Anil Narang, P.S Sandesh Kiran and Praveen Kumar
From the Neonatal Unit, Advanced Pediatrics Center,
Postgraduate Institute of Medical Education and Research, Chandigarh
160 012, India.
Correspondence to: Dr. Anil Narang, Professor and
Head, Department of Pediatrics, PGIMER, Chandigarh 160 012, India.
E-mail:
anilnarang2004@yahoo.com
Manuscript received: April 29, 2004; Initial review
completed: June 11, 2004,
Revision accepted: March 30, 2005.
Abstract
Background: The number of neonatal intensive
care units (NICUs) in India has increased substantially over the last
decade; yet many more are required. There is limited information on
the actual costs of setting up and running an NICU in India.
Objective: Systematic and comprehensive calculation and analysis
of the costs of neonatal intensive care in a tertiary care teaching
hospital. Methods: The costs were compiled by studying the
detailed records of various hospital departments and prospectively
documenting the costs of drugs, consumables and investigations for a
representative group of 30 babies. Results: The total cost of
establishing a 16 bed level III tertiary care NICU was Rs 3.78 crore (Rs.
37.8 million, US$ 860,000) (2003). Equipment cost formed two-thirds of
the establishment cost. The running cost of NICU care per patient per
day was Rs 5450 (US$ 125). NICU and ancillary personnel salary
comprised the largest proportion of the running costs. The average
total cost of care for a baby <1000 grams was Rs. 1,68,000 (US$ 3800),
Rs. 88,300 (US$ 2000) for babies 1000-1250 g. and Rs. 41,700 (US$ 950)
for those between 1250 -1500 g. The family had to bear only 25%; rest
was subsidized. Conclusions: Equipment and personnel salary
form the biggest proportion of establishment and running costs. The
costs of treatment for a baby in NICU should be seen in context with
costs of other types of health care and the number of useful life-
years gained.
Key words : Cost, Intensive care, Neonate, NICU.
In this era of increasing healthcare costs,
the role of economic evaluations of health-care interventions has
become increasingly important. Economic evaluations can be of many
types: cost-benefit studies where each outcome is equated in monetary
terms, cost-minimization studies where two different methods of
achieving the same goal are compared for their costs-utility analysis
in which the healthy years of life are taken as the outcome and
cost-effectiveness analysis in which natural units like life years
gained are used as outcomes(1,2).
Neonatal intensive care stays are among the most
expensive types of hospital-izations(3). However, there is little
information on the cost of neonatal intensive care from India and
other developing countries (4-9). The information would be useful for
(i) setting up of a new neonatal unit, (ii) guiding
families at the time of neonatal intensive care unit (NICU) admission,
and (iii) exploring means of utilizing the limited budget more
effectively(10). The aim of this study was to calculate and analyze
the actual costs and cost to the family of neonatal intensive care in
a tertiary care teaching hospital in northern India.
Subjects and Methods
This study was conducted in Nehru Hospital,
Postgraduate Institute of Medical Education and Research, Chandigarh -
a tertiary care teaching hospital catering to northern Indian states.
The neonatal unit has 40 beds, out of which, 16 are intensive care
beds. Eight of these intensive care beds have provision for mechanical
ventilation with separate multi-parameter monitoring equipment and
infusion pumps. However, these equipments are shared when the baby is
cared in a non-ventilation bed. All the beds have either an incubator
or an open care system. Nearly three fourths of equipments used in the
NICU are imported. A doctor to patient ratio of 1:5 and a nurse to
patient ratio of 1:2 (1:4 for non-ventilation beds) are maintained
round the clock.
Admission-discharge policy
Depending on the availability of beds, attempts are
made to admit babies weighing < 1750 g (or < 34 weeks gestation) and
all sick babies to NICU. After recovery, they are shifted to
intermediate care level II nursery and subsequently they are
discharged to home. The mean discharge weight and gestation of very
low birth weight (VLBW) babies is 1400 grams and 34 weeks
respectively. Discharges are often expedited because of shortage of
beds and inability of families to stay away from home for long
periods.
