SEARCH

SEARCH BY CITATION

Keywords:

  • empirical test;
  • explanatory theory;
  • Korea;
  • midwifery;
  • nursing;
  • postpartum fatigue;
  • structural equation modelling

Abstract

  1. Top of page
  2. Abstract
  3. What is already known about this topic
  4. Introduction
  5. Background
  6. The study
  7. Results
  8. Discussion
  9. Conclusion
  10. Funding
  11. Conflict of interest
  12. Author contributions
  13. Acknowledgements
  14. References

song j.-e., chang s.-b., park s.-m., kim s. & nam c.m. (2010) Empirical test of an explanatory theory of postpartum fatigue in Korea. Journal of Advanced Nursing66(12), 2627–2639.

Abstract

Aim.  This paper is a report of a study designed to test an explanatory theory of postpartum fatigue.

Background.  Postpartum fatigue is influenced by various factors and affects a mother’s performance. A full understanding of postpartum fatigue is very important for developing effective nursing strategies to reduce postpartum fatigue and enhance mothers’ performance.

Methods.  Healthy postpartum women were recruited from five medical centers and one midwifery office in urban area in Korea (n = 291) by convenience sampling. Data were collected at 4- to 8-week follow-up visits after childbirth in 2006, using a self-report questionnaire. The proposed fatigue theory incorporated postpartum fatigue, postpartum depression, sleep quality, childcare stress, unsatisfactory feeding, social support, infant difficulty and satisfaction with Sanhujori, the Korean traditional postpartum care provided for 3 weeks following delivery by non-professional caregivers. Structural equation modelling was used to test the explanatory theory of postpartum fatigue.

Results.  The modified fatigue theory showed good fit and high compatibility with the empirical data. In the final explanatory theory, postpartum depression and sleep quality directly affected postpartum fatigue, while childcare stress and the cultural phenomenon of Sanhujori had indirect effects on postpartum fatigue, via postpartum depression and sleep quality respectively.

Conclusion.  These findings suggest the potential role of comprehensive nursing focused on decreasing postpartum depression and improving sleep quality as a way to decrease postpartum fatigue. Also, nursing strategies for decreasing childcare stress and enhancing Sanhujori satisfaction may be helpful in reducing postpartum fatigue in Korean mothers.


What is already known about this topic

  1. Top of page
  2. Abstract
  3. What is already known about this topic
  4. Introduction
  5. Background
  6. The study
  7. Results
  8. Discussion
  9. Conclusion
  10. Funding
  11. Conflict of interest
  12. Author contributions
  13. Acknowledgements
  14. References
  • Postpartum fatigue is the most common subjective symptom that first-time mothers experience after childbirth.
  • Postpartum fatigue can be affected by various physiological, psychological, and situational factors and has negative effects on mothers’ performance.
  • Only direct relationships between fatigue and depression, sleep, feeding problems, social support, and infant temperament have been tested.

What this paper adds

  • Postpartum depression and sleep quality directly affected postpartum fatigue, as well as acting as mediators for other variables, and the psychological pathway via postpartum depression was stronger than the physiological pathway via sleep quality in inducing postpartum fatigue.
  • Childcare stress was statistically linked to the psychological pathway of postpartum fatigue via postpartum depression, while the cultural phenomenon of Sanhujori was significantly related to the physiological pathway of postpartum fatigue via sleep quality.

Implications for practice and/or policy

  • Nursing care to alleviate postpartum fatigue might be focused on reducing postpartum depression and childcare stress of the psychological pathway of postpartum fatigue for Korean women.
  • Nursing care to improve sleep quality and Sanhujori satisfaction of the physiological pathway should be considered to decrease postpartum fatigue for Korean women.

Introduction

  1. Top of page
  2. Abstract
  3. What is already known about this topic
  4. Introduction
  5. Background
  6. The study
  7. Results
  8. Discussion
  9. Conclusion
  10. Funding
  11. Conflict of interest
  12. Author contributions
  13. Acknowledgements
  14. References

Childbirth is a dramatic situation where a new life is created. However, enormous amounts of physical and psychological energy are consumed during parturition and adaptation to the new role as a mother. Therefore first-time mothers especially experience fatigue (Milligan 1989, Mercer & Walker 2006, Rychnovsky 2007). Although there are differences in the frequency of postpartum fatigue reported in the literature, postpartum fatigue is the most common unpleasant symptom following childbirth (McGovern et al. 2006, Rychnovsky 2007) and one of the major concerns among the postpartum women (Troy 2003, Taylor & Johnson 2008).

If a postpartum woman experiences fatigue, negative consequences may occur (Corwin & Arbour 2007). In addition to the risk of reduced self-care capacity and health problems related to delayed physical recovery, the ability to care for the newborn baby may also be negatively affected. Furthermore, postpartum fatigue can negatively affect family relationships by weakening the capacity for interpersonal relationships (Pugh & Milligan 1993). Considering that the goal of nursing for first-time mothers is to help with physical and psychological recovery as well as playing the role of mother successfully (Reeder et al. 1997, p. 664), a better understanding of the phenomenon of postpartum fatigue would be beneficial for nurses and these women.

Little research was available on postpartum fatigue until the 1980s (Rychnovsky 2007). After Milligan’s (1989) classic work on childbearing fatigue, the topic has been studied as a phenomenon distinct from general fatigue or cancer-related fatigue (Rychnovsky 2007). Pugh and Milligan (1993) first proposed ‘a framework for the study of childbearing fatigue’, which was developed to guide clinical research in preventing or reducing the impact of fatigue in childbearing women. Many physical, psychological and situational factors that may predispose women towards fatigue are described in this framework. Of these, feeding problems (Milligan 1989, Hantos 1993, Pugh & Milligan 1995, Wambach 1998, Fisher et al. 2004), infant temperament (Milligan 1989, Hantos 1993, Wambach 1998), postpartum depression (Milligan 1989, Gardner 1991, Webster 1994, Wambach 1998, Rychnovsky 2007), social support (Milligan 1989, Webster 1994) and sleeping problems (Hantos 1993, Wambach 1998, Lee & Zaffke 1999, Hunter et al. 2009, Rychnovsky & Hunter 2009) are supported by empirical research as predisposing factors for postpartum fatigue. However, previous studies have not provided integrated knowledge about the paths through which these variables contribute to postpartum fatigue.

In addition, even though Milligan (1989) suggested that the role strain and stress that can be experienced during maternal role attainment should be considered in relation to postpartum fatigue, previous studies have rarely included them. Therefore, the relationship between stress and fatigue in postpartum women needs to be clarified in considering relationships with other influencing factors.

