann et al 2009 (RR only provided for all lymphohematopoietic malignancies of lymphoid origin combined). In this meta-analysis, since Walrath and Fraumeni, 1984 had no lymphatic leukemia cases the relative risk was arbitrarily assigned a value of 0.5 and Byars approximation was used to estimate the 95% confidence interval. Assigning values lower than 0.5 had little effect. As seen in this table, the summary relative risk is higher for myeloid leukemia (RR = 1.55; 95% CI, 1.18-2.03) than for lymphatic leukemia (RR = 0.95; 95% CI 0.60-1.51). This provides some evidence that formaldehyde is more strongly associated with myeloid than lymphatic leukemia and suggests that combining these two subtypes would drive meta-analysis summary relative risks towards 1.0.
References for Supplementary Table 1
1. Pinkerton LE, Hein MJ, Stayner LT. Mortality among a cohort of garment workers exposed to formaldehyde: an update. Occup Environ Med 2004;61(3):193-200.
2. Beane Freeman LE, Blair A, Lubin JH, Stewart PA, Hayes RB, Hoover RN, et al. Mortality from lymphohematopoietic malignancies among workers in formaldehyde industries: the National Cancer Institute Cohort. J Natl Cancer Inst 2009;101(10):751-61.
3. Stroup NE, Blair A, Erikson GE. Brain cancer and other causes of death in anatomists. J Natl Cancer Inst 1986;77(6):1217-24.
4. Hauptmann M, Stewart PA, Lubin JH, Beane Freeman LE, Hornung RW, Herrick RF, et al. Mortality from lymphohematopoietic malignancies and brain cancer among embalmers exposed to formaldehyde. J Natl Cancer Inst 2009;101(24):1696-708.
5. Hayes RB, Blair A, Stewart PA, Herrick RF, Mahar H. Mortality of U.S. embalmers and funeral directors. Am J Ind Med 1990;18(6):641-52.
6. Walrath J, Fraumeni JF, Jr. Mortality patterns among embalmers. Int J Cancer 1983;31(4):407-11.
7. Walrath J, Fraumeni JF, Jr. Cancer and other causes of death among embalmers. Cancer Res 1984;44(10):4638-41.