Preventable maternal death: the three delays

A 24-year-old woman gives birth at home, with the help of a community midwife. One hour later, the bleeding does not stop. The family discusses whether to take her to the hospital, because the husband is away and the trip is expensive. When they decide to go, there is no transport nearby. She arrives at the hospital three hours later, pale and with low blood pressure, and the service has no blood available at that moment.

This story is fictional, but each stage of it appears in studies on maternal deaths in several countries. What kills, in this case, is not only the hemorrhage. It is the sum of delays between the problem and the treatment.

The three delays model organizes this reasoning into three questions: why the decision to seek care took long, why reaching the service took long and why adequate care took long after arrival. This text shows how the model works, what the studies found and what its limits are.

Key points

• The model divides the path to care into three delays: deciding to seek, reaching and receiving care.
• Many maternal deaths happen in health services, after the woman has already arrived.
• The delays usually add up, and more than one appears in the same case.
• Recognizing danger signs is the core of the first delay, and not only the woman's lack of information.
• New versions of the model include referral between services and continuity of care.

Where the model comes from

The three delays model was proposed by Thaddeus and Maine in 1994 and became one of the most used tools to analyze maternal deaths (Cheong and Baltazar, 2021). The central idea is that most deaths from obstetric complications can be avoided if treatment arrives in time.

The three phases

A study in Malawi summarizes the phases directly (Mgawadere and colleagues, 2017). The first delay is the delay in deciding to seek care. The second is the delay in reaching the health service. The third is the delay in receiving adequate care after arrival.

Each phase has different causes and calls for different responses. That is why the model is useful for case analysis: it forces whoever reviews the death to ask at what point along the path time was lost.

Why time matters so much

The main direct causes of maternal death, obstetric hemorrhage, hypertensive disorders of pregnancy, obstructed labor, sepsis and complications of unsafe abortion, account for almost two thirds of deaths in the world, according to an Egyptian study that reviewed the topic (Mohammed and colleagues, 2020). Several of them progress within hours.

A postpartum hemorrhage can lead to death quickly, and eclampsia can appear without warning in a pregnant woman with unrecognized pre-eclampsia. In these conditions, minutes and hours of delay change the outcome, and the question "where was time lost" has a concrete answer.

A global inequality

An Indian study recalls that 99 out of every 100 maternal deaths in the world happen in low and middle income countries (Sk and colleagues, 2019). But the model is not only for these contexts. As we will see, it has also been used to analyze deaths in high income countries.

The first delay: deciding to seek care

The first delay begins before any contact with the health system. It depends on recognizing that something is wrong and on having the power to decide to act.

Recognizing the danger signs

A qualitative study in Malawi reviewed 32 maternal deaths, with interviews with professionals, family members, neighbors and traditional midwives (Combs Thorsen and colleagues, 2012). Among the factors found, the lack of recognition of signs, symptoms and severity of the situation appeared repeatedly, along with low female schooling.

A study in Bangladesh on emergencies in general found the same pattern: the ability to recognize symptoms and the power of decision were the main factors in the decision to seek care (Shah and colleagues, 2020). Recognizing danger signs is a skill that can be taught to the community and to the family.

Who decides

The decision to seek care is not always the woman's. A study in Tanzania interviewed 27 men whose partners had given birth in the previous year (Greenspan et al., 2019). The authors describe that men often control financial resources and household decisions, and that a wrong judgment about danger signs or about the time to leave for work can delay care.

A study in Zambia broadened the view to other relationships (Kaiser et al., 2019). Husbands, female relatives and community health workers were named as the people who most influence the pregnant woman. The husband often gives the final permission to go to the service, and female relatives accompany the pregnant woman there.

Beliefs and trust in the service

The Indian study, in a region with a high burden of eclampsia, reviewed 317 maternal deaths and found the first delay as the largest contributor, present in 48.6 per 100 (Sk et al., 2019). The study's own title, which speaks of praying until death, sums up the weight of the sociocultural factors found in the verbal autopsies.

Trust in the service also matters. A study in Haiti with traditional birth attendants showed that the decision to refer the pregnant woman depends on the perception of quality and of the treatment at the service (Jean-Baptiste et al., 2023). A bad previous experience can delay the next search for care.

The second delay: reaching the service

Deciding to seek care is not enough. The pregnant woman needs to reach a service able to resolve the problem.

Distance and transport

In the Malawi study, the second delay was found in 59.6 per 100 of the women who died, and the distance to the service was the main problem (Mgawadere et al., 2017). In the Indian study, women who lived far away had more second delay, with an adjusted odds ratio of 1.7 (Sk et al., 2019).