Hospital and investigation charges
A nominal amount is collected from patients towards
hospital charges (Rs 325 per day in NICU and Rs 100 per day in level
II care nursery) and investigations separately. Since this is a state
sponsored tertiary care center, these charges are subsidized and fixed
arbitrarily without any intention of profit or recovery of running
cost.
Drugs and consumables
A limited amount of drugs and consumables are
provided by the hospital; most of them have to be purchased by the
family.
Methods of cost estimation
The establishment cost and running costs were
compiled by consulting various hospital departments, looking at their
records, reviewing previous year’s unit records and prospectively
documenting the cost of drugs, consumables and investigations.
Depreciation cost
A measure of loss of value of an asset due to
general wear and tear or obsolescence was determined to be around 13%
after evaluating the initial establishment cost and estimated life
span of each asset. Land value was not included, as it does not
depreciate.
Operational cost of equipments (replacement of spares)
The actual amount spent towards replacement of
spares for equipments over a period of one year was obtained from the
unit equipment maintenance records.
Ancillary services and ancillary personnel salary cost
Ancillary services include services provided by
other hospital departments for running of NICU. It involves
expenditure towards materials and supplies for ancillary services and
salary of ancillary personnel. Some of these ancillary services
include electricity, water, telephone, laundry, central sterilization,
radiology, laboratory, hospital administrative and engineering
branches, and hospital drug, medical and surgical stores. The actual
cost of ancillary services was arrived at by the method of
apportionment(4). The NICU has 1.26% of hospital beds and thus 1.26%
of the hospital’s expenditure on ancillary services and ancillary
personnel salary was taken as expenses for NICU.
Drugs, consumables and investigation cost
Use of drugs and requirement of consumables and
investigations were prospectively recorded in 30 babies admitted
consecutively to NICU in the month of March 2003, by meticulously
following them till discharge. The total consumables and investigation
cost involved in taking care of these 30 representative babies was
divided by the number of NICU patient days to obtain the cost per
patient per day in NICU. This amount was multiplied by 5480 (16
patients × 365 days) to obtain the annual drugs, consumables and
investigation cost of our 16-bedded NICU. The median length of stay,
incidence of prematurity, mean birth weight and gestation, and
proportion of babies requiring mechanical ventilation in this
representative study group were similar to the characteristics of
babies admitted to NICU during 2002(11). The differences were tested
using continuous variable independent ‘t’ test and discrete variable
Chi square test.
Cost of care in intermediate level II care nursery
The running cost during intermediate care nursery
stay (level II) was calculated by giving it 1:5.5 weightage in
comparison with NICU care as has been done in earlier studies(12-13).
Results
The present study was conducted between February
and April 2003. Table I shows the characteristics of the
patient population of the unit in the year 2002. The total cost of
establishing a 16 bedded level III tertiary care neonatal unit was Rs
3.78 crore (R 37.8 million, US$ 860,000) (2003). The establishment
cost for a ventilation bed was Rs 30.5 lacs (Rs 3.05 million, US$
70,000) and Rs. 16.9 lacs (Rs 1.69 million, US$ 38,500) for a
non-ventilation intensive care bed. Equipment cost formed the major
portion (two thirds) of the establishment cost (Table II). The
yearly running cost of Rs 3.18 crore (Rs 31.8 million, US$ 725,000)
includes the expenses involved in providing intensive care for 16
babies in NICU for 365 days (5840 patient days) . From this data the
cost of NICU care per patient per day to the unit was calculated as Rs
5450 (US$ 125) (Table III). NICU and ancillary personnel salary
was the largest expense accounting for 55%, followed by drugs and
consumables accounting for 15% of the annual running cost. To have a
quantitative measure of the extent of subsidization, the daily cost of
NICU care to the family was calculated. The median daily cost of NICU
care to the family was Rs 1350 (range Rs 876-1734) per patient per day
. This was approximately 25% of the actual daily cost of NICU care.
TABLE I
Patient Population Charactristics 2002).
