Delays in care for children with low anorectal malformations in Southwestern Uganda

Abstract

Introduction:

Disparities in anorectal malformation (ARM) outcomes between high- and low-income countries may be due to delayed diagnosis in the latter setting. The Three Delays model, comprising delays in seeking, accessing, and receiving care, provides a framework for exploring these challenges. We sought to examine the frequency and nature of the preoperative delays in children presenting for surgical correction of low ARMs.

Methods:

We conducted a cross-sectional study examining the delays in care among children with low ARMs in Southwestern Uganda between June 2021 and July 2023. Delayed diagnosis was defined as a diagnosis made > 48 hours of life. Potential associated factors such as caregiver, community, and aspect of healthcare system factors were examined. Statistical significance was set at p<0.05.

Results:

A total of 80 patients were included in the study. The median age at diagnosis was 29.2 days. In 82% of patients, the parents diagnosed the abnormality and 74% experienced delayed diagnosis. Among delays in seeking care, 23% of caregivers reported no knowledge of their child’s disease. For delays in reaching care, 37% encountered financial problems, and 28% lacked an appropriate diagnosis from a health center, contributing to delays in receiving care.

Conclusion:

Delays in care are frequent for children with low ARMs. Finances and caregivers’ and healthcare workers’ knowledge contribute significantly to these delays. To mitigate these delays, we recommend improving referral processes, prioritizing newborn screening examinations, advocating for a national child health insurance policy, and enhancing the training of primary healthcare providers.

Introduction

Anorectal malformation (ARM) is a prevalent visible congenital anomaly in low-income countries (LICs), accounting for 26% of all gastrointestinal congenital anomalies seen in this setting. Disparities exist in the morbidity and mortality of ARM between high-income countries (HIC) and LICs, which may be partially attributed to the many challenges that arise from the delayed presentation of these patients [1].

Delayed diagnosis of ARM, defined as a diagnosis made more than 24–48 hours of life [2–7], is a common challenge experienced by children with ARM in the Ugandan setting. The key to ARM recognition is early perineal examination, specifically – inspection and palpation of the perineum in the neonate soon after birth. However, this exam is often not completed, hence the diagnosis is missed [8]. An earlier study in our center showed that 63% of ARMs presented beyond 24 hours [9]. Even in HIC settings, 9–25% of ARM patients were diagnosed following discharge from the birth facility [10–13]. An exploration of the reasons for these delays is critical to identifying interventions able to improve health outcomes in LICs.

The Three Delays (3D_s) model (D1 – delay in seeking care, D2 – delay in accessing or reaching care, and D3 – delay in receiving care) provides a framework for understanding the nature of delays faced by children with ARMs in Uganda [14–19]. Unique to the ARM population is the presence of a visible congenital anomaly. If this anomaly is found on the newborn exam while the mother and baby are still in the hospital, it could obviate the first two delays (D1 and D2), as patients would be in a setting where care or hospital-to-hospital referral could be provided. Despite recent improvements in perinatal services in Uganda, there are concerns about the quality of this care [20]. A limited emphasis on newborn examination, especially of the perineum, typically results in parents/caregivers returning home undiagnosed, and therefore being the first to notice the perineal anomaly [21].

No studies have explored delays in the care of children with low ARMs utilizing the three-delays framework to identify intervenable areas for care improvement. We, therefore, sought to examine the frequency and nature of delayed diagnosis among children presenting for surgical correction of low ARM in Southwestern Uganda.

Methods

Study Design & Setting

We conducted a mixed-methods cross-sectional study to examine the association between delayed low ARM diagnosis, defined as age > 48 hours at diagnosis, and parental (caregivers), community, and healthcare factors. Parents/caregivers were interviewed to solicit all care-related data, as well as their perceived reasons for any delays in care using structured interviews. Convenience sampling was used to recruit participants between June 2021 and July 2023. All parents of patients who presented to a regional referral hospital with a new diagnosis of ARM and a clinical exam confirming perineal fistula, vestibular fistula, or anal stenosis types of ARM were included in the study. We excluded children with cloaca, ARM returning for subsequent procedures, and those whose caregivers could not recall the age at diagnosis. The study was conducted in the Pediatric Surgical Unit (PSU) of a regional referral hospital in Southwestern Uganda. Ethical approval was obtained from both the local university ethics review committee (MUST-2021–64) and the Uganda National Council of Science and Technology (UNCST) ethics board (#HS1632ES). Informed consent was obtained from all the participants

Variables & Data Sources

Data was collected directly from study participants at their first post-operative visit. The following information was solicited: child’s sex, ARM type (anal stenosis, perineal, or vestibular), age at diagnosis, diagnosing individual, location of diagnosis, reasons for late presentation to the hospital for care, age at first surgery, and type of ARM repair. Age of ARM diagnosis was defined as the child’s age when the parents or a healthcare worker identified an anatomical anomaly in the child. Demographic information was also collected on the family, including distance traveled for care, rurality, and caregiver education level. Patients were also contacted via telephone after this visit to obtain perinatal care details, including attendance of antenatal care, receipt of antenatal ultrasound examinations, delivery location, and receipt of postnatal maternal/child care. The data collected was coded with unique identifiers into a REDcap database [22].

