Maternal morbidity in the first year after childbirth in Mombasa Kenya; a needs assessment

In recent years, maternal health services in resource-constrained settings have increasingly focused on the importance of skilled birth attendants and the management of intrapartum complications [1]. Also, much efforts have been made to rationalise the package of services for antenatal care [2, 3]. Antenatal care coverage remains high in most of Africa and the proportion of births which occur within medical services is steadily increasing [4, 5]. By contrast however, few resources are allocated to delivering specific services for women in the year after childbirth. The essential package of postpartum services is poorly defined, and the optimum structure and number of visits for these services remains uncertain [6]. Given weaknesses in the current model for health services for women after childbirth, it is worthwhile to examine alternative configurations of service delivery. Most notably, contact with health services in the year after childbirth presents a vital opportunity to assist women return to full health, and to prevent sexual and reproductive health problems in this period.

The postpartum period, defined by WHO as beginning one hour after delivery of the placenta and continuing until six weeks after childbirth [7], is a period of substantial transition as women move from pregnancy and childbirth back to reproductive potential. Many studies in Africa refer to the postpartum period in more general terms, even including up to two years after childbirth [8‐12]. Some definitions of maternal mortality include late maternal deaths (deaths between six weeks and one year after childbirth) [13]. This acknowledges the fact that some physiological effects of pregnancy still occur in this period and that the health of women at this time warrants specific attention [14, 15]. The pre-pregnant state has not yet completely returned within six weeks of childbirth, lactation usually continues throughout the first year, often the menstrual cycle has not yet normalized and sexual activity may not have resumed. Further, with screening, detection and treatment of diseases at several time points in the year after giving birth, the safety of future pregnancies could be improved, as well as offering the inherent benefits of improvements in women's overall health status.

Though in many settings over half of maternal deaths occur after childbirth [16, 17], data on the health of women during this time are sparse. Moreover, reports usually focus on life-treating obstetric complications such as postpartum haemorrhage with a relative neglect of important common morbidities [7, 18‐20]. Thus, while several previous studies have highlighted the contribution of the postpartum period to maternal mortality, few have described the burden of more common chronic conditions such as anaemia and depression [21‐23]. In addition, most recent studies among postpartum women in Kenya have concentrated on HIV-infected women [8, 10, 11], and additional information is required about the burden of disease in women without HIV infection, who constitute the majority in these settings.

This study describes acute and chronic morbidity in the year after childbirth, which can be considered as either: conditions associated with pregnancy and childbirth; diseases specific to the postpartum period (such as those related to breastfeeding or the return of reproductive potential); or as intercurrent illnesses (such as cervical neoplasia). Further, the study sets out how patterns of sexual behaviour and morbidity vary with the length of time since delivery. Identifying the needs of women in the year after childbirth would inform selection of essential interventions to secure the health of women during this period, tailored to their changing needs over this time course. Provision of these interventions for women bringing their child for health care could achieve high population coverage as child-health services have good uptake in most countries [24].

At Coast Provincial General Hospital in Mombasa, Kenya, 500 women attending an immunization and acute care paediatric clinic participated in a cross-sectional survey to describe the levels of maternal morbidity in the year after childbirth. The survey, held in 2006, also investigated the feasibility of providing HIV testing and counselling, and the levels of HIV-related disease in this population [25, 26]. Study activities formed part of a project to improve maternal health services in Coast Province Kenya. From 2002 to 2005 this project focused on strengthening antenatal and intrapartum services; however as it appeared that few women attended postpartum services and their needs were ill defined, this survey was planned to assess women's needs after childbirth and to define a service package that could be provided for women bringing their child for health care. The study was approved by the Kenyatta National Hospital Ethics and Research Committee.

The study population consisted of consecutive women who were older than 16 years, biological mothers of the child, and between four weeks and one year after childbirth.

After completion of the child-health visit, women were approached and invited to participate in the study. Those willing to participate gave written informed consent and were interviewed using a pre-tested structured questionnaire administered in Swahili, the local language.

Besides information on demographic characteristics, data were collected on access to sexual and reproductive health services; family planning needs and sexual health status. Women were screened for mild and major depression (using ICD-10 definitions [27]) and for harmful alcohol use with the AUDIT tool [28]. Thereafter study nurses did a clinical examination and collected blood, urine, a cervical swab and a Pap smear. Full details of test procedures and the sample size calculation are provided elsewhere [26]. Urine dipstick detected nitrites and leucocytes. Blood samples were used for malaria microscopy, RPR test for syphilis and serial rapid HIV testing. Trichomonas vaginalis and candida were detected in a wet mount, and Nugent's criteria used for diagnosing bacterial vaginosis. Participants were advised to return for test results and, if required, received treatment according to local guidelines. For conditions such as syphilis, malaria and severe anaemia, women with positive results were contacted and asked to return to the clinic immediately. All participants were offered HIV counselling and same-day testing; those testing positive were enrolled in an HIV care and treatment clinic at the study site.

The study describes the burden and patterns of morbidity in the year after childbirth. Conditions associated with pregnancy, childbirth or breastfeeding predominate in the early and middle periods, while illnesses stemming from a return of sexual function and a transition to reproductive potential were commoner in the latter period. Broadly speaking, therefore, in the initial period after childbirth, interventions targeting breast health, the urinary tract and anaemia are most pertinent, while promotion of sexual health and prevention of unintended pregnancy have more importance in subsequent periods. It is, however, difficult to make a definitive distinction between needs of women in each time period and whether morbidities in these periods are related to pregnancy, return of reproductive potential or intercurrent illness. For example, throughout the first year, high levels of fever, abdominal pain and abnormal vaginal discharge make vigilance for sepsis important, a major cause of maternal mortality in Africa [16].

