Appropriateness of Hospital Admission and Length of Stay in the Pediatric Department of a Tertiary Care Hospital in West Bengal

Abstract

Background:

Appropriateness of admissions and inpatient length of stay (LOS) are important factors controlling health-care expenditure. The current study was conducted to assess the proportion of appropriate admissions among 29-day to 12-year--old aged patients admitted to pediatric wards of a tertiary health-care center in one month and evaluate inappropriate LOS among them.

Materials and Methods:

An observational longitudinal study was carried out for two months among patients admitted to the pediatric inpatient department using Pediatric Appropriateness Evaluation Protocol (PAEP)-II tool. All patients admitted to the institution during the study period were included in the study.

Results:

Among 495 children, 438 (88.5%) were appropriately admitted. Inappropriate admission was mostly found in children aged 1 to 5 years (29, 16.6%), those who lived >5 km from the hospital (45, 15.2%), and those who presented with other systemic illnesses (13, 31.7%). Among the appropriately admitted children, 333 (76%) met the criteria of appropriate LOS during their hospital stays. Multivariable logistic regression showed that the age of the children (AOR = 0.99; 95% CI = 0.98, 0.99), age-appropriate feeding (AOR = 2.12;95% CI = 1.30,3.47), and hematological disorder (AOR = 0.16; 95% CI = 0.05, 0.52) were found to be associated with the children being kept at the hospital inappropriately.

Conclusions:

The current study observed a high proportion of appropriate admission in children. However, a higher proportion of inappropriate LOS among them was seen. The younger age group, age-appropriate feeding, and type of illness were found as the primary predictors for inappropriate LOS.

INTRODUCTION

Evaluation of the admissions and patients’ length of stay (LOS) is important in hospital management to ease the burden on the health-care system and increase its operational efficiency. Inappropriate hospital admission leads to significant depletion of hospital resources and is a burning problem in developing countries across the world.[1-3] Inappropriate admissions also increase the chances of contracting nosocomial infections, especially in tertiary care hospitals dealing with the most critically ill patients. Because of the high expense of health-care services in Western countries, there is a greater emphasis on the appropriateness of hospital services utilisation. A study that evaluated the effect of inappropriate hospitalization in three countries through propensity score matching reported that the inappropriate group had excessive health service utilization compared with appropriate admissions, especially for internal diseases.[1] A cross-sectional study in Spain found that standardizing admission and discharge process can improve utilization, patient flow and waiting time.[4]

The three-tier public health-care system of India accounts for more than 40% of total inpatient health-care delivery to its 1.38 billion population.[5] However, with a hospital bed to population ratio of only 5.3 per 10,000, it is also one of the most overburdened health-care systems in the world.[6] Pediatric disease management is among the most resource-intensive disciplines of medicine, and children are most prone to suffer from the ill effects of inappropriate admissions. A possible way of avoiding this is by using a standardized protocol for admission and hospital stays. In the absence of information on the same, the present study was conducted to assess the appropriateness of admissions and LOS of pediatric patients in a tertiary care teaching hospital in Eastern India.

MATERIALS AND METHOD

Study design, study period, and setting

This observational, longitudinal study was conducted in the pediatric ward of North Bengal Medical College and Hospital, a tertiary care teaching hospital from 01.11.2019 till 31.12.2019 (including the follow-up period of last recruited participant).

Study population

Children aged 29 days to 12 years admitted to the pediatric wards from the emergency or outpatient department were included in the study. Children with non-consenting parents were excluded.

Sampling

During the period of study, a total of 504 children were admitted. After getting consent from their parents, 495 children were included.

Study tools

PAEP-II tool was used to collect data. It is a widely used validated tool to assess the appropriateness of hospital admissions and has been validated for the Indian health-care scenario.[7] The PAEP-India tool is available in two versions, one for neonates (≤28 days) and another for children (>28 days to 18 years). In this study PAEP-II for children was used. Section I assesses the appropriateness of admission with the following criteria:

• Severity of illness: 9 criteria

• Severe electrolyte/acid-base/CBC abnormality: 15 criteria

• Conditions not responding to out-patient management: 8 criteria

• Special pediatric problems: 5 criteria

• Intensity of services: 7 criteria

Section II assesses the length of hospital stay with the following criteria:

• Medical services: 9 criteria

• Nursing/life support services: 9 criteria

• Patient condition within 24 hours: 5 criteria

• Patient condition within 48 hours: 6 criteria

An admission was considered to be appropriate if any one of the criteria for section I was met. Similarly, meeting any one of the criteria for Section II represented that the LOS at the hospital was appropriate for the patient.