On the other hand, although it is common to offer special help and support for postpartum woman immediately after childbirth, Korea has a unique tradition of postpartum care called ‘Sanhujori’. There are six principles of Sanhujori: (1) augmentation of heat and avoidance of cold, (2) absolute resting without work such as housework, (3) eating well for recovery, (4) protecting the body from harmful strains, (5) keeping clean and (6) ‘handling with the whole heart’ i.e. receiving the utmost care (Yoo 1993). A parturient woman is usually taken care of by non-professional caregivers such as her mother or mother-in-law throughout the first 3 weeks (21 days) after childbirth to adhere thoroughly to these six principles. According to Korean beliefs, Sanhujori is related to women’s health straight after childbirth and is a critical factor in assuring their long-term health (Kim & Yoo 1998, Yoo 1998, Yoo et al. 1998, Chong & Yoo 1999, Ahn 2005). Postpartum Korean women have special expectations for doing Sanhujori well for 3 weeks after childbirth, and when such expectations are not satisfied, this can negatively influence the woman’s physical and psychological health (Yoo 1997, Ahn 2005). Therefore it is necessary to consider such cultural characteristics as an influencing factor in postpartum fatigue.

Background

  1. Top of page
  2. Abstract
  3. What is already known about this topic
  4. Introduction
  5. Background
  6. The study
  7. Results
  8. Discussion
  9. Conclusion
  10. Funding
  11. Conflict of interest
  12. Author contributions
  13. Acknowledgements
  14. References

Development of a theory and definition of postpartum fatigue

Based on Pugh and Milligan’s (1993) classic framework and a review of previous studies related to postpartum fatigue, we developed a theory of postpartum fatigue in Korean women, illustrated in Figure 1. In this study, the definition of fatigue is ‘an overwhelming sustained sense of exhaustion and decreased capacity for physical and mental work’, according to the work of Pugh and Milligan (1993), which is also the definition of the North American Nursing Diagnosis Association (NANDA) (Carpenito-Moyet 2006, p. 155). This definition of fatigue represents an unpleasant sensation in contrast to the satisfied feeling of having done a job well or feeling good after exercise (Carpenito-Moyet 2006, p. 156).

image

Figure 1.  Theoretical framework.

Download figure to PowerPoint

In the theoretical framework of this study, social support and infant temperament are exogenous variables, and postpartum fatigue, postpartum depression, childcare stress, sleep quality, breastfeeding difficulty and Sanhujori satisfaction are endogenous variables.

Relationships among the research variables

When reviewing previous studies, we found a controversy about the relationship between fatigue and depression. Although there is an opinion that postpartum fatigue is a predicting factor for postpartum depression (Beck 2001, 2008, Corwin & Arbour 2007, Runquist 2007), depression is also explained as one of the important psychological factors that cause fatigue in the framework for the study of childbearing fatigue (Pugh & Milligan 1993), the unpleasant symptom theory (Lenz et al. 1997), and many empirical studies related to postpartum fatigue (Milligan 1989, Gardner 1991, Wambach 1998, Rychnovsky 2007). Therefore we set up a path in which postpartum depression directly affects postpartum fatigue.

Sleep quality or problem is an important variable that is most frequently reported in studies on fatigue. Unsatisfactory sleep or segmented sleep can cause fatigue by interfering with the deep-sleep rapid eye movement stage, and has been significantly related to postpartum fatigue in various studies (Hantos 1993, Webster 1994, Wambach 1998, Lee & Zaffke 1999, Hunter et al. 2009, Rychnovsky & Hunter 2009). Therefore we set up a path in which sleep quality directly affects postpartum fatigue.

Feeding difficulties are one of the main concerns for first-time mothers (Wambach 1998, Haku 2007), and previous studies have shown a significant relationship between a feeding problem and postpartum fatigue (Milligan 1989, Hantos 1993, Pugh & Milligan 1995, Wambach 1998, Fisher et al. 2004). Therefore, we set up a path in which unsatisfactory feeding directly affects postpartum fatigue. Also, unsatisfactory feeding may increase a mother’s stress and anxiety (Fisher et al. 2004, Haku 2007) and the chance of waking more frequently during the night, which may lead to segmented or unsatisfactory sleep (Alley & Rogers 1986, Rychnovsky 2007, Hunter et al. 2009). Therefore we set up a path in which unsatisfactory feeding directly affects childcare stress and sleep quality, and indirectly affects postpartum fatigue through these variables.

Reports on the relationship between childcare stress and postpartum fatigue are limited. Although it is explained that the process of acquiring the maternal role causes tension and stress (Mercer & Walker 2006, Fontenot 2007), and stress during transition to motherhood may be related to postpartum fatigue (Milligan 1989, Wambach 1998), the relationship between childcare stress and postpartum fatigue has rarely been supported by empirical data. Considering that various researchers explain childcare stress as a significant predicting factor of postpartum depression rather than postpartum fatigue (Barnet et al. 1996, Beck 2001, Honey et al. 2003, Park et al. 2004, Leung et al. 2005), childcare stress was proposed as indirectly affecting postpartum fatigue through postpartum depression in this study.

Infant temperament is also considered a predicting factor of postpartum fatigue, such that mothers of fussier infants are more likely to experience fatigue (Milligan 1989, Hantos 1993, Wambach 1998). Therefore we set up a path in which difficult infant temperament directly affects postpartum fatigue. Also, fussy infant temperament has an impact on mothers’ childcare stress (Fisher et al. 2004), feeding difficulty (Fisher et al. 2004), inferior quality of sleep (Dennis & Ross 2005) and postpartum depression (Beck 2001, Honey et al. 2003, Fisher et al. 2004, Dennis & Ross 2005). Consequently, the quality of Sanhujori deteriorates because it is difficult to take sufficient rest (Yoo 1993, 1997, 1998). Therefore we set up paths in which infant temperament directly affects childcare stress, postpartum depression, unsatisfactory feeding, sleep quality and Sanhujori satisfaction, and indirectly affects postpartum fatigue through these variables.

