Diabetes can also contribute to many pregnancy-related complications for the Because the common feature of diabetes is high blood glucose, it is often …

Pacific Action for Health Project

Economic impact assessment of non-communicable diseases on hospital
resources in
Tonga, Vanuatu and Kiribati

Dr Christopher Doran BEc Hons PhD, Health Economist
National Drug and Alcohol Research Centre
University of New South Wales, Australia

Final report
September 2003

Correspondence
Dr Christopher Doran
Address: NDARC, UNSW, Sydney, 2052
Phone: 61 2 9385 0283
Fax: 61 2 9385 0333
Email: cdoran@unsweduau

This report has been funded under the Pacific Action for Health Project:
Secretariat of the Pacific Community, AusAID

Table of contents

Executive summary v

1 overview of project and definition of Non-communicable diseases 1

11 Introduction 1
12 International classification of diseases 2
13 What is a non-communicable disease? 3
14 Key NCDs considered in this report 3
15 Key risk factors for NCDs in this report as per project brief 8

2 MOST
RELIABLE SOURCES OF hospital data 11

21 Introduction 11
22 Health care provision and data collection methods used in Tonga 11
23 Health care provision and data collection methods used in Vanuatu
14
24 Health care provision and data collection methods used in Kiribati 17

3 resource use and cost associated with ncd treatment 19

31 Introduction 19
32 Resource use and cost associated with treatment of disease in Tonga
19
33 Resource use and cost associated with treatment of disease in Vanuatu
22
34 Resource use and cost associated with treatment of disease in Kiribati
25

4 Impact of tobacco and alcohol on hospital resources 35

41 Introduction 35
42 Impact of tobacco and alcohol on disease 35
43 Prevalence of alcohol and tobacco in each country 37
44 Quantifying harm caused by alcohol and tobacco in each country 43

5 economic projections of the NCD burden 48

51 Introduction 48
52 Future rise in NCDs 48
53 Consideration of demographic transition 50
54 Consideration of epidemiological transition 51
55 What can be done to reduce potential impact of NCDs? 53

6 Conclusions
and recommendations 55

61 Introduction 55
62 Limitations 55
63 Key findings 57
64 Conclusions 58
65 Recommendations 60

7 acknowledgments 63

8 References 64

Executive summary

The aim of this report is to conduct an assessment of existing hospital
inpatient data to estimate the economic burden of NCDs in the countries of
Tonga, Vanuatu and Kiribati The term NCD refers to heart disease, cancer,
diabetes, high blood pressure and chronic respiratory diseases, among
others Data were collected from MOH representatives during site visits
between March 2003 and May 2003 Data sought included information on
hospital morbidity, budget reports and prevalence reports on lifestyle
diseases These data have been used in the current project, together with
reasonable assumptions in lieu of missing or incomplete data, to consider
the impact, both current and future, of NCDs
on the burden of disease in
these countries As part of this assessment, estimates of treatment costs
are derived and applied to hospital admissions to calculate total and
average costs of treatment by NCD and non-NCD Further, an attempt has
been made to consider the extent to which tobacco and alcohol use
contributes to the total current, and future, NCD burden in Tonga, Vanuatu
and Kiribati

The results indicate that in each of the countries included, Tonga, Vanuatu
and Kiribati, NCDs currently account for 104, 58 and 81 of all
admissions, respectively Patients being admitted for a NCD are much older
than those presenting for a non-NCD, 16 years older in Tonga and 10 years
older in Kiribati Patients admitted for a NCD also stay in hospital
longer with ALOS for a NCD admission being 92 days, 75 days and 135 days
in Tonga, Vanuatu and Kiribati, respectively This compares with an ALOS
for a non-NCD admission of 49 days in both Tonga and Vanuatu and 93 days
in Kiribati Consistent with this finding, average treatment costs of NCDs
are considerably higher than non-NCD admissions Although NCDs account for
a relatively low proportion of all admissions, they account for
a
disproportionately larger share of all treatment expenditures In Tonga,
Vanuatu and Kiribati NCDs account for 196, 9 and 81, respectively, of
all treatment expenditures To put this in context, in Tonga for example,
one out of every 10 104 patients admitted to hospital are admitted for
a NCD, however, for every 5 dollars spent on treating all patients, one of
these dollars 20, is required to treat the patient with a NCD

The future impact of NCDs on hospital morbidity and treatment cost is
likely to be substantial The transition towards an older population in
these three countries, coupled with greater harm from current risk
behaviours tobacco and alcohol use, physical inactivity etc, suggests
that the burden of disease from NCDs will increase For example, it is
conservatively projected that the costs related to tobacco and alcohol use
alone, will increase from approximately 9 of the NCD treatment budget to
21 by the year 2020 These financial burdens which do not include any
hospital outpatient or social costs to the community eg time off work
etc, will have major impacts on the national economies of these countries
Additionally, as many of the strategic and evidence-based
preventive
approaches are outside of the MOH mandate eg financial, agricultural and
educational policies, it is important for all Government ministries to work
together to address these issues The following recommendations provide a
means by which Governments can begin to make a concerted effort and reduce
the future costs associated with NCD in their country as well as improve
efficiency in budget allocations

Recommendations
Recommendation 1
Improve the reliability of hospital data
Actions
Commit additional resources to MOH to improve entry and monitoring of
hospital data
How best achieved
Maintain adequate staff levels in Statistics Office
Ensure staff are well qualified and undergo ongoing training
Ensure computer hardware and software is state of the art and well-
maintained
Example
Vanuatu Comprehensive Reform Program

Recommendation 2
Develop a clinical costing system
Actions
Commit resources to MOH to link current hospital data collection with
current cost of treatment
How best achieved
Purchase expertise to develop system
Train staff to contribute to dynamic system
Example
Australian Diagnostic Related Groupings adds value to
data collected
at patient level by costing treatment received

Recommendation 3
Improve the monitoring of risk factors for disease
Actions
Commit additional resources to MOH to conduct regular surveys to better
understand the prevalence and behaviours of key risk factors for
disease
How best achieved
Add additional questions to Census
Regular population surveys to capture prevalence of risk factors
Example
Vanuatu non-communicable disease survey report of key risk factors for
NCD
Diabetes survey in Kingdom of Tonga
STEPS surveillance WHO

Recommendation 4
Improve the efficiency of MOH funding
Actions
To assess current spending, identify room for improvement and shift
resources accordingly
How best achieved
Commit additional resources to MOH to engage in Program Budgeting and
Marginal Analysis
Purchase expertise to review current MOH funding
Example
None present but scope exists to conduct for all Ministries

Recommendation 5
Develop a strategic NCD plan for each country
Actions
Based on the STEPS framework, develop an evidence-based strategic plan
to guide implementation on NCDs
How best achieved
Ensure
multi-sectoral participation and commitment across all
government ministries to the development of the NCD plan and its
implementation
Example
Tonga NCD Control Strategy

overview of project and definition of Non-communicable diseases

1 Introduction

The rapid rise of non-communicable diseases NCDs represents one of the
major health challenges to global development It is estimated that by 2020
over 70 of the global burden of disease will be related to NCDs
especially cancer, diabetes, cardiovascular diseases and chronic
respiratory diseases, mental health disorders and injuries This burden of
disease is disproportionately high in poor and marginalised populations and
is contributing to widening health gaps between and within countries,
resulting in enormous human suffering and increased burden on many
countries[1]

The Pacific Action for Health Program PAHP is an AusAID funded initiative
with the goal of contributing to the reduction of NCDs within the
populations of the Pacific Islands Specifically, the purpose of the
program is to enhance the capacity of the Ministries of Health and
communities of the Pacific Islands, initially Kiribati, Tonga and Vanuatu,
to
address NCD priority risk factors and provide health-promoting
environments for young people 10-19 years An important component of the
program is an initial assessment of existing hospital inpatient data to
estimate the current and projected economic burden of NCDs treatment in
these countries As the three participating countries are at varying
stages of both demographic and epidemiological transition, such evidence
will provide an opportunistic assessment of the current and likely future
impact of NCDs on the health system It will also provide a reliable basis
from which to develop and initiate relevant public health measures to
address NCDs, both in these three countries specifically, and other Pacific
Island Countries PIC

The report is structured according to the specific aims of the project
These aims are:
i In conjunction with Ministry of Health MOH representatives in each
country, identify the most reliable sources of hospital data for use in
the assessment of NCD
ii To assess bed-day and direct treatment data relating to the most
commonly occurring NCDs in the Pacific - cardiovascular diseases,
neoplasms, chronic obstructive pulmonary disease, diabetes
and
hypertension
iii Define, where practical, the impact of tobacco and alcohol usage on
subsequent bed day utilisation and treatment aetiological fractions
iv Where practical, provide economic projections on the NCD burden through
to 2020

As a prelude to specifically addressing these aims, the remainder of this
chapter is devoted to outlining the method of disease classification used
in this report, including the International Statistical Classification of
Diseases and Related Health Problems ICD-10; the main types of NCDs
considered in this report, by ICD code; and key risk factors for these
NCDs

2 International classification of diseases

A classification of diseases may be defined as a system of categories to
which morbid entities are assigned according to established criteria
There are many possible axes of classification and the one selected will
depend upon the use to be made of the statistics to be compiled A
statistical classification of diseases must encompass the entire range of
morbid conditions within a manageable number of categories

The Tenth Revision of the International Statistical Classification of
Diseases and Related Health Problems is the latest
in a series that was
formulated in 1893 as the Bertillon Classification or International List of
Causes of Death The major difference between the 9th and 10th versions is
the coding structure Although the traditional ICD structure has been
retained in the 10th version, an alphanumerical coding scheme replaces the
previous numeric one This provides a larger coding frame and leaves room
for future revision without disruption of the numbering system[2]

The main categories reported in ICD-10 are as follows: certain infectious
and parasitic diseases A00-B99; neoplasms C00-D48; diseases of blood
and blood forming organs and certain disorders involving the immune
mechanisms D50-D89; endocrine, nutritional and metabolic diseases E00-
E90; mental, behavioural disorders F00-F99; diseases of the nervous
system G00-G99; diseases of the eye and adnexa H00-H59; diseases of the
ear and mastoid process H60-H95; diseases of the circulatory system I00-
I99; diseases of the respiratory system J00-J99; diseases of the
digestive system K00-K93; diseases of the skin and subcutaneous tissue
L00-L99; disease of the musculoskeletal system and connective tissue M00-
M25; diseases of the genitourinary
system N00-N99; pregnancy, childbirth
and the puerperium O00-O99; certain conditions originating in perinatal
period P00-P96; congenital malformations, deformations, and chromosomal
abnormalities Q00-Q99; symptoms, signs and abnormal clinical and
laboratory findings, not elsewhere classified R00-R99; injury, poisoning
and certain other consequences of external causes S00-T98; external
causes of morbidity and mortality V01-Y98; and, factors influencing
health status and contact with health services Z00-Z99

3 What is a non-communicable disease?

The term NCD refers to a category of diseases that excludes communicable,
maternal and perinatal diseases, injuries and under-nutrition Examples
include heart disease, cancer, diabetes, high blood pressure and chronic
respiratory diseases, among others In the parlance of ICD-10, NCDs
predominantly comprise Group II conditions, as distinct from Group I
maternal and perinatal diseases and diseases related to under-nutrition,
and Group III injuries

4 Key NCDs considered in this report

Diseases of the circulatory system cardiovascular diseases
Cardiovascular disease
The term cardiovascular disease CVD is used to cover all diseases
and
conditions involving the heart and blood vessels The main underlying
problem in CVD is atherosclerosis, a process that clogs blood vessels with
deposits of fat, cholesterol and other substances that have built up in the
inner lining of the vessels It is most serious when it effects the blood
supply to the heart potentially causing angina or heart attack or to the
brain which can lead to stroke[3] A common set of key risk factors for
CVD include obesity, high blood pressure, diabetes, tobacco consumption,
high fat consumption and physical inactivity

ICD-10 codes for diseases of the circulatory system encompass: acute
rheumatic fever I00-I02; chronic rheumatic heart diseases I05-I09;
hypertensive diseases I10-I15; ischaemic heart disease coronary heart
disease I20-I25; pulmonary heart disease and diseases of pulmonary
circulation I26-I28; other forms of heart disease I30-I52;
cerebrovascular diseases stroke I60-I69; diseases of the arteries,
arterioles and capillaries I70-I79; diseases of veins, lymphatic vessels
and lymph nodes, not elsewhere classified I80-I89; and, other and
unspecified disorders of the circulatory system I95-I99 The most common
types of cardiovascular
disease are as follows

Hypertensive diseases
Hypertension is a condition in which an individual has a higher blood
pressure than is considered normal The elevated blood pressure is caused
by an increase in peripheral resistance resulting from constriction or
narrowing of peripheral blood vessels In the majority of patients, the
aetiology of the condition is unknown and develops without apparent cause
Hypertension can be benign slow progression, or malignant rapid
progression, a form accompanied by severe vascular damage[4] Hypertension
is a clinically silent condition and often goes undiagnosed until after it
becomes manifested in an acute clinical condition including heart attack,
stroke, kidney failure or blindness Risk factors for hypertension include
behavioral/lifestyle aspects such as high dietary salt and alcohol
consumption, stress, physical inactivity and obesity There is also a
hereditary component

Ischaemic heart disease coronary heart disease
Coronary heart disease often results in sudden death It consists mainly
of acute myocardial infarction heart attack and angina A heart attack
is a life-threatening emergency that occurs when a vessel supplying blood
to the
heart muscle suddenly becomes blocked by a blood clot Angina is
temporary chest pain or discomfort when the hearts own blood supply is
inadequate to meet extra needs[3]

Cerebrovascular diseases stroke
Stroke occurs when an artery supplying blood to the brain suddenly becomes
blocked or bleeds, often causing paralysis of parts of the body or speech
problems It includes ischaemic stroke blockage, haemorrhagic stroke
bleeding, transient ischaemic stroke and other cerebrovascular diseases
Evidence from Australia suggests that 75 of stroke victims suffered a
disability that required assistance with self-care, mobility or
communication[5]

Diseases of the respiratory system
Acute respiratory infections, which include influenza and pneumonia and
other upper and lower respiratory tract infections, are commonly related to
communicable diseases The specific ICD-10 codes for diseases of the
respiratory system encompass: acute upper respiratory infections J00-J06;
influenza and pneumonia J10-J18; other acute lower respiratory infections
J20-J22; other diseases of upper respiratory tract J30-J39; chronic
lower respiratory diseases J40-J47; lung diseases due to external causes
J60-J70; other
respiratory diseases principally affecting the
interstitium J80-J84; suppurative and necrotic conditions of lower
respiratory tract J85-J86; other diseases of pleura J90-J94; and other
diseases of the respiratory system J95-J99

Chronic obstructive pulmonary disease
The NCD related respiratory diseases can be classified under the umbrella
of chronic obstructive pulmonary disease COPD Chronic obstructive
pulmonary disease COPD is a permanent and typically progressive disease,
where damage to the lungs obstructs oxygen intake and causes breathlessness
with exertion and limitation of exercise capacity In severe cases,
breathlessness may occur with little or no exertion[3] Symptoms of copd
include shortness of breath, wheezing, coughing and mucus production, all
of which tend to become progressively worse over time Disease progression
is marked by a steady decline in lung function, punctuated with periodic
exacerbations, either through infection for which copd is a predisposing
factor or inflammation alone

