Trends in haemoglobin levels from 1968 to 2017 and association with hormonal contraceptives: observations from the population study of women in Gothenburg, Sweden

Figures & data

Table 1. Participation rate in the population study of women in Gothenburg.

Examination year	38-year-olds, n (%)	50-year-olds, n (%)	Total, n
1968–1969	372 (91)	398 (91)	770
1980–1981	122 (85)	355 (82)	477
1992–1993	69 (72)	99 (76)	168
2004–2005	207 (60)	293 (58)	500
2016–2017	263 (63)	310 (73)	573
Total	1033	1455	2488

Figure 1. (a) Level of Hb in different examination years, mean values and SD. (b) Mean Hb level (g/L) in 38-year-old women. (c) Mean Hb level (g/L) in 50-year-old women.

Table 2a. Differences in mean values in Hb between 38-year-olds in different examination years.

Examination year	1980–1981	1992–1993	2004–2005	2016–2017
1968–1969	4.6 (1.1)***	7.5 (1.4)***	4.5 (0.9)***	2.1 (0.8)
1980–1981		2.9 (1.6)	–0.1 (1.2)	–2.5 (1.1)
1992–1993			–2.9 (1.5)	–5.4 (1.4)**
2004–2005				–2.5 (1.0)

Numbers describe changes in mean values in Hb (g/L) between different examination years, standard error in parentheses. A negative value indicates the mean value for the examination year in the column is greater than the mean value in the row.

*p < .05.

** p < .01.

*** p < .001.

Table 2b. Differences in mean values in Hb between 50-year-olds in different examination years.

Examination year	1980–1981	1992–1993	2004–2005	2016–2017
1968–1969	4.0 (0.8)***	8.0 (1.2)***	4.1 (0.8)***	1.5 (0.8)
1980–1981		4.0 (1.2)*	0.06 (0.9)	–2.6 (0.8)*
1992–1993			–3.9 (1.3)*	–6.5 (1.3)***
2004–2005				–2.6 (0.9)*

Numbers describe changes in mean values in Hb (g/L) between different examination years, standard error in parentheses. A negative value indicates the mean value for the examination year in the column is greater than the mean value in the row.

* p < .05.

*** p < .001.

Table 3. Changes in Hb after adjustments for covariates, with examination 2016–2017 used as a reference.

Examination year	38-year-olds	50-year-olds
	B-coefficient (95% CI)	B-coefficient (95% CI)
2016–2017	0	0
2004–2005	–2.7 (–4.7 to −0.8)**	–2.3 (–4.0 to −0.6)**
1992–1993	–5.7 (–8.7 to −2.8)***	–6.6 (–9.2 to −4.1)***
1980–1981	–3.4 (–6.2 to −0.6)*	–2.8 (–4.8 to −0.8)**
1968–1969	1.7 (–0.7 to 4.2)	1.5 (–0.6 to 3.5)
Covariates
Hormonal contraceptives	2.4 (0.6–4.2)**	3.0 (0.8–5.2)**
BMI	0.4 (0.2–0.6)***	0.3 (0.2–0.5)***
Smoking	3.4 (1.8–5.1)***	4.7 (3.5–6.0)***
Alcohol
Wine	0.4 (–0.6 to 1.4)	1.0 (0.1–1.8)*
Beer	0.1 (–0.8 to 1.0)	–0.04 (–0.8 to 0.7)
Spirits	0.2 (–1.2 to 1.5)	–0.8 (–1.8 to 0.3)
Education	0.2 (–1.0 to 1.4)	–0.2 (–1.1 to 0.8)

Dependent variable: Hb. B-coefficient: the degree of change in the dependent variable for every unit of change of the independent variable.

* p < .05.

** p < .01.

*** p < .001.

Figure 2. (a) Trends of anaemia and elevated Hb in 38-year-olds. Results from logistic regression models show significant non-linear trends in anaemia and elevated Hb, p = .004 and .0002, respectively. (b) Trends of anaemia and elevated Hb in 50-year-olds. Results from logistic regression models show significant non-linear trends in anaemia and elevated Hb, p = .0004 and .01, respectively.

Figure 3. Use of hormonal contraceptives.

Table

	38-years-old		50-years-old
Examination year	n	Mean (SD)	n	Mean (SD)
1968	372	135.18 (10.26)	398	137.24 (11.03)
1980	122	130.58 (9.64)	355	133.21 (11.21)
1992	69	127.72 (12.89)	99	129.25 (10.18)
2004	207	130.66 (10.59)	293	133.15 (11.57)
2016	263	133.11 (10.15)	310	135.76 (9.44)
Total	1033	132.71 (10.66)	1455	134.58 (11.04)