Social support is also important in the study of postpartum fatigue. Postpartum fatigue has been shown to increase with an absence or only a small number of helpers (Gardner & Campbell 1991), and social support had been found to reduce postpartum fatigue along with sleep and rest (Webster 1994). Therefore we set up a path in which social support directly affects postpartum fatigue. Also Milligan (1989) reported that social support buffered breastfeeding difficulties at 6 weeks after delivery, and other studies have shown that support from significant others reduces breastfeeding difficulties (Freed & Fraley 1993, Freed et al. 1993, Haku 2007). Furthermore, social support has also been widely recognized as an influencing factor in depression (Barnet et al. 1996, Beck 2001, Honey et al. 2003, Park et al. 2004, Gao et al. 2009) and childcare stress (Hung & Chung 2001, Warren 2004, Hung 2005), and has also been reported to be associated with improved quality and satisfaction of Sanhujori in Korea (Yoo 1993, 1997, 1998). Therefore we set up paths in which social support directly affects unsatisfactory feeding, postpartum depression, childcare stress and Sanhujori satisfaction.

Lastly, Sanhujori, the traditional Korean postpartum care, is influential in affecting sleep quality, psychological and physical recovery, and general health of the parturient woman (Yoo 1993, 1997, 1998, Kim & Yoo 1998, Yoo et al. 1998, Chong & Yoo 1999, Ahn 2005). Therefore we set up paths in which Sanhujori satisfaction directly affects sleep quality, postpartum depression and postpartum fatigue.

The study

  1. Top of page
  2. Abstract
  3. What is already known about this topic
  4. Introduction
  5. Background
  6. The study
  7. Results
  8. Discussion
  9. Conclusion
  10. Funding
  11. Conflict of interest
  12. Author contributions
  13. Acknowledgements
  14. References

Aim

The aim of the study was to test an explanatory theory of postpartum fatigue.

Design

A cross-sectional correlation design was used and the data were collected in Korea during 2006 by a self-report questionnaire.

Participants

A convenience sample of 320 postpartum women was recruited from the outpatient department of obstetrics at five medical centers and one midwifery practice in urban area in Korea, and 291 women completed the instrument without missing data. They were all healthy primiparous women revisiting the clinical setting for routine check-up after childbirth. The inclusion criteria were (1) gave birth after 37 weeks gestation, (2) infant birth weight over 2500 g, and (3) within 4–8 weeks after delivery. Women with health problems such as bleeding or anaemia (Hgb <10·0 mg/dL), hyper- or hypothyroidism, postpartum infection, or a history of depression, or those who had delivered twins were excluded. The reason for selecting first-time mothers within 4–8 weeks after childbirth was to select participants within a puerperal period (i.e. 6–8 weeks after delivery) and exclude those during the Sanhujori duration (i.e. first 3 weeks after childbirth) in Korea (Yoo 1993), and because first-time mothers have been especially noted to experience high levels of fatigue in relation to the process of maternal role attainment (Milligan 1989, Mercer & Walker 2006, Rychnovsky 2007).

To determine the minimum sample size, we used the hypothesis-testing framework for root mean square error of approximation (RMSEA) as an indicator of statistical power. When the degrees of freedom (d.f.) of the data are 82, the statistical significance level (alpha) is 0·05, RMSEA (H0) is 0·00, and RMSEA (H1) is 0·05, then the required N is 179. Our study included 291 participants, which can be considered a sufficient sample size for SEM analysis (MacCallum et al. 1996, 2006).

Data collection

Data were collected using a self report questionnaire and completed during while waiting in the outpatient department. The time for completing a questionnaire was 10–15 minutes. The questionnaire consisted of demographic items and instruments for measuring the major concepts in the theoretical model, as presented in Table 1. The Cronbach’s alphas calculated from the current data are included in this table, showing that all had reasonably high internal consistency values.

Table 1.   Descriptive statistics for study variables (n = 291)
Study variablesPossible rangeObtained rangeMean ± sdInterpretation of higher scoresCronbach’s alpha
Postpartum fatigue10–4010–4022·00 ± 6·06Greater fatigue0·88
 Physical fatigue6–246–2413·80 ± 3·72
 Mental fatigue4–164–168·19 ± 2·98
Postpartum depression0–301–238·59 ± 4·51Greater depression0·80
Sleep quality0–200–197·18 ± 3·90Better sleep quality0·81
 Sleep satisfaction0–100–103·92 ± 2·17
 Sleep non-disturbance0–100–103·26 ± 2·10
Child care stress14–7014–6336·90 ± 9·30Greater childcare stress0·85
Unsatisfactory feeding0–200–2010·69 ± 5·10Less satisfactory feeding0·73
 Feeding difficulty0–100–105·56 ± 2·75
 Feeding concern0–100–105·13 ± 2·99
Sanhujori satisfaction7–2811–2820·87 ± 3·74Greater Sanhujori satisfaction0·82
 With contents6–2410–2418·00 ± 3·16
 With duration1–41–42·87 ± 0·81
Infant difficulty4–284–2715·73 ± 4·13Greater infant difficulty0·76
 Frequency of fussiness1–71–73·20 ± 1·22
 Difficulty of soothing1–71–74·47 ± 1·43
 Mood variability1–71–74·17 ± 1·22
 General fussiness1–71–73·89 ± 1·59
Social support21–8438–8472·99 ± 9·67Better social support 
 Emotional support8–3214–3227·24 ± 3·690·85
 Instrumental support13–5216–5245·75 ± 7·340·94
Postpartum fatigue

Postpartum fatigue was measured using the modified shortened 10-item fatigue scale developed by Milligan et al. (1997) and translated into Korean by Song (2007), with modification of the response format from dichotomous (yes/no) to a 4-point Likert scale (1–4). The validity of the Korean translation was evaluated by three bilingual clinical experts, and content validity and internal consistency reliability (Cronbach’s α = 0·82) were tested for use with Korean women (Song 2007). This inventory measures mental fatigue (6-items) and physical fatigue (4-items), with higher scores indicating higher degree of fatigue.

Postpartum depression

Postpartum depression was measured using the Edinburgh Postnatal Depression Scale (EPDS) developed by Cox et al. (1987) and translated into Korean by Kim (2003). The EPDS has been developed to assist primary healthcare professionals to detect women suffering from postpartum depression, and is one of the most widely used self-report tools for postpartum depression. This instrument consists of 10-items rated on a 4-point Likert scale (0–3), with higher scores indicating higher degrees of depression. The internal consistency reliability (Cronbach’s α = 0·79) and content validity have been established in Korean postpartum women (Kim 2003).

Sleep quality

Sleep quality was measured using two items with a 10-point numerical rating scale developed by Song (2007) to assess general satisfaction of sleep and degree of disruption during the past week. A rating of zero means non-satisfactory sleep or completely disruptive sleep, and a rating of ten describes greatest satisfactory sleep or non-disruptive sleep. The instrument’s internal consistency reliability (Cronbach’s α = 0·77) has been reported in Korean postpartum women (Song 2007).