Chronic bronchitis J42 and emphysema J43 are the two main components of
COPD Another category, other COPD J44, is included in ICD-10 coding
It is important to note that asthma
J45 is not considered as a COPD due
to the fact that lung function tends not to decline over time As such it
is not included in subsequent analyses Consequently, only disease
categories J42-J44 inclusive will be included in subsequent analysis
Neoplasms Cancer
Cancer is a diverse group of diseases in which some of the bodys cells
become defective, begin to multiply out of control, can invade and damage
the area around them, and can also spread metastasize to other parts of
the body to cause further damage Cancer impacts on almost all sites of
the body with lung, breast, bowel and prostate among the most commonly
diagnosed cancers in the western world[3]

The specific ICD-10 codes for neoplasms encompass: malignant neoplasms of
lip, oral cavity and pharynx C00-C14; malignant neoplasms of digestive
organs C15-C26; malignant neoplasms of respiratory and intrathoracic
organs C30-C39; malignant neoplasms of bone and articular cartilage C40-
C41; malignant neoplasms of skin C43-C44; malignant neoplasms of
mesothelial and soft tissue C45-C49; malignant neoplasms of breast C50;
malignant neoplasms of female genital organs C51-C58; malignant neoplasms
of male genital organs C60-C63;
malignant neoplasms of urinary tract C64-
C68; malignant neoplasms of eye, brain and other parts of central nervous
system CNS C69-C72; malignant neoplasms of thyroid and other endocrine
glands C73-C75; malignant neoplasms of ill defined, secondary and
unspecified sites C76-C80; malignant neoplasms of lymphoid,
haematopoietic and related tissue C81-C96; malignant neoplasms of
independent primary multiple sites C97; in situ neoplasms D00-D09;
benign neoplasms D10-D36; and, neoplasms of uncertain or unknown
behaviour D37-D48 All these disease categories will be included in
subsequent analysis

Endocrine, nutritional and metabolic disorders
ICD-10 codes for endocrine, nutritional and metabolic disorders encompass:
disorders of the thyroid gland E00-E07; diabetes mellitus E10-E14;
other disorders of glucose regulation and pancreatic internal secretion
E15-E16; disorders of the endocrine glands E20-E25; malnutrition E40-
E46; other nutritional deficiencies E50-E64; obesity and other hyper
alimentation E65-E68 and metabolic disorders E70-E90 In accordance
with the project brief, only diabetes mellitus E10-E14 will be included
in subsequent analysis

Diabetes mellitus
DM is a long term
chronic condition in which blood glucose levels become
too high because the body produces little or no insulin or cannot use
insulin properly Insulin is a hormone produced by the pancreas that helps
the body use glucose Over the course of the disease, diabetes can lead to
long term damage to various parts of the body, especially the heart and
blood vessels, eyes, kidneys and nerves Diabetes can also contribute to
many pregnancy-related complications for the mother and baby, both before
and after birth It is largely irreversible once established, and is a key
risk factor for CVD, including heart attack and stroke Because the common
feature of diabetes is high blood glucose, it is often mistakenly thought
to be a single disease

There are two main types of diabetes, Type 1 and Type 2, each with
different casual mechanisms[3] Type 1 diabetes insulin-dependent
diabetes mellitus or IDDM is marked by a total or near lack of insulin
It results from the body destroying its insulin-producing cells in the
pancreas People with this form of diabetes require daily insulin therapy
to survive It is the most common cause of childhood diabetes and accounts
for around 10-15 of all people
with diabetes Type 2 diabetes non-
insulin dependent diabetes mellitus or niddm is marked by reduced levels
of insulin, or the inability of the body to use insulin properly insulin
resistance

This disease is most common among people aged 40 years and over and
accounts for 85-90 of all people with diabetes Many people with this
form of diabetes eventually need insulin therapy to control their blood
glucose levels Risk factors for type 2 diabetes include obesity, poor
diet and lack of exercise niddm also has a strong genetic component in
some Pacific communities, with some ethnic groups being affected more than
others In contrast to iddm, the decline in insulin production and
effectiveness in NIDDM is progressive rather than absolute, and in some
cases this decline can be prevented or reversed with appropriate changes in
diet, weight and physical activity[4]

5 Key risk factors for NCDs in this report as per project brief

Tobacco and alcohol
Tobacco consumption
Cigarette smoking and other forms of tobacco consumption are important risk
factors for CVD, COPD, cervical cancer and lung cancer of which it is the
most important single cause Smoking prevalence is high in many PICs
and
has been increasing over the past 10 to 20 years In many PICs, smoking
prevalence is now greater than in Australia, New Zealand and the United
States Of particular concern is the uptake of smoking by young people
More data on smoking prevalence are discussed in Chapter 4 Due to the
fact that the risk of smoking-related diseases increases with how long an
individual has smoked, and since smoking is an addictive habit, these data
point to the likelihood of a substantial increase in smoking-related
diseases as these cohorts age[6]

Alcohol consumption
In general, higher overall levels of consumption in a population are
associated with higher levels of alcohol-related problems Alcohol
consumption has health and social consequences via intoxication
drunkenness, high dependence habitual, compulsive, long-term heavy
drinking and other biochemical effects Intoxication is a powerful
mediator for acute outcomes, such as motor vehicle accidents and domestic
violence, and is associated with many chronic health and social problems
There are casual relationships between average volume of alcohol
consumption and more than 60 types of disease and injury Most of these
relationships are
detrimental, but there is some evidence of beneficial
relationships with coronary heart disease, stroke and diabetes, provided
low average volume of consumption is combined with non-binge patterns of
drinking

Other lifestyle risk factors not factored into subsequent analysis
Obesity
Obesity is a risk factor for CVD, diabetes and hypertension, as well as
cancers of the breast, and colon It is defined on the basis of the body
mass index BMI, a figure obtained by dividing an individuals weight in
kilograms by the square of his or her height in meters Evidence suggests
the prevalence of overweight and obesity combined is greater than 50
percent in at least 10 PICs, and rates as high as 75 percent - among the
highest in the world - have been reported in Nauru, Samoa, Tonga, Cook
Islands and French Polynesia[4] These figures have been used to justify
the description of obesity as an epidemic in PICs Underlying risk
factors for obesity include poor diets and physical inactivity, both of
which are discussed separately below Also relevant is the issue of body
image In many PICs, cultural norms support the pursuit of increased
weight, and obesity is considered a sign of wealth or
social status
Obesity tends to be more prevalent in urban than rural populations and is
more common among women than men; although these trends are changing due to
increases in rural and male obesity in many PICs Obesity typically peaks
in the 20 to 45 year age group, though wealthier PICs have been
experiencing an increase in childhood obesity in recent years A lower
mean BMI is noted in older age groups, probably reflecting the better
chances of survival with a lower BMI than any trend to lose weight with
age[6]

Poor diet
Dietary factors are implicated in the origin of obesity, diabetes,
hypertension, CVD and certain cancers-including bowel, breast and prostate
cancers For a large percentage of people, the modern Pacific Island diet
has changed from one consisting primarily of root vegetables, coconut,
fresh fish and green leaves, to one with a lower vegetable content and a
higher intake of rice, bread, tinned fish and meat, sugar, salt and
alcohol As a result, the consumption of energy calories, salt, animal
fats, sugar and alcohol has increased, leading to a greater prevalence of
obesity due to excess energy from fats, sugar and alcohol, diabetes same
factors, hypertension
salt, alcohol and CVD all At the same time,
the consumption of fiber, antioxidants and trace minerals has generally
decreased, leading to an increase in bowel low fiber, high fat and other
cancers low antioxidants, high fat, high alcohol These changes have
been most evident in urban settings but are occurring in many rural
communities as well[6]

Physical inactivity
Physical inactivity is a risk factor for obesity, hypertension and CVD In
the case of obesity, a lack of physical activity, especially when combined
with a high calorie diet, leads to a surplus of calories that are
subsequently converted into fat as an energy storing mechanism Physical
inactivity can contribute to high blood pressure either directly or through
its effect on BMI since obesity increases blood pressure Its effects on
CVD are mediated through a similar set of pathways In comparison with
other NCD risk factors, relatively little is known about the extent to
which patterns of physical activity have changed in recent years in PICs,
although there is considerable anecdotal evidence that the level of
physical activity in PICs has declined with the advent of westernisation

MOST RELIABLE SOURCES OF
hospital data

1 Introduction

The specific aim of this chapter is to identify the most reliable sources
of hospital data for assessing the impact of NCDs on health systems Since
the reliability of the data depends, in part, on existing health care
delivery infrastructure, this chapter will also consider separately for
Tonga, Vanuatu and Kiribati: the health care services available; current
expenditure on health care; data collection methods used to track an
inpatients episode of care from admission to discharge; the clinical
costing systems used to track resource use for inpatients; and finally, an
assessment of the most suitable data on which to base subsequent economic
impact analyses

2 Health care provision and data collection methods used in Tonga

Health care provision
The Kingdom of Tonga has a population of approximately 100,000 spread over
36 inhabited islands[7] The country is divided into three main islands
Tongatapu, Vavau and Haapai and the health districts of the MOH are
aligned with the national structure The largest and most populous
district is Tongatapau, in which the capital, Nukualofa, is situated
This district also dominates the co-ordination and provision
of health
care It comprises the island of Tongatapau and the small adjacent island
of Eua[8]

The national referral hospital, Vaiola, is situated on the outskirts of
Nukualofa and the MOH is in the grounds of Vaiola hospital All
specialist services are provided through Vaiola hospital, including
outpatient and dental services Diagnostic, pharmaceutical and other
clinical support services are also provided at Vaiola hospital
Environmental health, health promotion and community health services are co-
ordinated from the MOH through the Public Health Division Villages on
Tongatapau are serviced by 7 community health centres staffed by a health
officer and public health nurses In addition, there are 19 Maternal and
Child Health MCH clinics run by public health nurses alone Health
services in Eua comprise one 16 bed hospital Niueiki and three MCH
clinics While Eua is officially linked with Tongatapau, it is managed as
a separate health district with the Chief Medical Officer reporting
directly to the Director of Health
The health districts of Vavau and Haapai are each serviced by a community
hospital, Prince Wellington Ngu 61 beds and Niuui 28 beds
respectively These hospitals
are supported by small pathology, radiology,
pharmacy and dental services, and both include an outpatient department
MCH, environmental health and community health services for these districts
are co-ordinated from the hospital Prince Ngu hospital is supported by 3
health centres and 5 MCH clinics on Vavau itself and is also responsible
for the two health centres on Niuafoou and Niuatoputapu Niuui hospital
is supported by 2 health centres and 3 MCH clinics and is responsible for
some isolated island communities[8]

Health care expenditure
Government expenditure on health has increased by 14 from 7372m in
1997/98 to 8414m in 2001/2002 The 2001/02 health budget was equivalent
to around 10 of the total government expenditure 872m for that year
and included an additional 25m made available from overseas aid funding
to supplement the health budget In the year 2001/02, expenditure on
salaries accounted for around 68 of the budget 57m, goods and services
consumed a further 26 22m and the remaining 6 was spent on
maintenance and other items[8]

Estimates from the WHO indicate that Tonga has the lowest per capita
expenditure on health US73 compared with other Pacific Nations[9]
In
spite of this, Tonga does spend a greater proportion of its gross
domestic product GDP on health than other countries across the region,
with the exception of the Marshall Islands and Kiribati

Data collection methods
Hospital patient records usually contain information about a patients
diagnosis and procedures performed during the hospital stay This is no
exception in the Kingdom of Tonga, with admission and discharge data forms
collecting data on patient admissions, ward occupancy, average length of
stay, diagnostic tests performed and medications provided Tonga uses ICD-
10 coding At Vaiola hospital, diagnosis is coded by the Medical Records
staff and then submitted to the Statistics Unit All other facilities
outside Vaiola send admission and discharge information to the Statistics
Unit at the MOH where staff code individual patient data The Health
Statistics Unit is responsible for entering the data from all notifications
provided by outlying areas and the Public Health Division, as well as data
analyses and feedback of statistical reports to all sections of the
Ministry

Clinical costing systems
In general, clinical costing systems provide a bottom up assessment
of
resource use which is measured and valued accordingly[10] Ideally,
resource use should monitor the spectrum of care provided, diagnostic tests
performed and medications provided In most developed countries, data from
admission / discharge forms are matched with the value of the resources
utilized in provision of care to a particular patient, to provide an
estimate of the cost of care In Australia, this classification system is
referred to as Australian Refined Diagnostic Related Groups AR-DRGs
This classification provides a means to determine the number and type of
patients admitted and treated in a hospital ie its case mix and
relating this information to the resources required by the hospital The
AR-DRGs, therefore, provide a standardized method for assessing the
performance of hospitals and facilitates future planning and funding in the
provision of hospital services

Data of this nature are not currently available in Tonga While data are
collected on disease classification and treatment procedures, only limited
patient data are entered into the database which impedes the ability to
track and compare resources used by patients, both between and within
disease
classifications Further, no medical benefits schedule is
currently in place to reliably ascertain the relative values of particular
health care procedures

In the absence of a complete clinical costing system, a general indication
of the costs associated with treatment provision can be obtained by
combining available data such as the number of hospital admissions per
disease with estimates based on reasonable assumptions, in lieu of missing
data This approach is adopted in this project with details provided in
Chapter 4

An alternate approach was adopted in an earlier study of the Economic costs
of NCDs in the Pacific[11] and reported in Khaleghian 2001[6] The
authors used a WHO estimate of the average cost per inpatient bed-day for
all diseases, not just NCDs at WST 235 USD 7750[12] This approach,
while reasonable, does not utilize available local data, and, hence, may
not provide an accurate assessment of local conditions

Most reliable data
At the time of site visit, in May 2003, the most accurate and reliable
hospital morbidity data available were for the year 2001 Budget estimates
and MOH staff lists were provided for the year 2002/03 These data sets
are used in
subsequent analyses

3 Health care provision and data collection methods used in Vanuatu

Health care provision
The majority of health care provided in Vanuatu is through the MOH The
MOH has been restructured under the Comprehensive Reform Program CRP and
now comprises four divisions: the Southern Health Care group including
Shefa and Tafea, the Northern Health Care group including Torba, Sanma,
Malampa and Penama; Public Health; and, Planning and Administration

The Vanuatu health system comprises a four tier structure:
i Aid posts 180 in the year 2001 Staffed by community volunteers
provided with six weeks of training, aid posts treat minor ailments and
promote family planning and health education Aid post workers are
trained to detect serious complaints and to refer patients as early as
possible to dispensaries
ii Dispensaries 79 in the year 2001 Each service a population of around
3,000 with basic facilities Some have 1-2 beds Services include
maternal and child health, family planning, immunizations, health
education, and detection and referral of serious diseases
iii Health centers 23 in the year 2001 With about ten beds, the centres
are
intended to be run by a nurse practitioner, with 1-2 nurses They
service a population of 3,000-8,000, supplementing the dispensaries and
providing beds for inpatients and birthing
iv Hospitals 5 in the year 2001 A hospital is located in each province,
except Torba Three of the five hospitals are designated as provincial
hospitals, Lenakel hospital in Tafea province, Norsup Hospital in
Malampa province, and Lolowai hospital in Penam province Two other
hospitals, Vila Central Hospital VCH in Shefa province and Northern
Region Hospital NRH located in Sanma province, provide referral
services for the provincial hospitals and are designated as referral
hospitals[13]

