DECLARATION · Web viewAs can be seen from Figure 4.3, the highest expenditure on essential...
115
FINANCING AND AVAILABILITY OF ESSENTIAL MEDICINES BEFORE AND AFTER INTRODUCTION OF THE NATIONAL HOSPITAL INSURANCE FUND CIVIL SERVANTS AND DISCIPLINED SERVICES MEDICAL SCHEME: A CASE STUDY OF WEBUYE DISTRICT HOSPITAL, WESTERN KENYA BY LUCY WINKIE MECCA, BPHARM U51/63523/2013 A thesis submitted in partial fulfillment of the requirements for the award of the degree of Master of Pharmacy in Pharmacoepidemiology & Pharmacovigilance DEPARTMENT OF PHARMACOLOGY AND PHARMACOGNOSY UNIVERSITY OF NAIROBI
Transcript of DECLARATION · Web viewAs can be seen from Figure 4.3, the highest expenditure on essential...
FINANCING AND AVAILABILITY OF ESSENTIAL
MEDICINES BEFORE AND AFTER INTRODUCTION OF THE
NATIONAL HOSPITAL INSURANCE FUND CIVIL SERVANTS
AND DISCIPLINED SERVICES MEDICAL SCHEME: A CASE
STUDY OF WEBUYE DISTRICT HOSPITAL, WESTERN
KENYA
BY
LUCY WINKIE MECCA, BPHARM
U51/63523/2013
A thesis submitted in partial fulfillment of the requirements for the
award of the degree of Master of Pharmacy in Pharmacoepidemiology
& Pharmacovigilance
DEPARTMENT OF PHARMACOLOGY AND
PHARMACOGNOSY
UNIVERSITY OF NAIROBI
NOVEMBER, 2014
DECLARATION
This thesis is my original work and has not been presented for a degree in any other university
Mecca Lucy Winkie Signature………………………. Date………………………
This thesis has been submitted with our approval as university supervisors.
Dr. Eric M. Guantai Signature………………………. Date………………………
Department of Pharmacology and Pharmacognosy
School of Pharmacy
College of Health Sciences
University of Nairobi
Dr. James Riungu Signature………………………. Date………………………
Senior Technical Advisor- Commodity Security
Jhpiego – Affiliate of Johns Hopkins University
KURHI/Tupange program
i
DECLARATION OF ORIGINALITY FORM
Name of student: Mecca, Lucy Winkie
Registration number: U51/63523/2013
College: Health Sciences
Faculty: Pharmacy
Department: Pharmacology and Pharmacognosy
Course Name: Master of Pharmacy- Pharmacovigilance and Pharmacoepidemiology
Title of work: Financing and availability of essential medicines before and after introduction of
the National Hospital Insurance Fund Civil servants and Disciplined services medical scheme: a
case study of Webuye District Hospital
DECLARATION
1. I understand what plagiarism is and I am aware of the University’s policy in this regard.
2. I declare that this thesis is my original work and has not been submitted elsewhere for
examination, award of a degree or publication. Where other people’s work or my work has
been used, this has properly been acknowledged and referenced in accordance with the
University of Nairobi’s requirements.
3. I have not sought or used the services of any professional agencies to produce this work.
4. I have not allowed, and shall not allow anyone to copy my work with the intention of passing
it off as his/her own work
5. I understand that any false claim in respect of this work shall result in disciplinary action, in
accordance with University Plagiarism Policy
Signature…………………...
Date.............................
ii
ACKNOWLEDGEMENT
I would like first and foremost like to thank God who performs all things for me.
I would like to thank my supervisors Dr. Eric M. Guantai and Dr. James Riungu for their
invaluable and gracious input.
I would also like to thank the staff at Webuye District Hospital, particularly the Medical
Superintendent-Dr. Caesar Bitta, pharmacists-Dr. Martha Mandale, Dr. Lidya Anyanzwa, Dr.
Ferdinand Ndubi and all pharmacy staff for their selfless assistance.
I am also grateful to my friend Mercy Mulaku for her generous input.
I would also like to acknowledge my classmates who were instrumental in bringing this thesis to
completion.