Childcare stress

Childcare stress was measured using the Childcare Stress Inventory (CSI) developed by Cutrona (1984) and translated into Korean by Cheon (1990). The CSI consists of 14-items rated on a 5-point Likert scale (1–5), with higher scores indicating higher degrees of childcare stress. The internal consistency reliability (Cronbach’s α = 0·87) and content validity have been established in Korean postpartum women (Kim 2003).

Unsatisfactory feeding

Unsatisfactory feeding was measured using two items with a 10-point numerical rating scale developed by Song (2007) to assess general difficulty of feeding and concern about feeding. A rating of zero means extremely easy or having no concerns at all about feeding, and a rating of ten describes extremely difficult or concerns about feeding. The instrument’s internal consistency reliability (Cronbach’s α = 0·76) has been reported in Korean postpartum women (Song 2007).

Sanhujori satisfaction

Sanhujori satisfaction was measured using the Sanhujori Satisfaction Tool (SST) developed by Song (2007). This tool consists of 6-items about the mother’s judgment of the degree of adherence to the six principles of Sanhujori, rated on a 4-point Likert scale (1–4), with higher scores indicating higher degree of Sanhujori satisfaction. Internal consistency reliability (Cronbach’s α = 0·79) and content validity of the SST have been established for use in Korean women (Song 2007). An additional item on mother’s satisfaction about sufficiency of Sanhujori duration was also measured.

Infant difficulty

Infant difficulty was measured using items from the Fussy-Difficult subscale of the Infant Characteristics Questionnaire (ICQ) developed by Bates et al. (1979) and translated into Korean by Lee (1992). Of the nine items in the Fussy-Difficult subscale, four were excluded as they were not applicable to infants under 3 months of age, based on previous research (Milligan 1989), and one item was deleted because of a small factor loading in confirmative factor analysis. Fit indices of measurement model for infant temperament were χ2 = 2·444, P = 0·295, GFI = 0·996 and NFI = 0·991 (Song 2007). The four items were rated on a 7-point Likert scale (1–7), with higher scores indicating more difficult temperament in the child.

Social support

Social support was measured as emotional support and instrumental support. Emotional support was measured using the 13-item Taylor inventory (Taylor 1974) translated into Korean by Lee (1992), and instrumental support was measured using the 8-item Social Support Inventory (SSI) developed by Lee (1992). Each inventory is rated on a 5-point Likert scale (1–5), with higher scores on both scales indicating higher degrees of social support. Internal consistency reliability (Cronbach’s α = 0·83, 0·90 respectively) of both instruments and content validity have been established for use with Korean postpartum women (Lee 1992).

Ethical considerations

The study received approval from the Human Research Review Board in the hospitals.

Data analysis

The data were analysed using spss 12.0. Cronbach’s alpha for each instrument, descriptive statistics and Pearson’s correlations between each pair of study variables were calculated. The Amos 5.0 program was used to evaluate the appropriateness of the postpartum fatigue theory using structural equation modelling (SEM) with the maximum likelihood method, because all the variables were normally distributed.

To test the appropriateness of the theory, nine indices of three aspects were evaluated. First, χ2, χ2/d.f. (χ2 statistics/degree of freedom), root mean square error of approximation (RMSEA), goodness of fit index (GFI) and adjusted goodness of fit index (AGFI) were evaluated as absolute fit indices, which were goodness of fit tests for compatibility based on predicted and observed covariance. Second, the normed fit index (NFI), non-normed fit index (NNFI) and comparative fit index (CFI) were evaluated to test goodness of fit by comparing the proposed model and null model. These indices represent the degree of improvement of the proposed model from the null model, having maximum chi-square statistics. Third, the parsimony normed fit index (PNFI) was evaluated to test the lack of parsimony of the proposed model (Hu & Bentler 1999, Garson 2009). The cut-off values of all above indices are presented in Table 4. To test the statistical significance of paths between the latent variables in the theory, standardized path coefficients were evaluated. These may be useful in comparing relative effects when the observed variables have very different units of measurement (Bollen 1989, p. 281–282), and are labelled ‘standardized regression weights’ in the AMOS results. Interpretation of standardized path coefficients is similar to standardized beta (β) in regression analysis (Garson 2009). To appraise the fit of separate equations in the structural model, squared multiple correlations (SMC) for each endogenous variables were evaluated. SMC is the percent variance explained by the exogenous variables in the model, and its interpretation is similar to R2 in regression analysis (Garson 2009).

Table 4.   Model fit measures of the preliminary and modified explanatory theory
Evaluation indicesPreliminary theoryModified theory
  1. χ2/d.f., χ2 Statistics/degree of freedom; RMSEA, Root Mean-Square Error of Approximation; GFI, Goodness of Fit Index; AGFI, Adjusted Goodness of Fit Index; NFI, Normed Fit Index; NNFI, Non Normed Fit Index; CFI, Comparative Fit Index; PNFI, Parsimonious Normed Fit Index.

Non- statistical significance of χ2 (P>0·05)χ2 = 155·144 (P < 0·001)χ2 = 96·124 (P = 0·136)
χ2/d.f. <31·8251·172
RMSEA <0·050·0530·024
GFI >0·900·9400·963
AGFI >0·900·9050·939
NFI >0·900·9000·938
NNFI >0·900·9310·986
CFI >0·900·9510·990
PNFI >0·600·6380·641

To improve the fit of the model, modification was done according to the standardized path coefficients and modification index (MI) under considering the theoretical implications (Garson 2009). The regression weight of the seven paths between the latent variables were fixed to ‘zero’ due to not being statistically significantly supported by the empirical data at the 0·05 level, and the theoretically reliable five paths between the measurement errors were added, because the modification indices were above 4·0. After modification, the overall fitness indices of the modified theory were re-evaluated.

Results

  1. Top of page
  2. Abstract
  3. What is already known about this topic
  4. Introduction
  5. Background
  6. The study
  7. Results
  8. Discussion
  9. Conclusion
  10. Funding
  11. Conflict of interest
  12. Author contributions
  13. Acknowledgements
  14. References

Participant demographics

The 291 study participants ranged in age from 20 to 39 years with a mean of 30 years (Table 2). More than half were university graduates (73%), in the middle economic class (67%), had a religion (57%), and were full-time housewives (56%). Most lived only with their husbands (86%) and had given birth vaginally (71%). Nearly half were using mixed feeding (48%), and slightly more participants (52%) had had a boy.