Statistical methods:

STATA 17.0 was used for all statistical analysis of quantitative data. Descriptive statistics were used to assess the frequency by which individual-level, community-level, and healthcare-level factors were associated with the delayed diagnosis of ARM. All categorical variables were assessed using frequencies and proportions, and comparisons were made with chi-square and Fisher’s exact tests, as appropriate. Continuous variables were assessed with Mann-Whitney U tests due to a lack of normal distributions. P-values less than 0.05 were considered statistically significant.

All expressed reasons for delays in care were qualitatively analyzed for common themes, using an inductive approach. Themes were then divided according to the associated delay in the three-delays model. Themes were aggregated and compared to whether or not there was a delay in diagnosis.

Results:

A total of 80 patients with low ARM were included in the study. The median age at diagnosis was 29.2 days (Q1-Q3, 2 – 121.7). Only 21 patients (26%) were diagnosed within 48 hours of birth, while the remaining 59 (74%) were classified as delayed diagnosis. In the no-delay (ND) group, the median age at diagnosis was 1 day (Q1-Q3, 1–2), compared to 60.8 days (Q1-Q3, 14–152) in the delayed diagnosis group. Patients traveled a median of 118 km (Q1-Q3, 42 – 195) for definitive care (i.e. to reach the study site).

There was no difference in the educational attainment of the primary caregiver, level of rurality, or distance from the family home to the hospital between those with and without a delay in diagnosis. In the vast majority of cases (80%), the parents themselves diagnosed the ARM in their children regardless of the group they were in (71% ND vs. 86% delayed group, p=0.18). Birth personnel made the diagnosis of ARM less frequently (5%) resulting in no delays to care (19% ND diagnosis by birth personnel vs. 0% delayed, p=0.004). Differences were also noted by patient sex, with a higher proportion of males than females diagnosed within 48 hours (62% vs 31%, p=0.011) (Table 1).

Table 1.

Demographic differences between those patients who experienced delayed diagnosis versus those who did not

	Total N=80	No Delayed Diagnosis N=21	Delayed Diagnosis N=59	p-value
Sex, n (%)				0.011
Male	31 (39%)	13 (62%)	18 (31%)
Female	49 (61%)	8 (38%)	41 (70%)
Distance Traveled (km), M (IQR)	80	175 (97–295)	118 (41–195)	0.40
Rurality, n (%)				0.23
Urban	23 (29%)	8 (38%)	15 (25%)
Rural	55 (69%)	12 (57%)	43 (73%)
Missing	2 (2.5%)	1 (4.8%)	1 (1.7%)
Level of Caregiver Education, n (%)				0.49
None	10 (13%)	2 (9.5%)	8 (14%)
Primary	45 (56%)	11 (52%)	34 (58%)
Secondary	11 (14%)	2 (9.5%)	9 (15%)
Tertiary	14 (18%)	6 (29%)	8 (14%)
Person who Made Diagnosis, n (%)
Parents	66 (83%)	15 (71%)	51 (86%)	0.18
Birth personnel	4 (5.0%)	4 (19%)	0 (0.0%)	0.004
Doctor	7 (8.8%)	2 (9.5%)	5 (8.5%)	1.00
Other	3 (3.8%)	0 (0.0%)	3 (5.1%)	0.56
Place Diagnosis Made, n (%)				0.004
Home	62 (78%)	11 (52%)	51 (86%)
Healthcare	18 (22%)	10 (48%)	8 (14%)
Type of Malformation, n (%)				0.005
Perineal Fistula	36 (45%)	15 (71%)	21 (36%)
Vestibular Fistula	44 (55%)	6 (29%)	38 (64%)
Type of Repair, n (%)				0.044
1-Stage	59 (74%)	12 (57%)	47 (80%)
2-Stage	21 (26%)	9 (43%)	12 (20%)

Forty patients were successfully contacted after discharge regarding their experience with perinatal care — 67% (n=14) of the ND group and 44% (n=26) of the delayed group. In this entire group, 95% reported attending antenatal care, 65% had prenatal ultrasound examinations, 90% delivered in a healthcare facility, and 93% attended mother-child postnatal visits. Despite this high attendance to perinatal care, all stated that no abnormalities were noted during the postnatal care visits. There were no significant differences in the attendance to perinatal care between the two groups (Table 2).

