The present cross-sectional observational study was carried out in the Department of Obstetrics and Gynaecology from October 2015 to December 2016 in a tertiary care centre and teaching hospital in northern India in New Delhi. Approval was taken from the Institutional Ethics Committee (IEC/VMMC/SH/Thesis/October/2015) for undertaking the observational study. Informed written consent in all near- miss cases were taken.

Inclusion criteria: Study population consisted of all women who were identified as MNM and MM occurring during the study period.

Exclusion criteria: Among these patients whose relatives were either not available or not willing or are unable to provide details to participate in the face -to face interview process were excluded from the study.

The three phases of delays i.e., Phase I- delay in deciding to attend a health care facility; Phase II- delay in reaching an adequate health care facility; and Phase III- delay in receiving adequate care at that facility; are subsequential and inter-linked [8].

Maternal Near-Miss (MNM): “A woman who has suffered from life-threatening conditions during pregnancy or within 42 days of pregnancy termination, either due to termination of pregnancy or childbirth and sustained oneself regardless of getting emergency medical/surgical interventions or otherwise [12]”.

A woman is identified as MNM, when at least 3 criteria (one from each category) are fulfilled [12]:

Maternal Mortality (MM): “A woman who died during pregnancy or within 42 days of termination of pregnancy, regardless of the duration and site of the pregnancy, due to any cause which is related to or is escalated by the pregnancy or its management, but not from accidental or incidental causes [13].”

For all the women enrolled, the causes of MNM and MM were identified. Facility based MNM-Review (MNM-R) and Maternal Death Review (MDR) proforma were filled. Attendants accompanying the patients mostly nearest kin who were able to give details of her health were interviewed. Details regarding the sequence of events that caused her severe morbidity were taken right from recognition of morbid status to landing up in the tertiary setup. Approximate duration of delay for each case of MNM and MM was assessed. Any delays in accessing or receiving medical care were recorded if available.

Data entry was done on Microsoft Excel spreadsheet. In data analysis, for qualitative data, proportions were calculated. Mean score with confidence interval was calculated for quantitative data. Test of significance of differences between proportions and means were calculated. Qualitative data was analysed by Chi-Square test and t-test was applied for quantitative data. Statistical analysis was done using Statistical Package for the Social Sciences (SPSS) software for Windows version 20.0.

There was a total of 31,925 deliveries during the study period out of which total number of live births were 31,111. There were 249 women who experienced a severe complication, nearly died but survived. A total of 131 women could not be saved despite all efforts. Analysis of demographic factors revealed women belonging to lower socioeconomic status and those who were unable to read and write were at high risk of dying from childbirth. Most of these women did not attend the minimum antenatal visits. Patients who went through referral from multiple health centres had higher chances of dying than those who came directly or with referral from single health centre. The referral from hierarchy of health institutions caused significant delay deteriorating the health condition of mothers [Table/Fig-1] [14].

Delay in women seeking help was observed in a total of 92.36% of cases in MNM group and 97.70% of cases in MM group. A total of 7.63% and 2.3% of patients in MNM and MM group respectively had no delay at any phase with significant difference (p=0.034). When only first phase delay was considered, the difference in both the groups was significant (p=0.033). However, most cases of MM could not be ascribed to a single phase of delay and more frequently a combination of delay in different phases was in play to cause deterioration in the health of women’s health. Most common delay observed was a combination of first delay and third phase delay in a total of 46.98% in the MNM group and 54.19% in MM [Table/Fig-2].

Among the direct causes gestational hypertensive disorder accounted for the highest proportion of deaths i.e., 34.35%. Among the women who survived, most of them were suffering from haemorrhage followed by hypertensive disorder and sepsis. In present study, patients with hypertensive disorders of pregnancy who reached the institution within six hours of first recognition of symptoms, all of them could be saved. Those reporting within 6 to 12 hours had prognosis depending on their stage of presentation. There were women who reached the institution with more than 12 hours delay. Despite all interventions and efforts, six unfortunate mothers survived the acute insult but were unable to cope with dysfunction of multiple organs and passed on after prolonged intensive care. Whereas in cases of PPH, delay >6 hours were observed with extremely poor prognosis. Although they underwent maximum number of interventions, 14 mothers could not be saved. A delay of <3 hours was associated with good prognosis and appropriate intervention undertaken at the proper time could save their lives. A delay for 3-6 hours leaves patients with organ dysfunction mostly [Table/Fig-3].

More than one phase of delay was present in most of the women contributing to her ill-health. Attendants accompanying the patients were interviewed and causes were elaborated [Table/Fig-4].

The huge difference in maternal deaths in high and low-income countries currently is primarily due to variation in the management of time while dealing with obstetric complications. Judicious use of the precious moments with skilled obstetric care can save mothers from dying. However, any delay in providing such can bring devastating effects which may range from long time morbidity to death. Delay in different phases hence can reflect standards of obstetric care in a community. According to an estimate by the World Health Organisation, 88-98% of maternal deaths can be prevented with effective referral systems providing prompt access to emergency obstetric care [15]. In the study, high incidence of Phase 1, Phase 3 and Phase 1+3 delay in both MNM and MM groups was observed. A combination of delays poses a greater health risk as the cumulative effect causes collation of risk factors for causing morbidity and mortality in mothers.