The qualitative study from Malawi describes delays in access to transport and the difficulty of long distances over difficult terrain (Combs Thorsen et al., 2012). Ambulance, road, travel cost and time of day are part of the clinical problem, even if they do not appear in the medical record.

Reaching the wrong service

The second delay also includes reaching a service that cannot resolve the case. A 2026 analysis from Madagascar shows that, in health systems organized as a pyramid, the pregnant woman often goes through a first service and needs to be referred to another (Franke and Iata Pasteur, 2026). Each referral repeats the cycle of deciding, reaching and receiving.

For this reason, the authors propose a model of six delays, which separates the delay in seeking, reaching and receiving initial care from the delay in seeking, reaching and receiving referral care (Franke and Iata Pasteur, 2026).

Legal and administrative barriers

A study in Malaysia with stateless people and migrants in an irregular situation shows another type of barrier (Cheong and Baltazar, 2021). Political and legal barriers delay the decision to seek the service, the arrival there and the care itself. Exclusion from one system, such as civil registration, increases the risk of exclusion from the other, health.

The third delay: receiving adequate care

The third delay happens inside the health service. It is the most uncomfortable for professionals, because it is the one closest to their control.

It is more common than it seems

In the Malawi study, 62.2 per 100 of maternal deaths occurred in a health service, and another 21.2 per 100 occurred at home after the woman had been through a service (Mgawadere et al., 2017). In these women, the third delay was the most important, present in 96.8 per 100 of cases.

A study in an Egyptian tertiary hospital reviewed 207 maternal deaths over 10 years (Mohammed et al., 2020). The third delay was present in 88.9 per 100 of deaths, and the second in 50 per 100, always together with others.

What makes up the third delay

The qualitative study from Malawi describes delays in emergency obstetric care and delays within the hospital (Combs Thorsen et al., 2012). The Bangladesh study points to lack of staff, lack of resources and lack of training as causes of delay in receiving care (Shah et al., 2020).

In practice, the third delay includes several failures that add up.

• Delay in triage and in recognizing severity.
• Lack of blood, medicines or equipment.
• Lack of a trained professional or of an available surgical team.
• Wrong or late diagnosis of the cause of the worsening.
• Late referral to a service of greater complexity.

The third delay also involves clinical reasoning

Part of the third delay is diagnostic error. A hidden hemorrhage, an unrecognized pre-eclampsia or sepsis treated late are failures of reasoning, not only of structure. Case review in morbidity and mortality sessions is one way to turn these failures into learning.

What the comparison between studies shows

The delays add up

One of the most consistent lessons from the studies is that delays rarely appear alone.

More than one delay in the same case

In the Malawi study, more than half of the women who died in a health service, 52.1 per 100, had more than one type of delay (Mgawadere et al., 2017). The first delay was present in 39.7 per 100 of the women and the second in 59.6 per 100.

The woman who took long to decide to seek care arrives sicker, and the woman who arrives sicker has less margin for any delay within the service. The delays multiply, they do not merely add up.

The weight of each delay changes with the context

The comparison between studies shows that the most important delay varies.

Study and context	Most frequent delay	What the authors highlight
Malawi, 151 deaths	Third, in those who reached the service	More than one delay in more than half
Egypt, tertiary hospital, 207 deaths	Third, in 88.9 per 100	Second delay always together with others
India, region with many eclampsia cases, 317 deaths	First, in 48.6 per 100	Cultural factors and distance
Sweden, 75 records	Substandard care factors	Communication barriers in women born abroad

Table source: Mgawadere et al., 2017; Mohammed et al., 2020; Sk et al., 2019; Esscher et al., 2014.

The change with expanded access

The authors of the Malawi study observe that, previously, most of the women who died had suffered the first and second delays (Mgawadere et al., 2017). With more births in health services, the third delay gained weight. The Egyptian study describes the same shift in pattern (Mohammed et al., 2020). Taking the pregnant woman to the service is necessary, but not enough.

The model in high-income countries

The three delays model was born with low-resource contexts in mind, but it has also been applied where maternal mortality is low.

The Swedish experience

A study in Sweden audited 75 records of maternal deaths, comparing women born in low- and middle-income countries with women born in the country (Esscher et al., 2014). The authors used a version of the model adapted for migrants.

Substandard care factors were associated with most of the deaths and were more frequent in women born abroad (Esscher et al., 2014). The main delays in seeking care included communication barriers, such as a different language and inadequate use of interpreters.

A lesson for any service

The Swedish experience shows that the first and third delays can appear even with well-equipped hospitals. The language barrier delays the decision to seek care and the team's understanding of the signs. The language used with the patient and listening to her history are part of prevention.