I would also like to appreciate my employer who gave me leave to do this work.
iii
Table of ContentsDECLARATION.......................................................................................................................................... i
DECLARATION OF ORIGINALITY FORM............................................................................................ii
ACKNOWLEDGEMENT.......................................................................................................................... iii
LIST OF TABLES.....................................................................................................................................vii
LIST OF FIGURES..................................................................................................................................viii
LIST OF APPENDICES............................................................................................................................. ix
OPERATIONAL DEFINITION OF TERMS..............................................................................................x
ABBREVIATIONS...................................................................................................................................xii
ABSTRACT..............................................................................................................................................xiii
CHAPTER 1: INTRODUCTION................................................................................................................1
1.1 BACKGROUND...............................................................................................................................1
1.2 PROBLEM STATEMENT................................................................................................................5
1.3 JUSTIFICATION..............................................................................................................................6
1.4 RESEARCH QUESTIONS...............................................................................................................6
1.5 NULL HYPOTHESES......................................................................................................................7
1.6 OBJECTIVES....................................................................................................................................7
1.6.1 Main Objective...........................................................................................................................7
1.6.2 Specific Objectives.....................................................................................................................7
CHAPTER 2: LITERATURE REVIEW.....................................................................................................8
2.1 INTRODUCTION.............................................................................................................................8
2.2 HEALTH INSURANCE...................................................................................................................8
2.2.1 Health Insurance and medicines.................................................................................................9
2.2.2 Moral Hazard..............................................................................................................................9
2.3 CASE STUDIES OF INSURANCE SCHEMES AND IMPACT ON MEDICINES.......................10iv
2.3.1 China........................................................................................................................................10
2.3.2 Thailand....................................................................................................................................10
2.3.3 Mexico......................................................................................................................................11
2.3.4 Tanzania...................................................................................................................................12
2.3.5 Nigeria......................................................................................................................................12
2.3.6 Ghana........................................................................................................................................13
2.3.7 Kenya........................................................................................................................................13
2.3 CONCEPTUAL FRAMEWORK....................................................................................................14
CHAPTER 3: METHODOLOGY.............................................................................................................16
3.1 STUDY DESIGN............................................................................................................................16
3.2 LOCATION OF STUDY.................................................................................................................16
3.3 DATA COLLECTION....................................................................................................................16
3.4 DATA ANALYSIS.........................................................................................................................20
3.5 LIMITATIONS...............................................................................................................................21
3.6 ASSUMPTION................................................................................................................................21
3.7 ETHICAL CONSIDERATIONS AND APPROVAL......................................................................22
CHAPTER 4: RESULTS...........................................................................................................................23
4.1 Allocation of FIF before and after the introduction of the NHIF civil servants scheme...................23
4.1.1. Quarterly allocation of funds for purchase of medicines from the FIF (Absolute amounts). . . .23
4.1.2 Proportion of total FIF allocated quarterly for purchase of medicines......................................25
4.2 Contributors to the Essential Medicines Budget..............................................................................25
4.2.1 Proportion of essential medicines procured through KEMSA, FIF and Other Facility.............26
4.2.2 Expenditure on essential medicines procured through KEMSA, FIF and Other Facility..........28
4.2.3 Efficiency.................................................................................................................................30
4.3 Availability of essential medicines..................................................................................................31
4.3.1 Monthly stock-out time (%)......................................................................................................31v
4.3.2 Stock-out rate per essential medicine........................................................................................34
4.4 Analysis for factors that influence stock-out rate.............................................................................36
4.4.1 Monthly data- Autoregressive Integrated Moving Average......................................................36
4.4.2 Total days out of stock- Negative binomial regression.............................................................37
CHAPTER 5: DISCUSSION....................................................................................................................39
CONCLUSION.........................................................................................................................................45
RECOMMENDATIONS...........................................................................................................................45
REFERENCES..........................................................................................................................................46
vi
LIST OF TABLES
Table 3.1 : Summary table for methods……………………………………………………..…....
….17
Table 4.1: Quarterly Percentage allocation of FIF for purchase of medicines……………............
…25
Table 4.2: Results from within-source comparison of proportion of essential medicine
procured….27
Table 4.3: Results from within- source comparison (expenditure).……………………………….
