Prevalence of discharge against medical advice and its associated demographic predictors among pediatric patients: A cross-sectional study of Saudi Arabia.

BACKGROUND: Discharge against medical advice (DAMA) occurs when the patient or their caretaker leaves the hospital against the recommendation of their treating physician. DAMA may expose the children to a high risk of inadequate treatment, which may result in readmission, prolonged morbidity, and mortality. The study aimed to identify the predictors of DAMA in the emergency department (ED) within the pediatric age group.
METHODS: This was a cross-sectional study. The study used the medical records of pediatric patients (n = 5609) that were admitted to the ED of King Abdullah Bin Abdulaziz University Hospital (KAAUH) in Riyadh, Saudi Arabia, during 2017 and 2018. Descriptive statistics, Chi-square, or Fisher's exact test were used. Unadjusted and adjusted odds ratios with their 95% CI were reported by performing logistic regression modeling.
RESULTS: A significant interaction between age and gender was observed in the multivariate analysis after adjusting for the other covariates. The odds of DAMA for a 5-year-old female child were 4.43 times higher than those of a 5-year-old male child (P < 0.1).
CONCLUSIONS: The public should be educated about the consequences of DAMA. Continued health education and the promotion of child survival strategies at the community level, combined with an improvement in the socioeconomic conditions of the population, may further reduce DAMA and improve the chances of survival for children. Future studies should assess the socioeconomic status of the patients and estimate the cost that is incurred by the patients. Copyright:

INTRODUCTION

Discharge against medical advice (DAMA) occurs when the patient or their caretaker leaves the hospital against the recommendation of their treating physician.[1]However, children are the most vulnerable population in such situations[2]because they are dependent on their parents or caregivers for decisions regarding health care, as they are usually not considered to have the emotional or cognitive maturity to make such decisions for themselves.[3]DAMA may put the children at a high risk of inadequate treatment, which may result in readmission, additional or prolonged morbidity, and mortality.[456]

The prevalence of DAMA varies according to the geographical area. The reported prevalence of DAMA among children from different parts of the world ranges from 1.2% to 31.7%;[24789]however, data from Middle Eastern nations such as the Kingdom of Saudi Arabia (KSA) are lacking.

There are several demographic and socioeconomic factors that may influence pediatric DAMA rates, including lower socioeconomic class,[810]younger age,[7811]and a lack of medical insurance.[1]There are several other parent-related and hospital-related reasons for DAMA that include the parents' perception of the child as being well; a preference for outpatient care; financial constraints; the high cost of hospital services; dissatisfaction and disagreements with the health-care professionals; the inconvenience of hospitalization (lost days at work, the inability to care for other children, etc.); a preference for traditional forms of treatment; and the hopelessness of the clinical situation.[312]

However, reasons for DAMA differ between developed and developing countries and it has been suggested that the DAMA rates in developing countries are two times higher than in developed countries.[1314]It has been observed that financial reasons are the predominant factors for DAMA in developing countries due to the lack of institutionally organized health insurance.[15]

Therefore, it is imperative that the risk factors of DAMA are understood. This will aid in the proactive management of those at risk of DAMA[16]and in taking preventive measures in advance to reduce DAMA, decreasing the morbidity and mortality among the pediatric population. Some of these preventive measures include the early identification of those at risk of DAMA; improved doctor–patient communication; increased availability of skilled health manpower; the improvement of hospital facilities; and a change in the attitudes of health workers towards patients.[17]

In previously published studies conducted in KSA, the prevalence of DAMA was 1.6% in neonatal intensive care units[18]and was 1% among adults in the emergency department (ED).[19]Although DAMA is of major importance to public health, especially for the pediatric age group, there is a dearth of information regarding the demographic factors that might be responsible for DAMA in this specific context. Therefore, it is essential to explore the risk factors of DAMA within the pediatric age group so that early intervention can be implemented to prevent unwanted consequences, i. e., higher morbidity, mortality, and health-care costs.

In light of the gap in the literature and in order to uncover the information needed to reduce DAMA in King Abdullah Bin Abdulaziz University Hospital (KAAUH), the objectives of this study are to determine the prevalence of DAMA among those of the pediatric age group presenting to the ED of a tertiary care hospital in Riyadh and to assess the demographic factors (such as age and gender) as well as their relationship to the DAMA of pediatric patients.