• |
Total live births: 3224 |
|
• |
Booked 33.2% |
|
• |
Gestation
<37 wks : |
1101 (34%) |
|
<33 wks : |
336 (10.4%) |
|
<28 wks : |
37 (1.1%) |
• |
Birth weight <2500 g :
|
1385 (43%) |
|
< 1500 g :
|
323 (10.1%) |
|
<1000 g :
|
80 (2.6%) |
• |
NICU admissions :
: |
733 |
• |
No. ventilated : |
161 (HMD 98, Pneumonia 26, MAS 11, others 26) |
• |
Total no. of ventilation days
: |
1256 |
• |
Birth asphyxia (5 min Apgar <6) : |
69 (2.1%) |
TABLE II
Establishment Cost of a 16-bed NICU.
Componen
(thousand) |
Cost in Rs
establishment |
% of total
cost |
Equipment
|
25,623
|
67
|
|
7,500
|
20
|
Building cost**
|
4,384
|
12
|
Furniture
|
320
|
1
|
Total |
37,827 |
|
* Real estate rates, Chandigarh city @ Rs 15000/ square yard.
** Estimate inclusive of civil, public health, water, electricity,
lab gases, air conditioning etc. @ Rs 1200 / square feet.
TABLE III
Annual Running Cost of a 16-bed NICU.
Component
|
Cost in Rs
(thousand) |
% of total running cost |
NICU personnel salary
|
9,150
|
30
|
Ancillary personnel salary
|
8,000
|
25
|
Drugs and consumables
|
4,882
|
15
|
Depreciation
|
4,251
|
13
|
Ancillary maintenance
|
3,600
|
11
|
Investigations
|
1,109
|
3
|
Operational (Replacement |
|
|
of spares)
|
839
|
3
|
Total cost per year
(5840 patient days) |
31,832 |
|
The total cost of neonatal care for surviving
babies in different birth weight categories is depicted in Table
IV. The median total cost of neonatal care for a baby with birth
weight of less than 1000 g was Rs 1,68,600. For each subsequent 250 g
increments in birth weight, the cost of care decreased by about 50%
amongst babies < 1500 g. For a baby weighing less than 1000 g at
birth, a family spent about Rs 39,000 on hospital charges, drugs,
consumables and investigations. This amounted to nearly 75% subsidy by
the hospital.
TABLE IV
Cost of Neonatal Care of Survivors According to Birth Weight.
|
Hospital stay
(median) (days) |
Total Cost
(NICU + Level II nursery)* |
|
Birth weight
(grams) |
NICU
|
Level II
nursery |
[median (range)]
|
Cost to the family
[median (range)] |
|
|
|
1,68,600 |
39,000 |
500-999 |
28 |
16 |
(77,150 - 419,300) |
(13,350-101,500) |
|
|
|
|
|
1000-1249 |
14 |
12 |
(16,450-254,450) |
(2,450-58,450) |
|
|
|
|
|
1250-1499 |
6 |
9 |
(10,450-179,450) |
(1,850-38,950) |
|
|
|
|
|
1500-1749 |
4 |
7 |
(9,450-375,150) |
(1,750-91,450) |
|
|
|
|
|
≥ 1750 |
3 |
5 |
(5,450-281,500) |
(1,350-68,400) |
* NICU cost was obtained by multiplying the median duration of NICU stay with the daily
NICU cost per patient per day. The daily cost of intermediate level II nursery care to the
hospital was estimated to be Rs. 1000 per patient per day on the basis of 1:5.5 weightage
in comparison with level III NICU care (9,11,13). The daily cost to the family during stay in
level II nursery was Rs 100 per patient per day.
Table V depicts the total cost of neonatal care for
babies needing mechanical ventilation versus those not requiring
mechanical ventilation. The total cost of caring for a baby weighing
less than 1000 g and needing ventilation was Rs 2,37,100. The
requirement of mechanical ventilation increased the total cost of care
by 1.6 times in babies weighing less than 1000 g. However, in babies
weighing more than 1000 g, this increase was 2.6 to 3.2 times.
