Delivery factors identified in the study by MacArthur et al (1991) as independent risk factors for long-term fatigue included multiple pregnancy, longer first stage of labour, inhalation anaesthesia and postpartum haemorrhage but not operative delivery. Milligan et al (1990) did show a significant association between caesarean section and fatigue in a group of 259 women surveyed before discharge from hospital, but when the same women were surveyed at 6 weeks and 3 months, no significant effect was seen. Findings from the other two large studies previously quoted are consistent with an early excess for caesarean section which diminishes over time. At 12–18 months fatigue was not associated with caesarean section in Glazener’s study, but there was an association in the reports of fatigue at 0–13 days and at up to 8 weeks (Glazener 1995). Tiredness was more common in women delivered by caesarean section and surveyed at 6–7 months in Brown & Lumley’s study (1998) but the effect was not statistically significant (Brown & Lumley 1998). The other studies have not reported on the other delivery factors.

Few studies have examined the physiological determinants of postnatal fatigue, but Paterson et al (1994) investigated the impact of a low (<10.5 g/dl) haemoglobin (Hb) on the postnatal mental and physical health of 1010 women. Hb results were obtained at ‘booking’, 34 weeks, third day postdelivery and at 6 weeks postpartum. Women were asked to complete questionnaires about their health at 10 days, 4 weeks and 6 weeks after delivery on their health. Full data including Hb were obtained from only 52% of the original sample. A low Hb on day 3 was more likely to be diagnosed in younger women (aged under 25); among primiparae; women who had had operative or instrumental delivery; women who had a low Hb at 34 weeks gestation; and those with a blood loss of over 250 ml recorded at delivery. Some of these variables are interrelated (for example, parity and mode of delivery) but statistical analysis to determine independence of effect was not undertaken. Women with a low Hb at day 3 were significantly more likely to report feeling low in energy in the questionnaire at day 10. They also were more likely to report being breathless; faint and dizzy; to have painful sutures and tingling of the fingers or toes. By 6 weeks no difference was apparent between groups with and without low Hb, but it is not clear from the paper what action was taken on the basis of the Hb result, though it is apparent that some of the women in the study were taking iron supplements.

Other physical problems worth considering if a woman reports extreme tiredness are infections and, though much less common, thyroid disorders and cardiomyopathy (Atkinson & Baxley 1994). CFS, a syndrome where persistent fatigue is felt and significant disability experienced without apparent cause, may affect women after childbirth, but no studies have reported rates of occurrence in this group. There is ongoing debate about the most appropriate diagnostic criteria for CFS, with two commonly cited definitions, the Oxford criteria and the US Centers for Disease Control and Prevention criteria, both stating that debilitating fatigue must be present for at least 6 months, that there is some functional impairment and that this is not associated with any other identifiable condition (Fukuda et al., 1994 and Sharpe et al., 1991). These definitions do differ, however, in the number and severity of symptoms which must be present. In a general population study of fatigue, Chen (1986) found heavier women more likely to be fatigued than lighter ones (based on body mass index).

Fatigue is a well-recognised symptom of depression though it may be difficult to clarify whether anxiety and depression are the cause or result of fatigue (Unterman et al 1990). Of 1065 postpartum women who reported depression in MacArthur et al’s 1991 study, 47% (496) also reported extreme tiredness. These 496 women constituted 35% of all women (1427) who reported extreme tiredness. In Brown & Lumley’s (2000) survey of 1336 Australian women at 6–7 months postpartum, tiredness was 3.4 times more likely to be reported by women with scores on the Edinburgh Postnatal Depression Score indicating probable depression. Ansara and colleagues (2005) undertook a study in Toronto to examine the extent and correlates of common physical health symptoms in women 2 months after giving birth. Women were recruited from six Toronto area hospitals and interviewed by telephone 8–10 weeks later. Of 332 women approached, 200 (60%) were interviewed. Most women (96%) reported at least one physical health problem at 2 months. Stepwise logistic regression analysis showed that a self-reported history of antenatal depression was a significant predictor of excessive fatigue after giving birth.

Gardner (1991) collected data by questionnaire at 2 days, 2 weeks and 6 weeks postpartum from 35 of 68 randomly selected American women. The sample were only mildly fatigued, as scored on the Rhoten Fatigue Scale, and fatigue and depression scores were significantly, but not strongly, correlated at 2 days and 2 weeks, but not at 6 weeks postpartum. Milligan et al (1990) found that comparatively little of the variation in fatigue at 6 weeks and 3 months postpartum was explained by other factors in models which controlled for depression.

The correlation between fatigue symptoms and psychiatric disorder has also been demonstrated in general population studies of patients with CFS. In a nested case–control study, 60% of 214 chronic fatigue patients were found to have a current psychiatric disorder compared to 19% of 214 matched controls (Wessely et al 1996).

It is difficult to determine causal association between fatigue and depression, although evidence from prospective studies indicates that sleep disruption may be a risk factor for postnatal depression. Dennis & Ross (2005) examined relationships between infant sleep patterns, maternal fatigue and the development of postnatal depression in women with no major depressive symptomatology at 1 week after giving birth; 505 women who had an EPDS score of <13 at 1 week after giving birth completed a questionnaire at 4 and 8 weeks. Those women who had EPDS scores of >12 at 4 and 8 weeks were significantly more likely to report that their babies cried often, that they were woken three times or more between 10pm and 6am, received less than 6 hours of sleep in a 24-hour period over the previous week, and considered their baby’s sleep pattern did not allow them to get a reasonable amount of sleep. The researchers concluded that infant sleep patterns and maternal fatigue were associated with the onset of depression after birth, and preventive interventions should be designed to reduce sleep deprivation in the first few weeks after birth. It is also possible that women experiencing depression may be more likely to overreport crying and waking in their babies.