Data collection

Patients were assessed for appropriateness of admission at the time of their admission to the pediatric ward. Only those found to be admitted appropriately were followed up till their discharge or in the event of their death. The LOS in the hospital, were assessed as being appropriate or inappropriate using the PAEP tool. If a patient admitted for less than 24 hours then LOS was counted as ‘0’ days. The data collection procedure is shown in Figure 1.

Figure 1.

Flowchart of the data collection procedure for the study

Data analysis

Collected data were entered into and cleaned using MS Excel (Microsoft Corp) and analyzed using the IBM Statistical Package for Social Sciences (SPSS v. 22). Sociodemographic and admission-related characteristics were presented as frequencies and percentages. For assessing the association between different patient characteristics and appropriateness of their LOS, a multiple logistic regression analysis was used. Bivariate logistic regression was performed and predictors with ≤0.10 level of significance were considered for backward stepwise multivariable logistic regression. A p value of <0.05 was considered to be statistically significant.

Ethical consideration

The present study was granted clearance by the Institutional Ethics Committee [NBMC/IEC/2019-20/116]. Written informed consent was obtained from the parents and/or guardians of the patients, and care was taken to ensure confidentiality and anonymity.

RESULTS

Of the total 504 children admitted to the pediatric ward during the study period, the parents of 9 children did not provide consent and were excluded (1.7% non-response rate). Most of the patients were less than 1 year of age (35.8%), with a median age of 24 months. Sixty percent of the admitted patients were male and were from families that lived more than 5 km away from the hospital. Of the children, 89.5% had age-appropriate immunization, and 67.1% had age-appropriate feeding. Of the admissions, 41.2% were for respiratory diseases, followed by 23.6% for hematological diseases [Table 1]. Among admitted children, 6 (1.9%) deaths were reported, while the others were discharged in stable condition.

Table 1.

Distribution of study population according to sociodemographic and other characteristics (n=495)

Variables		Frequency	Percentage (%)
Age group (in year)	<1	177	35.8
	1–5	175	35.4
	6–10	124	25.1
	>10 (up to 12)	19	3.7
Sex	Male	297	60
	Female	198	40
Religion	Hindu	317	64.0
	Muslim	153	30.9
	Others	25	5.1
Distance of residence from hospital	≤5 km	208	42.0
	>5 km	287	58.0
Age-appropriate immunization	Yes	443	89.5
	No	52	10.5
Age-appropriate feeding	Yes	332	67.1
	No	163	32.9
Diagnosis	Cardiological	7	1.4
	Gastrointestinal	53	10.7
	Haematological	117	23.6
	Neurological	28	5.7
	Renal disease	11	2.2
	Respiratory	204	41.2
	Other systemic diseases*	41	8.3
	Others**	34	6.9
Total		495	100.0

*Other systemic diseases include fever under evaluation, joint pain, diabetes. **Others include poisoning, vaccine preventable diseases, torticollis, admission for surgical procedures, skin infection, drug overdose, etc

Analysis of appropriateness of admission found that among the 495 participants, 57 (11.5%) were admitted inappropriately [Figure 2]. Inappropriate admission was more prevalent among children aged 1 to 5 years (16.6%), those who lived >5 km from the hospital (15.2%), and those (31.7%) who presented with other systemic illnesses (fever under evaluation, joint pain, diabetes) [Table 2].

Figure 2.

Appropriate admission and length of stay of patients in pediatric wards (n = 495)

Table 2.