METHODS

The study had a cross-sectional study design. All pediatric patients age 0–18 years admitted to the ED of KAAUH in Riyadh in the period between 2017 and 2018 were included. The KAAUH offers medical care to all students enrolled at Princess Nourah University, as well as employees of the university and the hospital and their families (approximately 70,000 individuals). The ED is composed of the acute care, triage, and resuscitation units. Patients are triaged according to a 5-level triage system (Canadian Triage and Acuity Scale [CTAS]), ensuring that patients are assessed and sorted according to acuity. All patients who are eligible and meet the scope or service of KAAUH are seen in the ED unless it comes in CTAS 2–5. All patients with CTAS 1, or with life or limb-threatening conditions (trauma, toxicology, burn, etc.), regardless of their eligibility status, only receive the stabilization management in the resuscitation unit of the ED and are then transferred straight to KAAUH for further management. ED staff are specialized emergency health-care providers who have advanced qualifications and experience. The medical records that were used were those of the 5609 pediatric patients admitted to the KAAUH-ED.

As per the regulations and policies of the hospital ED, all information related to patients presented is entered into the Health Information Management System-TrakCare®(InterSystems Corp, Cambridge, Massachusetts, USA). The information entered for each admitted patient includes the following: gender, age at admission, triage category, discharge date and time, bed, and discharge classification. Data from the system were extracted and entered into a Microsoft Office Excel (2010 for Windows [v14.0]) (Microsoft Corporation, Redmond, WA, USA) sheet for data cleaning and basic exploratory analysis before SPSS data entry and analysis.

The study was given an exemption from the Ethical Review Committee at Princess Nourah University along with approval from KAAUH to retrieve the needed data from patients' medical records.

An analysis was performed using IBM® SPSS® Statistics for Windows, version 22 (IBM Corp., Armonk, N. Y., USA). Descriptive statistics were computed for categorical variables (such as age, gender, and triage level) by computing their frequencies and percentages. Their relationship with discharge status was assessed using Chi-square or Fisher's exact test where appropriate. The quantitative variables (such as age) were computed by their median (interquartile range) and their relationships with discharge status were assessed using the Kruskal–Wallis test. Post hoc analysis was also performed. Unadjusted and adjusted odds ratios with their 95% CI were reported by performing logistic regression modeling. All plausible interactions were assessed. P ≤ 0.05 was considered statistically significant throughout the study.

RESULTS

Description of the population

Table 1 shows the description of the study participants. A total of 5609 pediatric patients presented to the ED of a tertiary care hospital in Riyadh. It was observed that a higher proportion of the patients presenting to the ED were younger than or exactly 5-year-old (59.3%) in comparison to those who were over 5-year-old (40.7%). The number of males who presented to the ED was higher, with 3102 patients (55.3%), whereas only 2507 (44.7%) female patients presented to the ED. A higher proportion of patients (94.3%) were sent home, followed by 4.25% who were admitted to the hospital. 0.6% were discharged against medical advice (DAMA), 0.6% were discharged with an outpatient appointment, and 0.25% were transferred to another health-care facility. Moreover, based on urgency, a higher proportion (68.22%) of patients were directed to the less urgent triage, followed by urgent (19.27%), nonurgent (12.01%), and emergent (0.5%).

Table 1

Description of the study participants

Variables	Frequency (n=5609), n (%)
Age, median (IQR)	3 (1-7)
Age (years)*
≤5	3326 (59.3)
>5	2283 (40.7)
Gender
Male	3102 (55.3)
Female	2507 (44.7)
Discharge classification
Admission to hospital	133 (4.25)
Discharged with outpatient appointment	18 (0.6)
Home	2953 (94.3)
Discharge against medical advice	19 (0.6)
Transfer to another health-care facility	8 (0.25)
Total	3131
Triage
Nonurgent	658 (12.01)
Less urgent	3736 (68.22)
Urgent	1055 (19.27)
Emergent	27 (0.5)
Total	5476

*Categorization on a cutoff age of 5 years was based on the distribution of children’s age admitted to the hospital. IQR: Interquartile range

Relationship of demographic factors and triage level with discharge status

Table 2 illustrates the relationship between discharge status and demographic factors among pediatric patients who visited the ED. There was a significant relationship between age and discharge status (P < 0.001). We observed that about 63.2% of children who were DAMA were ≤5-years of age versus 36.8% who were >5-year-old. The median age was significantly higher among those who were discharged with outpatient appointments (9.5) than among those who were admitted to the hospital (1; P < 0.001). Similarly, the median age was significantly higher among those who were sent home (3) than among those who were admitted to the hospital (1; P < 0.001). However, there was no difference in the median age between DAMA and other discharge statuses of the patients on post hoc analysis.