TABLE V
Cost of Neonatal Care of Ventilated and Non-ventilated Babies (Rs).
|
Ventilated |
Non-ventilated |
Birth weight
(grams)
|
Median length
of hospital stay
(days)
(NICU + Level II nursery) |
total cost
[median (range)]
|
Median length
of hospital stay
(days)
(NICU + Level II nursery) |
Total Cost
[median (range)]
|
500-999 |
68 |
237,100 |
36 |
147,250 |
|
|
(77,150-419,300) |
|
(125,000-170,500) |
1000-1249 |
48 |
177,050 |
21 |
65,500 |
|
|
(99,200-254,450) |
|
(16,450-152,700) |
1250-1499 |
30 |
101,200 |
15 |
37,250 |
|
|
(27,800-179,450) |
|
(10,450-97,300) |
1500-1749 |
20 |
77,850 |
11 |
24,350 |
|
|
(17,350-375,150) |
|
(9,450-77,300) |
>1750 |
14 |
51,050 |
8 |
20,350 |
|
|
(12,900-281,500) |
|
(5,450-75,400) |
Discussion
Although advances in neonatal techno-logy have
improved the survival prospects of premature infants significantly,
they have come at a high financial cost(3). Limited resources,
widespread poverty and absence of health insurance pose further
problems in third world countries(14). Knowledge of establishment and
running costs of NICU along with total cost of neonatal care can help
in optimizing the inputs required for organization of NICU. We were
able to do a comprehensive analysis of the various components of the
costs involved in providing NICU care which has not been earlier
reported from our country. Though we did not do any cost-effectiveness
or cost-utility analysis, this can form the ground for planning
further cost-minimization trials. Private set-ups are likely to differ
from government set-ups in many ways including salaries, efficiency,
use of more contract employees, hiring of equipment, linking of
staffing to patient-load etc.; hence our results may not be
generalizable to them.
The cost of establishing a ventilation bed is
higher than a non-ventilation bed because of additional monitoring
equipments required in managing a ventilated patient. The high
proportion of establishment cost due to equipments can be partly
attributed to the use of more expensive imported equipments. The
establishment cost of a 28-bedded NICU in a previous study from
Chennai(6) was reported as Rs 80 lacs in 1990 [(Rs 1.70 crore;
corrected for 2003 at 6% annual inflation rate(15)]. However data on
proximity to city center, design, staffing, number of ventilation beds
and type of neonatal equipments was not provided. The land cost which
formed 20% of establishment cost in our study could vary widely
depending on geographic location and so could the equipment cost.
The total cost of hospital care is the sum of
direct and indirect costs(4,16). Direct costs are those that can be
directly billed to patients e.g., investigations, consumables,
drugs etc. Indirect costs are those that cannot be billed
directly to patients e.g., administrative and clerical staff
salaries, cost towards materials and supplies of ancillary services.
Indirect cost allocations are made using arbitrary accounting rules
based on the square footage occupied by the unit or proportion of beds
out of total hospital beds(4). Thus, ancillary services and ancillary
personnel salary were calculated using the strategy of apportionment
(based on NICU beds as proportion of hospital beds)(4). However, NICU
beds may consume disproportionately larger resources as compared to
non-ICU wards. We used 1:5.5 weightage to calculate the level II
nursery stay costs as has been used in previous studies(12,13). This
method may have its limitations depending on the variations in
staffing and transfer policies between level III and II Units.
In our analysis, 55% of running cost was due to
personnel salary. Other studies have reported it to be 44% to 75% of
running cost(12,13). These may vary from place to place due to
differences in the staffing pattern and salary structure. The drugs
and consumables cost (15% of NICU running cost) may differ between
units depending on the birth weights, sickness profile and NICU
policies(17). The cost towards materials and supplies of ancillary
services contributed to 11% of NICU running cost, and was comparable
with other studies(17). The proportion of running cost of NICU spent
on investigations in our unit (3%) was lower when compared with other
studies [5% to 8.5%](4,5,9). This could be partly due to subsidized
charges in our hospital(5,7).