Table 2.

Maternal utilization and experience of perinatal care

	Total N=40	No Delay in Diagnosis N=14	Delayed Diagnosis N=26
Attendance of Antenatal care
No	2 (5.0%)	0 (0%)	2 (7.7%)
Yes	38 (95%)	14 (100%)	24 (92%)
Receipt of Antenatal Ultrasound
No	26 (65%)	8 (57%)	18 (69%)
Yes	14 (35%)	6 (43%)	8 (31%)
Delivery Location
Home	4 (10%)	1 (7.1%)	3 (11%)
Hospital/Health Center	36 (90%)	13 (93%)	23 (89%)
Receipt of Postnatal Care
No	3 (7.5%)	1 (7.1%)	2 (7.7%)
Yes	37 (93%)	13 (93%)	24 (92%)

A broad distribution was noted for the cohort in the age at surgery and the time between diagnosis and surgery. Age at surgery was higher for children with a delayed diagnosis (median age 42 days (Q1-Q3 7.0 – 152) vs 137 days (Q1-Q3 91.3 – 243.3), p=0.002). However, no difference was found in the time from diagnosis to surgical intervention (40 days (Q1-Q3 6.0 – 150.1) vs 37 days (Q1-Q3 10.5 – 113.2), p=0.95) as demonstrated by the distribution of patients according to (A) the age at the time of initial surgery (posterior sagittal anorectoplasty with(out) colostomy) and (B) the duration of time between the diagnosis and receipt of surgical care (Figure 1).

Figure 1.

Distribution of patients according to (A) the age at the time of initial surgery (posterior sagittal anorectoplasty with(out) colostomy) and (B) the duration of time between the diagnosis and receipt of surgical care

When asked about their perceptions of why delays in care happened, common themes were identified for every delay type. Significant differences were found in the proportion of patients who experienced delays seeking care between the ND group and delayed group (p=0.039), but not for reaching care (p=0.75) or receiving care (p=0.54) (Figure 2).

Figure 2.

Significant differences in the proportion of patients who experienced delays in seeking care between the ND group and delayed group (p=0.039), but not for reaching care (p=0.75) or receiving care (p=0.54) * p<0.05.

Perceived reasons for delays in surgical care according to whether or not the patient experienced a delay in diagnosis. In the ND group, 2 children in the no delays group experienced delays both in seeking and reaching care. In the delayed group, 16 patients experienced more than one type of delay. Of those who experienced a delay in seeking care, 6 experienced a delay in reaching care, and 3 experienced a delay in receiving care. Of those who experienced a delay in reaching care (but not seeking), 7 experienced a delay in receiving care. 2 patients experienced overlapping challenges within a delay type.

When the delay types were examined, for delays in seeking care, 23% of caregivers reported no knowledge of their child’s disease, 2.5% attempted traditional medical treatments, and 2.5% stated they were unaware that treatment options existed. For those who experienced delays in reaching care, 37% stated that they encountered financial problems, 18% lacked knowledge of where to go for treatment, and 3.8% had transportation challenges. Regarding delays in receiving care, 28% lacked a referral or a formal diagnosis, including 11% who sought care at other healthcare facilities without a referral. Other reasons for delays in receiving care included delayed surgical appointments (6.3%) and COVID restrictions (2.5%). Additionally, one patient required surgical intervention for a cardiac anomaly, while another was reluctant to undergo surgery given the potential need for a colostomy (Table 3).

Table 3.

Parental perceived reason for any delay in receipt of care for the child’s anorectal malformation

	Total N=79	No Delay in Diagnosis N=20	Delayed Diagnosis N=59
Delay in Seeking Care
Lack of knowledge of disease	18 (23%)	2 (10%)	16 (27%)
Attempt at traditional treatment/non-treatment	2 (2.5%)	0 (0%)	2 (3.4%)
Unaware treatment existed	2 (2.5%)	0 (0%)	2 (3.4%)
Delay in Reaching Care
Unaware of where treatment could be obtained	14 (18%)	5 (25%)	9 (15%)
Financial Problems	29 (37%)	6 (30%)	23 (39%)
Transportation challenges	3 (3.8%)	0 (0%)	3 (5.1%)
Delay in Receiving Care
Lack of referral/diagnosis	22 (28%)	4 (20%)	18 (31%)
Delayed appointment	5 (6.3%)	3 (15%)	2 (3.4%)
COVID Lockdown	2 (2.5%)	1 (5%)	1 (1.7%)
Other	2 (2.5%)	1 (5%)	1 (1.7%)

Discussion

In our study, 74% of children with low ARMs experienced delayed diagnosis, a higher incidence than that previously reported for all ARM types in our LIC setting (63%) [9], and significantly higher than the 7 – 25% rate reported in HICs [10,11,23]. This increase is likely due to the non-obstructed nature of the malformation, with most babies passing stool spontaneously[12]. All of the 3 delays (seeking care, reaching care, and receiving definitive care) limited access to care for our patients.