Phase 1 delay may be the event most perplexing to settle as it deals with the conceptual and behavioural phenomenon of patient and family members [16] and have been discovered in many studies in India even in Tertiary health centres [17-19]. In present study, first phase delay was observed in 67.84% in MNM cases and 79.38% in MM cases either in isolation or in combination with other phases of delay. The whole process of answering a call and hence making a decision is coaction of observation and impression influenced adequately by social, cultural practice and economic rearing [16,20]. In present study, the reasons found from conducting intensive individual interviews to explore their perspectives on coming to health facilities consequent to morbidity appraisal were heedless and unsuspecting of the symptoms, financial strain and unattended kids [Table/Fig-4]. In a study from Uttar Pradesh, India, users insisted on conduction of delivery at nearby First Referral Unit (FRU) or private facilities mostly due to the issue of distance, lack of money and transport, and being unaccustomed to the surroundings of the health centre [10]. In a study in less developed district of Kerala, out of 14 maternal deaths, first delay occurred in five cases where poverty, economic constraints, apprehension of the health system and gender issues were primary causes for such mishaps [11]. In present study, mothers haemorrhaging from ectopic pregnancy were high among MNM cases. This could be due to non-recognition of the entity and hence its complications, thereafter mothers were died suffering from complications of raised Blood Pressure (BP). Most of the relatives of these unfortunate mothers could not assess the seriousness of raised BP or were not aware of the morbid condition. In a country where doctor (modern medicine) and population ratio is 0.77:1,000 [21] as per population estimate of 1.33 billion, one on one counselling may not be possible for all cases. Instilling counsellors for these high-risk cases can help mothers and relatives to recognise early danger signs and report early.

Phase II delay was observed in 13.63% in MNM cases and 3.81% in MM cases in the present study either in isolation or in combination with other phases of delay. In a study in Kerala, this was observed in 50% of maternal deaths which was attributed to the difficult terrain of the region [11]. Due to better condition of roads of Delhi, the primary causes of second phase delay remained high costs of travelling and long distance to the tertiary centre which is located in the southern part of Delhi [Table/Fig-4]. These were the cases where referral health care units could not provide for vehicles due to multiple referrals at a time or the patients were too poor to access private mode of transport due to odd hours of the day when public transport was not at ease to find.

The most detrimental cumulative effect was evident with Phase III delay in both the groups. With health care personnel at primary level being unable to recognise and procure adequate diagnosis of disease, the delayed referral is the adverse sequel. The poor state of affairs may exacerbate with lack of funds, shortage of blood products, technical incompetence among health care personals, disagreeable attitude towards patients [22-24], particularly in low income countries with a few obstetric tertiary units receiving huge caseloads of critically ill women. Third phase delay was observed in a total of 70.25% and 72.50% of patients in the MNM and MM group, respectively either in isolation or in combination with other phases of delay. The main reasons ascribed for third phase delay which are also causes of referral in this study were non-availability of blood bank, operation theatre and competent staff. The institution is well equipped to provide comprehensive emergency obstetric care but with swarming loads of referrals in critical conditions, it is on rare occasions difficult to meet the demands of blood products and vacant operation theatre. Though the cases are pursued with strict adherence to saving critical mothers, decisions are biased in situations of concurrently occurring obstetric emergencies. This is though quite rare and only five cases in total (2 MNM and 3 MM) had to wait for blood products or operation theatre in the institution. Rare blood groups pose a challenging task due to non-availability and consumption of time as a result of subsequent outsourcing for procurement.

In present study, three patients despite no delays could not be saved. One of them was haemorrhaging from placenta previa and the other two were suffering from complications of hypertensive disorders of pregnancy when they reported to the hospital. Kasongo Project team while screening for fetopelvic dystocia found that it is difficult to screen women and predict the committal of Emergency Obstetric care and hence primary prevention is a myth in deterring the cases of MM [25]. With no recognisable determinants of clinical risk factors in most of the cases of MNM and MM, it can be deduced that a large proportion of serious complications remain largely unpredictable as validated by other studies as well [7,26].

This study was conducted in this institution for a period of one year, hence representing the data of a single tertiary centre. Large-scale state-wide representative data on all MNMes and deaths will provide a better understanding of the causes of delay and also guide to provisions of amendments in health policy. The study failed to evaluate the standards of antenatal care which has a huge confounding effect on the prognosis and survival of pregnant mothers.

Delay in taking decisions to seek healthcare being ignorant of the warning signs of the ensuing complication are a major causes of MM. Non-availability of blood bank and operation theatre facilities remains the most common causes for referral causing third phase delay calling for improvement in health infrastructure of the country. There is a significant difference between multiple deferral and referral than direct referral to an adequate health facility contributing immensely in terms of prognosis and survival of mothers.