The causes behind the delays

The three delays model does not replace the question about the cause of death; it complements it. A review of 32 maternal deaths in Lilongwe district, Malawi, found 16 deaths from direct obstetric causes, with sepsis and hemorrhage at the top, and 16 from indirect causes, mainly anemia, followed by HIV and heart disease (Combs Thorsen and colleagues, 2012). In this series, the most common delay was the third, linked to delays in referral, missed diagnoses, lack of blood and medicines, and inadequate care (Combs Thorsen and colleagues, 2012).

In case analysis practice, this means crossing two columns: what the cause was and where time was lost. The same delay does not carry the same weight for all causes, and an indirect disease such as anemia calls for a look at the whole pregnancy, not only the final hours. Without this joint reading, the analysis risks blaming transport for a death that depended on a diagnosis that was never made.

Wider access does not solve everything

A national survey in Malawi started from an observation: previously, most women who died had suffered the first and second delays, and, with wider service coverage, the authors wanted to measure the relative weight of each one (Mgawadere and colleagues, 2017). Of the 151 deaths, 62.2 per 100 occurred inside a health facility, and the third delay appeared in 96.8 per 100 of the women who died at the facility or at home after seeking care (Mgawadere and colleagues, 2017). Among the causes of this delay were long waits before treatment, delays in admission, lack of medicines, and the absence or lack of preparation of qualified professionals (Mgawadere and colleagues, 2017). The authors' conclusion is that most women try to reach the service when the emergency arises, and that improving the quality of care within it is the next step (Mgawadere and colleagues, 2017).

This shift has a direct consequence for those who work inside the service. The question is no longer only why the woman took so long, and now includes what happened between the entrance door and the first effective action. Time to assessment, to testing, to blood and to surgery become questions the team can ask in every review.

Those at the front line of care

Many of a pregnant woman's decisions involve people close to her. In focus groups in rural Zambia, participants named husbands, female relatives and community health workers as the most influential figures during maternity (Kaiser and colleagues, 2019). Community health workers and traditional birth attendants appeared as educators about danger signs and the benefits of giving birth at the service, and the birth attendants also referred the pregnant woman when complications arose (Kaiser and colleagues, 2019). When these people did not fulfill their role, this became a barrier to deciding to give birth at the service, the first delay, or to reaching it, the second (Kaiser and colleagues, 2019).

For the student, the lesson is that the analysis of a maternal death must look at the whole chain, from the pregnant woman's home to the operating room, and that the time lost at each link has causes that can be identified and corrected.

Limits of the model and use in case analysis

The limits of the model and the new versions

Like any model, the three delays model simplifies reality. Recent studies propose expansions.

Continuity of care

A study with 76 verbal autopsy narratives in Pakistan and Mozambique suggests that the model needs to be complemented now that most births take place in health services (Vidler and colleagues, 2023). The narratives showed temporary improvements after seeking care, quality failures in emergencies, confusing referrals and arrival at the final service in critical condition.

The authors propose focusing on continuity of care and the critical connection points: between the phases of pregnancy, between families and professionals, and between health services (Vidler and colleagues, 2023). Maternal death, in this view, is not a single journey but a series of connections that can fail.

Referral in multiple stages

The three delays model, already mentioned, responds to the same problem (Franke and Iata Pasteur, 2026). In systems where the pregnant woman goes through several services, each passage has its own risk of delay, and the original model puts everything into a single box.

What the model does not show

The model describes where time was lost, but it does not measure the quality of care itself. It relies on verbal autopsy and on medical records, subject to gaps in memory and in documentation. It can attribute to the woman and her family a delay that is, in fact, the system's. And it does not capture well the indirect causes, such as anemia and chronic diseases, which require a look at the whole pregnancy, and not only at the final hours.

Using the model in the analysis of a case

The three delays model can be used as a script to review a maternal death or a near miss.

1. Reconstruct the timeline, from the first symptom to the outcome.
2. Identify when the danger signs appeared and when they were recognized.
3. Ask who decided to seek care and what weighed in the decision.
4. Measure the time to reach the first service and the barriers along the way.
5. Check whether the first service could handle the case and how long the referral took.
6. Analyze what happened inside the service: triage, diagnosis, resources and team.
7. Separate what was avoidable from what was not, without looking for individual blame.

Back to the woman in the case

The 24-year-old woman suffered the three delays. The family took time to decide, because of cost and the absence of whoever made the decisions, which recalls the role of men and of close relationships described in studies in Tanzania and Zambia (Greenspan et al., 2019). Transport was lacking, the second delay (Mgawadere et al., 2017). And the hospital had no blood, the third delay (Combs Thorsen et al., 2012).