The running cost of NICU per patient per day (Rs
5,450) was higher in our study when compared to studies from
Chennai(6) and Delhi(8) where it was Rs 3960 and Rs 3241 per patient
per day respectively [corrected to 2003 at 6% annual inflation
rate(15)]. The cost of NICU care to the family was nearly one fourth
of the actual running cost to the unit. Thus about Rs 4100 per patient
per day was being subsidized by the hospital. The cost of neonatal
care varied inversely with birth weight (Table IV). This
reflects the differences in the length of stay and intensity of
treatment needed for different weight categories. The cost of neonatal
care in non-survivors was however not different for different birth
weight categories. Similar patterns have been observed in a study from
Vermont Oxford network(5).
Increment in the cost of neonatal care due to
requirement of mechanical ventilation was lower in babies weighing
less than 1000 g, when compared with babies weighing more than 1000 g
(Table V). Babies weighing less than 1000 g need more intensive
monitoring, investigations and numerous consumables irrespective of
the requirement of ventilation. However babies weighing more than 1250
g who do not need mechanical ventilation, may not require that
intensive degree of care.
Personal expenditure of patient attendees is
generally not accounted for, when cost estimates of NICU is made.
However this can be an additional substantial drain on the family
apart from the threat of losing one’s job.
Can we reduce NICU costs?
Equipment formed two thirds of the establishment
cost. By reducing the imported component of equipment, a significant
cost reduction may be achievable. However till recently, Indian made
equipments have generally suffered from lack of quality and
reliability. To decrease the running cost several approaches can be
considered. Since personnel salary forms 55% of running costs,
rationalization and variation of staff strength in relation to
changing patient numbers in NICU may be useful. Minimum staffing
pattern for each clinical unit and cross training of nursing staff for
utilization on higher patient load days may significantly reduce the
costs(18). The ancillary services may sometimes be cheaper and more
efficient when run on private contracts. Depending on patient load,
procuring equipments like ventilator on hire basis may be more
economical and efficient.
Regionalization of health care especially perinatal
care has been shown to be an effective way of reducing costs(19,20).
Promotion and development of more level II care units and developing
linkages with them so that babies needing simple supportive care
(temperature regulation, gavage feeing, antibiotic completion) and
basic monitoring can be back-transported, should help in cutting
costs. Currently, the numbers of such units are less, the linkages are
weak and public confidence in them is lacking.
As per our annual neonatal statistics of 2002(11)
three factors viz., requirement of mechanical ventilation,
occurrence of sepsis and apnea were found to have a significant
association with prolongation of hospital stay across all birth weight
categories. Strategies to reduce sepsis including initial ‘extra’
investment on buying adequate disposables and chemical disinfectants
should reduce the overall costs. Similarly, strategies like optimum
use of antenatal steroids and prevention of asphyxia may decrease the
need for mechanical ventilation. Tolerance of higher PaCO2 and lower
pH to reduce pulmonary barotrauma should also decrease the duration of
mechanical ventilation and hence neonatal intensive care costs.
To conclude, neonatal intensive care is expensive.
Some cost-cutting can be done by reorganization of staffing and using
indigenous equipments. However, to get the real perspective, one
should compare the costs of NICU care with the costs and outcomes of
other critical care procedures. For example, coronary artery bypass
graft costs Rs 1 to 1.5 lacs and has an average life expectancy of
5-10 years. Whereas neonatal intensive care for a 1000 to 1500 g baby
costs Rs 65,000 and is followed by a life expectancy of 50-70 years.
Acknowledgements
We are thankful to the officials of various
hospital departments and sections especially finance and engineering,
for cooperating and sharing their records with us. We are also
grateful to the families of the babies in NICU who readily shared the
information about expenses being incurred by them under various heads.
Contributors: AN conceived the study, planned
the methodology and reviewed the manuscript. He will act as guarantor.
SK did the data collection, analysis and wrote the initial draft. PK
helped and supervised the data collection, analysis, interpretation
and critical revision of‘ the manuscript.
Competing interests: None.
Funding: None.
Key Messages |
• Equipment and personnel salary respectively, form the biggest
proportion of establishment and running costs of a neonatal
intensive care unit .