Table 2.   Demographic data (n = 291)
Characteristicsn (%)Mean ± sd
Age (years)
 ≤30183 (63)30 ± 3·50
 >30108 (37) (range 20–39)
Education level
 High school79 (27) 
 College or more212 (73) 
Economic status
 Low46 (16) 
 Middle195 (67) 
 High50 (17) 
Religion
 Yes167 (57) 
 No124 (43) 
Employment
 Yes129 (44) 
 No161 (56) 
Family type
 Couple only251 (86) 
 With parents40 (14) 
Type of delivery
 Normal spontaneous  vaginal delivery207 (71) 
 Cesarian section84 (29) 
Type of feeding
 Breast feeding only112 (39) 
 Mixed feeding141 (48) 
 Artificial feeding only38 (13) 
Sex of baby
 Boy152 (52) 
 Girl139 (48) 

Descriptive statistics and correlations among study variables

The mean score for overall postpartum fatigue was 22·00 (sd = 6·06), and scores for the subscales were 13·80 (sd = 3·72) for physical fatigue and 8·19 (sd = 2·98) for mental fatigue. The mean score for postpartum depression was 8·59 (sd = 4·51), which is comparatively low, considering that scores above 13 are considered indicative of depression. The mean scores for other variables are presented in Table 1.

The correlations among the variables are presented in Table 3. Positive correlations were found between fatigue and depression, childcare stress, unsatisfactory feeding, and infant difficulty, while negative correlations were noted between fatigue and sleep quality, social support, and Sanhujori satisfaction. All correlations were in the anticipated direction and indicate that most variables appeared to be statistically related to each other.

Table 3.   Pearson’s correlations among the observed variables (n = 291)
 Y1Y2Y3Y4Y5Y6Y7Y8Y9Y10X1X2X3X4X5
  1. Y1: physical fatigue, Y2: mental fatigue, Y3: sleep satisfaction, Y4: sleep non-disturbance, Y5: postpartum depression, Y6: feeding difficulty, Y7: feeding concern, Y8: childcare stress, Y9: Sanhujori satisfaction with contents, Y10: Sanhujori satisfaction with duration, X1: frequency of fussiness, X2: difficulty of soothing, X3: mood variability, X4: general fussiness, X5: emotional support, X6: instrumental support.

  2. *< 0·05, **< 0·001.

Y20·60**              
Y3−0·33**−0·22**             
Y4−0·27**−0·19**0·66**            
Y50·37**0·44**−0·29**−0·22**           
Y60·13**0·13**−0·20**−0·37**0·21**          
Y70·10*0·12**−0·18**−0·29**0·23**0·60**         
Y80·30**0·28**−0·33**−0·42**0·49**0·49**0·43**        
Y9−0·17**−0·16**0·28**0·25**−0·27**−0·12**−0·15**−0·22**       
Y10−0·15**−0·08*0·22**0·18**−0·21**−0·04−0·07−0·12**0·66**      
X10·20**0·14**−0·22**−0·25**0·21**0·12**0·15**0·36**−0·09*−0·06     
X20·12**0·11**−0·23**−0·30**0·22**0·37**0·30**0·57**−0·02−0·030·35**    
X30·16**0·17**−0·22**−0·26**0·15**0·22**0·18**0·38**−0·07−0·010·34**0·38**   
X40·08*0·08*−0·24**−0·29**0·17**0·24**0·23**0·44**−0·13**−0·020·39**0·55**0·49**  
X5−0·11**−0·13**0·14**0·09*−0·34**−0·07−0·08*−0·23**0·34**0·25**−0·10**−0·07−0·06−0·09* 
X6−0·06−0·12**0·12**0·07−0·22**−0·04−0·02−0·10*0·28**0·25**−0·060·030·050·020·44**

Testing the preliminary theory and constructing the modified theory

According to the test results of the preliminary theory, as shown in Table 4, some indices of model fit statistics satisfied the suggested values (χ2/d.f. = 1·825, GFI = 0·940, AGFI = 0·905, NFI = 0·900, NNFI = 0·931, CFI = 0·951 and PNFI = 0·638), while some did not satisfy the recommended values (χ2 = 155·144, P < 0·001, RMSEA = 0·053). Thus, the preliminary theory was modified after considering the theoretical implications and statistical significance of the parameter estimates. After modification, the postpartum fatigue theory showed improved overall fitness indices and a higher degree of compatibility with the empirical data (χ2 = 96·124, P = 0·136, χ2/d.f. = 1·172, RMSEA = 0·024, GFI = 0·963, AGFI = 0·939, NFI = 0·938, NNFI = 0·986, CFI = 0·990 and PNFI = 0·641). Therefore, the modified postpartum fatigue theory can be considered a good one for explaining the phenomenon of postpartum fatigue. Fit indices for the preliminary and modified theory and cut-off values of fit statistics are shown in Table 4.

Direct and indirect effects among study variables in the modified theory

The path diagram of the modified theory is shown in Figure 2, which also shows standardized estimates (β) and SMCs. In the modified theory, all estimated path coefficients between the latent variables were in the anticipated direction. Negative path coefficients between the latent variables were the result of variables scaled in opposite directions. In the postpartum fatigue theory, there were two paths to induce postpartum fatigue – a psychological path via childcare stress and postpartum depression, and a physiological path via sleep quality. In addition, postpartum depression and sleep quality had direct effects on postpartum fatigue, while childcare stress, unsatisfactory feeding, Sanhujori satisfaction, infant difficulty, and social support had indirect effects on postpartum fatigue, through postpartum depression and sleep quality. These variables explained 43·6% of the variance of postpartum fatigue.

image

Figure 2.  Path coefficients for the modified explanatory theory. *< 0·05, **< 0·001, +Squared Multiple Correlation (SMC). Y1: physical fatigue, Y2: mental fatigue, Y3: sleep satisfaction, Y4: sleep non-disturbance, Y5: postpartum depression, Y6: feeding difficulty, Y7: feeding concern, Y8: childcare stress, Y9: Sanhujori satisfaction with contents, Y10: Sanhujori satisfaction with duration, X1: frequency of fussiness, X2: difficulty of soothing, X3: mood variability, X4: general fussiness, X5: emotional support, X6: instrumental support.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. What is already known about this topic
  4. Introduction
  5. Background
  6. The study
  7. Results
  8. Discussion
  9. Conclusion
  10. Funding
  11. Conflict of interest
  12. Author contributions
  13. Acknowledgements
  14. References

