This study investigated the prevalence and characteristics of minor/major and major PPDS among mothers in the Canadian provinces and territories. The national prevalence rates of minor/major PPDS and major PPDS were found to be 8.46% and 8.69% respectively. The analysis revealed an association between total household income and both minor/major and major PPDS, which was found to decrease as household income increased. Immigration status, delivery at a young age, and a prior diagnosis of depression were all found to be positively associated with both forms of PPDS. Yet, the amount of stress during pregnancy and the lack of availability of support postpartum had the highest direct association with both minor/major and major PPDS.
This study found the national prevalence rate for major PPDS to be lower (8.69%) compared to a previous report conducted by Statistics Canada in 1998 (10-15%) . The lower national prevalence rate for major PPDS may partially be attributed to the timing of the EPDS survey in this study, which was administered 5 to 14 months postpartum. A meta-analysis that analyzed several international studies found, at 6 weeks postpartum, the mean American prevalence rate of PPD to be 15.4% and the mean prevalence rate in the United Kingdom to be 12.8% . The higher prevalence rate of PPDS observed in the Canadian territories may partially be attributed to the population being comprised of a greater proportion of aboriginal people. Studies have shown that aboriginal people are at increased risk of suffering from depression . The low minor/major and major PPDS prevalence rates found in the Atlantic provinces may be related to the reported high levels of social support .
The association between total household income and both minor/major and major PPDS were found to be higher with a decrease in lower income group, which has been seen previously in the literature. A meta-analysis based on 59 studies, also found that a decreased household income was associated with a greater risk for PPD . This may partly be attributed to the increased amount of stress placed on a mother due to the availability of limited financial means necessary for raising an infant [2, 26]. Variation in the external environment (urban/rural) in conjunction with socioeconomic status may also affect the association found.
Immigrants were at increased odds of experiencing minor/major PPDS (OR: 1.84, 95% CI: 1.41-2.40), and major PPDS (OR: 2.35, 95% CI: 1.77-3.13) compared to non-immigrants. Consistent with this finding, a Canadian study conducted with participants from the Calgary Health Region between 2001 and 2004, report that having been born outside of Canada was associated with an increased risk of 1.87 of developing minor/major PPD (95% CI: 1.17-3.00) . However, such an association may have been moderated by time since immigration, which may account for the effects of acculturation that could be present among immigrants who have spent substantial periods of time in Canada. The added stresses that accompany living in new surroundings among an unfamiliar culture, may compound the pressures that coincide with being a parent of a newborn.
Within the maternal characteristics, mothers between the ages of 15 and 19 years, a prior diagnosis of depression or past use of prescription antidepressants, smoking during the 3rd trimester, and a mother's stress level during pregnancy were all associated with experiencing PPDS. Although adolescent mothers were found to be associated with major PPDS (OR: 2.03, 95% CI: 1.09-3.78), the literature is not conclusive when it comes to the association between maternal age and PPD; subsequently it is regarded as a possible predictor of PPD [2, 5]. However, concerning the adolescent population, higher prevalence rates of PPD have been reported [5, 7, 33].
A prior diagnosis of depression or past use of prescription antidepressants was associated with a higher odds of experiencing both minor/major and major PPDS. This substantial higher risk of PPDS is in concordance with previous literature regarding depression history independent of childbirth [2, 21, 34]. A mother's stress level during pregnancy was significantly and substantially associated with experiencing symptoms of both minor/major and major PPDS. The current literature is in agreement with the findings of the present study [2, 5, 35]. Although occupation during pregnancy, living with a husband/partner, and planned pregnancy did not remain significant in the adjusted model, their relationship with PPDS may have been partially accounted for with the significantly higher associations observed between stress and the risk of experiencing minor/major and major PPDS (OR: 3.59, 95% CI: 2.58-5.00 and 6.98, 95% CI: 4.99-9.77 respectively). Accompanying the significant results for stress during pregnancy are significant results regarding the amount of support available to the mother postpartum. Using the referent category "most of the time" for the variable of support after pregnancy garnered a significantly stronger positive relationship with minor/major and major PPDS when support was available "some of the time" compared to "none of the time". This association is seen when a lack of necessary social support is present in the form of family and friends, as well as professionals [26, 36].
This was the first national study that analyzed the national, provincial, and territorial prevalence rates of minor/major and major PPDS. An extensive list of characteristics of PPDS was examined among a diverse and representative sample of Canadian women. Conducting this study with a large sample size increased the statistical power. The current study was based on a cross-sectional survey. Information bias may be present due to the self-report nature of the MES. A few of the characteristics assessed may be affected by recall bias, such as a mother's stress level during pregnancy, whether she adhered to the pregnancy weight gain guidelines, and recalling a past history of depression and/or treatment with antidepressants. However, recall bias was limited for the measurement of PPDS, due to the fact that the EPDS refers to the past seven days. A limitation of the study was the timing of the administration of the EPDS. Since the criteria for the date of the births was different for the provinces and the territories, it created a sampling period that ranged from 5 to 14 months postpartum, which ultimately garnered conservative minor/major and major PPDS prevalence rates. Although symptoms of PPD can last up to 14 months , there is a chance that lower prevalence rates were observed in our study, due to the fact that symptoms of PPD may have resolved by the time the participants were surveyed. In an attempt to account for this, the adjusted model controlled for the age of the infant. As well, since the MES was conducted during the winter months in the territories, seasonal variations in prevalence of PPDS may have been present , which may have inflated the prevalence rates seen in the territories. The fact that a confirmatory instrument was not used in this study presents a limitation; hence the results of the study refer to postpartum depression symptomatology. A lack of evidence regarding the validity of the EPDS when administered by telephone, may result in a misclassification bias. Another factor that may have affected the results of this study was the differences between the respondents and the non-respondents. However, in order to decrease the non-response bias, weighting adjustments designed by Statistics Canada were applied to all variables in the MES. Information about the mothers' knowledge of symptoms of PPD prior to delivery, as well as knowledge of various treatment options and accessibility to them, would have benefited the current study.