Each hospital provides the following services:
Lenakel provides both outpatient and inpatient facilities including
maternity and delivery, female, male, pediatric, tuberculosis TB;
laboratory, pharmacy, theatre, radiology, dental; and, general
administrative support services which include office administration,
maintenance, kitchen, laundry and cleaning;
Norsup provides the same services as Lenakel, with the addition of
providing a surgical ward
Lolowai
provides similar outpatient and administration services but
inpatient services comprise only a general and maternity ward
Vila Central Hospital VCH is the main referral hospital, providing
all major health care facilities: outpatient, maternity, pediatric,
surgical, medical, psychiatry, theatre, laboratory, radiology,
pharmacy, ear/nose/throat, physiotherapy, orthotics, dental and general
administration The central medical store is also located within the
hospital compound
Northern Region Hospital NRH provides similar services to VCH with
the exception of physiotherapy and orthotics[14]

In addition to government run health services, private health delivery is
expanding A number of doctors practice western medicine, predominantly in
Port Villa and Luganville There are also numerous traditional
practitioners and birth assistants in rural and urban areas Private
doctors typically provide services for tourists, expatriates and an
increasing number of ni-Vanuatu Consultation fees are of the order of
Vt1,500 while traditional pharmacies can charge more The growth of these
suggests that many people are prepared to pay for what they perceive to be
a
more efficient or effective service

Health care expenditure
The MOH budget for 2002/03 was Vt9163m[15] The proportion of the
national budget spent on health has not changed greatly since 1990, though
the trend over the past four years has been downwards from a peak of 124
in 1997 to 103 in 2001[13] Expenditure per capita Vt4909 per person
can be calculated by dividing expenditure on health Vt9163m with
population estimates 186,680[16]

Data collection methods
Admission forms capture data pertaining to patient characteristics,
clinical diagnosis and discharge status Three of the five hospitals
complete all admission forms and send them to a National Statistics Officer
for coding and data entry Two of the hospitals, VCH and NRH, enter the
data themselves and forward complete data sets to a National Statistics
Officer for data checking All data is then entered into the Vanuatus
Health Information System It has been documented, however, that the
coding and entry of hospital statistics has, in the past, been very poor
For example, in the Second National Health Development Plan report, it was
noted that for the NRH, the officer appointed to be responsible for
hospital statistics
was not well qualified, necessitating immediate
replacement of the officer with a more qualified person[14] This implies
that the admission forms processed are likely to represent an underestimate
of the true healthcare services utilization rate, due to possible leakages
in data collection, coding and entry However, the CRP appears to have
improved the efficiency and reliability of data collection such that the
database will provide an adequate overall picture of NCD admittances in
Vanuatu

Clinical costing systems
The situation of clinical costing in Vanuatu is similar to that in Tonga
That is, although each hospital separation is tracked using a detailed
admission form that captures, among other things, patient length of stay,
diagnosis and procedures performed, not all of this information is stored
electronically for each patient Data were consistently available on
length of stay and diagnosis by ICD-10, but not for resource use by
particular disease classifications

As for Tonga, in the absence of a complete clinical costing system, a
general indication of the costs associated with treatment provision can be
obtained by the use of available data such as the number of
hospital
admissions per disease, complemented by estimates for missing data, based
on reasonable assumptions Details of this method are provided in Chapter
4

Most reliable data
At the time of the site visit in April 2003, the most accurate and reliable
hospital morbidity data available were for the year 2002 Budget estimates
and MOH staff lists were provided for the year 2002/03 These data sets
are used in subsequent analyses

4 Health care provision and data collection methods used in Kiribati

Health care provision
The Republic of Kiribati comprises 33 small and scattered atolls in the
central South Pacific Ocean The total land area of 811 square kilometers
is spread over 35 million square kilometers of ocean The economy is
primarily based on copra, fish and workers remittances On the Human
Development Index, Kiribati ranks 11th out of the 14 countries in the
Pacific region and 129th in the world In terms of infant mortality and
child morbidity, per capita GDP and living conditions, particularly in
regard to water and sanitation, Kiribatis development indicators are among
the lowest in the Pacific region

The population of Kiribati numbers around 84,000 The eastern islands
are
very sparsely populated, making delivery of services difficult and costly
because of communication and transport difficulties Most people live on
the western islands of Kiribati, with over one third of the national
population living on the capital atoll of South Tarawa[17]

The MOH is responsible for the health care system in Kiribati The overall
outcome of the MOH is to achieve the improvement of health to all I-
Kiribati This will be approached through: a focus on primary health care
as the lead mechanism for delivering health services; rationalization of
curative services through greater reliance on cost recovery, manpower
planning and maintenance of facilities; involvement of churches, local
government and community organizations in health care delivery; and
encouraging development of a private health care market

Health care expenditure
According to the Republic of Kiribati 2002/03 budget report, operating
expenditure for the financial year 2002/03 was AUD77,911,230 of which
expenditure to the MOH accounted for 116 or 9,043,867[18] With an
estimated population in 2000 of 84,494[17], this represents per capita
expenditure on health of 107

Data collection
methods
Information from inpatient admission forms is transferred to hospital
discharge forms, where relevant ICD coding occurs Hospital discharge
forms collect data on, among other things, patient characteristics, ward
admitted, date of admission and discharge, principle diagnosis and
procedures performed A medical record officer enters the relevant ICD
code and forwards completed discharge forms to the Statistics Office for
electronic entry

Clinical costing systems
The situation of clinical costing in Kiribati is similar to that in Vanuatu
and Tonga That is, although each hospital separation is tracked using a
detailed admission form that captures relevant information, not all of this
information is stored electronically for each patient Available data
includes length of hospital stay and diagnosis by ICD-10, but excludes
resource use by particular disease classification The approach taken to
cost clinical activity for Kiribati is detailed in Chapter 4

Most reliable data
At the time of the site visit in March 2003, although staff were using the
ICD-10 coding system, the most accurate and reliable hospital morbidity
data available were for the year 1997, which was coded
using ICD-9 In
terms of MOH expenditure, a copy of the proposed 2002/2003 budget was
sighted and used in these analyses, but was not made available in hard
copy Although this budget had not been ratified by the Government, it
provides a good approximation of likely MOH expenditure, as the estimates
do not vary substantially from previous years

resource use and cost associated with ncd treatment

1 Introduction

The purpose of this chapter is to address the second aim of this project:
to assess bed day and direct treatment data relating to the most commonly
occurring NCDs in the Pacific - cardiovascular diseases, neoplasms, chronic
obstructive pulmonary disease, diabetes and hypertension In order to
achieve this aim it is useful, in the first instance, to consider hospital
morbidity and cost data by disease classification and secondly, to
elaborate on specific NCD categories The quantification of resource use
and cost associated with the treatment of disease will be presented
individually for each country All costs presented are expressed in local
currency

2 Resource use and cost associated with treatment of disease in Tonga

Hospital morbidity and cost by disease and
NCD classification
Data obtained from the Kingdom of Tonga Hospital inpatient database for the
year 2001 was sorted by ICD-10 code Table 1 provides an overview of these
data and presents results for each disease category in terms of: the total
number of separations or episodes of care; the proportion of all
separations; average age; separation by gender; ALOS; and, average cost
Table 2 follows a similar presentation style to Table 1 in considering
hospital morbidity and cost data by NCD The derivation, and discussion,
of average cost is outlined in the following section quantifying the cost
of treatment

As shown in Table 1, a total of 8,776 admissions were registered on the
provided database for Tonga The results suggest that for the year 2001,
the main diagnoses for hospital admission were: pregnancy, childbirth and
the puerperium 344 of total; diseases of the respiratory system 117
of total; diseases of the digestive system 82 of total; and injury,
poisoning and external causes 80 of total Diseases of the circulatory
system and neoplasms accounted for 46 and 18, respectively, of total
admissions The average age of all patients was 30 years of age while the
ALOS across all
separations was 49 days Patients diagnosed with
endocrine, nutritional and metabolic disorders had the longest ALOS 127
days, followed by mental, behavioural disorders 101 days and neoplasms
96 days

Table 2 shows a total of 911 NCD-related admissions, accounting for 104
911 / 8776 of all admissions in 2001 Diabetes was the most prevalent
NCD with 273 admissions representing 30 273 / 911 of all NCD admissions
Diseases of the circulatory system accounted for 44 403 / 911 of NCD
admissions The two most common circulatory conditions, other forms of
heart disease 140 admissions and hypertensive disease 105 admissions,
together comprised 61 245 / 403 of all circulatory diseases
Neoplasms were the next most common NCD representing 173 of all NCD
admissions The most common neoplasms were malignant neoplasms of
digestive organs 34 admissions - 11 for stomach cancer, 11 for liver
cancer and 7 for colon cancer and benign neoplasms 27 admissions - 11 for
benign lipomatous neoplasm Less common neoplasms were breast cancer 19
admissions, lung cancer 8 admissions, and prostate cancer 8
admissions There were a total of 75 admissions for unspecified COPD
comprising 82 of all NCD admissions
The average age of all patients
diagnosed with an NCD was 46 years of age with an ALOS of 92 days, almost
twice that for all disease categories 49 days Within the NCD group,
patients with diabetes had the longest ALOS at 147 days followed by
neoplasms 96 days and unspecified COPD 8 days

Quantifying the cost of treatment
As alluded to earlier, in the absence of a sophisticated clinical costing
system, a decision was made to estimate the cost of a hospital separation
using available information, predominantly budget and staff records The
process for deriving this estimate involved three main steps

First, budget estimates had to be separated for inpatient and outpatient
stay; as well as hospital and non-hospital care In addition, although
management and support staff such as cleaners, maintenance and drivers
may be paid from separate program accounts, a certain portion of their
budget needs to be apportioned to hospital care Management and support
staff provide services that ensure the hospital, as well as the MOH in
general, is a well-functioning, well-supervised and well-maintained service
to the community The 2002/03 budget was used as a starting point and
augmented or
reduced with apportioned costs from, or to, budgets currently
attributable to other divisions For example, medical records are
currently part of the MOH budget As the function of medical records would
be required in a hospital, a proportion of the budget has been attributed
to Vaiola hospital Rates of apportionment are, generally, consistent with
those used in the recent Kingdom of Tonga Health Sector Reform Project[19]
The Health Sector Reform project focused on Vaiola hospital as a whole,
therefore, further apportioning was required to exclude outpatient activity
from budget estimates, as well as include inpatient focused activity from
other hospitals

Second, using the MOH employment staff list for the year 2002/03, the total
number of staff working in the medical services division 444 excluding
emergency and outpatient activity 21, leaving predominantly inpatient
activities 423, was divided by the total number of staff currently
employed by the MOH 695, to estimate the proportion of MOH staff engaged,
predominantly, in inpatient activity 61 This rate was then applied to
budget estimates that were not apportioned in the Health Sector Reform
project, notably, general
medical, laboratory, X-Ray and radiology,
pharmacy, maintenance, medical administration, other islands with hospitals
and the development fund

As a consequence of these two steps, the budget estimate for inpatient
activity for all hospitals is estimated at 6,740,788 or 52 of the total
MOH budget 12,898,852 for the year 2002/03 Given that the medical
division employs approximately 63 of all MOH staff and accounted for close
to 44 of the total MOH budget for the same year, the estimate of 52 of
the total budget for inpatient activity including administrated support
appears reasonable

Finally, in order to cost an inpatient stay it is necessary to calculate an
inpatient cost per day Dividing total inpatient cost 6,740,788 with
total inpatient length of stay of 42,885 days, results in a cost per
inpatient day of 157 This compares favourably with WHO estimates of
average cost per inpatient bed-day for all diseases, not just NCDs at WST
235 USD 7750 The value of 157 is applied in this study to estimate
the resource implications associated with NCD admissions compared with
admissions for other illnesses/diseases

Cost associated with treatment
The last column in Table 1 and Table 2
provide the average cost of
treatment by disease and NCD category respectively The average total cost
of a hospital admission for any disease is estimated at 768 Table 1
compared to 1,447 for a NCD admission Table 2 The average cost of the
most prevalent diagnoses for admission: pregnancy, childbirth and the
puerperium 481; diseases of the respiratory system 687; and, diseases
of the digestive system 640 are all substantially lower than the average
cost of the most prevalent NCDs: diseases of the circulatory system 884,
diabetes 2,306, unspecified COPD 1,262 and neoplasms 1,501 This
suggests that the average hospital admission costs associated with NCDs are
approximately two to six times higher than those attributable to the most
prevalent non-NCDs

Total cost of treatment by disease or NCD may be estimated by multiplying
the number of admissions for a specific disease by the respective average
cost For example, the total cost of neoplasms can be estimated at
237,189 158 admissions x 1,5012 The application of this method to
NCD and all admissions suggest that total treatment costs for all NCDs was
1,318,136 while, treatment costs for all conditions was 6,740,788
Numbers
presented in the Tables have been rounded so estimates derived by
the reader may differ slightly from these totals

These data suggest that NCDs accounted for 104 911 / 8776 of all
hospital admissions while the treatment of these diseases represented 196
1,318,136 / 6,740,788 of all inpatient hospital-based expenditures

—–Tables 1 and 2 about here—–

3 Resource use and cost associated with treatment of disease in Vanuatu

Hospital morbidity and cost by disease and NCD classification
Data obtained from the National Statistics Officer for the year 2002 were
analysed in an identical way to data from Tonga Table 3 provides an
overview of these data and presents, for each disease category, results in
terms of: total number of separations; proportion of all separations;
proportion of patients over the age of 15 years; separation by gender;
ALOS; and, average cost Table 4 follows a similar presentation style to
Table 3 in considering hospital morbidity and cost data by NCD The
derivation, and discussion, of average cost is outlined in the next section
quantifying the cost of treatment

Table 3 shows a total of 10,216 admissions were registered on the
provided
database for Vanuatu for the year 2002 The main diagnoses for hospital
admission were: pregnancy, childbirth and the puerperium 288 of total;
certain infections and parasitic diseases 176 of total; diseases of the
respiratory system 139 of total; and, injury, poisoning and external
causes 98 of total Diseases of the circulatory system and neoplasms
accounted for 3 308 / 10216 and 11 115 / 10216, respectively, of
total admissions Seventy-four percent of all patients were over 15 years
while the ALOS was 49 days Patients diagnosed with endocrine,
nutritional and metabolic disorders had the longest ALOS 117 days,
followed by certain conditions originating in the perinatal period 111
days

Table 4, shows a total of 595 NCD-related hospital admissions accounting
for 58 595 / 10216 of all admissions in 2002 Diseases of the
circulatory system accounted for 52 308 / 595 of all NCD admissions
Table 4 The most two most common diseases of the circulatory system,
hypertensive disease 163 admissions and other forms of heart disease 60
admissions, were together responsible for 724 223 / 308 of all
admissions for circulatory system disease Neoplasms were the next most
prevalent
NCD category with 115 admissions representing 193 of all NCD
admissions The most common neoplasms were: benign neoplasms 27
admissions; neoplasms of an uncertain or unknown behaviour 27
admissions; and malignant neoplasms of female genital organs 15
admissions Less common cancers were malignant neoplasms of digestive
organs 11 admissions of which 7 were related to liver cancer, breast
cancer 5 admissions and lung cancer 1 admission There were 50
admissions for unspecified COPD representing 84 50 / 595 of all NCD
admissions Ninety-six percent of all patients admitted for an NCD were
over 15 years while the ALOS was 75 days This compares with the ALOS of
all separations of 49 days For the key NCD groups, patients with
diabetes had the longest ALOS at 133 days followed by neoplasms at 81
days