Regarding the delay in seeking care, despite high attendance for perinatal care, 93% of those surveyed reported adherence, and no anomalies were diagnosed. ARM diagnosis is clinical, identified during the newborn exam [8,24]. Studies reveal a gap in immediate postnatal care, stressing the need for integrated maternal and newborn services, awareness, and advocacy [20]. A related study showed that half of the Infants and children with late diagnosis of ARM were detected by paediatricians [11]. Parents can identify ARM when healthcare workers miss the diagnosis. Though antenatal care and facility-based deliveries have increased, the quality of care is insufficient [25–28]. A lack of healthcare workers hinders appropriate care provision; however, the physical exam is an essential tool for identifying ARM, and it takes only a few minutes to complete. Greater resources are needed to ensure emphasis on newborn examinations and possible visible anomalies that may be identified during perinatal care visits to ensure timely intervention [22–25].

This study found a significant disparity in delayed diagnosis between male and female neonates with anorectal malformations. The delayed group had a higher proportion of vestibular fistulas (64%) than the ND group (29%). This may be due to anatomic differences between the males and females. An abnormal anal opening or misplaced anus in boys could be more easily recognized since only a single opening is expected. Related studies have found slightly high female predominance at delayed diagnosis of low-ARM. Therefore, regular examination of the female perineum for all patients is necessary to address this challenge [6,10,29].

Delays in reaching care are a significant challenge for many in Uganda. In this study, 37% reported financial problems, 18% lacked knowledge of where to go for treatment, and 3.8% had transportation challenges. These delays can be compounded by societal factors like poverty, transportation logistics, social stigma, and lack of education [19,30]. Improvement of the existing referral system and greater education of front-line providers could help address many of these delays. Despite the growth of ambulance services in Uganda, patients still have to pay for fuel during referrals or transfers to tertiary centers, mostly located in urban areas, which can be expensive. Investing in capacity building could help reduce the financial burdens that occur when seeking care from multiple centers before reaching the right place, one of only 2 public hospitals in Uganda that can provide definitive care [4,14,15,31–35]. Our team is working with collaborators on these initiatives to improve access to care

Regarding the delays in receiving care, 28% of patients faced a lack of referral or formal diagnosis, with 11% seeking care at other healthcare facilities without a referral. In Uganda, mobile electronic applications have been deployed to support congenital anomaly surveillance systems and diagnose visible anomalies [36,37] However, none of these efforts tie diagnostic information with treatment options or referral sources. The lack of incorporation of surgical services in child health initiatives remains a concern. To address this, the Pediatric Emergency Surgery Course (PESC) has been providing regional hospital training, resulting in earlier neonatal referral rates [38,39]. More funding support is needed for this training program to continue building local capacity. A lack of surgical infrastructure and trained healthcare workers worsens the situation. More nurses with specialty skills in pediatric surgery and anesthesia providers capable of handling children are needed [4,15].

Our study has several limitations. First, the evaluation of perinatal care experiences was limited by missingness in this data due to the inability to contact families for feedback on their perinatal care experiences. Although the missing data did not appear to bias the comparison between the delay and non-delay groups, it may have affected the overall understanding of perinatal care quality. The uncontacted individuals might represent a demographic with limited resources, potentially skewing the representation of missed antenatal visits. Second, parental recall bias could have influenced the accuracy of reported care processes, particularly for diagnoses made significantly earlier. Memorable events, such as hospital presentations without referrals, might disproportionately influence reported delay types. Third, Sampling bias and small sample size, it is highly possible that those families that were not surveyed had different responses. Finally, the absence of documented presentation dates at Mbarara restricted our ability to assess the interval between diagnosis and care initiation, which limited our analysis. Nonetheless, the observed duration from parental recognition of an anomaly to surgical intervention remains significant.

Conclusion

Delays in care are frequent problems for children with low ARMs. Finances and caregivers’ and healthcare workers’ knowledge contribute significantly to these delays. To mitigate these delays, we recommend improving referral processes, prioritizing newborn screening examinations, advocating for a national child health insurance policy, and enhancing the training of primary healthcare providers.

Highlights -.

What is currently known about this topic?

Delayed diagnosis and treatment is a common challenge for all types of anorectal malformations (ARM) in low-income countries (LICs).

What new information is contained in this article?

We described these delays for low-ARM to provide insights of appropriate perineal examinations of newborns by birth personnel that may help address delays.