Appropriate admission related characteristics of the participants (n=495)

Variables	Appropriate admission (%)	Inappropriate admission (%)	Total
Sex
Female	173 (87.4)	25 (12.6)	198 (100.0)
Male	265 (89.2)	32 (10.8)	297 (100.0)
Age group
<1 year	160 (90.4)	17 (9.6)	177 (100.0)
1–5 year	146 (83.4)	29 (16.6)	175 (100.0)
5–10 year	116 (93.5)	8 (6.5)	124 (100.0)
>10 year	16 (84.2)	3 (15.8)	19 (100.0)
Diagnosis
Cardiological	6 (85.7)	1 (14.3)	7 (100.0)
Gastrointestinal	41 (77.4)	12 (22.6)	53 (100.0)
Haematological	113 (96.6)	4 (3.4)	117 (100.0)
Neurological	26 (92.9)	2 (7.1)	28 (100.0)
Renal disease	8 (72.7)	3 (27.3)	11 (100.0)
Respiratory	183 (89.7)	21 (10.3)	204 (100.0)
Other systemic diseases*	28 (68.3)	13 (31.7)	41 (100.0)
Others**	33 (97.1)	1 (2.9)	34 (100.0)
Distance from residence
≤5 km	186 (93.4)	12 (5.05)	198 (100.0)
>5 km	252 (84.8)	45 (15.2)	297 (100.0)
Total	438 (88.5)	57 (11.5)	495 (100)

*Other systemic diseases include fever under evaluation, joint pain, diabetes. **Others include poisoning, vaccine preventable diseases, torticollis, admission for surgical procedures, skin infection, drug overdose, etc

In the 438 appropriately admitted children, the median LOS at the hospital was 2.5 days (IQR-4, average- 3.3 ± 3.2, range 0–23). Of them, 333 (76%) children met the criteria of appropriate LOS during their hospital stays and their median LOS was 2 days (IQR-3, average- 2.7 ± 2.6). Inappropriate LOS were found to be higher in males (26%) and those living at a distance of >5km from the hospital (25.4%). Majority of the children with inappropriate LOS also didn’t have age-appropriate feeding (34.5%) [Table 3]. Furthermore, patients who were admitted with gastrointestinal complications were admitted for longer periods (39%). It was seen that the majority of the inappropriate LOS was due to hospital-related factors like lack of availability of primary health-care services in the nearby areas and delay in receiving results of laboratory investigations (22.8%).

Table 3.

Factors associated with inappropriate length of stay (number of days) at the hospital (n=438)*

Characteristics	Appropriate LOS, n (%)	Inappropriate LOS, n (%)	Crude odds ratio	Adjusted Odds Ratio** (95% CI)	P
Age (completed months)	47.58±43.36	24.08±29.22	0.98 (0.98, 0.99)	0.99 (0.98, 0.99)	0.03a
Sex
Female	137 (79.2)	36 (20.8)	0.74 (0.47,1.18)	0.74 (0.46,1.23)	0.25
Male	196 (74.0)	69 (26.0)	1	1
Age-appropriate feeding
No	95 (65.5)	50 (34.5)	2.28 (1.45, 3.58)	2.12 (1.30,3.47)	0.003a
Yes	238 (81.2)	55 (18.8)	1	1	-
Distance of residence from hospital
≤5 km	145 (78.0)	41 (22.0)	0.83 (0.53,1.30)	0.97 (0.59,1.59)	0.91
>5 km	188 (74.6)	64 (25.4)	1	1
Age-appropriate immunization
No	39 (86.7)	6 (13.3)	0.45 (0.18,1.11)	1.35 (0.47,3.87)	0.58
Yes	294 (74.8)	99 (25.2)	1	1
Type of illness
Cardiovascular	5 (83.3)	1 (16.7)	0.36 (0.03, 3.52)	0.25 (0.02, 2.61)	0.25
Gastrointestinal	25 (61.0)	16 (39.0)	1.15 (0.43, 3.12)	1.29 (0.47, 3.59)	0.61
Hematological	107 (94.7)	6 (5.3)	0.10 (0.03, 0.31)	0.16 (0.05, 0.52)	0.002a
Neurological	21 (80.8)	5 (19.2)	0.43 (0.12, 1.48)	0.37 (0.10, 1.32)	0.12
Respiratory	123 (67.2)	60 (32.8)	0.87 (0.38, 2.02)	0.84 (0.36, 1.98)	0.69
Renal	6 (75.0)	2 (25.0)	0.60 (0.10, 3.55)	0.92 (0.15, 5.69)	0.93
Other diseases	28 (84.8)	5 (15.2)	0.32 (0.09, 1.09)	0.52 (0.14, 1.85)	0.31
Other systemic diseases	18 (64.3)	10 (35.7)	1	1	-
Total	333 (76.0)	105 (24.0)

