INTRODUCTION

It is now generally recognized that medical outcomes cannot be explained fully by an evaluation of clinical outcomes and clinical indicators alone. Factors such as pain, anxiety, and financial circumstances may influence health status, and also, subjective feelings of well-being affect the health of an individual.[1] Facts known only by physicians need to be complemented by values known only by patients. Outcomes of research have referred to the importance of the viewpoint of the patient on the objectives of medical care; measuring quality of life is one such ‘patient - centered’ outcome.[2]

Quality of life (QoL) is complex and comprehensive, and may be affected by many factors, including physical, psychological, emotional, social, sexual, and spiritual parameters.[3]

Over the last few decades, the importance of QoL has increased as an objective of medical care, and also methods of measurement of QoL have improved.[4] The concept of patient-reported QoL assessment is an emphasis on the fact that it is the patient, not the physician, who evaluates his or her QoL.[2] Hence, medical care aspires to narrow the gap between what the patient expects from the care she or he received, and what has been achieved in reality.[5]

Physical and psychological postnatal morbidity is an important health-care problem that requires better understanding.[6] There are validated tools for assessing postnatal QoL. As an example, the Maternal Postpartum Quality of Life (MAPP-QoL) tool measures QoL during the early postpartum period.[7, 8] Like most QoL tools, MAPP-QoL involves asking a structured set of QoL questions to the target group. In addition, the mother-generated index (MGI) has been developed and validated for assessing postnatal QoL. The instrument allows postnatal women to select the QoL issues that are most important to them (hence ‘mother generated’), and thus avoids a top-down predefined approach to the assessment of QoL.[9]

The MGI allows the woman to determine both the content of the form and the overall score. This approach tries to assess QoL issues from the viewpoint of the woman herself.[10] In various studies, the MGI has correlated well with a range of other measures of postnatal index, the Edinburgh Postnatal Depression Scale (EPDS), the Maternal Adjustment and Maternal Attitudes Scale (MAMA), the Short Form-12, and general health questionnaire (GHQ-12), and was found to be acceptable and helpful.[3, 11, 12, 13, 14, 15, 16, 17]

Although some studies have generally evaluated postpartum QoL in Iran,[18] no previous study has used a nongeneric specific QoL tool for postnatal issues in Iran. To date, no specific QoL instrument has been validated for use in the Farsi language for postnatal issues. The aim of this study was to examine whether the Iranian version of the MGI is a valid measure to assess the QoL of postnatal women. When comparing MGI and QoL scores, we hypothesized that a higher MGI score would be associated with less morbidity and higher QoL scores on generic QoL instruments.

METHODS

RESULTS

One hundred and twenty-four postnatal women were approached and five were excluded, two aged 15 years, two were severely ill, and one whose baby had died. Of the 119 remaining women, 96 (81%) mothers declared their consent to participate in the study. The main demographic variables among the study population are shown in Table 1. The average age of the participants was 27.7 years (SD: 6.1) and 54.2% were primiparous. None of the participants smoked cigarettes. Overall, 67.7% women had undergone a cesarean section, and 39.6 and 37.5% of the mothers had Fars and Azeri ethnic backgrounds, respectively. Also, the self-reported economic status of the household has been shown in Table 1. The participants commented on two to six (mean: 4.6; SD: 1.4) aspects of their life that had been affected since having a baby. Some comments were in single words and some were in short phrases. The mean of primary MGI scores (the scores given to the comments) was 5.38 [95% confidence interval (CI): 4.95-5.81].

Table 1. Demographics of postnatal women

Of the comments made, 42% (152/364) were positive, 49% (182/364) were negative, and 9% (30/364) were both/neither. The mean of primary scores for positive comments was 8.37 (95% CI: 8.22-8.52); for negative comments it was 2.37 (95% CI: 2.20-2.54); for those that were labeled ‘both/neither’ it was 5.85 (95% CI: 5.62-6.08). The mean of the respondents’ comment aggregate scores was 0.11 (95% CI: −0.30-0.52).

