Indicator 14: Direct obstetric case fatality rate

The DOCFR provides an overall picture of whether care provided to women with obstetric complications is effective in averting preventable maternal deaths. It is a crude measure of the quality of emergency obstetric care. In the Revised EmONC Guide, the DOCFR is accompanied by other measures of consistency and effectiveness of EmONC and should be interpreted alongside those indicators (see Exploring quality of care with the EmONC indicators).

The WHO defines a maternal death as: “the death of a woman while pregnant or within 42 days of termination of pregnancy, irrespective of the duration and site of the pregnancy, from any cause related to or aggravated by the pregnancy or its management but not from unintentional or incidental causes.”¹

Maternal deaths can be from either direct or indirect causes. Maternal deaths from direct obstetric causes are those “resulting from obstetric complications of the pregnant state (pregnancy, labour and puerperium), and from interventions, omissions, incorrect treatment, or from a chain of events resulting from any of the above.” Direct obstetric causes of death include: antepartum and postpartum hemorrhage (including retained placenta), severe pre-eclampsia and eclampsia, postpartum sepsis, severe abortion complications, uterine rupture, prolonged/obstructed labor and ectopic pregnancy. Deaths that result from complications of anesthesia or cesarean section are also considered to be from direct obstetric causes.² Maternal deaths from indirect causes are those deaths that result “from previous existing disease or disease that developed during pregnancy, and that were not due to direct obstetric causes but were aggravated by the physiologic effects of pregnancy.” Indirect obstetric causes of death include: existing cardiac or renal disease, or infectious diseases such as HIV, TB, and malaria.³⁴ This indicator, unlike Indicator 13: Institutional maternal mortality ratio, focuses only on maternal deaths from direct obstetric causes.

Maternal deaths occur either inside or outside the health system. This indicator, like Indicator 13: Institutional maternal mortality ratio, focuses on maternal deaths that occur in health facilities. Maternal deaths that occur outside of health facilities (i.e., in communities) are not included in this measure.

This indicator captures women admitted to a facility with direct obstetric complications, and those who develop such complications after admission. The numerator is the number of women who died in the facility as the result of those direct obstetric complications. The denominator is the number of women with direct obstetric complications recorded in facility registers who were not referred out (this is the same as the numerator for Indicator 8 met need for emergency obstetric care). The numerator and denominator come from the same set of facilities.

The direct obstetric complications that are included in the calculation of this indicator are: hemorrhage (antepartum and postpartum), severe pre-eclampsia and eclampsia, ruptured uterus, prolonged and obstructed labor, ectopic pregnancy, sepsis/maternal peripartum infection and complications of abortion. The operational definitions of these complications can be used to standardize data collection.

Data sources for obstetric complications include: registers in the maternity unit, operating theater, female or gynecological unit, and the emergency, inpatient and outpatient departments. Patient records can also be used for further investigation of complications. Some countries record direct obstetric complications in their HMIS and if the data are determined to be good quality (see ‘Supplementary studies – met need for emergency obstetric care’), they can be used in place of reviewing individual facility registers.

Data on maternal deaths are ideally captured by the civil registration and vital statistics (CRVS) system in a country. However, few low-and middle-income countries have CRVS systems that produce high-quality cause of death data.⁵ Therefore, data for this indicator most often come from the HMIS and originate from health facility registers in a range of locations: the maternity unit, the operating theater, the inpatient women’s unit, other relevant wards and the morgue. Data on maternal deaths can also be captured from maternal death reviews and other types of clinical audits.

This indicator is expressed as a percentage.

The benchmark is <1% of women with direct obstetric complications in health facilities die. The ultimate goal should be ending all preventable maternal deaths.

The stability of any rate depends on a large enough number in the numerator and accurate denominators. WHO cautions the use of case fatality rates when the numerator is small; this will affect the frequency of calculation. Even if data on maternal deaths and obstetric complications are collected routinely in the HMIS, the DOCFR is most usefully calculated and analyzed on an annual basis.

When the DOCFR is higher than 1%, it suggests that the obstetric care delivered was not effective. But, it is important to interpret this indicator cautiously and with the following in mind:

• The medical condition of some women on admission at a facility can be precarious, leaving providers little opportunity to save a woman’s life. Thus, not all deaths can be attributed to poor quality of facility-based care, but rather to weaknesses in the larger health system such as a poorly performing and ineffective network of services for referral, “dumping” practices (when one facility takes a critically ill woman to another facility in order to keep casualties low at the first facility), where trust in the health system is low, or where other barriers to timely care exist. Sometimes these circumstances are beyond the control of facility managers. If few facilities provide Basic and Comprehensive EmONC, women with complications are likely to arrive at the hospital after a long journey, reducing the likelihood of survival.
• When a facility or region begins to track the DOCFR, one of the early signs of improved data quality is the increased reporting of obstetric complications as well as an increase in the reporting of maternal deaths. This can lead to change in a DOCFR that is attributable to improved reporting. If monitoring is persistent and data quality improves, the impact of improved reporting diminishes.

