Assessing respectful maternity care in a fragile, conflict-affected context: Observations from a 2016 national assessment in Afghanistan

Abstract

Evidence on experiences and perceptions of care in pregnancy and childbirth in conflict-affected settings is limited. We interviewed 561 maternity care providers and observed 413 antenatal care consultations, 671 births, and 393 postnatal care consultations at public health facilities across Afghanistan. We found that healthcare providers work under stressed conditions with insufficient support, and most women receive mixed quality care. Understanding socio-cultural and contextual factors underpinning acceptance of mistreatment in childbirth, related to conflict, insecurity, gender and power dynamics, is critical for improving the quality of maternity care in Afghanistan and similar fragile and conflict affected settings.

The number of women and children affected by armed conflict has grown steadily since 2000; during this period, at least 630 million women and children—10% of women and 16% of children worldwide—have either been displaced by conflict or living dangerously close to armed conflict events (Bendavid et al., 2021). With this shifting global landscape, there is growing attention to quality of health services in crisis-affected settings, with efforts to review evidence on health system quality (Jordan et al., 2021), understand context-specific challenges to delivery of quality care (Munyuzangabo et al., 2021), and develop operational guidance for improving quality of care (Syed et al., 2020) in areas affected by conflict and fragility.

As authors of the Lancet Maternal Series in 2016 emphasize: Every woman, every newborn, everywhere has the right to appropriate, timely, evidence-based, respectful care (The Lancet, 2016). Mistreatment during childbirth has been highlighted as a significant obstacle in reducing maternal and newborn mortality around the globe – it contributes to negative user experiences and can be a significant deterrent to the uptake of facility-based childbirth (Kruk et al., 2018; Sen et al., 2018). The vision of the World Health Organization (WHO) Quality of Maternal and Childcare Network for ‘Quality, Equity, Dignity’ is garnering momentum (WHO, 2018). The WHO Quality of Care framework for maternal and newborn healthcare sets out the provision of respectful, dignified care as a tenet of quality – on par with other quality domains such as provider competence and health system infrastructure. A woman’s “experience of care” (effective communication, respect and dignity, and emotional support) are critically important during childbirth (WHO, 2016).

Respectful maternity care (RMC) is even more important in humanitarian settings, where weakened and fragile health systems struggle to provide quality care and all women, but especially the most marginalized and vulnerable, may avoid care seeking as there are no perceived benefits in attending a health facility (Sacks & Peca, 2020). Systematic evidence on how women’s health can be affected by armed conflict is sparse, however researchers have reported on poor mental health (e.g. anxiety and depression) and poor sexual and reproductive health, with increased risk of morbidity and mortality from direct violence and deterioration of local conditions.(Bendavid et al., 2021). According to United Nations Population Fund (UNFPA), an estimated 500 women and girls die from complications due to pregnancy and childbirth in fragile and crisis-affected countries every day (UNFPA, 2015). The lifetime risk of maternal death for women living in fragile states is 1 in 54 versus 1 in 4900 in developed countries (Every Woman Every Child, 2016). Globally, of the ten countries with the worst neonatal mortality rates, five – one of which is Afghanistan - are experiencing humanitarian emergencies (UNFPA, 2015). Dignity is essential to human rights, and childbirth needs to be established as a safe zone.

Advancing respectful maternity care in Afghanistan

Despite ongoing conflict and poverty, Afghanistan has made reasonable progress in its reproductive, maternal, newborn, and child health indicators over the last decade and the government has committed to the ambitious goals outlined in the Global Strategy for Women’s, Children’s’ and Adolescents’ Health, protecting rights to the highest attainable standard of health and ensuring that women and children can survive and thrive (Every Woman Every Child, 2016). Equitable access to healthcare remains a challenge (Akseer et al., 2016), however, and the maternal mortality ratio (MMR) remains high at 661/100,000 livebirths (Ministry of Public Health, 2017).

Afghanistan’s health system is affected by continuous conflict and terrorism, which has intensified since 2010, and more and more people now live in areas affected by high levels of insecurity (Chopra & Arur, 2018). Primary healthcare services are provided by non-governmental organizations contracted to manage public health facilities; and available services are considered of questionable quality and not “user friendly” (Ministry of Public Health, 2016). Post-traumatic stress disorder is widespread, not only affecting those in combat; many women, children, and adolescents also live with depression and anxiety (The Lancet, 2020). Only 48% of births take place in a health facility and 51% of births are assisted by a skilled provider (Central Statistics Organization et al., 2017). Mistreatment during childbirth is normalized (Arnold et al., 2015) and researchers identify it as a major barrier for women in seeking and receiving adequate care at health facilities (Rahmani & Brekke, 2013).

Mistreatment of women during childbirth at health facilities is complex and highly contextual (White Ribbon Alliance, 2015). Little is known about how armed conflict and health system fragility affect drivers of mistreatment or effectiveness of strategies to promote RMC. In 2016, Jhpiego was commissioned to conduct a National Maternal and Newborn Health Quality of Care Assessment to inform the Ministry of Public Health (MOPH) and development partners’ efforts to improve quality of health services and reduce preventable maternal and newborn mortality and morbidity in Afghanistan. In this paper, we reflect on intentional efforts to include examination of both providers’ and clients’ experiences in this assessment, and share lessons learned for future efforts to measure and promote RMC in Afghanistan and similar fragile settings.