The comments were categorized into twelve ‘themes’. Table 2 demonstrates the frequency of comments under each theme and mean scores of the comments by themes. The themes of ‘labor pain’, ‘sleep pattern’, ‘personal time’, ‘educational life’ and ‘financial problems’ generated universally negative comments. Over 75% of the comments under each of the themes of ‘relationship with family’, ‘body image’ and ‘feeling about their baby’ were positive. Many of the other themes contained both positive and negative comments (e.g., social life and working life). Of 37 comments on ‘social life’, 16 were positive (e.g., ‘good relationship with friends’), 13 were neither/both, and eight were negative (e.g., ‘deficiency in social life’). Relationship with family included the most number of positive comments, whereas physical problems, working life, and sleep patterns included the most number of negative comments.

Table 2. The themes, the frequency of comments under each theme, and the mean scores assigned to the comments for positive, negative, or neither/both

In Step 3, the comments included under the theme ‘relationship with family’ received the mean of highest weighted scores of 7.6 out of 20 (SD: 4.2). ‘Sleep patterns’ and ‘labor pain’ received the lowest weighted scores, respectively, 2.8 (SD: 1.2) and 2.9 (SD: 1.5) indicating that although many women considered sleep pattern and labor pain as negative aspects of their postnatal QoL, these were not considered as major factors.

Correlation between aggregate scores of comments and the primary scores was high (r = 0.68, P < 0.01). The mean of the secondary index score was 6.47 (95% CI: 5.60-7.33). The secondary MGI score was not significantly correlated with the primary score or the aggregate score of comments.

The mean score of physical function on the SF-36 was 52.81 (SD: 29.54), role limitation of physical function, 44.53 (SD: 37.9), role limitation of emotional function, 61.45 (SD: 37.56) vitality, 59.21 (SD: 24.11), mental health, 65.70 (SD: 21.18), body pain, 53.85 (SD: 23.63) and general health perception, 61.28 (SD: 24.59). The primary MGI scores were moderately and significantly correlated with all the subscales of the SF-36 (for all of subscales: Pearson r > 0.4; P < 0.0001). There were no such significant correlations between the SF-36 subscales and the secondary MGI scores (except for one subscale) and the aggregate scores of comments [Table 3].

Table 3. Correlation coefficients between SF.36 subscale scores and primary MGI score, secondary MGI score, and aggregate score of comments

There was no significant association between the primary or secondary MGI score, and SF-36 scores for age, educational attainment, and the number of children. The analysis of findings about type of delivery showed that women after vaginal delivery scored slightly higher on the primary MGI score compared to mothers who experienced cesarean sections (5.6 vs. 4.8, 95% CI for mean difference: −0.16 to 1.6; P = 0.09). Also, women after vaginal delivery scored nonsignificantly higher on the SF-36 compared to mothers after cesarean section (60 vs. 56.2, 95% CI for mean difference: −12.69 to 5.63; P = 0.08). Women who had more than one comorbidity had significantly lower primary MGI scores than women without comorbidity [Table 4]. Also women who had experienced more than one postpartum medical complication had significantly lower primary MGI scores than women without complications [Table 4].

Table 4. Mean score of Step 2 (primary index score) by number of comorbidities and complications

DISCUSSION

Previous studies have demonstrated that the MGI was a workable tool for assessing postnatal QoL. In this study, we assessed the validity of the tool in an Iranian context. This is the first study assessing postnatal QoL in Iran using a nongeneric, specific, and relevant index while documenting the limitations of the methods used. Our findings suggest that the MGI primary score and aggregate score of the comments may provide valid estimates of postnatal QoL in Iran. This conclusion is based on the high correlation between the ‘aggregate scores of comments’ and the primary MGI scores (suggesting internal consistency), and the significant correlations between the scores and the SF-36 scores (SF-36 was previously validated for use in Iran). We also noted that the MGI scores were significantly associated with the reported number of postnatal complications as well as patient comorbidities, suggesting criterion validity of the MGI scores [Figure 1-2].

