Background: Obstructed labour is still a major cause of maternal morbidity and mortality and of adverse outcome for newborns in low-income countries. The aim of this study was to investigate the role of individual and health facility factors and the risk for obstructed labour and its adverse outcomes in south-western Uganda.Methods: A review was performed on 12,463 obstetric records for the year 2006 from six hospitals located in south-western Uganda and 11,180 women records were analysed. Multivariate logistic regression analyses were applied to control for probable confounders.Results: Prevalence of obstructed labour for the six hospitals was 10.5% and the main causes were cephalopelvic disproportion (63.3%), malpresentation or malposition (36.4%) and hydrocephalus (0.3%). The risk of obstructed labour was statistically significantly associated with being resident of a particular district [Isingiro] (AOR 1.39, 95% CI: 1.04-1.86), with nulliparous status (AOR 1.47, 95% CI: 1.22-1.78), having delivered once before (AOR 1.57, 95% CI: 1.30-1.91) and age group 15-19 years (AOR 1.21, 95% CI: 1.02-1.45). The risk for perinatal death as an adverse outcome was statistically significantly associated with districts other than five comprising the study area (AOR 2.85, 95% CI: 1.60-5.08) and grand multiparous status (AOR 1.89, 95% CI: 1.11-3.22). Women who lacked paid employment were at increased risk of obstructed labour. Perinatal mortality rate was 142/1000 total births in women with obstructed labour compared to 65/1000 total births in women without the condition. The odds of having maternal complications in women with obstructed labour were 8 times those without the condition. The case fatality rate for obstructed labour was 1.2%.Conclusions: Individual socio-demographic and health system factors are strongly associated with obstructed labour and its adverse outcome in south-western Uganda. Our study provides baseline information which may be used by policy makers and implementers to improve implementation of safe motherhood programmes. © 2011 Kabakyenga et al; licensee BioMed Central Ltd.
A retrospective review of obstetric records was conducted in six hospitals namely Mbarara Regional Referral which also doubles as a university teaching hospital, Kitagata, Ishaka Adventist, Comboni, Ibanda, and Rushere Community located in five neighbouring districts of south-western Uganda. Table Table11 shows the hospitals and the district of location, locality (urban/rural), category/ownership and deliveries for the year 2006. Kitagata, Ishaka Adventists and Comboni hospitals were at the time of conducting the study located in Bushenyi district (split into 5 districts since 1st July, 2010). While the other 3 hospitals Mbarara, Ibanda, Rushere are located in Mbarara, Ibanda and Kiruhura districts respectively. Isingiro district did not have a hospital of its own and comprehensive emergency care services were sought from neighbouring Mbarara hospital. Hospitals in the study by district, location, category/ownership, and total deliveries for 2006 ºBushenyi was administratively split into 5 districts with effect from 1st July 2010 *PNFP = Private Not For Profit The districts of Mbarara, Bushenyi, Ibanda, Kiruhura, and Isingiro with a population of about one and a half million people share borders and have overall similar socio-economic and cultural conditions and use the local dialect of Runyankore as the common language. The Uganda’s health care system is structured in such way that there are corresponding health units or services at different levels of the administrative structure. The village health team (VHT) is the lowest level while a national referral hospital is the highest level of care [19,20] as shown in Table Table22. Comprehensive emergency obstetric care services, especially operative delivery and blood transfusion, are available in all general, regional referral and national referral hospitals. According to our knowledge all the hospitals in the study were by structure able to offer a full range of comprehensive emergency obstetric care services at all times. Services offered in public hospital are officially free of charge although due to frequent shortages of drugs/supplies, patients/clients are requested to procure missing items from private pharmacies. Mbarara and Ibanda hospitals were the two hospitals with specialists (Obstetricians) while the other hospitals had general doctors as their highest ranked clinicians. Structure of Uganda national health system Adapted from Government of Uganda Health Strategic Plan II, 2005/06-2009/10 [19] & Rutebemberwa et al., 2009 [20] Twelve thousand four hundred and sixty three (12,463) obstetric records of women who were admitted in the maternity wards of the six hospitals (Mbarara, Kitagata, Ishaka, Comboni and Ibanda) from January 1 through December 31, 2006 were reviewed. The data collectors were midwives proficiently trained to collect data from women’s obstetric files or charts and to validate the diagnosis of obstructed