Study limitations

As the study was based on cross-sectional data, there is a limitation in explaining the relationships among variables. It is especially difficult to tease out the relationships among depression, sleep quality and fatigue, because these three variables might all be symptoms of same problem and all three are thought to be linked to each other (Groer et al. 2005). With regard to this matter, prior to this study, we tried to test a preliminary theory of two-way arrows between theses three variables, but this theory was did not fit and was not supported by the data. Therefore there is some limitation in explaining the relationships among these three variables, and we cannot exclude the possibility of other fatigue theories with other pathways among variables. Nevertheless, all the paths in our fatigue theory were theoretically constructed based on previous empirical findings (Milligan 1989, Hantos 1993, Webster 1994, Wambach 1998, Rychnovsky 2007, Rychnovsky & Hunter 2009), the framework for the childbearing fatigue (Pugh & Milligan 1993) and theoretical considerations (Lenz et al. 1997). Furthermore, the theoretical framework for this study was highly compatible with the empirical data, contributing to better understanding of postpartum fatigue. Another limitation is that the concepts may have been insufficiently measured. Although there were more valid and reliable instruments available, only a limited number of items or shortened formats were used to avoid response fatigue.

Despite these limitations, the postpartum fatigue theory is the first work on postpartum fatigue of primiparas in Korea, and includes not only physiological, psychological, situational factors, but also the cultural component of postpartum care, i.e. Sanhujori. Also, this study is meaningful in that empirical evidence for psychological and physiological paths inducing postpartum fatigue, as well as the direct and indirect effects among postpartum fatigue and study variables, are explained.

Explanatory variables for postpartum fatigue

In this study, postpartum depression presents as the most important explanatory variable to affect postpartum fatigue directly, and was found to have a stronger effect than sleep quality. This finding is in agreement with the results of previous studies showing depression as the most important influencing factor in postpartum fatigue (Milligan 1989, Gardner 1991, Wambach 1998, Rychnovsky 2007). We also identified that childcare stress directly affects postpartum depression and thereby has indirect effects on postpartum fatigue. This explains the psychological path in which postpartum fatigue is induced via childcare stress and postpartum depression. This supports the position of previous researchers that postpartum fatigue has a significant correlation with psychological variables such as depression and anxiety, which are not naturally mitigated by rest or sleeping but subsequently may require intervention (Milligan 1989, Pugh 1990). Therefore, we suggest that the psychological path for postpartum fatigue in the early postpartum period might be preferentially considered in relation to risk of postpartum fatigue.

Second, sleep quality affected postpartum fatigue directly, suggesting a physiological path in which postpartum fatigue is induced. This finding supports various previous studies that explain direct relevance between sleep and fatigue (Hunter et al. 2009, Rychnovsky & Hunter 2009). However, the effect of sleep on postpartum fatigue was smaller than the direct effect of postpartum depression. This can be interpreted as indicating that although sleep problems are an important factor in the induction of postpartum fatigue, the overall sleep quality of parturient women is low because of taking care of a newborn, having an irregular sleeping pattern, and therefore there is no great difference between individuals. Nevertheless, this study reaffirms that sleep is of high priority in the study of postpartum fatigue, and is also important in relation to the physiological path for postpartum fatigue.

Third, childcare stress was verified as an important factor contributing indirectly to postpartum fatigue through postpartum depression. This result is consistent with previous studies identifying childcare stress as a significant predictor of postpartum depression (Beck 2001, 2008, Honey et al. 2003, Leung et al. 2005). Therefore, reducing childcare stress for new mothers may contribute to decreasing the risk of postpartum depression and subsequently decreasing postpartum fatigue.

Fourth, in the hypothetical theory of this study we set up a direct path in which unsatisfactory feeding increases postpartum fatigue, based on previous studies (Milligan 1989, Wambach 1998, Rychnovsky 2007); however, this path was not supported. Based on this finding, feeding itself does not appear to make a postpartum woman get tired, but rather difficulty during feeding appears to cause decreased sleep quality and increased childcare stress, thereby inducing postpartum fatigue indirectly.

Fifth, of the three direct paths in which Sanhujori satisfaction was suggested to affect postpartum depression, sleep quality, and postpartum fatigue based on previous studies (Kim & Yoo 1998, Yoo 1998, Chong & Yoo 1999, Ahn 2005), only one path linked to sleep quality was significant, which suggests that Sanhujori is a phenomenon that has greater physiological effects than psychological effects. This could be due to the periodical characteristic of Sanhujori, when care is provided at a time when physical recovery is required (mainly for 3 weeks right after delivery), thus maximizing the effects of promoting physical recovery and contributing to improving sleep quality. Therefore, it can be suggesated that the current Sanhujori culture in Korea is a phenomenon which promotes physiological recovery and sleep quality.

Although we initially set up infant temperament as an exogenous variable that affects all variables except social support, infant temperament was only significantly connected with feeding difficulty, childcare stress and sleep quality. This finding is in agreement with results of previous studies explaining the relevance between infant temperament and feeding problems (Fisher et al. 2004), childcare stress (Fisher et al. 2004), and sleep quality (Dennis & Ross 2005). However, it is inconsistent with previous studies showing a direct relationship between infant character and postpartum fatigue (Milligan 1989, Wambach 1998), and postpartum depression (Milligan 1989, Wambach 1998). We believe that the seeming incongruence with previous results may be related to lack of inclusion of childcare stress and feeding factor in previous studies. In other words, infant temperament affects childcare stress and unsatisfactory feeding, which are preceding variables of postpartum fatigue and depression, rather than directly affecting postpartum fatigue or postpartum depression.

Lastly, of the various hypothetical paths of social support suggested, social support contributes the most to Sanhujori satisfaction, which is consistent with reports on the relationship between social support and Sanhujori (Yoo 1998). Social support also directly affected childcare stress, postpartum depression, and feeding difficulty, and these paths are consistent with previous findings on the relationships between social support and childcare stress (Hung & Chung 2001), postpartum depression (Beck 2001, Gao et al. 2009), and feeding difficulty (Haku 2007). From these results, we conclude that social support is a very important concept and has effects on various psychological and physiological phenomena relating to postpartum fatigue.