*Inappropriate LOS of stay was only observed among participants who were admitted appropriately. **Adjusted for age of the children, sex of the children, age-appropriate feeding, age-appropriate immunization, distance of residence from hospital, and type of illness. ^aStatistically significant

A multivariate logistic regression model [Table 3] showed that keeping other factors constant, the age of the children (AOR = 0.99; 95% CI = 0.98,0.99; P value = 0.03), age-appropriate feeding (AOR = 2.12;95% CI = 1.30,3.47; P value = 0.003), and hematological disorder (AOR = 0.16;95% CI = 0.05, 0.52; P value = 0.002) were found to be associated with the children being kept at the hospital inappropriately. Higher odds of inappropriate LOS was seen among children with gastrointestinal disorders, but no statistical significance was found.

DISCUSSION

Appropriate admission and length of hospital stay indicate whether the severity of illness or type of clinical or diagnostic services justify in-hospital care for a given patient.[1] Inappropriate admission rates vary from country to country and even region to region. The inappropriate admission rate of 11.5% found in the current study is similar to that reported in Australian hospitals (13%) but lower than what has been reported by a study conducted in a British hospital (32%).[8-10] In Delhi it was reported as 2.2% whereas in South India as 1.1%.[6] These differences might be due to the different sets of criteria utilized in the other hospitals regarding admission of patients; however, it might also be due to the different patient pools that these hospitals serve. The current study was conducted in a tertiary care centre that catered to a large population hailing from some of the remotest areas of Eastern India, who have to travel long distances to reach the hospital. The feeling of need to admit the children by the parents might have contributed significantly to the observed inappropriate admission rate, as evidenced by the significant association observed between the distance of the home of the patient from the hospital and the prevalence of inappropriate admission in them. This has also been reported by Campbell.[11] Lack of accessibility to health-care services at lower levels and lack of investigation facilities might also contribute to these inappropriate admissions.

Even among appropriately admitted children, the prevalence of inappropriate LOS in the hospital was found to be high (24%). Lack of required infrastructure meant that the children, after being discharged from the tertiary care setup, could not be adequately monitored in their local hospitals, thus necessitating increased days of stay at the tertiary setups.[12] Delays due to pending investigations and results of investigations and the distance of hospital from home prompted parents to keep their child at the hospital till they could arrange for transportation.

In the current study the lesser the age of the child, the higher the odds of a prolonged LOS. This is in contrast to reports from Europe or North America.[13-15] The high prevalence of morbidity and mortality among infants and young children makes doctors overly cautious, leading to inappropriately longer LOS in them.[16] Age-appropriate feeding especially during the first 2 years of childhood has paramount importance in growth and health. In its absence nutritional deficits can affect recovery after hospitalization and in turn can increase the hospital stay.[17] Other systemic diseases and gastrointestinal diseases were associated with an increased risk of longer LOS, while children suffering from haematological illnesses were the least likely to have inappropriate LOS. This finding conforms with those reported by Bianco et al. in Italy.[18] Most systemic diseases are chronic in nature, and gastrointestinal diseases present with very vague and non-specific symptoms, that need time to work up.[19] However, hematological conditions are comparatively much easier to diagnose and require far fewer tests to assess and have more specific management protocols.[20]

CONCLUSION AND RECOMMENDATION

The current study observed a high proportion of appropriate admission in children. However a higher proportion of inappropriate LOS among them were seen. The younger age group, age-appropriate feeding, and type of illness were found as the primary predictors for inappropriate LOS. Proper assessment before admission using standard protocols can help to mitigate these problems.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.