Over one third reported health problems and overall perceived health status was low. Among women in the early postpartum in Zambia, 84% reported at least one health problem [31]. A large study in India found 23% had health problems in the early postpartum [32] with even higher levels reported in Sri Lanka [33]. Taken together, it is thus clear that health problems are common among women after childbirth, but health care services are not commiserate with these levels of ill health.

Prevalence of treatable reproductive tract infections was high, similar to previous findings in Kenya [34]. The association between bacterial vaginosis and postpartum endometritis is potentially important in this population, but also its links with complications in a future pregnancy such as preterm delivery and preterm rupture of membranes [35‐39]. STIs generally increased with each period since childbirth, concurring with the reported increase in sexual activity over time and the low condom use throughout. Low condom use, comparable to levels in a similar population in Zimbabwe [9], may be due to women perceiving themselves at low risk for pregnancy in the postpartum [40]. Absence of syphilis infection within two months of childbirth likely reflects the high coverage of antenatal care and quality of these services. As reported previously [25], acceptability of HIV testing was high among study participants and around four fifths received their results. Women with HIV infection have high morbidity and mortality in the year after childbirth [11] - for example, in our population, two thirds had anaemia and 20% reported current febrile symptoms [26]. These women require specific HIV care and treatment in addition to the needs of women described in this paper. Among the whole study population, levels of violence were remarkably high, though comparable with previous prevalence estimates in the general population in eastern Africa [41].

More than half of recent pregnancies were unplanned, and despite a desire to postpone pregnancy, more than a third of women had unmet need for contraception. Adequate birth spacing confers many benefits for women and their children and WHO recommends at least two to three years between pregnancies [42]. Women attending child health clinics need to be made aware of the benefits of adequate birth spacing, supported in their fertility choices, and to be offered contraception and dual protection. With adequate family planning services, uptake of hormonal contraception is high among postpartum women in Kenya [10], as much as 72% in a study [8].

Over half the women, and three quarters of some high-risk groups (such as those who had a caesarean section), had anaemia. Anaemia may be exacerbated by pregnancy and childbirth, and current WHO recommendations for populations with a severe prevalence of anaemia (>40%, as in this population) are that preventative and therapeutic iron supplementation should be given to all women of childbearing age and routinely in the postpartum [43]. Giving women attending child health clinics haematinic supplements, which are highly effective in reducing anaemia [44], may prevent anaemia in subsequent pregnancies.

That demographic characteristics of women in each of the three time periods were similar, suggests that health-seeking patterns were not strongly influenced by socio-demographic characteristics. These characteristics were, however, important in a study in rural Zambia [45]. In our population, differences between the study groups in experiences of sexual violence and depression indicates that psychosocial experiences may affect the timing and number of visits to child health centres.

However, perhaps women with severe postpartum depression in early period were less likely to bring their child for immunizations, spuriously lowering depression in this group. There is a biological explanation for higher rates of abnormal Pap smears shortly after childbirth. False positive smears can be caused by cervical inflammation related to the birth process, which regresses gradually over time [46, 47]. Overall, coverage of cervical cancer screening in Kenya remains notably low and this cancer the most common malignancy among women in the country [48].

This study has several limitations. While factors such as sexual activity, rate of caesarean section and prevalence of HIV are broadly comparable to those in urban areas of Kenya [49, 50], it is difficult to generalise the findings to rural areas, or to lower-level health facilities. In addition, women who do not attend child-health clinics, women whose child has died, or the substantial proportion who declined study participation may differ in important ways from the study population. Rates of morbidity are also likely to vary markedly between populations, in for example postpartum clinics and child-health clinics, or between community and hospital-based samples. Moreover, self-reported symptoms and behaviour incur recall and social desirability bias, though can be useful indicators of perceived health status.

Evaluation of morbidity in the whole year since childbirth makes it difficult to compare our findings with studies that included only women in the conventional six-week postpartum. However, including a longer period is a strength of this study, as it acknowledges the high risk women have for mortality between six weeks and a year after childbirth [14, 15]; is cognisant of the need for a continuum of care throughout this time; and presents evidence of the potential benefits of optimising the interface between women and health workers during child health visits to address the substantial burden of disease women have throughout this period.

Although most of the common problems women experienced were not life-threatening, they likely have marked influence on their wellbeing and health status in the long run. Most conditions could be addressed through provision of health promotion and preventive interventions, which is highly complementary to the intrapartum focus on acute severe problems.

The year after childbirth, an important part of women's lifecycle, is a neglected component of the continuum of care and support for women. Women in this period require interventions such as prevention and treatment of anaemia and reproductive tract infections, but also provision of family planning counselling and services, mental health services and distribution of condoms. Though many indicated they might benefit from home visits, this intervention requires additional study [51, 52]. Interface with women at child-health clinics, however, offers an opportunity to provide or facilitate entry to sexual and reproductive health services, and to encourage their long-term use.

The potential benefits of extending postpartum services beyond the conventional six-week period after childbirth may necessitate revision of the classic approach to postpartum care. The findings of this study indicate that consideration is warranted of recommending an early visit around four to six weeks postpartum and later visits at three, and six to nine months after childbirth, which concurs with the immunization schedule in many countries. Adequate care for women in the year after childbirth, possibly provided in child-health centres, could make an important contribution to international commitments to improve maternal health [53].