Materials and methods

The 2016 Afghanistan Maternal and Newborn Health Quality of Care Assessment was a cross-sectional assessment examining readiness to provide routine care and address major obstetric and newborn complications at 286 health facilities across all 34 provinces of Afghanistan (Jhpiego, 2017). In this analysis, we use a subset of findings to assess the extent to which care provided at hospitals in Afghanistan meets international standards for respectful maternity care.

Study setting and sample

Our dataset includes a census of all accessible public hospitals with an average of five or more births per day reported in the national health management information system in 2015: 40 district hospitals, 27 provincial hospitals, 5 regional hospitals and 5 specialized hospitals. Two additional district hospitals reported an average of five or more births per day in 2015 but were not accessible due to insecurity at the time of data collection.

Data collection

We used two data collection methods to assess the quality of MNH services both in terms of compliance with global clinical practice standards, as well as the manner and environment in which care is provided: (1) semi-structured interviews with skilled birth attendants, and (2) direct observation of antenatal care, intrapartum care, and inpatient postnatal care before discharge after childbirth.

We based data collection tool content on WHO guidelines and tools used in conducting quality of care assessments in other countries (Bartlett et al., 2015); as well as tools used in Demographic and Health Survey (DHS), Service Provision Assessments (Demographic and Health Surveys Program), and EmONC assessments supported by the Averting Maternal Death and Disability (AMDD) program (Demographic and Health Surveys Program). We added additional interview questions to examine maternity care provider views on drivers and barriers to RMC and their working conditions, based on a review of tools from RMC studies in other settings (Bohren et al., 2015; Filby et al., 2016; Maung et al., 2020). All tools were developed in English, translated into Dari and Pashto, and pre-tested during data collector training at a public hospital in Kabul (Jhpiego, 2016).

Data collectors were 32 experienced female midwives and doctors who received technical updates on maternal and newborn healthcare and training on data collection techniques, with a focus on clinical observation, data quality assurance, research ethics and use of CommCare software. They completed data collection during 2-3 day visits to each facility between 14 May and 3 August 2016. The number of interviews conducted at each facility (1-8 interviews) was dependent on the number of maternity staff (all female) assigned to that facility type and available at the time of data collection. Data collectors aimed to observe up to five antenatal care (ANC) consultations, five vaginal births and five pre-discharge postnatal care (PNC) examinations at each district hospital and up to five antenatal care consultations, 10 vaginal births (five during a day shift and five during a night shift) and five postnatal examinations at each provincial, regional and specialized hospital.

Data collection was conducted using CommCare software loaded on Android tablets, with paper tools used as backup in sites where use of tablets was considered a security risk or unacceptable to care providers or women. When paper tools were used, data from completed checklists were entered into the software when data collectors were in a safe location with internet access.

Data analysis and interpretation

We used descriptive statistics to analyze patterns in interview responses and quality of care provided, and Chi2 tests to measure differences in provider-client communications and mistreatment by hospital type. Stata software version 15 was used for all statistical analysis.

Observations of antenatal care, intrapartum care and postpartum ward rounds were analyzed as qualitative case narratives. Completed observation checklists were exported from CommCare to Excel for data cleaning (removal of duplicates and incomplete observations, correction of checklist responses based on qualitative notes from data collectors at the end of each observation form), and then uploaded to NVivo 12 for thematic content analysis guided by a codebook based on the typology of mistreatment of women during childbirth developed by Bohren and colleagues, and barriers to quality midwifery care provision developed by Filby and colleagues (Bohren et al., 2015; Filby et al., 2016). Line-by-line coding of data collector observations was conducted by the second author (LN); summary memos and outputs for specific codes were reviewed and further analyzed by SC, LN and HT. Throughout the analysis process, the study team considered questions related to reflexivity, including the influence of their preexisting assumptions on what constitutes respectful care and mistreatment, as well as influence of their ongoing involvement in maternal and newborn health programming in Afghanistan upon their interpretation of the study results.

Ethical considerations

The 2016 Afghanistan National Maternal and Newborn Health Quality of Care Assessment protocol was approved by the ethical review boards of the Afghanistan Ministry of Public Health (361533) and the Johns Hopkins Bloomberg School of Public Health in Baltimore, Maryland (6799). Written permission for data collection was obtained from the in-charge of the hospitals, and oral informed consent was obtained from all participating healthcare providers and women (or women’s next of kin when they were too ill to provide informed consent directly).

Results

Interviews were conducted with a total of 561 skilled maternity care providers (333 at provincial, regional and specialized hospitals; 228 at district hospitals with an average of five or more births per day) and a total of 413 ANC consultations, 671 births and 393 PNC consultations were observed. Not all births were observed from admission; some observations began as labor progressed, and others during childbirth.

Standards of care observed

A woman’s “experience of care” (effective communication, respect and dignity, and emotional support) is critically important during childbirth. The majority of women observed during antenatal, intrapartum and postnatal care received mixed quality care, meeting some international standards for RMC and not others.

