Table 3 Effect estimates of the differences in immune cell counts in blood by recent night-shift work (past three days) compared to non-shift workers.

From: Immunological effects of shift work in healthcare workers

Cells/μl blood

Night-shift workers with no night shift in past three days (n = 196) B (95%-CI)a,b

Night-shift workers with night shift in past three days (n = 57) B (95%-CI)a,b

Monocytes

1.15* (1.05–1.27)

1.14* (1.01–1.28)

Granulocytes

1.05 (0.94–1.18)

0.97 (0.84–1.12)

Lymphocytes

1.03 (0.94–1.12)

1.12*c (1.01–1.26)

NK cells

0.90 (0.76–1.05)

0.89 (0.72–1.08)

NKT cells

1.02 (0.80–1.30)

1.10 (0.82–1.48)

B cells

1.03 (0.89–1.20)

1.16 (0.96–1.39)

T cells

1.04 (0.94–1.14)

1.16*c (1.03–1.31)

CD4 T cells

1.00 (0.90–1.11)

1.13c (0.99–1.29)

   CD4 effector memory T cells

1.07 (0.93–1.23)

1.12 (0.94–1.34)

   CD4 central memory T cells

1.02 (0.91–1.13)

1.13c (0.99–1.30)

   CD4 naive T cells

0.94 (0.80–1.11)

1.13c (0.92–1.38)

   CD4 effector T cells

0.83 (0.62–1.12)

0.89 (0.61–1.29)

CD8 T cells

1.12 (0.98–1.27)

1.23* (1.05–1.45)

   CD8 effector memory T cells

1.19 (0.95–1.49)

1.24 (0.94–1.65)

   CD8 central memory T cells

1.14 (0.97–1.35)

1.30* (1.05–1.61)

   CD8 naive T cells

1.05 (0.89–1.23)

1.25*c (1.02–1.52)

   CD8 effector T cells

1.02 (0.81–1.29)

0.96 (0.72–1.29)

CD4/CD8 T cell ratio

0.90 (0.79–1.01)

0.92 (0.79–1.07)

  1. Reference group: non-shift workers (n = 57).
  2. B, regression coefficient; CI, confidence interval.
  3. T cell subsets: Effector memory T cells: CD27−/CD45RO+, Central memory T cells: CD27+/CD45RO+, Naive T cells: CD27+/CD45RO−, Effector T cells: CD27−/CD45RO−.
  4. *Statistically significant difference (p < 0.05) between night-shift workers and non-shift workers tested using linear regression analysis for the log-transformed outcomes.
  5. aThe regression coefficients are ratios between geometric means of night-shift workers and non-shift workers.
  6. bAdjusted for age, gender, CMV status, occupation, educational level, general perceived health, smoking status, influenza vaccination status, and recent infection.
  7. cStatistically significant difference (p < 0.05) between night-shift workers with night shift in past three days and night-shift workers with no night shift in past three days tested using linear regression analysis for the log-transformed outcomes.
Back to article page