• The average total cost of care for a baby <
1000 grams was Rs. 1,68,000 and Rs. 88,300 for babies 1000-1250
grams. Of this actual cost , the family had to bear only 25%;
rest was subsidized.
• To get the real perspective, the costs of NICU care should
be compared with the outcomes, number of life-years gained and
costs of other critical care procedures.
|
|
1. Drumond MF. Guidelines for authors and peer
reviewers of economic submissions to the BMJ. BMJ 1996; 313:
275-283.
2. Brosnan CA, Swint JM. Cost analysis: Costs
and application. Public Health Nursing 2001; 18: 13-18.
3. U.S. Congress, Office of Technology
Assessment, Neonatal Intensive Care for Low Birthweight Infants:
Costs and Effectiveness (Health Technology Case Study 38),
OTA-HCS-38 (Washington, DC: U.S. Congress, Office of Technology
Assessment, December 1987).
4. Rogowski J. Measuring the cost of neonatal
and perinatal care. Pediatrics 1999; 103 (Suppl): 329-335.
5. Rogowski J. Using economic information in a
quality improvement collaborative. Pediatrics 2003; 111: e411-418.
6. Shanmugasundaram R, Padmapriya E, Shyamala
J. Cost of neonatal intensive care. Indian J Pediatr 1998; 65:
249-255.
7. Oostenbrink JB, Buijs-Van der Woude T, van
Agthoven M, Koopmanschap MA, Rutten FF. Unit costs of inpatient
hospital days. Pharmacoeconomics 2003; 21: 263-271.
8. Kannaraj V, Paul VK, Gupta S , Sarma RK,
Anand K, Singh M. Cost of neonatal intensive care. In XVI Annual
convention of National Neonatology Forum, 1996, November 1-3,
Chandigarh, India.
9. Cheah IG, Soosai AP, Wong SL, Lim TO.
Cost-Effectiveness NICU Study Group. Cost-effectiveness analysis
of Malaysian neonatal intensive care units. J Perinatol 2005; 25:
47-53.
10. Rogowski JA, Horbar JD, Plsek PE, Baker LS,
Deterding J, Edwards WH, et al. Economic implications of
neonatal intensive care unit collaborative quality improvement.
Pediatrics 2001; 107: 23-29.
11. Annual inborn statistics 2002. Neonatal
Unit, Advanced Pediatric Center, PGIMER, Chandigarh, India.
12. Sandhu B, Stenenson RC, Cooke RWI, Pharoah
POD. Cost of neonatal intensive care for very-low-birth weight
infants. Lancet 1986; 1: 600-603.
13. Tubman TRJ, Halliday HL, Normand C. Cost of
surfactant replacement for severe neonatal respiratory disease: a
randomized controlled trial. BMJ 1990; 301: 842-845.
14. Moazam F, Lakhani M. Ethical dilemmas of
health care in the developing nations. J Pediatr Surg 1990; 25:
438-441.
15. Financial year average indices and
inflation rates. Monthly economic scenario, May 2003. Office of
the Economic Adviser to the Government of India Ministry of
Commerce and Industry. Available from: URL: http://eaindustry.nic.in/mes/inf_rate.htm
; accessed on December 15, 2003.
16. Rosenstein AH. Fixed vs variable costs of
hospital care. JAMA 1999, 18; 282: 630.
17. Newns B, Drummond MF, Durbin GM, Culley P.
Costs and outcomes in a regional neonatal intensive care unit.
Arch Dis Child 1984; 59: 1064-1067.
18. Foster GP, Nelson J. 2002 Dane county
budget issue, Badger prairie health care center deficit reduction,
Dane county department of human services. June, 2001 Available
from: url: http://www.co.dane.wi.us/humanservices/reports/2002/;
accessed on December 15, 2003.
19. Rosner F. Allocation or misallocation of
limited medical resources. Cancer Invest 2004; 22: 810-812.
20. Committee on Perinatal Health: Toward improving the outcome
of pregnancy: Recommendations for the regional develop-ment of
maternal and perinatal health services and beyond. White Plains,
NY, The National Foundation-March of Dimes, 1976.
|