Effective communication

Effective communication between a woman and her healthcare provider can reduce anxiety and contribute to making childbirth a positive experience. The woman should receive information about her care and feel involved in decisions regarding her management. As shown in Table 1, 223 (54%) of women received an explanation of what the health worker was doing during ANC; 172 (34.3%) during the first stage of labor and 131 (30.8%) women received an explanation during PNC examination. Only 129 (31.2%) women were asked if she had any questions during ANC examinations and 85 (21.6%) during PNC. Gaps in communication extended to only 480 (75%) of women being informed of the sex of their baby at birth.

Table 1. Provider-client rapport and communications recorded in observation checklists.

	All relevant public facility types	Provincial, Regional & Specialty Hospitals	District Hospitals with 5 or more deliveries per day	p-value
Antenatal care	n = 413	n = 206	n = 207
Percent of women receiving explanation of what health worker is doing during antenatal care examinations	54.0% (223)	52.4% (108)	55.6% (115)	0.466
Percent of women asked if she has any questions during antenatal care examinations	31.2% (129)	34.5% (71)	28.0% (58)	0.219
Labor and Delivery
Admission	n = 431	n = 385	n = 276
Percent of women receiving explanation of what health worker is doing during initial examinations upon admission	40.1% (173)	39.6% (97)	40.9% (76)	0.954
First stage of labor	n = 502	n = 284	n = 218
Percent of women told during the first stage of labor what will happen during labor and birth	34.3% (172)	32.8% (93)	36.2% (79)	0.581
Percent of women who were encouraged to ambulate and assume different positions during labor	37.1% (186)	33.8% (96)	41.3% (90)	0.157
Immediately postpartum	n = 637	n = 367	n = 270
Percent of women informed of sex of baby	75.4% (480)	74.4% (273)	76.7% (207)	0.442
Postpartum care	n = 393	n = 209	n = 184
Percent of women receiving explanation of what health worker is doing during postpartum ward round examinations	30.8% (121)	24.4% (51)	38.0% (70)	0.018
Percent of women asked if she has any questions during postpartum examinations	21.6% (85)	16.3% (34)	27.7% (51)	0.016

Respect and dignity

Meeting professional standards of care is a minimum requirement of healthcare providers and there were examples of good care, for example:

All the process was followed in order. The pregnant woman was under respectable treatment. Patient’s privacy was respected; BP was checked, abdominal exam was performed; fetal heart sounds were examined. Guidance regarding vaccination, time of next visit and Folic Acid was given to the patient; (…) the patient was discharged very kindly; she happily went home. (A49_C4)

However, in many cases, little attention was paid to women in labor when they first arrived, the progress of birth was not monitored regularly and preparations for the birth were not completed on time. Only 340 (52.7%) women observed during labor and birth had a companion present during childbirth. There were observations of women giving birth on the floor, bed or washroom by themselves. In several cases, essential care for newborns was not provided. Partographs were usually completed after the birth.

Inequitable treatment and discrimination were noted frequently in data collector observations, often linked to poor women or those with no ‘connections’ in the facility.

The patient who has an intermediary or belongs to a commander or hospital’s head, she will receive ideal health care services respectfully and on time. But other patients have to spend money to do their examination in private hospitals and buy drugs as well. Most of them are very poor and not able to do this … (A76_B10)

A common observation was neglectful or substandard postpartum/postnatal care, such as no monitoring, no support with breastfeeding, keeping newborns and mother separate/giving newborns to relatives, no counseling on maternal and newborn health such as danger signs. In some cases, women themselves requested early discharge (due to security or transportation challenges) and even removed their IV cannulas and left.

The patient waited in the birth room for around two hours but she was not transferred to the PNC room. The midwife didn’t encourage her to stay for post-birth observation and she directly left the hospital. (A22_E2)

Mistreatment. Many types of mistreatment defined in Bohren’s typology were observed. Extreme cases of abuse were relatively rare, but other forms of mistreatment - including stigma and discrimination, failure to meet professional standards of care, and poor rapport – were pervasive. As shown in Table 2, observations revealed that 44 (6.6%) women were shouted at, insulted or threatened at any time during labor or after. 12 (1.8%) of newborns were slapped by the health worker and 24 (3.9%) of newborns held upside down. Various unindicated or potentially harmful practices were observed during labor and birth, for example, 56 (8.4%) women had fundal pressure to hasten birth of baby or placenta and 76 (11.3%) had manual exploration of uterus after birth.

Table 2. Abuse and potentially harmful practices recorded in observation checklists.