Quantifying the cost of treatment
A three-step procedure similar to that used to estimate hospital cost per
day for Tonga was applied for Vanuatu The 2002/03 budget was used in
conjunction with the chart of accounts in an attempt to net out the cost of
providing inpatient care The chart of accounts provided detail on each
health-related program, activity and account range The programs
included:
Southern Health Care SHC; Northern Health Care NHC; Public Health; and,
Planning and Administration Within each program a number of activities
had been costed For example, the SHC program contained cost breakdowns
for the Director, Vila Central Hospital VCH, Lenakel Provincial Hospital,
Shafea province and Tafea province As a consequence, expenditure data
were available on all referral and provincial hospitals in Vanuatu Using
details of staff employed at MOH contained in the 2002/2003 budget
estimates, the number of staff working in either a referral or provincial
hospital N289 was divided by the total number of MOH salaried staff
N540, to estimate the proportion of MOH staff engaged, predominantly, in
hospital activity 54 This rate was then applied uniformly across all
programs and their respective activities to provide a proxy for the cost of
inpatient hospital care As a consequence of this method, the budget
estimate for inpatient activity for all hospitals is estimated at
Vt490,400,889, or 54 of the total MOH budget Vt490,400,889 for the year
2002/2003 Although there are obvious limitations to this method
discussed in Chapter 6, this estimate 54 of total
budget attributable
to inpatient care is similar to the estimate derived for Tonga, which was
developed using a more rigorous procedure

A cost per inpatient stay was calculated by dividing total inpatient cost
Vt490,400,889 with total inpatient length of stay of 49,620 days,
resulting in a cost per inpatient day of Vt9,883 The value Vt9,883 is
applied in this study to estimate the resource implications associated with
NCD admissions compared with admissions for other illnesses/diseases

Cost associated with treatment
The last column in Table 3 and Table 4 provide the average cost of
treatment by disease and NCD category, respectively The average cost of a
hospital admission for all diseases is estimated at Vt48,003 compared to
Vt73,799 for a NCD admission The average costs for the majority of the
most prevalent diagnoses for admission: pregnancy, childbirth and the
puerperium Vt34,124; diseases of the respiratory system Vt41,922; and,
injury, poisoning and external causes Vt38,579, are all substantially
lower than the average cost of the most prevalent NCDs: diseases of the
circulatory system Vt57,951 and neoplasms Vt80,252 with the exception
of infectious and parasitic
diseases Vt63,928

The total treatment cost for all NCD admissions is estimated at
Vt43,909,845 Vt7379806 x 595 while treatment expenditure for all
conditions was Vt490,400,889 Numbers presented in the Tables have been
rounded so estimates derived by the reader may differ slightly from these
totals

These data suggest that NCDs accounted for 58 595 / 10216 of all
hospital admissions while the treatment of these diseases amounted to the
equivalent of 9 Vt43,909,845 / Vt490,400,889 of all inpatient hospital
based expenditures

—–Tables 3 and 4 about here—–

4 Resource use and cost associated with treatment of disease in Kiribati

Hospital morbidity and cost by disease and NCD classification
Data obtained from the Health Information Officer for the year 1997 were
analysed in a similar way to the data for Tonga and Vanuatu Given these
data were coded according to ICD-9 coding, the numerical coding of disease
categories differed marginally Nevertheless, the divergence between ICD-9
and ICD-10 is not significant and similar result tables were able to be
generated Table 5 provides a summary of the findings in terms of main
disease category as per ICD-9 code,
total number of separations,
proportion of all separations, average age, separation by gender, ALOS and
average cost Table 5 follows a similar presentation style to Table 4 in
considering hospital morbidity and cost data by NCD The derivation, and
discussion, of average cost is outlined in the following section
quantifying the cost of treatment

Table 5 shows a total of 3,872 admissions were registered on the provided
database for Kiribati The results suggest that for the year 1997, the
main diagnoses for hospital admission were: pregnancy, childbirth and the
puerperium 301 of total; certain infections and parasitic diseases
131 of total; diseases of the respiratory system 131 of total; and
diseases of the digestive system 92 of total Diseases of the
circulatory system and neoplasms accounted for 39 151 / 3872 and 2 79
/ 3872, respectively, of total admissions The average age of all
patients was 29 years, while the ALOS was 93 days Excluding mental and
behavioural disorders, the diseases with the longest ALOS were certain
infections and parasitic diseases 159 days, diseases of the skin and
subcutaneous tissue 135 days and diseases of the circulatory system
134 days

Table 6
shows a total of 313 NCD-related hospital admissions accounting for
81 313 / 3872 of all admissions in 1997 Diseases of the circulatory
system accounted for 482 151 / 313 of all NCD admissions Table 6
The three most prevalent diseases of the circulatory system other forms of
heart disease 46 admissions, hypertensive disease 38 admissions and
stroke 21 admissions were together responsible for 70 105 / 151 of all
diseases of the circulatory system Neoplasms were the next most prevalent
NCD category with 79 admissions representing 25 of all NCD admissions
The most common cancers were: malignant neoplasms of genitourinary organs
22 admissions; malignant neoplasms of bone, connective tissue, skin and
breast 11 admissions; neoplasms of unspecified nature 11 admissions,
and malignant neoplasms of digestive organs 9 admissions There were 60
admissions for diabetes and 23 admissions for bronchial COPD The average
age of all patients diagnosed with a NCD was 39 years of age with an ALOS
of 135 days This compares with the ALOS of all diseases of 93 days

Quantifying the cost of treatment
Detailed data on budget estimates and MOH staff were not available at the
time of the site visit
Subsequent effort to procure this information was
not successful In light of a complete lack of reliable data from which to
derive estimates of the cost of inpatient hospital care, a decision was
made to use the average proportion of inpatient to MOH expenditure derived
for Tonga and Vanuatu For Tonga, the budget estimate for inpatient
activity for all hospitals is estimated at 6,740,788 or 52 of the total
MOH budget 12,898,852 for the year 2002/03 For Vanuatu, the budget
estimate for inpatient activity for all hospitals is estimated at
Vt490,400,889, or 54 of the total MOH budget Vt490,400,889 for the year
2002/2003 Therefore, the midpoint of these two estimates, 53, is used as
a proxy for expenditure on hospital inpatient care in Kiribati As a
consequence of this method, the budget estimate for inpatient activity for
all hospitals is estimated at 4,793,250 of the total MOH budget
9,043,867 for the year 2002/2003

A cost per inpatient stay was calculated by dividing total inpatient cost
4,793,250 with total inpatient length of stay of 35,939 days, resulting
in a cost per inpatient day of 134 The value 134 is applied in this
study to estimate the resource implications
associated with NCD admissions
compared with admissions for other illnesses/diseases

Although not included in this report, it is important to note that in the
period between admission as an inpatient following initial treatment and
any subsequent admissions/treatments, patients referred from the outer
islands and their caregivers that normally accompany them, are provided
with food, shelter and personal hygiene items at the hospital Maneaba
Currently the hospital Maneaba will accommodate about 60 families and an
additional Maneaba is being constructed due to lack of space at the
existing site While these are not technically a direct inpatient cost,
they are strongly related, as these funds are derived from the MOH budget
Hence the financial imposts presented represent a conservative estimate of
costs

Cost associated with treatment
The last column in Table 5 and Table 6 provide the average cost of
treatment by disease and NCD category, respectively The average cost of a
hospital admission for any disease is estimated at 1,238 compared to
1,806 for a NCD admission The average costs for the majority of diseases
were relatively close to the averages The three exceptions
were:
pregnancy, childbirth and the puerperium 560; congenital anomalies
785; and, mental, behavioural disorders 10,170 The most expensive
NCDs to treat were: malignant neoplasms of lip, oral cavity and pharynx
average cost of 3,501; benign neoplasms average cost of 3,152 and
malignant neoplasms of digestive organs average cost of 3,097

Given the extreme ALOS 763 days and average costs 10,170 of mental
and behavioural disorders Table 5, this category was deleted from the
database This resulted in reduction of 01 ALOS days and a 9 decrease in
average cost Since this category impacts relatively little on total ALOS
and average cost due to the small number of patients, N4, it was
retained in the analysis

The total treatment cost for all NCD admissions is estimated at 565,363
180627 x 313 while treatment expenditure for all conditions was
4,793,250 Numbers presented in the Tables have been rounded so estimates
derived by the reader may differ slightly from these totals These data
suggest that NCDs accounted for 81 313 / 3872 of all hospital
admissions while the treatment of these diseases amounted to the equivalent
of 118 565,363 / 4,793,250 of all inpatient hospital
based
expenditures

—–Tables Table 1: Hospital morbidity |Category | |Diseases of the respiratory |system |Diseases of the digestive |system |Injury, poisoning and external|702 |causes |Certain infections and |parasitic diseases |Symptoms, signs and abnormal |524 |clinical and laboratory |findings, not elsewhere |classified |Diseases of the circulatory |system |Diseases of the genitourinary |352 |system |Endocrine, nutritional and |metabolic disorders |Certain conditions originating|236 |in perinatal period |Diseases of the skin and |subcutaneous tissue |Disease of the musculoskeletal|178 |system and connective tissue | | |Neoplasms |Diseases of the nervous system|105 |Congenital malformations, |deformations, and chromosomal | | |abnormalities |Diseases of blood and blood |48 |forming organs and certain | | |disorders involving the immune| | |mechanisms |Factors influencing health |status and contact with health| | |services |Mental, behavioural disorders |29 |Diseases of the ear and |mastoid process |Miscellaneous |External causes of morbidity |20 |and mortality |Diseases of the eye and adnexa|11 |TOTAL 5 and 6 about here—–
and cost data by disease category, Tonga 2001
|Total | of |Av |M |F |ALOS |Av |
| |total |age | | | |cost|
|1029 |117 |23 |582 |447 |44 |687 |
| | | | | | | |
|720 |82 |27 |412 |308 |41 |640 |
| | | | | | | |
|80 |23 |505 |197 |58 |909 |
| | | | | | | |
|535 |61 |24 |268 |267 |44 |693 |
| | | | | | | |
|60 |27 |262 |262 |41 |644 |
| | | | | | | |
| | | | | | | |
| | | | | |
| |
|403 |46 |57 |209 |194 |56 |884 |
| | | | | | | |
|40 |38 |103 |249 |59 |927 |
| | | | | | | |
|348 |40 |57 |107 |241 |127 |1993 |
| | | | | | | |
|27 |0 |110 |126 |79 |1237 |
| | | | | | | |
|225 |26 |28 |138 |87 |75 |1182 |
| | | | | | | |
|20 |36 |116 |62 |88 |1377 |
| | | | | |
|158 |18 |54 |64 |94 |96 |1501 |
|12 |29 |61 |44 |80 |1265 |
|56 |06 |6 |38 |18 |61 |954 |
| | |
| | |
| | | | | | | |
|05 |53 |13 |35 |63 |989 |
| | | | | |
| | | | | |
| | | | | | | |
|32 |04 |27 |6 |26 |33 |526 |
| | | | | |
| | | | | | | |
|03 |29 |17 |12 |101 |1594 |
|26 |03 |16 |13 |13 |41 |641 |
| | | | | | | |
|23 |03 |28 |6 |17 |73 |1148 |
|02 |23 |11 |9 |24 |369 |
| | | | | | | |
|01 |41 |6 |5 |75 |1186 |
|8776 |1000 |30
|3062 |5714 |49 |768 |

Expressed in local currency
Table 2: Hospital morbidity and cost data by NCD, Tonga 2001
|NCDs |Total |Av |M |F |ALOS |Av |
| | |age | | | |cost|
|Neoplasms |158 |54 |64 |94 |96 |1501 |
|Malignant neoplasms of digestive |34 |64 |17 |17 |121 |1900 |
|organs | | | | | | |
|Benign neoplasms |27 |35 |9 |18 |56 |879 |
|Neoplasms of uncertain or unknown |23 |56 |7 |16 |114 |1797 |
|behaviour | | | | | | |
|Malignant neoplasms of breast |19 |46 |0 |19 |116 |1820 |
|Malignant neoplasms of lymphoid, |11 |51 |5 |6 |90 |1415 |
|haematopoietic and related tissue | | | | | | |
|Malignant neoplasms of ill defined, |10 |64 |5 |5 |67 |1053 |
|secondary and unspecified sites | | | | | | |
|Malignant neoplasms of respiratory |9 |64 |8 |1 |114 |1799 |
|and intrathoracic organs
| | | | | | |
|Malignant neoplasms of male genital |8 |72 |8 |0 |128 |2004 |
|organs | | | | | | |
|Malignant neoplasms of female genital|7 |55 |1 |6 |61 |966 |
|organs | | | | | | |
|Malignant neoplasms of thyroid and |3 |48 |0 |3 |70 |1100 |
|other endocrine glands | | | | | | |
|In situ neoplasms |2 |67 |0 |2 |95 |1493 |
|Malignant neoplasms of lip, oral |1 |69 |1 |0 |30 |472 |
|cavity and pharynx | | | | | | |
|Malignant neoplasms of skin |1 |64 |0 |1 |30 |472 |
|Diabetes mellitus |273 |59 |83 |190 |147 |2306 |
|Diseases of the circulatory system |403 |57 |209 |194 |56 |884 |
|Other forms of heart disease |140 |56 |79 |61 |74 |1155 |
|Hypertensive diseases |105 |62 |37 |68 |50 |793 |
|Ischaemic heart disease coronary |66 |58 |44 |22 |54 |848 |
|heart disease
| | | | | | |
|Cerebrovascular disease stroke |52 |66 |26 |26 |85 |1333 |
|Acute rheumatic fever |14 |15 |7 |7 |87 |1370 |
|Chronic rheumatic heart diseases |11 |31 |7 |4 |89 |1400 |
|Diseases of veins, lymphatic vessels |9 |46 |5 |4 |71 |1118 |
|and nymph nodes, not elsewhere | | | | | | |
|classified | | | | | | |
|Other and unspecified disorders of |3 |49 |2 |1 |40 |629 |
|the circulatory system | | | | | | |
|Diseases of the arteries, arterioles |2 |64 |2 |0 |55 |865 |
|and capillaries | | | | | | |
|Other COPD |75 |69 |60 |15 |80 |1262 |
|Bronchitus COPD |2 |37 |1 |1 |15 |236 |
|TOTAL OF NCDs |911 |46 |417 |494 |92 |1447 |

Expressed in local currency

Table 3: Hospital morbidity and cost data by disease category, Vanuatu 2002
|Category |Total| total| |M |F |ALOS |Av
|
| | | |age | | | |cost |
| | | |15 | | | | |
|Certain infections and |1796 |176 |55 |936 |860 |65 |63938 |
|parasitic diseases | | | | | | | |
|Diseases of the respiratory |1418 |139 |44 |819 |599 |42 |41922 |
|system | | | | | | | |
|Injury, poisoning and external|1000 |98 |74 |610 |390 |39 |38579 |
|causes | | | | | | | |
|Diseases of the genitourinary |457 |45 |93 |151 |306 |48 |47290 |
|system | | | | | | | |
|Diseases of the digestive |420 |41 |71 |241 |179 |47 |46025 |
|system | | | | | | | |
|Diseases of the skin and |403 |39 |59 |230 |173 |55 |54404 |
|subcutaneous tissue | | | | | | | |
|Symptoms, signs and abnormal |361 |35 |78 |177 |184 |36 |35111 |
|clinical and laboratory | | | | |
| | |
|findings, not elsewhere | | | | | | | |
|classified | | | | | | | |
|Diseases of the circulatory |308 |30 |94 |155 |153 |59 |57951 |
|system | | | | | | | |
|Disease of the musculoskeletal|183 |18 |81 |101 |82 |61 |60309 |
|system and connective tissue | | | | | | | |
|Endocrine, nutritional and |180 |18 |79 |82 |98 |117 |115792 |
|metabolic disorders | | | | | | | |
|Factors influencing health |152 |15 |88 |63 |89 |80 |79410 |
|status and contact with health| | | | | | | |
|services | | | | | | | |
|Neoplasms |115 |11 |97 |42 |73 |81 |80251 |
|Certain conditions originating|104 |10 |0 |56 |48 |111 |109788 |
|in perinatal period | | | | | | | |
|Diseases of the nervous system|94 |09 |70 |59 |35 |62 |61633 |
|Mental, behavioural disorders |78
|08 |97 |36 |42 |28 |28015 |
|Diseases of the eye and adnexa|68 |07 |81 |40 |28 |38 |37498 |
|Diseases of the ear and |53 |05 |47 |29 |24 |51 |50385 |
|mastoid process | | | | | | | |
|Diseases of blood and blood |50 |05 |56 |16 |34 |82 |80884 |
|forming organs and certain | | | | | | | |
|disorders involving the immune| | | | | | | |
|mechanisms | | | | | | | |
|Congenital malformations, |33 |03 |15 |14 |19 |58 |57202 |
|deformations, and chromosomal | | | | | | | |
|abnormalities | | | | | | | |
|TOTAL |10216|1000 |74 |3857 |6359 |49 |48003 |