Table 2

Relationship of discharge status and demographic factors among the pediatric population

Variables	Admission to hospital (n=133), n (%)	Discharged with outpatient appointment (n=18), n (%)	Home (n=2953), n (%)	Discharge against medical advice (n=19), n (%)	Transfer to another health-care facility (n=8), n (%)	P
Age (years), median (IQR)	1 (0-4)	9.5 (3.75-12)	3 (1-7)	2 (0-6)	2 (1-11.5)	<0.001*
Age
<5	104 (78.2)	5 (27.8)	1740 (58.9)	12 (63.2)	6 (75)	<0.001**
≥5	29 (21.8)	13 (72.2)	1213 (41.1)	7 (36.8)	2 (25)
Gender
Male	73 (54.9)	11 (61.1)	1639 (55.5)	14 (73.7)	6 (75.0)	0.429
Female	60 (45.1)	7 (38.9)	1314 (44.5)	5 (26.3)	2 (25.0)
Triage
Nonurgent	3 (2.3)	5 (27.8)	391 (13.6)	3 (18.8)	0	<0.001***
Less urgent	35 (26.9)	10 (55.6)	2002 (69.5)	4 (25)	3 (42.9)
Urgent	82 (63.1)	3 (16.7)	481 (16.7)	8 (50)	4 (57.1)
Emergent	10 (7.7)	0	7 (0.2)	1 (6.3)	0

*Significant at P<0.05 using Kruskal-Wallis test/fisher’s exact test, **Significant at P<0.05 using Chi-square, ***Significant at P<0.05 using fisher’s exact test. IQR: Interquartile range

There was no significant relationship between gender and the discharge status of the patients. However, the results showed a significant relationship between triage level and the discharge status of the patients (P < 0.001). We observed that the majority of the patients who were DAMA were of urgent triage (50%), followed by less urgent (25%), nonurgent (18.8%), and emergent (6.3%).

Relationship of demographics factors and triage level with discharged against medical advice

Table 3 shows the relationship between DAMA and the demographic predictors among pediatric patients. We observed that the median age was lower among those who were DAMA than those who did not get DAMA, and most of the patients were 5-year-old or younger. However, there was no significant relationship between age and DAMA (P > 0.05). A higher proportion of females were DAMA than males, but there was no significant relationship between gender and DAMA (P = 0.112). However, there was a significant relationship between triage level and DAMA, with a higher proportion of patients who got DAMA presenting to urgent care (50%) than those who did not get DAMA (18.8%). Most of the patients who did not get DAMA presented to less urgent care (65.9%).

Table 3

Relationship of discharge against medical advice with demographic factors among the pediatric population

Variables	DAMA		P
	Yes (n=19), n (%)	No (n=3112), n (%)
Age (years), median (IQR)	2 (0-6)	3 (1-7)	0.176
Age
≤5	12 (63.2)	1855 (59.6)	0.753
>5	7 (36.8)	1257 (40.4)
Gender
Male	14.0 (73.7)	1723 (55.6)	0.113
Female	5.0 (26.3)	1383 (44.4)
Triage
Nonurgent	3 (18.8)	399 (13.1)	<0.001*
Less urgent	4 (25.0)	2050 (67.5)
Urgent	8 (50.0)	570 (18.8)
Emergent	1 (6.3)	17 (0.6)

*Significant at P<0.05 using fisher’s exact test. DAMA: Discharge against medical advice, IQR: Interquartile range

Univariate and multivariable analyses

Table 4 shows the univariate and multivariable analyses for assessing the relationship between DAMA and demographic factors. In the univariate analysis, age and triage level were significant (at P < 0.2). It was observed that with every 1 unit increase in age, the odds of DAMA increased by 9%. In the multivariable analysis, after adjusting for the other covariates, there was a significant interaction between age and gender. It was observed that the odds of DAMA for a 5-year-old female child were 4.43 times higher than those of a 5-year-old male child (P < 0.1).