	All relevant public facility types (n = 671)	Provincial, Regional & Specialty Hospitals (n = 395)	District Hospitals with 5 or more deliveries per day (n = 276)	p-value
Physical Abuse
Percent of women slapped, hit or pinched the woman during labor or after	0.8% (5)	0.8% (3)	0.7% (2)	0.959
Percent of newborns slapped by health worker	1.8% (12)	2.0% (8)	1.5% (4)	0.580
Percent of newborns held upside down by health worker	3.9% (24)	3.3% (13)	4.0% (11)	0.634
Verbal Abuse
Percent of women shouted at, insulted or threatened at any time during labor or after	6.6% (44)	6.4% (25)	6.9% (20)	0.678
Failure to Meet Professional Standards of Care
Percent of women subjected to unindicated or potentially harmful practices during labor and delivery
Use of enema	3.1% (21)	2.3% (9)	4.3% (12)	0.130
Fundal pressure to hasten delivery of baby or placenta	8.4% (56)	6.8% (27)	10.5% (29)	0.091
Lavage of uterus after delivery	14.5% (97)	14.2% (56)	14.9% (41)	0.806
Stretching of perineum	16.0% (107)	16.5% (65)	15.2% (42)	0.666
Restricting food and fluids in labor without indication	6.1% (41)	5.1% (20)	7.6% (21)	0.176
IV line started without indication	16.1% (108)	15.7% (62)	16.7% (46)	0.669

Case-by-case analysis of care received during ANC, labor and birth, and immediate PNC pointed to inconsistencies in objective and subjective quality of care observations, suggesting that sub-optimal standards of care are normalized and accepted. For example, observers noted that care was respectful but providers did not adhere to clinical standards, or overall case management was described as ‘excellent’ when infection prevention measures were ignored. Table 3 presents several examples of mistreatment and sub-standard care observed, using Bohren’s typology as an organizing framework.

Table 3. Mistreatment recorded in observer field notes.

General theme*	Specific themes observed in 2016 National Maternal and Newborn Health Quality of Care Assessment*	Illustrative examples of mistreatment
Physical abuse (use of force, physical restraint)	Women slapped	“…her privacy wasn’t respected; the patients make noise between them and since the midwives were very tired they couldn’t tolerate it and eventually one of them slapped the patient.” (A2_D1)
Verbal abuse (harsh language, threats and blaming)	Harsh and rude language Judgmental comments Blaming Shouting	“The attitude of the doctor was annoying. She was frequently humiliating the customs and living conditions of the patient. The doctor laughed on the patient as she had golden teeth and the doctor was irritating her that she was not worthy to have golden teeth.” (A36_E2)
Stigma and discrimination (discrimination based on socio-demographic characteristics or medical conditions)	Discrimination based on medical condition, family/clan relations, socio-economic status, and distance of residence from hospital	“The patient was Hepatitis B positive and nobody wanted to assist her delivery because they said nobody had risk incentives, saying ‘They send us infected patients and the doctors do not visit and they leave everything for the midwives’. The patient was alone in the room until nearly the fetal head was out. Then there was noise and chattering and finally the midwife came and there were no preparations for the delivery and partograph was not filled. After the child was born the patient was left alone and no guidance was given and her vital signs were not checked and the newborn wasn’t weighed.” (A1_D5)
Failure to meet professional standards of care (lack of informed consent and confidentiality, physical examinations and procedures, neglect and abandonment)	Lack of informed consent process before ANC, labor and delivery and PNC Neglect in provision of ANC, labor and delivery, and PNC procedures including no or partial history taking; routine physical examination; discussion of danger signs, vaccination, nutrition during pregnancy and lactation; developing and discussing delivery plans with the mothers; monitoring of labor; routine care during delivery for both mother and baby; support with breastfeeding; health education and family planning counseling Keeping babies and mothers separate after birth or giving babies to relatives Abandonment and long delays in ANC, labor and delivery and PNC; including delay in transfer of mothers to the delivery room, delay in provision of care because of unavailability of gloves, early discharge of mothers after delivery or leaving mothers alone until they took off the IV cannulas themselves and left the health facility Recording ‘made-up’ information in registers/ patient files/partographs after delivery or end of a mother’s stay at the facility Healthcare provider absence at time of delivery resulting in mothers delivering on floor, corridor, washroom or delivery room by themselves Provision of limited information to mothers/relatives and performance of unconsented surgical operations Refusal to provide pain relief Unindicated and potentially harmful practices including use of enema, stretching of perineum, fundal pressure to hasten delivery of baby or placenta, lavage of uterus after delivery, restricting food and fluids in labor without indication, use of episiotomy and vacuum extraction without indication, starting IV line without indication Poor infection prevention measures including not washing hands, no use of gloves for required physical examinations or use of the same syringe, pair of gloves, vacuum extractor or delivery set for multiple deliveries happening at the same time	“…The patient was unattended for hours, despite having anemia and high blood pressure and the patient’s contractions increased… eventually the patient delivered in the room. Her visitor helped her and pulled the baby but placenta remained, she shouted and moments later a health worker came and took out the placenta; the patient was given serum and was discharged minutes later.” (A14_D3) “… while the health worker was preparing the instruments and taking care of safety precaution, mother delivered the baby and the baby slipped from the delivery table and the baby’s head first hit the table then fell down to the ground and the cord was ruptured and the placenta was pulled out. The health worker clamped the cord then dried the baby on the floor and covered it with a piece of clothing. The health worker did not inject the oxytocin for the active management of third stage of delivery. She did not seek help from others also.” (A33_D6) “… there was a complication which was urethral rupture the midwife repaired it without giving anesthesia. The poor patient was screaming, even I (the observer) cried for her….” (A23_D10) “… so diagnosis was CPD (cephalo-pelvic disproportion) and she (the mother) was carried to the surgery ward for cesarean section and the relatives didn’t know the reason…. The specialist doctor is a male doctor he did the operation without permission of the patient or her relatives and after 5 min he came out of the delivery room and he says ‘did you see I finished the process in 5 min?’. He asked me to ask the Ministry to give him an appreciation letter…all patients are excited by his permission, he make all staff under his control, he has very bad behavior with his patients…” (A72_D1) “…in all patients after the placenta is detached they didn’t give the patient uterine massage; only uterine lavage is given but in its worst condition as patients’ cries and sobbing are heard out of the building. (A26_D1) “Because, the number of patients and their relatives are too much and there is low number of midwives. One midwife is not able to service respectfully to all admitted patients. We use one pair gloves for several patients "we don’t have the time to change our shoes" (an Afghan expression), if we take off our glove to change, the next baby will fall down in the ground or the cervix would lacerate.” (A23_B1)
Poor rapport between women and providers (ineffective communication, lack of supportive care, lack of autonomy)	Healthcare providers do not greet, listen to, ask questions from, and try to understand mothers/relatives, and limited/no attention to their concerns Healthcare providers and mothers speak different languages Healthcare providers humiliate mothers/relatives for having several children, yellow teeth, and delivering the baby in washroom No help to prepare woman for physical examination No explanation of procedures, physical examination, lab results, and pregnancy progress, treat mothers as passive participants during childbirth, and no emotional support during and after delivery Not informing mothers/relatives of the sex of the baby Bribery and detainment in facilities	“The patient didn’t know about the post-delivery care. She humbly asked the midwife to be examined and the midwife just checked her blood pressure. She also asked her child’s about vaccination. No information was given regarding breastfeeding and the methods of lactation. The danger signs were not checked and no information were given on vaccination. The doctor and midwife were in the room while the doctor was busy with the registration.” (A31_E4) “… when the doctor was asking the midwife to evaluate the pattern of breastfeeding by the mother, she responded angrily that this is not her duty to make the people understand everything. They left the room and soon after that, the patient and her relative left the center. Even their vital signs were not checked and recorded.” (A64_E4)