Expressed in local currency

Table 4: Hospital morbidity and cost data by NCD, Vanuatu 2002
|NCDs |Total| age |M |F |ALOS |Av |
| | | 15 | | | |cost |
|Neoplasms |115 |97 |42 |73 |81 |80252
|
|Benign neoplasms |27 |96 |8 |19 |53 |52784 |
|Neoplasms of uncertain or unknown |27 |96 |9 |18 |77 |76028 |
|behaviour | | | | | | |
|Malignant neoplasms of female genital|15 |100 |1 |14 |114 |112669 |
|organs | | | | | | |
|Malignant neoplasms of digestive |11 |100 |9 |2 |49 |48518 |
|organs | | | | | | |
|Malignant neoplasms of lymphoid, |9 |100 |6 |3 |45 |44914 |
|haematopoietic and related tissue | | | | | | |
|Malignant neoplasms of ill defined, |8 |100 |1 |7 |198 |195193 |
|secondary and unspecified sites | | | | | | |
|Malignant neoplasms of breast |5 |100 |0 |5 |68 |67206 |
|Malignant neoplasms of thyroid and |4 |100 |1 |3 |68 |66712 |
|other endocrine glands | | | | | | |
|Malignant neoplasms of lip, oral |2 |100 |2 |0 |90 |88949 |
|cavity and pharynx | |
| | | | |
|Malignant neoplasms of male genital |2 |50 |2 |0 |65 |64241 |
|organs | | | | | | |
|Malignant neoplasms of respiratory |1 |100 |1 |0 |100 |98832 |
|and intrathoracic organs | | | | | | |
|Malignant neoplasms of bone and |1 |100 |0 |1 |220 |217430 |
|articular cartilage | | | | | | |
|Malignant neoplasms of skin |1 |100 |1 |0 |90 |88949 |
|Malignant neoplasms of mesothelial |1 |100 |0 |1 |70 |69183 |
|and soft tissue | | | | | | |
|Malignant neoplasms of urinary tract |1 |100 |1 |0 |180 |177898 |
|Diabetes mellitus |103 |99 |55 |48 |133 |131677 |
|Diseases of the circulatory system |308 |94 |155 |153 |59 |57952 |
|Hypertensive diseases |163 |99 |69 |94 |52 |51557 |
|Other forms of heart disease |60 |93 |32 |28 |73 |71851 |
|Ischaemic heart disease coronary |23 |100 |17 |6 |35 |34936 |
|heart disease
| | | | | | |
|Cerebrovascular disease stroke |21 |100 |13 |8 |93 |91726 |
|Diseases of veins, lymphatic vessels |21 |67 |13 |8 |64 |63159 |
|and lymph nodes, not elsewhere | | | | | | |
|classified | | | | | | |
|Acute rheumatic fever |17 |65 |10 |7 |49 |48894 |
|Chronic rheumatic heart diseases |2 |100 |1 |1 |45 |44475 |
|Diseases of the arteries, arterioles |1 |100 |0 |1 |160 |158131 |
|and capillaries | | | | | | |
|Other COPD |50 |100 |30 |20 |49 |48428 |
|Bronchitus COPD |18 |100 |5 |13 |40 |39973 |
|Emphysema |1 |100 |1 |0 |130 |128482 |
|TOTAL OF NCDs |595 |96 |288 |307 |75 |73798 |

Expressed in local currency

Table 5: Hospital morbidity and cost data by disease category, Kiribati
1997
|Category |Total| |Av |M |F |ALOS |Av |
| | |total
|age | | | |cost |
|Certain infections and |509 |131 |22 |279 |230 |159 |2122 |
|parasitic diseases | | | | | | | |
|Diseases of the respiratory |503 |130 |18 |264 |239 |94 |1252 |
|system | | | | | | | |
|Diseases of the digestive |358 |92 |35 |202 |156 |103 |1376 |
|system | | | | | | | |
|Injury and poisoning |200 |52 |28 |152 |48 |111 |1477 |
|Diseases of the genitourinary |182 |47 |35 |57 |125 |90 |1205 |
|system | | | | | | | |
|Diseases of the nervous system|156 |40 |31 |91 |65 |87 |1160 |
|and sense organs | | | | | | | |
|Diseases of the circulatory |151 |39 |44 |93 |59 |134 |1786 |
|system | | | | | | | |
|Diseases of the skin and |144 |37 |32 |91 |53 |135 |1806 |
|subcutaneous tissue | | | | | | |
|
|Symptoms, signs and |143 |37 |31 |58 |85 |83 |1109 |
|ill-defined conditions | | | | | | | |
|Endocrine, nutritional and |93 |24 |41 |39 |54 |123 |1635 |
|metabolic disorders, and | | | | | | | |
|immunity disorders | | | | | | | |
|Disease of the musculoskeletal|86 |22 |32 |51 |35 |115 |1528 |
|system and connective tissue | | | | | | | |
|Neoplasms |79 |20 |46 |29 |50 |126 |1675 |
|Diseases of blood and blood |42 |11 |22 |21 |21 |76 |1016 |
|forming organs and certain | | | | | | | |
|disorders involving the immune| | | | | | | |
|mechanisms | | | | | | | |
|Congenital anomalies |35 |09 |14 |23 |12 |59 |785 |
|Certain conditions originating|23 |06 |8 |13 |10 |93 |1241 |
|in perinatal period | | | | | | | |
|Mental, behavioural disorders |4
|01 |21 |4 |0 |763 |10170 |
|TOTAL |3872 |100 |29 |1485 |2388 |93 |1238 |

Expressed in local currency

Table 6: Hospital morbidity and cost data by NCD, Kiribati 1997
|NCD |Total |Av |M |F |ALOS |Av |
| | |age | | | |cost |
|Neoplasms |79 |46 |29 |50 |126 |1675 |
|Malignant neoplasms of genitourinary|22 |47 |1 |21 |110 |1467 |
|organs | | | | | | |
|Malignant neoplasms of bone, |11 |40 |3 |8 |61 |812 |
|connective tissue, skin and breast | | | | | | |
|Neoplasms of unspecified nature |11 |45 |3 |8 |179 |2389 |
|Malignant neoplasms of digestive |9 |53 |6 |3 |232 |3097 |
|organs | | | | | | |
|Malignant neoplasms of lip, oral |4 |44 |2 |2 |263 |3501 |
|cavity and pharynx | | | | | | |
|Malignant neoplasms of other and |4 |53 |4 |0
|98 |1300 |
|unspecified sites | | | | | | |
|Malignant neoplasms of lymphoid, |3 |17 |1 |2 |147 |1956 |
|haematopoietic and related tissue | | | | | | |
|Benign neoplasms |3 |45 |7 |6 |233 |3112 |
|Malignant neoplasms of respiratory |2 |67 |2 |0 |125 |1667 |
|and intrathoracic organs | | | | | | |
|Diabetes mellitus |60 |52 |24 |36 |143 |1907 |
|Diseases of the circulatory system |151 |44 |93 |59 |134 |1786 |
|Other forms of heart disease |46 |43 |31 |15 |154 |2053 |
|Hypertensive diseases |38 |48 |21 |17 |119 |1593 |
|Cerebrovascular disease stroke |21 |56 |14 |7 |90 |1207 |
|Ischaemic heart disease coronary |14 |50 |12 |2 |42 |562 |
|heart disease | | | | | | |
|Chronic rheumatic heart diseases |13 |26 |5 |8 |71 |944 |
|Diseases of the arteries, arterioles|7 |47 |4 |3 |163 |2172 |
|and capillaries
| | | | | | |
|Diseases of veins, lymphatic vessels|7 |26 |5 |2 |189 |2515 |
|and lymph nodes, not elsewhere | | | | | | |
|classified | | | | | | |
|Bronchitus COPD |23 |15 |15 |8 |160 |2128 |
|TOTAL OF NCDs |313 |39 |161 |153 |135 |1806 |

Expressed in local currency

Impact of tobacco and alcohol on hospital resources

1 Introduction

The purpose of this chapter is to address the third aim of this project: to
define, where practical, the impact of tobacco and alcohol usage on
subsequent bed day utilisation and treatment aetiological fractions In
order to achieve this aim it is necessary to: first, consider the available
evidence regarding the impact of tobacco and alcohol use on burden of
disease / hospital admissions; second, consider the prevalence of tobacco
and alcohol use in the community; and third, apply these rates OR develop
aetilogical fractions for each country

2 Impact of tobacco and alcohol on disease

An aetiological fraction, also known as an attributable proportion
or
attributable risk, is a form of indirect quantification of morbidity and
mortality due to a specified risk factor Indirect methods involve the
estimation of a probability measure of the likelihood of causation by the
risk factor which is then applied to the total number of deaths, illnesses
or injuries resulting from a specific cause[20] In this particular case,
the risk factor is the consumption of alcohol or tobacco

There are two main sources of aetiological fractions for alcohol/tobacco
related illnesses and injuries First, the fractions can be estimated
directly from a series of cases where each case can be identified as caused
or not caused by the drug in question For example, from a representative
series of fire deaths subjected to review, an estimated 17 were caused by
cigarette smoking Thus an aetiological fraction of 17 can be applied to
all fire deaths in the population to estimate the number attributable to
cigarette smoking The second, and more common, source of aetiological
fractions is from studies of the comparative rates of death, illness or
injury in groups of people exposed and not exposed, or exposed at varying
levels, to the drug in question These
types of studies are more common in
developed countries and provide a reasonable understanding of the type of
illnesses usually attributable in whole or in part to the consumption of
alcohol and tobacco The following example illustrates this method Table
7 provides Australian derived aetiological fractions of cigarette smoking
for stroke and pneumonia[20] The table indicates that 416 of all stroke
deaths in Australian males in the age group 35-39 are casually associated
with tobacco Among females in the same age group the percentage 378
is slightly lower Much lower proportions of pneumonia deaths are
attributable to smoking 18 for males aged 35-39 and 159 for females

An aetiological fraction which is positive but less than one indicates that
the particular medical condition has more than one cause Table 7, for
example, indicates that smoking is not the only cause of strokes or
pneumonia Occasionally these fractions can be negative, indicating that
the drug in question has a protective effect against the medical condition
under study

Calculating an aetiological fraction using the second method requires two
fundamental pieces of information: i the relative risk measuring the
casual relationship between exposure to the risky drug and the condition
being studied; and, ii prevalence measuring the proportion of the
relevant population engaging in risky activity In Australia, researchers
and clinicians have conducted many clinical trials examining the impact of
tobacco and alcohol on various diseases in order to collect data on
relative risks and prevalence estimates The culmination of this research
is presented in Table 8 which lists all the conditions which Australian
researchers and their predecessors have concluded are casually linked to
the use of tobacco and alcohol[20] It is important to note that the
impact that either tobacco or alcohol has on a particular condition will
undoubtedly vary according to demographic and epidemiological transitions
discussed in Chapter 5 Further, for many NCDs the presence of multiple
risk factors in addition to tobacco and alcohol has a synergistic
effect
in causing disease, in other words, the risk of disease may actually
multiply when more than one risk factor is present This is important in
estimating the epidemiological impact of risk factors and diseases, and in
planning interventions against them

3 Prevalence of alcohol and tobacco
in each country

The WHO estimates that about 50 of men and 8 of women in developing
countries are smokers About 800 million of the worlds estimated 11
billion smokers 73 of the total are in developing countries Further,
the Western Pacific Region has shown the highest rise in tobacco use of any
WHO region The prevalence rates are 623 for males and 58 for females,
equivalent to 422 million smokers, or 344 of the total number of smokers
in the world[21]
Given the well-known harms associated with smoking, the current, and
rising, rates of smoking will be associated with significant future
morbidity and mortality unless current smoking trends can be reversed
While data on the prevalence of smoking appears reasonably reliable for
developed, and to a lesser extent, developing countries particularly more
so since the introduction of guidelines by WHO for conducting smoking
surveys[22], data on the prevalence of alcohol use is somewhat limited,
particularly for Kiribati, Tonga and Vanuatu The purpose of this section
is to provide an overview of research that has examined the prevalence of
alcohol and tobacco use in these countries To this extent, the review is
not meant to be
comprehensive but rather informative of the general
magnitude of use

Tonga
Woodward et al, 1994 measured the prevalence of smoking in Tonga in
1991[23] The questionnaire was based on the WHO guidelines for tobacco
smoking surveys [22] As indicated in Table 9, current smoking, defined as
smoking in the three months prior to the survey, was reported by 62 of
males and 14 of females The highest proportions of smokers were found
amongst young and middle-aged men, but the prevalence of smoking tended to
increase with age amongst women

Table 9: Prevalence of current smokers, by sex and age group, Tonga 1991
|Age group |Current smokers |Current smokers |
| |Males |Females |
|20-24 |604 |98 |
|25-29 |607 |99 |
|30-34 |644 |165 |
|35-39 |748 |149 |
|40-44 |589 |129 |
|45-49 |638 |115 |
|50-54
|680 |171 |
|55-59 |641 |148 |
|60-64 |664 |163 |
|65-69 |545 |261 |
|70-74 |609 |155 |
|75-79 |434 |146 |
|80 and over |439 |259 |
|All ages |624 |142 |

A recent study by UNICEF called, State of Health Behaviour and Lifestyle of
Pacific Youth Survey[24], found that substantial percentages of young
people, both school pupils and school leavers, had tried alcohol, tobacco
and illegal drugs Table 10 summarises the findings from this study This
survey was carried out in all parts of Tonga, but the report does not show
differences between island groups Moreover, although substance use was
reported by both males and females, the overall percentages by gender were
shown only for frequency of use by Out-of-School Youth The table
indicates, however, that both male school pupils and male Out-of-School
Youth
were more likely to have ever used or regularly used substances than
were their female counterparts There was also the expected pattern of
increasing ever use of substances with increasing age A striking feature
of these data is that almost half of the At-School Youth had tried smoking,
and around one in five had tried alcohol Among Out-of-School Youth, almost
70 had tried smoking, and more than 75 of boys and almost 20 of girls
were regular smokers More than 55 of Out-of-School Youth had tried
alcohol, with almost 70 of boys and almost 30 of girls using it
regularly Although only a few Out-of-School girls had tried other
substances, substantial percentages of boys were regular users of solvents,
marijuana and methylated spirits