The review of this case should not end with "postpartum hemorrhage" as the cause of death. It should point out the three points where time was lost and what could change in each one. Careful recording of this analysis in the medical record and in the service's sessions is the first step.

How NEXO helps train this reasoning

In NEXO, the case of the day is fictional, with a real clinical presentation and up to six progressive clues, and you propose the diagnosis by searching for the ICD-10 code.

The limits of use of the game are described on the NEXO economy page. If you research the teaching of clinical reasoning, the page for researchers explains how to get in touch. Professionals who write or review cases find information in curation and in creators, and institutions can learn about the proposal on the partnerships page. Training differential diagnosis with clues and searching for the right ICD-10 code complete this training.

Frequently asked questions

What is the three delays model?

It is a way of analyzing maternal deaths by dividing the path to care into three phases: deciding to seek it, reaching the service and receiving adequate care (Mgawadere et al., 2017).

Which delay is the most important?

It depends on the context. In studies in Malawi and Egypt, the third delay was the most frequent (Mohammed et al., 2020), while in a region of India the first was the largest contributor (Sk et al., 2019).

Does the model apply to high-income countries?

Yes. A study in Sweden used a version of the model and found communication barriers and suboptimal care, more frequent in women born outside the country (Esscher et al., 2014).

Is the first delay the woman's fault?

No. It depends on recognizing danger signs, on who has decision-making power in the family and on trust in the service (Kaiser et al., 2019).

Is the model still sufficient?

It remains useful, but recent studies propose including referral between services and continuity of care (Vidler et al., 2023).

Take-home messages

• The model divides the path to care into deciding to seek it, reaching it and receiving care.
• Many maternal deaths happen after the woman has already reached the service.
• The delays add up and, often, more than one appears in the same case.
• Recognizing danger signs involves the woman, the family and the community.
• Language and documentation barriers delay care in high-income countries too.
• New versions include referral and continuity of care.

References

1. Cheong AR, Baltazar MAK. Too precarious to walk: an integrated "three delays" framework for modeling barriers to maternal health care and birth registration among stateless persons and irregular migrants in Malaysia. Genus. 2021. PMID 34493875.
2. Combs Thorsen V, Sundby J, Malata A. Piecing together the maternal death puzzle through narratives: the three delays model revisited. PLoS One. 2012. PMID 23284882.
3. Esscher A, Binder-Finnema P, Bødker B, et al. Suboptimal care and maternal mortality among foreign-born women in Sweden: maternal death audit with application of the 'migration three delays' model. BMC Pregnancy Childbirth. 2014. PMID 24725307.
4. Franke MA, Iata Pasteur G. The Six Delays Model: expanding the three delays model with evidence from Madagascar for maternal referrals in LMICs. BMJ Glob Health. 2026. PMID 41702636.
5. Greenspan JA, Chebet JJ, Mpembeni R, et al. Men's roles in care seeking for maternal and newborn health: a qualitative study applying the three delays model to male involvement in Morogoro Region, Tanzania. BMC Pregnancy Childbirth. 2019. PMID 31409278.
6. Jean-Baptiste M, Millien C, Pognon PR, et al. Reframing the Three Delays framework: factors influencing referrals to facilities by matrones in rural Haiti. BMJ Glob Health. 2023. PMID 37967901.
7. Kaiser JL, Fong RM, Hamer DH, et al. How a woman's interpersonal relationships can delay care-seeking and access during the maternity period in rural Zambia: An intersection of the Social Ecological Model with the Three Delays Framework. Soc Sci Med. 2019. PMID 30500609.
8. Mgawadere F, Unkels R, Kazembe A, et al. Factors associated with maternal mortality in Malawi: application of the three delays model. BMC Pregnancy Childbirth. 2017. PMID 28697794.
9. Mohammed MM, El Gelany S, Eladwy AR, et al. A ten year analysis of maternal deaths in a tertiary hospital using the three delays model. BMC Pregnancy Childbirth. 2020. PMID 33023523.
10. Shah B, Krishnan N, Kodish SR, et al. Applying the Three Delays Model to understand emergency care seeking and delivery in rural Bangladesh: a qualitative study. BMJ Open. 2020. PMID 33361169.
11. Sk MIK, Paswan B, Anand A, et al. Praying until death: revisiting three delays model to contextualize the socio-cultural factors associated with maternal deaths in a region with high prevalence of eclampsia in India. BMC Pregnancy Childbirth. 2019. PMID 31455258.
12. Vidler M, Kinshella MW, Sevene E, et al. Transitioning from the "Three Delays" to a focus on continuity of care: a qualitative analysis of maternal deaths in rural Pakistan and Mozambique. BMC Pregnancy Childbirth. 2023. PMID 37872504.