* Second and third order themes presented in Bohren’s typology of mistreatment; ** First order themes presented in Bohren’s typology of mistreatment.

Bribery is common under the name of ‘giving a sweet’ [‘shrini’ is a local practice of informal gifts of money, sweets or similar], even the patients can’t get out of the center without paying a certain amount of money under the name of sweet. I even witnessed a case where the patient was physically harmed in order to receive some money. The attitude was terrible …. (A64_E5)

Quality of care was not good, despite having supplies they were not used. The patient’s birth took a long time so the midwife became angry and shouted at the patient. After birth the newborn was dried with the clothing brought by the mother. They wouldn’t use chlorhexidine. The birth was carried out with a single pair of very dirty and old gloves. The patient was not kept in the hospital and the neonate wasn’t breastfed. (A36_D3)

They behaved respectfully with the patient; however, she asked for an examination multiple times but due to the midwife’s neglect she gave birth in the washroom and was lying on the cold floor in a very bad condition. Her vital signs were not checked; the patient was scorned by the midwife for delivering in the washroom. (A38_D3)

Healthcare provider experiences and working environment

Analysis of provider interviews and data collector observations show that maternity care providers in Afghanistan are aware of women’s rights to quality care but face many structural, socio-political and contextual challenges in providing respectful care (Table 4). Some providers shared health system barriers that affect their motivation and performance, many linked to the enabling environment and facility readiness including infrastructure, human resources, drugs, supplies and equipment. While a number of providers denied the presence of any environmental barriers to provision of quality care, others talked openly about little attention being paid to them at work, inadequate staff, high client load/workload, long shifts, lack of transportation and child care support, persistent shortage of supplies and equipment, and inadequate infrastructure such as no water for handwashing.

Table 4. Working conditions and constraints documented in provider interviews and observer field notes.