Table 4

Univariate and multivariable analysis for assessing relationship of discharge against medical advice with demographics

Variables	Unadjusted OR (95% CI)	Adjusted OR (95% CI)
Age (years)	1.09 (0.95-1.25)*	_
Age (years)
≤5	1.16 (0.45-2.96)	1.1 (0.40-1.18)**
>5 (reference)	1	1
Gender
Female	2.24 (0.81-6.23)	0.706 (0.19-2.57)
Male (reference)	1	1
Triage
Nonurgent (reference)	1	-
Less urgent	3.85 (0.85-17.28)*
Urgent	0.54 (0.41-2.03)
Emergent	0.13 (0.01-1.29)
Age gender*
Female	-	4.43 (−2.89-11.76)***
Male (reference)

*Significant at P<0.25, **Significant at P≤0.05, ***Significant interaction at P<0.1. OR: Odds ratio, CI: Confidence interval

DISCUSSION

This study reported a DAMA of 0.6% among pediatric patients presenting to the ED in a tertiary level care hospital in Riyadh, which was lower than that of other studies.[478]The lower prevalence of DAMA in the Saudi population could be attributed to the lack of decision-making authority of children and parents' worry for their children's health. Consequently, the parents try to solve financial problems and continue the treatment of their children until they fully recover and are discharged based on physicians' orders. Moreover, the rate of DAMA among children was less than among adults because children do not work and are less affected by family problems.[19]

Studies from the literature report a slightly higher rate of DAMA among girls[1020]and younger age groups.[7811]In our study, DAMA was about 4 times higher among girls who were 5 years old than among boys of the same age group. The possible reason for this might be that in a patriarchal society like that of KSA, male children are given preference over female children, as well as the fact that the father is often the sole custodian of his family's resources and decides whether or not the family can bear the cost of the child's treatment.[21]Thus, empowering females (mothers) in keeping with World Health Organization's (WHO) fifth sustainable development goal (gender equality) may contribute to reducing the incidence of DAMA, especially among female children.[14]Moreover, the parents' perception that their child is better, personal reasons, and financial insecurity are among other reasons for DAMA, as reported in the literature.[7]

It was also observed that half of the patients with DAMA were categorized as urgent care, indicating the severity of illness within this age. The WHO[22]has reported that severe diseases, such as severe anemia, diarrheal diseases, and acute lower respiratory infections, are the leading causes of child morbidity and mortality. The severity of these diseases may lead to prolonged hospital stays, which would lead to increased costs on the parents, making it difficult for them to bear the financial burden, and ultimately leading to DAMA.

This study's strength is that it is the first study conducted in KSA evaluating the demographic factors of DAMA in the ED among pediatric patients. Since this tertiary care level hospital caters to people from different demographics, our study results can be generalized to the EDs of all the tertiary care hospitals in KSA.

However, the study did have certain limitations. Since the data had been collected from the medical records at the hospital, there was some missing information. For example, we were not able to stratify the clinical environments and severity of medical conditions that patients with DAMA came from. Moreover, there was no information on the socioeconomic factors (such as the educational status of the children and their parents, the occupational factors of the parents, and the monthly income of parents) or reasons for signing DAMA that might be important predictors of DAMA at the ED.

CONCLUSIONS

The study results concluded that DAMA was lower in KSA than in other countries. DAMA was more common among females than males in similar age groups. The DAMA among pediatric patients is a serious hidden health problem, affecting both children and health services.

To reduce the occurrence of pediatric DAMA, governments should implement health insurance policies. Moreover, hospital management should allow a 24-h grace period before insisting on payment as well as strengthening social work departments to identify and help truly deprived patients.

The public should be educated on the consequences of DAMA. Continued health education, and the promotion of child survival strategies at the community level, combined with an improvement in the socioeconomic conditions of the population, may further reduce DAMA and improve the chances of survival for children. Future studies should assess the socioeconomic status of the patients and estimate the cost that is incurred by the patients.

Research quality and ethics statement

This study was approved by the Institutional Review Board / Ethics Committee at 2King Abdullah Bin Abdulaziz University Hospital and Princess Nourah University (Approval # H-01-R-059). The authors followed the applicable EQUATOR Network (http://www.equator-network.org/) guidelines, during the conduct of this research project.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.