Table 10: Summary of UNICEF findings on substance use
|Substance |At School Youth |Out of School Youth |
| |Ever Used |Use Weekly or |Ever Used|Use Weekly or More |
| | |More Often | |Often |
| | | | |Boys |Girls |
|Tobacco |476 |171 |680 |759 |199 |
|Kava |299 |84
|569 |751 |541 |
|Alcohol |213 |No data |555 |696 |283 |
|Solvents |136 |43 |200 |289 |29 |
|Marijuana |56 |27 |272 |381 |59 |
|Methylated |40 |20 |238 |347 |24 |
|spirits | | | | | |
|Mushrooms |37 |21 |122 |165 |32 |
|Fafangu |16 |12 |54 |71 |21 |
|Datura | | | | | |

A recent study undertaken on behalf of the Pacific Action for Health
Project of SPC, working in conjunction with the Tonga Family Health
Association, investigated the nature and patterns of substance use among
young people in Tonga using focus groups[25] The study found that smoking
and alcohol consumption among Tongan youth was widespread The most common
ages to start smoking were 11 to 13 years, while 16 and 17 were the most
common ages to start drinking Smoking is often a daily activity, but
alcohol is most often consumed at weekends, especially on Friday nights
Younger children smoke in secret but older
youths smoke in public The
usual places for alcohol consumption are night clubs, huts and secret
locations Binge drinking seems to be the norm Narcotics are usually taken
in secret locations Some reported episodes of alcohol and narcotic
consumption were life-threatening Cigarettes are widely used, the
majority of respondents having tried commercial brands such as Winfield
Spirits hot stuff such as Bounty Rum were the most popular beverage
while local and imported beer were less common than spirits There was
considerable variation in the number of cigarettes smoked and the
quantities of alcohol consumed The majority of smokers smoked less than 10
cigarettes a day, with the youngest smokers likely to smoke only 1-3 a day
Those who smoked least tended to smoke irregularly While smokers were
likely to smoke everyday, drinking tended to be related to paydays or the
acquisition of money in other ways Most drinkers reported drinking mainly
on Fridays, and sometimes Saturdays as well, or opportunistically if one of
their peers had obtained some alcohol[25]

Vanuatu
One of the first NCD surveys conducted in Vanuatu occurred in 1984[26]
This survey was limited, however, to examining the
prevalence of obesity,
diabetes and hypertension In 1985, a joint effort by the MOH, SPC and
WHO, was undertaken to determine the prevalence of NCDs and their risk
factors amongst adults in Vanuatu, as well as to assess the impact of
increasing urbanization on those diseases[27] A cross-sectional
prevalence survey of adult ni-Vanuatu aged 20 years and over was conducted,
drawn from three populations presumed to be at different levels of
modernisation: Vila urban, Nguna intermediate and Middle Bush Tanna
rural Among other things, measurements were made of tobacco, alcohol and
kava use Summary findings for tobacco, alcohol and kava use are presented
in Table 11 Use of tobacco, kava and alcohol was much higher in men than
in women Tobacco use was lower in male urban civil servants than in the
rural population The use of alcohol and kava was about the same in males
for all locations although volume consumed may have been different, but
alcohol and kava use was very infrequent amongst women in Tanna[27]
Table 11: Summary statistics, Vanuatu NCD survey 1985
|Substance |Urban |Intermediate |Rural |
|Use of tobacco | |
| |
|Males |54 |80 |77 |
|Females |5 |3 |6 |
|Use of Kava | | | |
|Males |68 |78 |73 |
|Females |6 |9 |1 |
|Use of alcohol | | | |
|Males |74 |82 |70 |
|Females |19 |16 |3 |

In 1986 a food and nutrition policy was developed in Vanuatu with one of
its objectives being to prevent an increase in the prevalence of NCDs[28]
Since that time there have been ad hoc NCD surveys conducted in 1990 using
the same three sites as the 1985 survey and in 1993 amongst civil servants
in Vanuatu[29] Although these reports remain incomplete and unpublished,
they suggest that NCDs had increased further in urban areas but the
evidence in the reports was not strong enough to provide the basis for
national policy and programmes Additionally, as there had been no
national data collected on
the prevalence of NCDs since the 1985 survey,
the real picture of NCDs in Vanuatu remained unknown

In 1993, the Vanuatu National Nutrition Surveillance system was
established The system identified the need to assess the extent of
obesity, hypertension and diabetes in selected sites and to monitor the
change in these conditions over time[29] The main NCDs reported by the
HIS in Vanuatu were diabetes, hypertension, heart disease, cancer, stroke
and overweight/obesity One recommendation of the HIS was to ensure
efficient monitoring of the NCD situation by conducting sentinel site
surveys every five years to measure the trends[29] As a consequence of
this recommendation, in 1998, the Department of Public Health undertook an
assessment of the prevalence of non-communicable diseases and their risk
factors in Vanuatu[30] The survey was administered in five of the six
provinces of Vanuatu, based on the presence of provincial hospitals
Participants were selected using a cluster sampling methodology and only
household members aged 20 years and over were interviewed Questions on
alcohol and smoking consumption patterns were collected Table 12 reports
the findings for questions
relating to alcohol, smoking and kava use The
findings showed that over 50 of males and 11 of the females surveyed
reported consuming alcohol Reported frequency of alcohol consumption not
shown in Table 12 was low, with nearly 70 total male and 78 females
reporting consuming alcohol yearly Around 50 of males and 5 of females
reported smoking with the highest prevalence of smoking 61 found among
the 20-29 year old age cohort The majority of male 491 and female
564 of smokers consumed between 1 and 5 cigarettes per day Kava use
was reported by 67 of all males and 15 of females surveyed Overall 86
of males and 45 females consumed Kava at least weekly

Table 12: Distribution of respondents by alcohol consumption, smoking and
kava use, gender and age group
|Gender/age group |Alcohol |Smoking |Kava |
|Males | | | |
|20-29 |722 |609 |714 |
|30-39 |523 |462 |641 |
|40-49 |429 |417 |673 |
|50-59 |309 |442 |684 |
|60
|163 |420 |617 |
|All |504 |491 |672 |
|Females | | | |
|20-29 |132 |47 |113 |
|30-39 |109 |23 |136 |
|40-49 |119 |57 |157 |
|50-59 |75 |25 |163 |
|60 |57 |157 |300 |
|All |111 |50 |149 |
|All respondents |308 |272 |413 |

Kiribati
In a report by Collins and Lapsley 1997[31], data on the prevalence of
smoking in Kiribati were reported Table 13 presents the key findings of
this survey As indicated in the table, that prevalence of smoking was
high across all groups with prevalence highest among urban males aged 25-
34 In general, the prevalence of smoking was highest amongst males than
females, and among urban settings than rural settings

Table 13: Smoking prevalence in Kiribati, by age, gender, urban/rural, 1981
|Age |Urban male
|Urban female |Rural male |Rural female |
|20-24 |95 |63 |85 |57 |
|25-34 |90 |75 |83 |68 |
|35-44 |91 |78 |84 |73 |
|45-54 |84 |71 |87 |66 |
|55 |83 |82 |79 |71 |

Using data collected for a 1999 WHO report, Khaleghian 2001 reports that
82 of all males and 65 of females in Kiribati are current smokers[6]
Similar patterns were reported by Stanton 2001, who, using data collected
for the 1997 WHO Tobacco or Health: a global status report, concluded
that Kiribati had the highest rate of women smokers in the world with 74
of females regularly using tobacco[32]

The most recent data on the prevalence of smoking and alcohol use in
Kiribati comes from the report of the 2000 census of population[17]
Respondents were asked whether any member of the household smoked or drunk
alcohol Eighty one percent of households indicated a member smoked while
38 had consumed alcohol

4 Quantifying harm caused by alcohol and tobacco in each country

As
discussed previously, in order to calculate aetiological fractions two
fundamental pieces of information are required: the relative risk
measuring the causal relationship between exposure to the risky drug and
the condition being studies; and, prevalence measuring the proportion of
the relevant population engaging in the risky activity Although the data
bank on the prevalence of alcohol and tobacco use in Tonga, Vanuatu and
Kiribati is growing, the lack of detailed epidemiological data on relative
risks suggests that, at this stage, it is not possible to develop specific
tobacco and alcohol aetiological fractions for the three countries of
interest

Although Australian aetiological fractions have been developed and
comprehensively list all the conditions causally linked to the use of
tobacco and alcohol Table 8, these fractions are very country specific
For instance, in relation to tobacco, given the long delay between the
onset of smoking and development of tobacco-related illnesses, the patterns
of tobacco-related harm experienced in Australia today are a reflection of
smoking practices many years ago Further, many of the health effects of
smoking depend on the exposure
history, including the age at which smoking
began, the number of cigarettes smoked per day, the degree of inhalation
and cigarette characteristics such as tar and nicotine content or presence
of a filter Therefore, the conditions identified in Australia as being
attributable to tobacco and alcohol may not currently apply to Tonga,
Vanuatu or Kiribati However, a caveat of this argument is that it is
plausible to assume that these conditions are likely to be applicable in
the near future, as these populations undergo epidemiological transition,
resulting in more tobacco and alcohol-related harm

In the World Health Report 2002, region specific population attributable
fractions have been developed by age and gender for particular risk
factors[1] These fractions for tobacco and alcohol use are outlined in
Table 14 Included in the table are estimates for low mortality developing
and developed countries The low mortality developing category includes
the Western Pacific Region Group B of which Kiribati, Tonga and Vanuatu
are members The developed country category includes Australia and is
included for comparative purposes, to illustrate variations in the
magnitude of attributable
fractions between countries

Given that these estimates are the most reliable to date, they are used in
this document to consider the impact of tobacco and alcohol usage on
subsequent bed -day utilization and treatment costs in each country Using
data reported in Tables 1-6, the attributable fractions are applied to the
particular disease of interest or injury in the case of alcohol to
estimate the number of admissions and associated cost attributable to
smoking and alcohol These estimates are then expressed as a proportion of
NCDs and all diseases

Table 14: Population attributable fractions by risk factor, age and gender,
2000
|Risk factor and condition |Low mortality |Developed countries|
| |developing | |
| |Males |Females |Males |Females |
|Tobacco | | | | |
|COPD |35 |14 |79 |57 |
|Mouth and oropharynx cancers|0 |0 |0 |0 |
|Trachea/bronchus/lung |57 |20 |90 |69 |
|cancers | | |
| |
|Other cancers |12 |1 |19 |2 |
|Other medical conditions |7 |1 |16 |6 |
|CVD |12 |2 |32 |10 |
|Alcohol | | | | |
|Cirrhosis of the liver |45 |13 |63 |49 |
|Drowning |10 |6 |43 |25 |
|Epilepsy |27 |13 |45 |36 |
|Falls |8 |3 |21 |8 |
|Haemoohagic stroke |21 |2 |26 |0 |
|Homicide |28 |16 |41 |32 |
|Ischaemic heart disease |5 |0 |2 |-3 |
|Ischaemic stroke |3 |0 |5 |-16 |
|Unipolar depressive |3 |0 |7 |2 |
|disorders | | | | |
|Liver cancer |32 |11 |36 |28 |
|Mouth and oropharynx cancers|28 |10 |41 |28 |
|Oesophagus cancer |42 |16 |46
|36 |
|Other cancers |5 |2 |11 |8 |
|Self-inflicted injuries |10 |5 |27 |12 |
|Poisoning |11 |7 |43 |26 |
|Other intentional injuries |20 |11 |32 |19 |
|MVA |25 |8 |45 |18 |
|Other unintentional injuries|15 |6 |32 |16 |

Tonga
The number of hospital admissions and treatment costs attributable to
tobacco and alcohol for Tonga 2001 are presented in Table 15 The table
presents the impact of tobacco and alcohol according to NCDs, other
treatment and all diseases Although the attributable risk of tobacco is
generally confined to NCD, the attributable risk of alcohol is associated
with both NCD and other diseases predominantly injury The proportion of
total risk is estimated by dividing tobacco and alcohol risk by totals
derived from Tables 1 and 2 The data suggests that for the year 2001,
tobacco and alcohol were responsible for an estimated 99 90 / 911 of
all NCD admissions and 21 186 / 8776 of total admissions The cost of
treating patients with a tobacco- or
alcohol-related disease or injury is
estimated to account for 91 120,563 of the total cost of treating all
NCDs and 31 208,090 of the cost of treating all diseases

Table 15: Number of hospital admissions and treatment costs attributable to
tobacco and alcohol, Tonga 2001
|Attributable to |Tobacco |Alcohol |Tobacco and|Total from| |
|tobacco and alcohol | | |alcohol |Tables 12|Total |
|conditions | | | | | |
|Hospital admissions | | | | | |
|NCD |72 |18 |90 |911 |99 |
|Other diseases |0 |96 |96 |7865 |12 |
|All diseases |72 |114 |186 |8776 |21 |
|Cost of treatment | | | |
|NCD |92358 |28204 |120563 |1318136 |91 |
|Other diseases |0 |87528 |87528 |5422652 |16 |
|All diseases |92358 |115732 |208090 |6740788 |31 |

Expressed in local currency

Vanuatu
The number of hospital admissions and treatment costs attributable to
tobacco and alcohol for Vanuatu
2002 are presented in Table 16 The table
presents the impact of tobacco and alcohol according to NCDs, other
treatment and all diseases Although the attributable risk of tobacco is
generally confined to NCD, the attributable risk of alcohol is associated
with both NCD and other diseases predominantly injury The proportion of
total risk is estimated by dividing tobacco and alcohol risk by totals
derived from Tables 3 and 4 The data suggests that for the year 2002,
tobacco and alcohol were responsible for an estimated 99 59 / 595 of
all NCD admissions and 18 187 / 10216 of total admissions The cost of
treating patients with a tobacco or alcohol related disease or injury is
estimated to account for 85 Vt3,746,959 of the total cost of treating
all NCDs and 18 Vt8,679,144 of the cost of treating all diseases

Table 16: Number of hospital admissions and treatment costs attributable to
tobacco and alcohol, Vanuatu 2002
|Attributable to tobacco|Tobacco |Alcohol |Tobacco and|Total from| |
|and alcohol conditions | | |alcohol |Tables 34|Total |
|Hospital admissions | | | | | |
|NCD |49 |10 |59
|595 |99 |
|Other diseases |0 |128 |128 |9621 |13 |
|All diseases |49 |138 |187 |10216 |18 |
|Cost of treatment | | | | | |
|NCD |3087118 |659841 |3746959 |43909845 |85 |
|Other diseases |0 |4932184 |4932184 |446491044 |11 |
|All diseases |3087118 |5592026 |8679144 |490400889 |18 |

Expressed in local currency

Kiribati
The number of hospital admissions and treatment costs attributable to
tobacco and alcohol for Kiribati 1997 are presented in Table 17 The table
presents the impact of tobacco and alcohol according to NCDs, other
treatment and all diseases Although the attributable risk of tobacco is
generally confined to NCD, the attributable risk of alcohol is associated
with both NCD and other diseases predominantly injury The proportion of
total risk is estimated by dividing tobacco and alcohol risk by totals
derived from Tables 5 and 6 The data suggests that for the year 1997,
tobacco and alcohol were responsible for an estimated 98 31 / 313 of
all NCD admissions and 15 59 / 3872 of total
admissions The cost of
treating patients with a tobacco or alcohol related disease or injury is
estimated to account for 11 62,183 of the total cost of treating all
NCDs and 16 104,540 of the cost of treating all diseases