Constraint type	Specific constraint	Illustrative examples
Structural (facility readiness, professional barriers, etc)	Suboptimal facility infrastructure including small building and rooms; delivery rooms not equipped to ensure mothers’ privacy; washrooms far from the PNC rooms; limited number of beds; broken or worm-infested delivery beds; no water for the delivery rooms, toilets or the health facility as a whole; and absence of waiting area for relatives Shortage of medications, supplies and equipment; and old or broken equipment Staffing shortages resulted in one skilled birth attendant responsible for 5-8 deliveries at the same time or running between the delivery room, labor room and ANC room to provide care in all three locations Shortage of female health providers and/or specialist doctors in daytime and sometimes no female health worker in night shifts High number of clients/patients resulted in crowded health facilities Heavy workload Long shifts Restricted authority limitation in practice of midwives Skilled birth attendant absenteeism and leaving work early to attend private clinics Limited appreciation/recognition of skilled birth attendants Low salary Bad food Limited to no vacation Limited training opportunities No transportation and child care	“The delivery room is near to the pharmacy. The men can hear the voice and screaming of the women. Lack of a large and good building is the main problem here.” (A12_B5) “The midwives of this center complain that there is not enough personnel and the workload is too much. After night shift we must stay here till after noon. Hard working for 24 h makes us very tired and exhausted. Therefore, how it is possible we treat patients respectfully and kindly? There is not anyone who appreciates our hardworking. Even they do not let us to go for vacation once in a year. “(A8_B1) “Because, the number of patients and their relatives are too much and there is low number of midwives. One midwife is not able to service respectfully to all admitted patients. We use one set of gloves for several patients "we don’t have the time to change our shoes" (an Afghan expression), if we take off our glove to change, the next baby will fall down in the ground or the cervix would lacerate.” (A23_B1) “… the ob/gyn doctor’s private clinic was located near the hospital and most of the patients were referred for ultrasound exam there …. She would leave the hospital after 12′o clock and would give authority to the midwife to write prescriptions but the pharmacy wouldn’t accept prescriptions without the doctor’s signature.” (A73_C2) “The midwife said the doctor do not let us prescribe the patients Folic Acid or urine exam; the doctors say, the MOPH has told us that the midwives are only allowed to check patient’s BP and fill cards.” (A73_C3) “…. There is no fair salary and food. There is no encouragement for our hard work. Even we are insulted.” (A40_B6) “The head of the hospital does not count on midwife. He sends for workshops whomever he likes. I have been working here for four years but he has never sent me for workshop. Another midwife went to a workshop on the second day of her recruitment….” (A20_B1)
Social/political (social hierarchies, family ties, normative practices)	Disrespect of women/female health providers/female coworkers Very high expectations from female health providers Threatened with discontinued employment if overworked Skilled birth attendants discuss being tired/ exhausted Head of hospital disrespect skilled birth attendants and turn them against each other Disrespect of a newcomer Disrespect of trainer/doctor/midwife/nurse to midwife, doctor to nurse and doctor to doctor in front of mothers/relatives Clients/client families argue with, blame, beat, and threaten health providers verbally /by guns; or limit provision of healthcare services to other clients. Favoritism Blame of skilled birth attendants by other healthcare providers /mother’s relatives because of deteriorated health of a mother	“The doctor tells the midwives how dare you do episiotomy and how dare you to use vacuum. The one who has higher position, insults the others in front of the patients. Despite our good behavior and respect, they don’t allow the midwives to conduct examinations. The work environment doesn’t let us to be independent. Yes, when you don’t have the authority, your attitude will be bad with the patients.” (A31_B1) “The trainers disrespect the midwife. The internal medicine doctor insults the nurses. There is verbal abuse. They send the nurses either in obstetrics and operation theater. When they express their problems in the meetings. The doctor doesn’t value and make fun of our requests and tells, ‘Do I buy an airplane for you? If you cannot tolerate, you can resign’. There is too much discrimination.” (A21_B4)
Contextual (security, health system, other)	Insecurity resulted in SBAs feeling in danger at home, on the way to health facility and during service provision; feeling physically ill and less energetic to talk with clients; and fear of not returning home safely. Health system constraints and lack of response from health facility authorities and Ministry of Public Health to complaints made related to infrastructure, human resources, drugs, supplies and equipment.	“The environment is not secure. Even they threatened me against coming to work. But I am coming. Because I need money and also I think about mothers. Most of the people don’t count a woman as a human. They don’t let them go out. There are kidnapping cases of children. The women are scared their children will be kidnapped if they go out.” (A24_B4) “The problems are security of unreasonable people, the shortage of equipment. The patient does not come to hospital and her family requests us to go their home for management of the labor. Due to insecurity, we deny their request. They say,” you are staff of the government and have salary, I know what to do with you.” (A79_B3) “The workload is too much. They cannot afford to serve all patients. They requested to increase the number of night shift midwives to two persons. However, the head of the hospital did not accept.” (A23_B4) “.… All of the instruments and materials are old and even delivery tables are full of worms…. We complained 10 times to the representatives of the ministry. They just noted and forgot our problems. We do not have sets in the operation room. We use minimal supplies. The anesthetic machine needs repair, we use ambu bag in operations. The lamps are damaged and we repair every day. We do not have any incubator and facilities to care the neonates. I gently request you to deliver our voice to the ministry and related administrations…. we have so many problems, so what does respectful maternity mean for us?” (A31_B3)

The ‘carers’ were stressed by other factors including security issues yet demonstrate a level of resilience despite the conditions. When asked if, and to what extent, working conditions allow for provision of respectful maternity care, some providers shared examples of misogyny, abuse, and daily exposure to violence.

I am an experienced midwife. One time, a woman died in our ward. Her family abused me physically to the extent that I had bruises all over my body. I complained to the head of the hospital. He said: “go away, it’s normal to be slapped by people”. (A19_B8)

When the patient’s condition becomes worse, they blame the midwife and talk offensively. Yes, male and female health providers are threatened by guns. The people have guns here, and they speak with guns not with tongues. (A78_B2)

Insecurity resulted in providers feeling in danger at home, when traveling to the health facility and during service provision. Some also described feeling physically ill, with low energy to care for clients, and fear of not returning home safely. Participants mentioned health facilities being closed by insurgents, missiles hitting the health facility, client’s relatives (Taliban or civilians) demanding care for specific clients and limiting care to others, being threatened by weapons, and working in conflict for several years. This resulted in providers feeling chronically stressed and unsupported.