The aggregated statistic at a national level does not identify which facilities are contributing most to the DOCFR. Therefore, disaggregation by sub-national area, level of facility, managing authority of facility and EmONC classification can be used to understand patterns and to determine where to focus attention.

Unlike some of the other indicators, this indicator can also be used for individual facility level monitoring.

Results tend to vary across levels of care and characteristics of the catchment area. In the example below (Table 1), we see consistent differences across levels of care. In the periodic national EmONC assessment data in Table 1, we see: 1) lower DOCFRs at lower levels of care (non-hospitals) compared to higher levels of care (hospitals), the exception being Zambia, and 2) higher levels of DOCFRs among EmOC facilities compared to all facilities. These results likely reflect the patient mix at different levels of care and referral (or by-passing) patterns. Women with the most severe complications are most likely to be found at the end-points of the referral hierarchy.

Calculating cause-specific direct obstetric case fatality rates (DOCFRs) can help identify which obstetric complications are the most lethal. Cause-specific DOCFRs are the proportion of women with a specific direct obstetric complication in facilities who die before discharge. They are calculated by disaggregating both the numerator and denominator of the DOCFR by cause. Each cause-specific DOCFR is expressed as a percentage.

There is no benchmark for cause-specific DOCFRs, but analyzing them can reveal weaknesses in the health system’s ability to manage specific complications. Direct obstetric causes with high fatality rates may be good candidates for quality improvement interventions focused on prevention, identification, and management of those complications. Tracking cause-specific DOCFRs over time can help show if such interventions are improving fatality rates for the targeted complications.

High direct obstetric case fatality rates indicate problems with clinical management and/or facility readiness but do not, by themselves, identify corrective actions. They are, however, a good beginning for further studies.

Case studies of women’s condition on admission¹³

Information on the condition of women with major complications at the time of admission (e.g., pulse, blood pressure, and temperature) can be collected, for women who survive and those who do not. Better understanding of patients’ condition on admission would help differentiate the effect of condition on arrival from the quality of care after arrival.

Delays in diagnosis or treatment¹⁴

There are many possible reasons for delayed diagnosis or treatment once a woman has reached a facility. For example, women’s families may have to buy drugs and medical supplies from local pharmacies because the hospitals do not have enough or are not open 24h/7d. Other causes of delays can vary from back-ups in the emergency room, to a gatekeeper who demands a tip, to electricity failures.¹⁵

Studies of ‘the third delay’ (once women have reached health facilities) and the ‘client flow analysis’¹⁶ exercise in the “Tool Book for Improving the Quality of Services” are useful models for this type of supplementary study; they systematize the observation and measurement of delays and allow researchers to identify at what stage they are most frequent.¹⁷ The exercises are based on evidence-based standards and expert opinion to determine what constitutes a delay. Another approach is to collect data on the interval between the time a woman with a complication is admitted and when she receives definitive treatment. Good-quality monitoring reveals which delays are the longest and most dangerous, and the direct obstetric case fatality rate can be lowered by reducing those delays.

Calculating adherence to selected standards of care indicators

When facilities have high case fatality rates, it is important to identify why and then implement and monitor solutions. One effective way to improve outcomes is to assess whether the provision of EmONC has been performed according to WHO standards of care. A set of indicators for this purpose, along with instructions for their use, can be found in Adherence to selected standards of care indicators.

Reviewing cases of women who survive life-threatening complications (‘near misses’)¹⁸

An alternative, more positive and sometimes less threatening approach to improving quality is to study systematically the care given to women with life- threatening obstetric complications who are saved by the health facility (‘near misses’). One benefit of this method is that near misses occur more frequently than maternal deaths and therefore provide more opportunities for studying the quality of care. Another benefit is that such a review provides an occasion to look at what health professionals did correctly to save the woman rather than focus on the problems. This helps to create a more supportive environment in which to discuss aspects of care that could be improved. The WHO publication “Evaluating the quality of care for severe pregnancy complications – the WHO near-miss approach for maternal health” provides instructions on how to conduct near-miss audits.¹⁹

The direct obstetric case fatality rate for women referred in from other facilities

The direct obstetric case fatality rate for women who were referred in from other facilities can be compared with the direct obstetric case fatality rate for women who only received care at the facility (without any interfacility referrals).

• Maternal and perinatal death and surveillance and response: https://www.who.int/publications/i/item/9789240036666
• Evaluating the quality of care for severe pregnancy complications: https://www.who.int/publications/i/item/9789241502221