Table 17: Number of hospital admissions and treatment costs attributable to
tobacco and alcohol, Kiribati 1997
|Attributable to tobacco|Tobacco |Alcohol |Tobacco and|Total from| |
|and alcohol conditions | | |alcohol |Tables 56|Total |
|Hospital admissions | | | | | |
|NCD |23 |8 |31 |313 |98 |
|Other diseases |0 |28 |28 |3559 |08 |
|All diseases |23 |36 |59 |3872 |15 |
|Cost of treatment | | | | | |
|NCD |44356 |17827 |62183 |565363 |110 |
|Other diseases |0 |42357 |42357 |6175425 |07 |
|All diseases |44356 |60184 |104540 |6740788 |16 |

Expressed in local currency

economic projections of the NCD burden

1 Introduction

The purpose of this chapter is to address the
fourth aim of this project:
to provide, where practical, economic projections on the NCD burden through
to 2020 While a full appraisal of the social economic cost of NCDs is
beyond the brief of this report, it is useful to consider the spectrum of
resources usually involved in such an appraisal The economic costs of
NCDs and for that matter any disease can be described as having two
components: direct and indirect costs Direct costs are those specifically
incurred for the diagnosis, treatment and palliation of a disease
Examples include the cost of a diagnostic test, a home visit from a nurse
or a course of medication and include the cost of labour, materials and
facilities involved in the production of each Indirect costs are those
caused by the loss of productive life These include the cost of lost
productivity due to illness, disability or premature death on the part of
disease sufferers, as well as the cost of forgone economic opportunities
for family members and informal caregivers, who spend time caring for an
affected individual

Given this report has focused entirely on the treatment of NCDs within
hospitals, it is appropriate to limit discussion of future economic
costs
of NCDs to this setting Specifically, attention will be directed to
examining the direct resource implications of NCDs, with a focus on tobacco
and alcohol as risk factors for disease As such, no consideration will be
given to the indirect economic implications of tobacco or alcohol related
morbidity or mortality

2 Future rise in NCDs

As discussed in the introductory chapter, the rapid rise of NCDs represents
one of the major health challenges to global development It is estimated
that by 2020 over 70 of the global burden of disease will be caused by
NCDs especially cancer, diabetes, cardiovascular diseases and chronic
respiratory diseases, mental health disorders and injuries[1] Table 18
provides an overview of attributable mortality by selected risk factor, by
gender, for developing countries such as Tonga, Vanuatu and Kiribati and
developed countries, 2000[1] The risk factors listed in this table
indicate the current divergence in mortality between developing and
developed countries The largest divergence in risk factor attributable
death is seen for tobacco Tobacco-related disease causes 263 and 93
of all male and female deaths, respectively, in developed
countries,
compared to 122 and 29 of male and female deaths, respectively, in
developing countries The key message from this table is that, given the
current prevalence of risk factors for NCDs in developing countries, it is
just a matter of time until developing countries see similar patterns of
mortality

Table 18: Attributable mortality by selected risk factor, level of country
development and sex, 2000
| |Developing countries |Developed countries |
|Risk factor |Males |Females |Males |Females |
| | of total | of total | of total | of total|
| |deaths |deaths |deaths |deaths |
|Diet-related risks and physical activity | | |
|Blood pressure |127 |151 |201 |239 |
|Cholesterol |51 |56 |145 |176 |
|Overweight |42 |56 |96 |115 |
|Low fruit and vegetable |50 |48 |76 |74 |
|intake | | | | |
|Physical inactivity |28 |32 |60
|67 |
|Addictive substances | | | | |
|Tobacco |122 |29 |263 |93 |
|Alcohol |85 |16 |80 |-03 |
|Illicit drugs |06 |01 |06 |03 |

The escalation of NCDs in all countries of the world, and in particular
developing countries, can be attributed, predominantly, to what is know as
the health transition There are two components to the health transition:
a demographic component, also known as the demographic transition and an
epidemiological component, the epidemiological transition As countries
develop, the age structure of their population changes due mainly to
improvements in infant and adult survival and the proportion of elderly
individuals tends to increase This is the demographic component of the
health transition At the same time, changes in lifestyle factors, such as
diet and levels of physical activity, along with improvements in health
care and environmental sanitation, lead to an increase in the prevalence of
NCDs and a corresponding decline in communicable, perinatal and maternal
conditions As a result,
NCDs begin to displace communicable diseases as
the primary cause of ill-health and death This change represents the
epidemiological component of the health transition The combined effect of
these demographic and epidemiological transitions is a shift from a young
population characterized by a high level of communicable, maternal and
perinatal diseases, to an older population characterized by a greater
prevalence of NCDs[1]

This health transition has been under way in PIC for the past 30 years, as
indeed in most developing countries around the world In most countries,
the health transition is not characterized by sharp changes in
epidemiological and demographic profiles Instead, the process unfolds
gradually, and the overall disease profile is, for an extended period,
characterized by the coexistence of communicable and NCDs, the balance
between them being determined by a number of factors including levels of
environmental sanitation, the availability of health services and the age
structure of the population Different countries, therefore, are at
varying points in the health transition, from those who have just entered
the transition to those who have nearly completed it

3
Consideration of demographic transition

World Health Report 2002 population indicators for Tonga, Kiribati and
Vanuatu are presented in Table 19[1] Although each country has undertaken
a demographic census, the World Health Report standardizes these results
for the year 2001 Table 19 presents results in terms of the demographic
shift in population for each country Although annual population growth
rates have remained positive, fertility rates have fallen with the
exception of a 02 rise in Kiribati which imply a relatively neutral
population growth The table also presents life expectancy at birth, with
Tonga having the longest life expectancy at 694 years and Kiribati the
shortest at 636 years Although not presented in Table 19, life
expectancy is increasing in all three countries and is likely to continue
to increase[1] In summary, then, with relatively neutral population
growth and increased life expectancy, it appears that the demographic
transition for Tonga, Vanuatu and Kiribati is changing from a more
traditional societal pattern of relatively young populations, to those with
increasing numbers of middle-aged and elderly

Table 19: Basic indicators of population for
Tonga, Vanuatu and Kiribati,
2001
|Country|Population|Annual growth| pop aged 60|Total |LE at |
| |000 |rate |years |fertility rate|birth |
| | | | | |years |
| |2001 |1991-2001 |1991 |2001 |1991 |2001 |2001 |
|Vanuatu|201 |28 |52 |48 |49 |44 |656 |
|Kiribat|84 |14 |60 |69 |44 |46 |636 |
|i | | | | | | | |

The exploration of the potential economic impact of this demographic
transition on the NCD burden requires an understanding of both population
changes and economic indicators Using the data obtained in Table 19,
crude population projections are able to be made However, as suggested
above, population changes are likely to be minimal given current indicators
of population change Further, in the absence of future economic indicator
data, any attempt to quantify the potential impact of the demographic
transition is likely to be contentious with estimates subject to
considerable uncertainty

However, before moving on to consider the epidemiological transition, it
is
useful to reflect on the current age profile of disease morbidity in each
of the countries and how an ageing population will see a natural increase
in the proportion of the population with a NCD The average age of
patients admitted for any disease in Tonga is 30 years of age Table 1,
while the average age of patients admitted with an NCD is 46 years In
Kiribati, a similar picture unfolds: average age for all diseases is 29
years Table 5 while for NCDs it is 39 years of age Table 6 Average
ages of patients admitted to hospital were not available for Vanuatu With
an ageing population comes a demographic shift toward the middle-aged and
elderly, where the prevalence of NCDs is much higher This suggests that
even in the absence of population growth, the proportion of patients with
an NCD will inevitably increase and, as it does, so too will the cost of
treating these diseases The next section elaborates on this point in more
detail

4 Consideration of epidemiological transition

Chapter 44 used WHO attributable fractions Table 14 for the Western
Pacific Region to consider the impact of tobacco and alcohol usage on
hospital admissions and treatment costs for Tonga, Vanuatu and
Kiribati
Also reported in Table 14 were attributable fractions for developed
countries, including Australia, where the epidemiological transition is
advanced The fractions derived for developed countries are used in the
present analysis to examine the potential impact tobacco and alcohol use
may have on disease burden

Tonga
Using the same hospital morbidity and cost data as discussed in earlier
chapters, Table 20 presents the results of the number of hospital
admissions and treatment costs attributable to tobacco and alcohol using
developed country aetiological fractions The data in Table 20 suggest
that, after applying revised aetiological fractions, tobacco and alcohol
related NCD admissions have increased from 99 Table 15 to 242 This
implies that nearly 1 in 4 admissions for NCDs are attributable to tobacco
or alcohol The cost of treating patients with a tobacco or alcohol
related disease or injury has increased from 91 Table 15 of the total
cost of treating all NCDs to 22 The costs of these admissions represent
73 of all hospital related expenditures, up from 31 as reported in
Table 15

Table 20: Projected impact of tobacco and alcohol on hospital resources,
Tonga
2001
|Attributable to |Tobacco |Alcohol |Tobacco and|Total from| |
|tobacco and alcohol | | |alcohol |Tables 12|Total |
|conditions | | | | | |
|Hospital admissions | | | | | |
|NCD |192 |28 |220 |911 |242 |
|Other diseases |0 |15 |15 |7865 |02 |
|All diseases |192 |43 |235 |8776 |27 |
|Cost of treatment | | | |
|NCD |241889 |47062 |288951 |1318136 |219 |
|Other diseases |0 |200075 |200075 |5422652 |37 |
|All diseases |241889 |247137 |489026 |6740788 |73 |

Expressed in local currency

Vanuatu
The data in Table 21 suggests that, after applying revised aetiological
fractions, tobacco and alcohol related NCD admissions have increased from
99 Table 16 to 253 Similar to Tonga, this implies that over 1 in 4
admissions for NCDs are attributable to tobacco or alcohol The cost of
treating patients with a tobacco or alcohol related disease or injury has
increased from 85
Table 16 of the total cost of treating all NCDs to
216 The costs of these admissions represent 43 of all hospital
related expenditures, up from 18 as reported in Table 16

Table 21: Number of hospital admissions and treatment costs attributable to
tobacco and alcohol, Vanuatu 2002
|Attributable to tobacco|Tobacco |Alcohol |Tobacco and|Total from| |
|and alcohol conditions | | |alcohol |Tables 34|Total |
|Hospital admissions | | | | | |
|NCD |134 |17 |150 |595 |253 |
|Other diseases |0 |296 |296 |9621 |31 |
|All diseases |134 |313 |446 |10216 |44 |
|Cost of treatment | | | | | |
|NCD |8261551 |1211342 |9472892 |43909845 |216 |
|Other diseases |0 |11382309 |11382309 |446491044 |25 |
|All diseases |8261551 |12593650 |20855201 |490400889 |43 |

Expressed in local currency

Kiribati
The data in Table 22 suggests that, after applying revised aetiological
fractions, tobacco and alcohol related NCD admissions have
increased from
98 Table 17 to 179 The cost of treating patients with a tobacco or
alcohol related disease or injury has increased from 11 Table 17 of the
total cost of treating all NCDs to 205 The costs of these admissions
represent 31 of all hospital related expenditures, up from 16 as
reported in Table 17

Table 22: Number of hospital admissions and treatment costs attributable to
tobacco and alcohol, Kiribati 1997
|Attributable to tobacco|Tobacco |Alcohol |Tobacco and|Total from| |
|and alcohol conditions | | |alcohol |Tables 56|Total |
|Hospital admissions | | | | | |
|NCD |43 |13 |56 |313 |179 |
|Other diseases |0 |64 |64 |3559 |18 |
|All diseases |43 |77 |120 |3872 |31 |
|Cost of treatment | | | | | |
|NCD |84488 |31194 |115682 |565363 |205 |
|Other diseases |0 |96377 |96377 |6175425 |16 |
|All diseases |84488 |127571 |212059 |6740788 |31 |

Expressed in local currency

5 What can be
done to reduce potential impact of NCDs?

The future impact of NCDs on hospital morbidity and associated treatment is
likely to be significant The demographic transition implies that as the
age profile of a population changes, so to does the pattern of disease
Consistent with the WHO predictions, countries of Tonga, Vanuatu and
Kiribati are all likely to see a reduction in the more prevalent
communicable diseases that impact predominantly on younger people and an
increase in NCDs that impact predominantly on the middle-aged to elderly
The epidemiological transition is also inevitable The wealth of data from
developed countries has established the link between the prevalence of
certain risk factors and the delayed onset of NCDs The magnitude of the
transition which unfolds and the extent to which NCDs impact on disease
burden is, however, something governments are able to influence For
example, governments have the ability to devote resources into health
promoting programs that aim to reduce the prevalence of key risk factors
for NCDs, of which tobacco smoking is arguably the best example
Governments of Tonga, Vanuatu and Kiribati can learn much from the tobacco
control movement
of developed countries, in terms of implementing cost-
effective strategies aimed at reducing the prevalence of smoking While
the recommendations of specific interventions to alleviate or more
appropriately minimize the future burden of NCDs is beyond the scope of
this project, carefully planned and implemented interventions are likely to
be cost-beneficial

Conclusions and recommendations

1 Introduction

The primary aim of this report has been to conduct an assessment of
existing hospital inpatient data to estimate the economic burden of NCDs in
the countries of Tonga, Vanuatu and Kiribati Site visits were conducted
within each country during the period March - May 2003 During this time
data were collected from MOH representatives relating to hospital
morbidity, budget reports and prevalence reports on lifestyle diseases in
these countries These data have been used in the current project,
together with reasonable assumptions in lieu of missing or incomplete data,
to consider the impact, both current and future, of NCDs on the burden of
disease in these countries As part of this assessment, estimates of
treatment costs were derived and applied to hospital admissions
to
calculate total and average costs of treatment by NCD and non-NCD
Further, an attempt has been made to consider the extent to which tobacco
and alcohol use contributes to the total current, and future, NCD burden in
Tonga, Vanuatu and Kiribati

The key findings, conclusions and recommendations of this report need to be
considered in the context of the limitations imposed by the quality of the
data obtained, as well as the key assumptions made during data analyses

2 Limitations

Data quality
At the time of hospital data collection are a number of salient issues were
made apparent First, as for any data entry process, it is likely that a
number of admissions /discharge forms may have been misplaced en-route to
the statistics office and, hence, may not have been entered into the
relevant database Second, the relatively lack of expertise, and resources
available for training staff in appropriate coding, may lead to incorrect
codes being assigned to particular episodes of care The process of coding
is a difficult task compounded by the fact that certain diseases for
example, diabetes do not fit neatly into specific ICD categories Third,
and a related point, is that resources for
data entry and manual checking
are simply not available in the countries of interest, which often results
in delays in data entry and the use of unchecked databases This point is
particularly relevant for Kiribati Fourth, no comprehensive clinical
costing system is currently available to provide detailed information of
the type of care and cost associated with patients admissions The
development of such a system would facilitate a more reliable assessment of
hospital care by disease category Finally, although the data bank on the
prevalence of alcohol and tobacco use in Tonga, Vanuatu and Kiribati is
growing, the lack of detailed epidemiological data on relative risks
suggests that, at this stage, it is not possible to develop specific
tobacco and alcohol aetiological fractions for the three countries of
interest

Data assumptions
Consistent with most projects of this nature, a number of assumptions have
been necessary in deriving estimates of hospital cost per day as a proxy
for the value of resource use associated with treatment This estimate has
been derived by dividing total inpatient cost with total inpatient length
of stay to provide a cost per inpatient day For Tonga
and Vanuatu
reasonable data were available to guide the separation of inpatient and
outpatient components of the MOH budgets For Kiribati, however, no such
data were available The estimates of total length of stay derived from
hospital databases may also be conservative given the above issues
surrounding data quality Further, using an estimate of a hospital cost
per day does not take into account the fact that some admissions may cost
more per day than others, either because they involve more expensive
technologies or because they consume more staff time For example, an
admission for simple pneumonia is less expensive on a per-day basis than an
admission for surgery to remove bowel cancer, for which there may be an
added cost of operating time, surgical instruments and staff time eg
nurses, anesthetists and surgeons These differences are not captured in
an aggregated estimate of cost per day If NCD admissions are, on average,
more resource intensive than those for other conditions, then the overall
estimate of inpatient cost will underestimate the true cost of NCD
admissions using this method