Discussion

Mistreatment during childbirth is complex, sensitive and prevalent in many countries. In Afghanistan, challenges in understanding and addressing this complex problem are exacerbated by the backdrop of more than four decades of conflict and insecurity as well as a conservative cultural context. A study exploring the culture of care in one large Kabul hospital noted that staff had the capacity for kindness and altruism but also for neglect and cruelty. Sharing findings of a study conducted between 2010 and 2012, Arnold et al. highlight the complexity of factors leading to suboptimal care and mistreatment including weak working environments, poor human resource management (maldistribution of staff is a persistent bottleneck), lack of accountability across all levels and a dysfunctional institutional culture (Arnold et al., 2019).

Our interviews with more than 500 skilled birth attendants and observations of antenatal, intrapartum and immediate postnatal care experiences of more than 1,000 women at 77 hospitals across the country show that these challenges are persistent, and extend nationwide. Across Afghanistan, women lack voice and agency both as users and providers. Low levels of health awareness, education and decision-making power mean many women are unaware of what to expect or accept as quality dignified care. Only 48% of married women (aged 15-49) reported participating in decisions regarding their healthcare (Central Statistics Organization et al., 2017).

Mistreatment and poor quality maternity care are normalized, with constraints of a fragile, conflict-affected health system affecting both women and their care providers; Afghan women – including both healthcare providers and their clients, are exposed to multiple forms of political violence as well as harmful gender and social norms involving disrespect and abuse of women and violence as a form of dispute resolution (Corboz et al., 2019; Jewkes et al., 2019). Frequent exposure to traumatic experiences, which include violence, disruption of family structure, and social disintegration contribute to mental health issues in conflict-affected populations (Bendavid et al., 2021). Healthcare delivery in protracted conflicts can lead to ‘conflict blindness’ (Arnold et al., 2018) and substandard maternity care is normalized and even accepted in war. The daily struggle to survive and for many female providers, support their families whilst working, further limits providers capacity to provide empathic care to others (Thommesen et al., 2020).

Transforming health systems to provide respectful, dignified, quality care requires multi-faceted interventions and commitments to social change. Health systems and facilities must be structured and resourced in a way that enables quality maternity care provision, with respect for maternity care providers and adequate infrastructure and organization of the maternity ward (Bohren et al., 2020). An enabling environment especially for female healthcare providers requires complex and interacting elements including effective inter-professional teamwork, strong professional identity (especially for midwives), safety and security, sufficient resources, gender equity, and government and community support (4th Global Midwifery Symposium, 2019). In conflict-affected settings such as Afghanistan, where mistreatment triggers are common and women and their providers are especially vulnerable, greater attention to using rights based frameworks which ensure RMC is framed as a gender and equity issue is a priority (WHO, 2015). Within a health facility, management staff may support quality of care by setting up quality improvement teams, guided by client experiences and community interests, training and/or mentoring providers in interpersonal communication and attitude transformation, psychosocial support and mediation for carers, and monitoring provider adherence to best practices and women’s experiences of care (Bohren et al., 2020; Sacks & Peca, 2020).

Some of these efforts are already underway in Afghanistan, though continued investment and greater accountability are needed. For example, at the hospital level, accountability and responsiveness can be enhanced through a mechanism that ensures the hospital is accountable for their performance vis a vis standards, as part of a national quality assurance/accreditation process. This could include engaging more closely with the community they serve, for example, providing timely, transparent investigations and responses to clients’ complaints. The MoPH recognize that changing providers’ normalized disrespectful attitudes requires multifaceted and multi-layered approaches which engage all stakeholders (community, healthcare providers, MOPH, and highest level government leadership) some of which are detailed in the country’s key strategic documents (Ministry of Public Health, 2015, 2017). However, they are well aware of the challenges including conflict, political instability, corruption, nepotism, and gender inequity that affect health workers and clients alike (Ministry of Public Health, 2016). Addressing gender inequality and its adverse outcomes predominantly on women’s health should underpin all these efforts. Notably, to build trust and more equitable healthcare the Government of Afghanistan needs to stress that public health and healthcare services are not ‘political’ and anti-government organizations, such as the Taliban must ensure the safety and security of healthcare services providers.

Implications for future studies of quality of care

So what did we learn about integrating RMC measures in observational facility assessments? Defining mistreatment is complex but can be considered as interactions or facility conditions that local consensus deems to be humiliating or undignified, and those interactions or conditions that are experienced as, or intended to be humiliating or undignified (Freedman et al., 2014). Mistreatment is normalized to the extent that many aspects of sub-optimal care were not consistently identified as examples of mistreatment during observation of care. Although there were challenges in standardizing documentation of observations of the ‘softer areas’ of care such as interpersonal communication, we gained valuable insights from qualitative data collectors notes. Data collectors noted their own responses to observing mistreatment (e.g. women birthing alone) ranging from anger and frustration to sadness and understanding. However, they also expressed empathy for the providers noting health system constraints. Participating in the study appeared to positively change the observers’ perspectives and personal values on mistreatment and as they were also part of the system, may lead to changes in their individual attitudes as providers and potential advocates for systemic change.