[Figure ID: F1] Figure 1. Mother Generated Index questionnaire; Source: Symon A, MacDonald A, Ruta D. Postnatal Quality of Life Assessment: Introducing the Mother Generated Index. Birth 2002;29:40 6.

[Figure ID: F2] Figure 2. An example demonstrating how the secondary Mother-generated index scores were calculated is shown in the additional File 2

The mean primary MGI score in our sample was 5.38 out of 10, whereas the average secondary index score was 6.47. This is similar to the findings from the original MGI study in Scotland,[24] in which the mean of the lowest quartile was 4.6 and that of the highest quartile was 8. In another study by Nagpal et al. in India, the overall mean primary index score was 3.6 and the average secondary index score was 2.9.[14] The overall higher scores in our study in comparison with other studies such as the Indian study may be related to the primarily positive nature of the areas cited by mothers or could reflect a higher QoL in our sample. Obviously, to generalize the findings to the Iranian population requires larger studies of representative samples.

We modified the way the weights for the comments were obtained and the secondary MGI scores were calculated, because of the difficulties experienced in the Step 3 scoring. These changes inevitably limit the comparability of our results to other settings. This may be a reason for the observed nonconcordance between the secondary MGI scores and the primary and aggregate scores in our study. Previous studies have also failed in demonstrating the validity of the secondary MGI scores.[24] As such, further work on the methods used for the estimation of the secondary scores is warranted. Still, it is noteworthy that the primary MGI score is the MGI's main indicator of QoL. Our study had other limitations: Our sample size of 96 reduced our power of observing other significant relationships.

The face validity of the Farsi version of the MGI questionnaire appeared to be satisfactory. The majority of the comments raised by the participants were clearly related to the experience of childbirth (e.g., comments relevant to ‘labor pain’, ‘feeling about baby’). Other comments were more linked to the experience of motherhood (e.g., ‘relationship with family’, ‘social life’ and ‘working life’). These suggest that the MGI is capable of generating comprehensive comments on important aspects of a woman's life as they are affected by giving birth and motherhood. As such, the MGI may provide health-care professionals with the opportunity of appreciating all aspects of a woman's life that are affected with the experience of having a baby.

Specific measures of QoL can enhance the detection of small, clinically important aspects in QoL related to specific areas of interest,[25] and as such, we believe that the use of the MGI is justified. Moreover, health-related QoL can be a good tool for reflecting service needs and thus, it is useful information for physicians.[26] For this reason, the MGI could be a very useful tool in clinical practice to provide optimal management of women's needs. Another interesting aspect of the MGI is its ability to capture other (nonhealth) aspects of women's lives that are affected by the experience of having a baby (e.g., ‘financial problems’). Such aspects are not measured by health-related QoL measurement tools such as SF-36. This may help the professionals and policy makers to have a more holistic understanding of the postnatal experiences of women.

In our study, women who underwent vaginal delivery scored higher (better) on both the SF-36 and MGI, although the differences were statistically nonsignificant. Cesarean section is not just a type of childbirth; it is also an operation, and like any form of surgery, can lead to problems resulting from hospitalization. This might justify differences observed between the two groups. In Iran, data published in 2005 suggested that cesarean section constituted 47% of all deliveries, 52% of the deliveries in Tehran, and 64% of the deliveries in the private sector,[27, 28] and this may have increased in recent years. Other studies have reported even higher estimates of the rates of cesarean sections.[18, 29, 30] Our observed higher rates of cesarean sections may reflect the increasing trend over what was reported in 2005, or the peculiarities of the hospitals in which we sampled our participants.

CONCLUSIONS

In general, the Farsi version of the MGI performed well in an Iranian context and the findings suggest that it is a valid measure to assess QoL among postnatal women.

The MGI is an appropriate tool to assess holistic perinatal care for mothers. The instrument identifies the most important areas of a mother's life apart from the clinical aspects of care. This instrument can help to clarify the problems that may affect the mother's QoL that may not be immediately apparent to a doctor or midwife. As such, further assessment of its application to maternal care in Iran (and similar countries) is recommended.