In considering the impact of tobacco and alcohol usage on hospital
admissions and
treatment costs, WHO estimates of attributable risk have
been used These estimates of risk are region specific and may not
necessarily reflect the true burden of harm in particular countries For
example, the WHO estimates consider a fraction of the conditions identified
by Australian researchers that are attributable to tobacco and alcohol use
It may be possible that these conditions represent a true reflection of
current harm attributable to tobacco and alcohol however, it is at least
equally plausible to assume that there is a dearth of adequate data in
these countries to facilitate a comprehensive assessment

3 Key findings

Table 23 presents the key findings of this report In each of the
countries included, Tonga, Vanuatu and Kiribati, NCDs currently account for
104, 58 and 81 of all admissions, respectively Patients being
admitted for a NCD are much older than those presenting for a non-NCD, 16
years older in Tonga and 10 years older in Kiribati Patients admitted for
a NCD also stay in hospital longer with ALOS for a NCD admission being 92
days, 75 days and 135 days in Tonga, Vanuatu and Kiribati, respectively
This compares with an ALOS for a non-NCD admission of 49 days
in both
Tonga and Vanuatu and 93 days in Kiribati Consistent with this finding,
average treatment costs of NCDs are considerably higher than non-NCD
admissions Although NCDs account for a relatively low proportion of all
admissions, they account for a disproportionately larger share of all
treatment expenditures In Tonga, Vanuatu and Kiribati NCDs account for
196, 9 and 81, respectively, of all treatment expenditures To put
this in context, in Tonga for example, one out of every 10 104 patients
admitted to hospital are admitted for a NCD, however, for every 5 dollars
spent on treating all patients, one of these dollars 20, is required to
treat the patient with a NCD

Tobacco and alcohol are two risk factors for NCDs, the use of which are
prevalent in Tonga, Vanuatu and Kiribati Indeed, like most PIC, the rates
of smoking in the population are some of the highest in the world Using
data from a number of sources, it is estimated that approximately 10 of
all current NCD admissions are the result of tobacco and alcohol
attributable conditions in Tonga, Vanuatu and Kiribati A similar
proportion of money 10 is devoted to treating these tobacco and alcohol
related NCDs The extent
to which tobacco and alcohol will play in future
disease is clear The transition towards an older demography in these
three countries, coupled with an epidemiological transition towards greater
harm from current risk behaviour suggests that the burden of disease from
NCDs will increase Based on epidemiological transition towards the
attributable fractions of more developed countries, it is estimated that up
to 25 of all NCD admissions, and 20 of all NCD treatment cost, may be
attributable to tobacco and alcohol consumption use

Table 23: Summary of key findings for each country
| |Tonga |Vanuatu |Kiribati|
|Hospital data | | | |
|Number of admissions |8776 |10216 |3872 |
|Number of NCD admissions |911 |595 |313 |
| of NCD to all admissions |104 |58 |81 |
|ALOS all admissions days |49 |49 |93 |
|ALOS NCD admissions days |92 |75 |135 |
|Average age of all admissions years |30 |NA |29 |
|Average age of
NCD admissions years |46 |NA |39 |
|Cost data | | | |
|MOH budget millions |1290 |Vt91632 |904 |
|Inpatient hospital millions |674 |Vt490 |479 |
| inpatient hospital to MOH |52 |54 |53 |
|Inpatient hospital cost per day |157 |Vt9883 |133 |
|Average hospital cost per admission |768 |Vt48003 |1,238 |
|Average hospital cost per NCD admission |1,447 |Vt73799 |1,806 |
| total cost NCD to total cost all admissions |196 |9 |81 |
|Tobacco and alcohol | | | |
|Current: NCD admissions attributable to |990 |990 |980 |
|tobacco and alcohol | | | |
|Current: NCD cost attributable to tobacco |910 |850 |1100 |
|and alcohol | | | |
|Future: NCD admissions attributable to |2420 |2530 |1790 |
|tobacco and alcohol | | | |
|Future: NCD cost attributable to
tobacco and|2190 |2160 |2050 |
|alcohol | | | |

4 Conclusions

A major strength of this report is the use of a range of local data in
estimating the extent to which NCDs currently impact, and will impact in
the future, on hospital morbidity and treatment costs in Tonga, Vanuatu and
Kiribati Overall, the results indicate that:
NCDs are a significant contributor to the burden of disease in each of
these countries;
Patients admitted for a NCD are older and are likely to require more
care than patients admitted for other diseases;
The cost of treating patients admitted for a NCD is considerably higher
than the cost of treating the average patient;
Tobacco and alcohol use are high and are both significant risk factors
for NCD, accounting for approximately 10 of all NCD admissions;
In the future, NCDs will account for a greater proportion of the
disease burden, with tobacco and alcohol related attributable disease
estimated to account for up to 25 of all NCD treatment costs

While this project has focused on the impact of tobacco and alcohol as risk
factors for NCD, there are other
important risk factors including obesity,
poor diet and physical inactivity that contribute to the current, and will
contribute to the future, burden of NCD Further, it is important to note
that for many NCDs, the presence of multiple risk factors has a synergistic
effect in causing disease In other words, the risk of disease may
actually disproportionately increase when more than one risk factor is
present This is important in estimating the epidemiological impact of
risk factors and diseases, and in planning comprehensive interventions
against them

Given evidence of the increasing emergence of the NCD epidemic in these
countries,: what can be done about it? The PAPH project is an excellent
example of the need to make a concerted effort to clearly identify
behavioural and health problems, collect local level data to define the
magnitude and exact nature of health problems, and work with local people
to design and implement intervention strategies that educate people about
risk factors and promote a change towards healthy lifestyle behaviours
However, PAPH cannot work alone There needs to be a concerted effort by
Governments of PICs to alleviate the current and future burden
of harm
associated with NCDs Such a commitment requires the allocation of
resources to improve the capacity to understand burden of disease patterns
and identify potential solutions to bring about change The Government of
Vanuatu have indicated a commitment to improve the health of its people
through the development of the Comprehensive Reform Program Among other
things, this program seeks to improve data collection surveillance and
monitoring which is essential in understanding where the harm lies and
using this information to design interventions which will have the biggest
impact on reducing the burden of disease, particularly in relation to NCDs
In learning from the experiences of the more developed countries in the
South Pacific region, PICs have an opportunity to implement relatively
easier and cheaper interventions aimed at preventing the onset of
disease, rather than focusing disproportionately on treatment

The recommendations of this report provide a means by which Governments can
begin to make a concerted effort and reduce the future costs associated
with NCD in their countries as well as improve efficiency in budget
allocations

5 Recommendations

Recommendation 1

Improve the reliability of hospital data
Actions
Commit additional resources to MOH to improve entry and monitoring of
hospital data
How best achieved
Maintain adequate staff levels in Statistics Office
Ensure staff are well qualified and undergo ongoing training
Ensure computer hardware and software is state of the art and well-
maintained
Example
Vanuatu Comprehensive Reform Program

Recommendation 2
Develop a clinical costing system
Actions
Commit resources to MOH to link current hospital data collection with
current cost of treatment
How best achieved
Purchase expertise to develop system
Train staff to contribute to dynamic system
Example
Australian Diagnostic Related Groupings adds value to data collected
at patient level by costing treatment received

Recommendation 3
Improve the monitoring of risk factors for disease
Actions
Commit additional resources to MOH to conduct regular surveys to better
understand the prevalence and behaviours of key risk factors for
disease
How best achieved
Add additional questions to Census
Regular population surveys to capture prevalence of risk factors
Example
Vanuatu
non-communicable disease survey report of key risk factors for
NCD
Diabetes survey in Kingdom of Tonga
STEPS surveillance WHO

Recommendation 5
Develop a strategic NCD plan for each country
Actions
Based on the STEPS framework, develop an evidence-based strategic plan
to guide implementation on NCDs
How best achieved
Ensure multi-sectoral participation and commitment across all
government ministries to the development of the NCD plan and its
implementation
Example
Tonga NCD Control Strategy

acknowledgments

I would to like to acknowledge the support provided by the following
people
In Kiribati: Jason Hayward-Jones, Dr Kabwea Tiban, Tebau Barantarawa,
Dr Burentau, Loelu Tatapu, Ariane Kienene, Dr Tibwerate, Kireata
Ruteru, Vas Banschikov,
Peter Tong, Nuntaake Tokamauea, Dr Ueantabo
Neemia Mackenzie
In Vanuatu: Theto Wimae Moses, Maturine Tary, Yvannah Taga, Peter Toa,
Morgan Armstrong, Leon Stevens, Myriam Abel, Ben Wota
In Tonga: Siu, Dr Viliami Puloka, Dr Niklas Danielson, Sione Hutanga,
Dr Lynleigh Evans, Bhavani Peddintil, Dr Taniela Palu, Sunia Soakai,
Siusiana Fisiinava, Elisiva Naati, Meleseini Lomu, Tom Wilson, Siosaia
Faletau, Aholotu Saafi Palu, Maba Palouka, Lasini and Melenaite

Special thanks must go to Dr Tony Lower and Marie-Ange Roberts from SPC for
their assistance throughout this project; Dr Anthony Shakeshaft for his
editorial comments; and, Megan Bakewell for her support and encouragement
Without funding from AusAID this project would not have been possible

References

1 World Health Organisation, The World Health Report 2002: Reducing
risks, promoting healthy life, ed C Murray and A Lopez 2002,
Geneva: WHO
2 World Health Organisation, International Statistical Classification
of Diseases and Related Health Problems, 10th Revision ICD-10 1992,
Geneva: WHO
3 Australian Institute of Health and Welfare, Australias Health 2002
2002,
Canberra: AIHW
4 Coyne, T, Lifestyle diseases in Pacific communities, ed R Hughes
and S Langi 2000, Noumea, New Caledonia: SPC
5 Australian Bureau of Statistics, Disability, ageing and carers:
summary of findings, Australia, 1998 ABS Cat No 44300 1999, ABS:
Canberra
6 Khaleghian, P, Non-communicable diseases in Pacific Island
Countries: disease burden, economic cost and policy options 2001, SPC
and the World Bank: Noumea, New Caledonia
7 Statistics Department, Tonga population census 1996: demographic
analysis 1999, Nukualofa: Statistics Department
8 Ministry of Health, Tongas Health 2000 2000, Nukualofa, Tonga:
MOH
9 World Health Organisation, World Health Report 2000 2001, Geneva:
WHO
10 Collins DJ, Lapsley H, Counting the cost: estimates of the social
costs of drug abuse in Australia in 1998-99 National Drug Strategy
Monograph Series No 49 2002, Canberra: Commonwealth of Australia
11 Dalton, A and Crowley, S, Economic costs of NCDs in the Pacific
Unpublished findings 2001, Monash University: Melbourne
12 Hu, S, WHO Mission Report 1999, WHO: Geneva
13 Asian Development Bank, Vanuatu
Economic Performance and Challenges
Ahead 2002, Manilla: Asian Development Bank
14 Health Planning Unit, Republic of Vanuatu: Second National Health
Development Plan 1997-2001 1997, Port Villa: Health Planning Unit,
Ministry and Department of Health,
15 Department of Planning, 2003-2004 Budget Report 2003, Port Villa:
Department of Planning, Health and Administration,
16 National Statistics Office, Vanuatu National Population and Housing
Census 1999 Main Report 2000, Port Villa: National Statistics Office
17 Statistics Office Ministry of Finance, Report of the 2000 census of
population 2002, MOF: Bairiki, Tarawa, Republic of Kiribati
18 National Economic Planning Office, 2002/03 budget 2003, MOF and
Economic Planning: Bairiki
19 World Bank, Kingdom of Tonga health sector reform project: hospital
recurrent costing 2002: Nukualofa, Tonga
20 Ridolfo, B and C Stevenson, The quantification of drug-caused
mortality and morbidity in Australia, 1998 2001, Australian Institute
of Health and Welfare Drug Statistics Series No 7 AIHW Cat No
PHE 29: Canberra
21 Carrao, M, et al, Building the evidence base for
global tobacco
control Bulletin of the World Health Organisation, 2000 787: p
884-90
22 World Health Organisation, Guidelines for the conduct of tobacco
smoking surveys of the general population Report of a WHO meeting
held in Helsinki Finland 29 November to 4 December 1982 1983, WHO:
Geneva
23 Woodward , A, et al, Smoking in the Kingdom of Tonga: report from a
national survey Tobacco Control, 1994 3: p 41-45
24 UNICEF, The State of Health Behaviour and Lifestyle of Pacific Youth:
Kingdom of Tonga Report 2001, Suva: UNICEF Pacific
25 McMurray, C, Kingdom of Tonga Qualitative Study of Youth Substance
Abuse, 2003, Pacific Action for Health Project, Secretariat of the
Pacific Community: Noumea
26 Finlayson, P, et al, Diabetes, obesity and hypertension in Vanuatu
PNG Medical Journal, 1988 31: p 9-18
27 Montaville, B, et al, Vanuatu/WHO/SPC non-communicable disease
survey 1985 1987, Technical Paper No 192: Noumea, New Caledonia
28 Vanuatu Department of Health and FAO, Vanuatu Food and Nutritional
Policy 1986, MOH: Port Villa, Vanuatu
29 Marks, G and M Lund-Adams, Development of the Vanuatu National
food
and nutrition surveillance system 1993: Save the Children Fund,
Australia
30 Carlot-Tary, M, R Hughes, and MC Hughes, Vanuatu non-communicable
disease survey report 2000, Noumea: Secretariat of the Pacific
Community
31 Collins, D and H Lapsley, The economic impact of tobacco smoking in
Pacific Islands 1997, Pacific Tobacco and Health Project, Adventist
Development and Relief Agency: Sydney
32 Stanton, H, Smoking and development in the Pacific Development
Bulletin, 2001 54: p 67-68

———————–

Source:sph.umich.edu

del.icio.us:Diabetes can also contribute to many pregnancy-related complications for the Because the common feature of diabetes is high blood glucose, it is often ...

digg:Diabetes can also contribute to many pregnancy-related complications for the Because the common feature of diabetes is high blood glucose, it is often ...

spurl:Diabetes can also contribute to many pregnancy-related complications for the Because the common feature of diabetes is high blood glucose, it is often ...

newsvine:Diabetes can also contribute to many pregnancy-related complications for the Because the common feature of diabetes is high blood glucose, it is often ...

blinklist:Diabetes can also contribute to many pregnancy-related complications for the Because the common feature of diabetes is high blood glucose, it is often ...

furl:Diabetes can also contribute to many pregnancy-related complications for the Because the common feature of diabetes is high blood glucose, it is often ...

reddit:Diabetes can also contribute to many pregnancy-related complications for the Because the common feature of diabetes is high blood glucose, it is often ...

fark:Diabetes can also contribute to many pregnancy-related complications for the Because the common feature of diabetes is high blood glucose, it is often ...

Y!:Diabetes can also contribute to many pregnancy-related complications for the Because the common feature of diabetes is high blood glucose, it is often ...

Diabetes Treatments

Diabetes can also contribute to many pregnancy-related complications for the Because the common feature of diabetes is high blood glucose, it is often …

Recent Diabetes Treatments Articles

Categories