In reflecting on our experiences conducting this assessment, we identified considerations for conducting similar studies in conflict-affected/fragile settings. Preparing observers to note all aspects of care giving, the good and the bad, including mistreatment, needs special attention so that they can ‘be a fly on the wall’ but also be cognizant of their own values and possible biases. This can be achieved, for example, by using role plays of typical behaviors and communication practices during observer training followed by reflection and debriefing; observing and critiquing videos; reviewing accounts of care to gain consensus and a shared understanding of the issues arising. As study tools are often translated into multiple local languages, it is important to ensure questions are phrased to ensure words and expressions that ‘makes sense’ to the respondents will help embed experiences, especially of suffering and pain within larger narratives of violence, conflict and social and economic deprivation (Ventevogel & Faiz, 2018). Most importantly, it is important to ensure participant’s perceptions and experiences of birth capture socio-contextual issues and are embedded in the reality of the setting (Christou et al., 2019).

Providers demonstrated various levels of resilience and the more nuanced aspects of how persistent insecurity and stress impacts their daily lives can easily be overlooked in analyses, as these are also normalized as part of everyday reality in conflict-affected settings. Consideration of mental health issues among providers and studies exploring how strong gender norms/inequalities impact caring behaviors could deepen understanding of obstacles to quality improvement. More evidence is needed on how to offer more empathic and person-centred care and the importance of supporting the healthcare providers themselves as part of building more resilient health systems, improving quality of care and reducing burnout. The phenomenon of normalization of mistreatment (with women and providers) in conflict-affected and fragile settings merits a ‘deeper dive’ so that strategies to address mistreatment target the root causes.

The strengths of this study lie in its design as a national assessment, representative of all medium-high volume public health facilities across Afghanistan (having an average of at least five birth per day) and being the first large-scale study in Afghanistan to include direct observation of childbirth. The qualitative accounts from observers field notes encouraged reflection on normalized practices and captured more nuanced aspects of care, expanding on, clarifying and in some cases correcting what was reported in structured observation checklists. However, like all studies, our analyses are not without limitations. Provider interview and case observation data was not matched at a facility-level, thus we could not analyze the relationship between provider characteristics and specific care practices or outcomes. Second, we observed client experiences but did not capture client perspectives. Although we provide insights into women’s experiences of care during pregnancy and childbirth, we recognize that our findings would be enhanced by direct interviews with clients.

In the future, development and context-specific validation of measurement methods for use in routine audit/feedback or quality improvement initiatives would support improved tracking and accountability for RMC. Bringing together insights from multiple disciplines – including political science, behavioral science (Smith et al., 2020), and gender studies (Betron et al., 2018; Maung et al., 2020) - can lead to a more nuanced understanding of drivers of poor quality care and lead to different, complementary solutions which can help transform the experience of childbirth for both providers and clients in conflict-affected settings like Afghanistan.

Conclusions

Most women giving birth in Afghanistan will have lived their whole life through war, endemic violence and insecurity as well as experiencing persistent gender inequalities especially related to reproductive health rights. Despite high level commitments to advancing respectful maternity care, healthcare providers work under stressed conditions with insufficient support, and many forms of mistreatment are normalized as a consequence of years of conflict. Most women receive mixed quality care, and some endure severe mistreatment and abuse at health facilities. The 2016 National Maternal and Newborn Health Quality of Care Assessment reinforced lessons from other settings on the value of using multiple data collection and analysis methods to examine issues related to quality of care, and provided insights on the importance of considering the potential impact of conflict and context on measurement. Improving quality of care to cultivate a caring health system needs systemic change which addresses persistent violence and insecurity, unequal gender related barriers and the power dynamics which underpin mistreatment.

Acknowledgements

The authors would like to thank UNICEF and USAID for funding the study and technical collaboration, the central and provincial MoPH for their guidance and support, the nongovernmental organizations implementing Afghanistan’s healthcare services, and those professionals who participated in the study. Particular appreciation is owed to Ariel Higgins-Steele, Sharmina Sultana and Shirin Varkey from UNICEF; Sher Shah Amin from USAID; Sayed Attaullah Saeedzai, MoPH General Director of Evaluation and Health Information Systems; Catherine Todd, Reproductive, Maternal, Newborn Child Health Division, and Patsy Bailey, Senior Scientist, Health Services Division, FHI360; Nasrat Ansari, Faridullah Atiqzai, Partamin Manali, Zahra Sultani, Sayed Esmati, Ahmad Eklil Hussain, Mahmood Azimi, Enayatullah Mayar, Abdul Qader Rahimi, Raouf Saidzadah, and Matiuallah Noorzad, Jhpiego Afghanistan; facility staff participating in the study; and women allowing observation of care. Thanks to all the midwives and doctors who served as data collectors in very difficult situations.

Disclosure statement

No potential conflict of interest was reported by the authors.

Data availability statement

Quantitative data are available from the MoPH upon request. Requests should be directed to the MoPH’s Evaluation and Health Information Systems Department ([email protected]). Qualitative data are available from the corresponding author upon reasonable request and signature of a data sharing agreement.