Conclusion

  1. Top of page
  2. Abstract
  3. What is already known about this topic
  4. Introduction
  5. Background
  6. The study
  7. Results
  8. Discussion
  9. Conclusion
  10. Funding
  11. Conflict of interest
  12. Author contributions
  13. Acknowledgements
  14. References

The modified postpartum fatigue theory tested in this study showed higher fit indices and was supported by empirical data. This study explains that postpartum fatigue is induced through the physiological path via sleep quality and more significantly, the psychological path via childcare stress and postpartum depression. The Korean cultural phenomenon of Sanhujori for postpartum women appears to contribute to the physiological path of postpartum fatigue by directly affecting sleep quality. Therefore, from these results, we conclude that a priority in nursing care for women with postpartum fatigue in the first few months following childbirth might be focused on childcare stress and postpartum depression, as well as sleep quality. Supportive care to promote Sanhujori satisfaction was also found to be a potential contributor to reduced postpartum fatigue. Based on these results, we suggest a prospective study to verify these relationships among study variables with larger sample in Korea.

Author contributions

  1. Top of page
  2. Abstract
  3. What is already known about this topic
  4. Introduction
  5. Background
  6. The study
  7. Results
  8. Discussion
  9. Conclusion
  10. Funding
  11. Conflict of interest
  12. Author contributions
  13. Acknowledgements
  14. References

JES, SBC and SMP were responsible for the study conception and design. JES performed the data collection. JES and CMN performed the data analysis. JES and SK were responsible for the drafting of the manuscript. JES, SMP and SK made critical revisions to the paper for important intellectual content. CMN provided statistical expertise.

Acknowledgements

  1. Top of page
  2. Abstract
  3. What is already known about this topic
  4. Introduction
  5. Background
  6. The study
  7. Results
  8. Discussion
  9. Conclusion
  10. Funding
  11. Conflict of interest
  12. Author contributions
  13. Acknowledgements
  14. References

We wish to thank the respondents for participating in the study. We also appreciate to Dr Il-Young Yoo and Ji-Won Park for critical advice on the study and Dr Joohyung Kim for statistical advice.