labour using admission, delivery and theatre registers. Data was recorded in case record forms developed by the researchers and pre-tested on 200 maternity records for the year 2007. The case record was modified to correct observed inconsistencies. The case record form was designed to collect data on socio-demographic variables, labour, delivery and post-delivery periods. Computer data entry was performed using Epidata (Epidata Association, Denmark). The criteria we used for diagnosing obstructed labour in this study was admission to a hospital with a pregnancy of a gestational age of 28 weeks or more and having a clinical diagnosis of obstructed labour in the patient chart or having an operative intervention (i.e. vaginal or abdominal) for failed progress of labour due to cephalopelvic disproportion, malpresentation or malposition. Women for whom the diagnosis of obstructed labour could not be ascertained were classified as non-obstructed labour and still included in the study sample. Women who were admitted post-partum (n = 114) were excluded from the sample, as well as women whose gestational age was < 28 weeks upon admittance or were recorded as abortion (n = 482 women), and women who were discharged before delivery (n = 687 women). This reduced the number of records included in the sample from 12,463 to 11,180. Obstructed labour was classified as: "with obstructed labour" or "no obstructed labour" Cause of obstructed labour was classified as: "cephalopelvic disproportion", "malpresentation" or "malposition" as stated in obstetric file or chart. Neonatal outcome was classified as: "live birth" or "stillbirth" Maternal outcome was classified as: "alive" or "died in hospital" Maternal complications: coded as "Yes" (if a woman had at least one complication during labour or childbirth) otherwise coded "No" Perinatal mortality rate was defined as "stillbirths and deaths in the first week per 1000 total births (live births plus stillbirths)". Age of the woman was divided into 3 age groups: 15-19, 20-29 and ≥30 years. The age range was 15 to 49 years. The age group 20-29 years was taken as the reference age group. Parity was classified into 4 groups: "0" (nulliparous), "1", "2-4", "≥5". Parity 2-4 was taken as reference category, as it is considered to be the one with minimal risk of obstructed labour. Place of residence was defined as districts of residence, which were: "Mbarara", "Bushenyi", "Ibanda", "Isingiro", "Kiruhura" and other districts (districts other than the ones specified). Bushenyi district was used as a reference district since at the time of the study it was the only district with 3 hospitals, thus providing more accessibility to emergency comprehensive services required in cases of obstructed labour. Occupation of mother was classified as: "salaried employee", "subsistence farmer", "housewife". Salaried employee group was used as a reference group on the assumption that salaried women would be in the position to have financial resources to access health care services faster. In this study only occupation is used as a proxy for socio-economic status as most of the records were missing education information regarding levels attained by women. Health facility attended: Comboni hospital, Ibanda hospital, Ishaka Adventist hospital, Kitagata hospital, Mbarara regional referral hospital, Rushere Community Hospital. Mbarara regional referral hospital being a university teaching hospital was expected to offer a high level of care and thus was therefore used as a reference hospital in this study. Women seeking obstetric services in one hospital are expected to be more similar than women who visit other hospitals for the same service. Therefore SVY routines (Stata Version 10 Software) for handling correlated data were utilised to estimate proportions (%) and 95% confidence intervals of women who had obstructed labour and its outcomes (perinatal death and maternal complications). Crude Odds Ratios (COR), Adjusted Odds Ratios (AOR) and their 95% confidence intervals (CI) were calculated by means of multivariate logistic regression analysis taking into consideration the clustering of women at the hospital level. The mixed effects model was applied to determine the adjusted effects of age, parity, and residence on perinatal death and maternal complications among women who experienced obstructed labour assuming a random intercept. It was assumed that hospitals would have different intercepts due to differences in the level of care and some unmeasured health system factors. However the effect of the studied covariates was expected to be similar across hospitals. Cases with missing values were excluded from these analyses. Data was analysed using STATA version 9 (STATA Corporation, College Texas USA). Ethical clearance was applied and granted from Uganda National Council of Science and Technology and from Lund University, Sweden. Permission to access obstetric records was obtained from respective medical superintendents of the hospitals included in our study.
N/A