References

  1. Top of page
  2. Abstract
  3. What is already known about this topic
  4. Introduction
  5. Background
  6. The study
  7. Results
  8. Discussion
  9. Conclusion
  10. Funding
  11. Conflict of interest
  12. Author contributions
  13. Acknowledgements
  14. References
  • Ahn S.H. (2005) Canonical correlation between Korean traditional postpartum care performance and postpartum health status. Journal of Korean Academy of Nursing 35(1), 3746.
  • Alley J.M. & Rogers C.S. (1986) Sleep patterns of breast-fed and non breast-fed infant. Pediatric Nursing 12(5), 349351.
  • Barnet B., Joffe A., Duhha A.K., Wilson M.D. & Repke J.T. (1996) Depressive symptoms, stress, and social support in pregnant and postpartum adolescents. Archives of Pediatrics & Adolescent Medicine 150(1), 6469.
  • Bates J., Freeland C. & Lounsbury M. (1979) Measurement of infant difficulties. Child Development 50, 794803.
  • Beck C.T. (2001) Predictors of postpartum depression: an update. Nursing Research 50(5), 275285.
  • Beck C.T. (2008) State of the science on postpartum depression: what nurse researchers have contributed – part 1. The American Journal of Maternal Child Nursing 33(2), 121126.
  • BollenK.A. (1989) Structural Equations with Latent Variables. Wiley, New York.
  • Carpenito-Moyet L.J. (2006) Handbook of Nursing Diagnosis. Lippincott Williams & Wilkins, Philadelphia.
  • Cheon J.A. (1990) The Effect of Postpartum Stress and Social Network on the Postpartum Depression. Unpublished master’s thesis, Korea University, Seoul.
  • Chong Y.M. & Yoo E.K. (1999) A Study on the Relationship between the Present Physical Symptom Distress and Experience of Sanhujori, the Traditional Postpartal Care in Korea – Centered on Women of arthritis. Korean Journal of Women Health Nursing 5(1), 89105.
  • Corwin E.J. & Arbour M. (2007) Postpartum fatigue and evidence-based interventions. American Journal of Maternal Child Nurse 32(4), 215220.
  • Cox J., Holden J. & Sagovsky R. (1987) Detection of postnatal depression: development of 10 item Edinburgh postnatal depression scale. The British Journal of Psychiatry 150, 782786.
  • Cutrona C.E. (1984) Social support and stress in the transition to parenthood. Journal of Abnormal Psychology 93, 378390.
  • Dennis C.L. & Ross L. (2005) Relationships among infant sleep patterns, maternal fatigue, and development of depressive symptomatology. Birth 32(3), 187193.
  • Fisher J., Feekery C. & Rowe H. (2004) Treatment of maternal mood disorder and infant behavior disturbance in an Australian private mothercraft unit: a follow-up study. Archives of Women’s Mental Health 7(1), 8993.
  • Freed G. & Fraley J.K. (1993) Effect of expectant mother’s feeding plan on prediction of father’s attitudes regarding breastfeeding. American Journal of Perinatology 10, 300303.
  • Freed G., Fraley J.K. & Schanler R.J. (1993) Accuracy of expectant mother’s prediction of fathers’ attitudes regarding breastfeeding. The Journal of Family Practice 37, 148152.
  • Fontenot H.B. (2007) Transition and adaptation to adoptive motherhood. Journal of Obstetric, Gynecologic, and Neonatal Nursing 36(2), 175182.
  • Gao L., Chan S.W. & Mao Q. (2009) Depression, perceived stress, and social support among first time Chinese mothers and fathers in the postpartum period. Research in Nursing & Health 32, 5058.
  • Gardner D.L. (1991) Fatigue in postpartum women. Applied Nursing Research 4(2), 5762.
  • Gardner D.L. & Campbell B. (1991) Assessing postpartum fatigue. The American Journal of Maternal Child Nursing 16, 264266.
  • Garson G.D. (2009) “Structural Equation Modeling”, from Statnotes: Topics in Multivariate Analysis. Retrieved from http://faculty.chass.ncsu.edu/garson/pa765/statnote.htm on 1 January 2010
  • Groer M., Davis M., Casey K., Short B., Smith K. & Groer S. (2005) Neuroendocrine and immune relationships in postpartum fatigue. American Journal of Maternal Child Nursing 30(2), 133138.
  • Haku M. (2007) Breastfeeding: factors associated with the continuation of breastfeeding, the current situation in Japan, and recommendations for further research. The Journal of Medical Investigation 54, 224330.
  • Hantos B. (1993) Fatigue Patterns of Mothers During the First Eight Weeks Postpartum. Master thesis. University of Alberta, Edmonton, Canada.
  • Honey K.L., Bennett P. & Morgan M. (2003) Predicting postnatal depression. Journal of Affective Disorder 76(1–3), 201210.
  • Hu L. & Bentler P.M. (1999) Cutoff criteria for fit indexes in covariance structural analysis: conventional criteria versus new alternatives. Structural Equation Modeling 6(1), 155.
  • Hung C.H. (2005) Women’s postpartum stress, social support, and health status. Western Journal of Nursing Research 27(2), 148159.
  • Hung C.H. & Chung H.H. (2001) The effects of postpartum, stress and social support in postpartum women’s health status. Journal of Advanced Nursing 36(5), 676684.
  • Hunter L.P., Rychnovsky J.D. & Yount S.M. (2009) A selective review of maternal sleep characteristics in the postpartum period. Journal of Obstetric, Gynecologic, and Neonatal Nursing 38(1), 6068.
  • Kim J.Y. (2003) Psycho-Social Predicting Factors Model of Postpartum Depression. Master thesis, Sung-shin Women’s University, Seoul.
  • Kim T.K. & Yoo E.K. (1998) A Correlational Study on the level of Importance & performance of Postpatal Care and its Relationship with Women’s Health Status – Centered on the women who experienced delivery-. Korean Journal of Women Health Nursing 4(2), 145161.
  • Lee H.K. (1992) A Structural Model of Maternal Role Attainment of Primipara in Postpartum Period. Unpublished doctoral dissertation, Seoul National University, Seoul.
  • Lee K.A. & Zaffke M. (1999) Longitudinal changes in fatigue and energy during pregnancy and the postpartum period. Journal of Obstetric, Gynecologic, and Neonatal Nursing 28(2), 183191.
  • Lenz E.R., Pugh L.C., Milligan R.A., Gift A. & Suppe F. (1997) The middle-range theory of unpleasant symptoms: an update. Advances in Nursing and Science 19(3), 1427.
  • Leung S.S., Martinson I.M. & Arthur D. (2005) Postpartum depression and related psychosocial variables in Hong Kong Chinese women: findings from a prospective study. Research in Nursing and Health 28(1), 2738.
  • MacCallum R.C., Browne M.W. & Sugawara H.M. (1996) Power analysis and determination of sample size for covariance structure modeling. Psychological Methods 1(2), 130149.
  • MacCallum R.C., Browne M.V. & Cai L. (2006) Testing differences between nested covariance structure models: power analysis and null hypotheses. Psychological Methods 11(1), 1935.
  • McGovern P., Dowd B., Gjerdingen D., Gross C.R., Kenny S., Ukestad L., McCaffrey D. & Lundberg U. (2006) Postpartum health of employed mothers 5 weeks after childbirth. Annals of Family Medicine 4, 159167.
  • Mercer R.T. & Walker L.O. (2006) A review of nursing interventions to foster becoming a mother. Journal of Obstetric, Gynecologic, and Neonatal Nursing 35(5), 568582.
  • Milligan R.A. (1989) Maternal Fatigue During the First Three Months of the Postpartum Period. Unpublished doctoral dissertation, University of Maryland at Baltimore, USA.
  • Milligan R.A., Park P., Kitzman H. & Lenz E. (1997) Measuring women’s fatigue during the postpartum period. Journal of Nursing Measurement 5(1), 316.
  • Park Y.J., Shin H.J., Ryu H.S., Cheon S.H. & Moon S.H. (2004) The predictors of postpartum depression. Journal of Korean Academy of Nursing 34(5), 722728.
  • Pugh L.C. (1990) Psychophysiological Correlates of Fatigue During Childbirth. Unpublished doctoral dissertation, University of Maryland at Baltimore, USA.
  • Pugh L.C. & Milligan R. (1993) A framework for the study of childbearing fatigue. Advances in Nursing Science 15(4), 6070.
  • Pugh L.C. & Milligan R. (1995) Patterns of fatigue during childbearing. Applied Nursing Research 8(3), 140146.
  • Reeder S.J., Martin L.L. & Koniak-griffin D. (1997) Maternity Nursing (18th). Lippincort, New York.
  • Runquist J.J. (2007) A depressive symptoms responsiveness model for differentiate fatigue from depression in the postpartum period. Archives of Women’s Mental Health 10(6), 267275.
  • Rychnovsky J.D. (2007) Postpartum fatigue in the active duty military women. Journal of Obstetric, Gynecologic, and Neonatal Nursing 36(1), 3846.
  • Rychnovsky J. & Hunter L.P. (2009) The relationship between sleep characteristics and fatigue in healthy postpartum women. Women Health Issues 19(1), 3944.
  • Song J.E. (2007) Predictors of postpartum fatigue between early and late postpartum period in parturient women – divided by 3/4 weeks of postpartum period. Korean Journal of Women Health Nursing 13(4), 299309.
  • Taylor J.W. (1974) Measuring the outcome of nursing care. The Nursing Clinics of North America 9(2), 337348.
  • Taylor J. & Johnson M. (2008) How women manage fatigue after childbirth. Midwifery 26(3), 367375.
  • Troy N. (2003) Is the significance of postpartum fatigue being overlooked in the lives of women? American Journal of Maternal Child Nursing 28, 252257.
  • Wambach K.A. (1998) Maternal fatigue in breastfeeding primiparae during the first nine weeks postpartum. The Journal of Human Lactation 14(3), 219229.
  • Warren P.L. (2004) First time mothers: social support and confidence in infant care. Journal of Advanced Nursing 50(5), 479488.
  • Webster B.A. (1994) Maternal Fatigue During the Postpartum Period. Master thesis, Macgill University, Montreal, Canada.
  • Yoo E.K. (1993) An Ethnographic Study About Sanhujozi inline image, the Phenomenon of Korean Postpartum Care. Doctoral dissertation, University of California, San Francisco.
  • Yoo E.K. (1997) The relationship between the women’s health status and experience of Sanhujori. The Journal of Korean Nurse 36(5), 7490.
  • Yoo E.K. (1998) The cultural demand of postpartum care in Korean women: reconsideration of meaning of Sanhujori and nursing implication. The Journal of Korean Nurse 37(3), 2740.
  • Yoo E.K., Lee S.H. & Kim M.H. (1998) A Study on the Relationship between the Experience of Sanhujori, the Traditional Postpartal Care in Korea and Present Health Status of Chronic Arthritis Female Patient. Korean Journal of Women Health Nursing 4(2), 217230.