…29
Table 4.4: Efficiency in number of units per Ksh - Summary statistics……………..…………...
….30
Table 4.5: Monthly stock-out time (%) - Summary statistics……………………………….........
….31
Table 4.6: Results from comparison of stock out rate of different months in a quarter…..……...
….33
Table 4.7: Significant p-values for Wilcoxon Signed Rank Test for differences in stock out rates
for different classes of essential medicines………………………………………….……..……...
…….33
Table 4.8: 20 essential medicines with highest and lowest stock-out rates in 2010-2013….
………..35
vii
Table 4.9: Significant p values for independent variables regressed with log of monthly % stock-
out time…………………………………………………..…………………………………..…...
….…..37
Table 4.10 Significant p-values for independent variables regressed with total days out of
stock………………………………………………………………………………………………
….38
viii
LIST OF FIGURES
Figure 1.1 The Access Framework…………….……………………………………………..…….
…2
Figure 4.1 Quarterly FIF allocation for purchase of medicines-summary statistics ………..
……….24
Figure 4.2 Mean proportion (%) of essential medicines procured and source……………...
……….26
Figure 4.3 Mean expenditure on essential medicines by source………..………..………….
……….28
Figure 4.4 Monthly average stock-out time (%)…………………………………….……….
………32
ix
LIST OF APPENDICES
APPENDIX I- Data Collection Forms…………………………..……………………...
……………52
APPENDIX II- Letter of Ethical Approval………………………………………………….
………60
x
OPERATIONAL DEFINITION OF TERMS
Essential Medicines- as defined by the World Health Organization are those medicines that
satisfy the priority health care need of the population. They are selected with due regard to
disease prevalence, evidence on efficacy and safety, and comparative cost-effectiveness.
For the purpose of the study, essential medicines are those that are common to the hospital draft
formulary and the Kenya Essential Medicines List.
Capitation-a fixed sum per person paid in advance of the coverage period to a healthcare entity
in consideration of its providing, or arranging to provide, contracted healthcare services to the
eligible person for the specified period
Cost Sharing- that portion of health care costs not borne by the funding agency (the
government). It includes all contributions, including cash, which a recipient makes towards
health care.
District Hospital - A public hospital gazetted as such by the government, also known as a level
4 or 5 hospital. It is usually the referral hospital in a district.
Facility Improvement Fund - Cost sharing revenues which are additional to budget allocations
from the Treasury.
Fee-for-service: separate payment to a health-care provider for each medical service rendered to
a patient.
Fill-rate: Percentage of total order (by cost) that was supplied
Low, Middle and High Income Country-Economies are divided according to 2012 Gross
National Income per capita, calculated using the World Bank Atlas method. The groups are: low
xi
income, $1,035 or less; lower middle income, $1,036 - $4,085; upper middle income, $4,086 -
$12,615; and high income,$12,616 or more.
Out-of-pocket payments-include any direct payments made by patients to health care providers
at the point of use. These are also amounts which users are required to pay for health care that
are separate from any contributions to voluntary or mandatory insurance or through general
taxation
Quarterly Pharmacy Budget: A quarterly expenditure plan for both non essential and essential
medicines at a district hospital.
Universal coverage-usually refers to a health care system which provides health care and
financial protection to all its citizens. It is organized around providing a specified package of
benefits to all members of a society with the end goal of providing financial risk protection,
improved access to health services, and improved health outcomes
xii
ABBREVIATIONSAIE- Authority to Incur Expenditure
DDD-Daily defined dose
EEC- Executive expenditure committee
FIF-Facility Improvement Fund (Cost sharing)
HMT- Hospital Management Team
KEML- Kenya Essential Medicines List
KEMSA-Kenya Medical Supplies Agency
KSHS- Kenya Shillings
LMIC-Low and Middle Income Countries
MDG- Millennium Development Goal
MEDS- Mission For Essential Drugs and Supplies
MOH-Ministry of Health
NGO-Non-governmental organization
NHIF- National Hospital Insurance Fund
NHIS- National Hospital Insurance Scheme
NSHI-National Social Health Insurance Fund
UN-United Nations
WHO- World Health Organizationxiii
ABSTRACT
Introduction
About 30% of the world’s population is estimated not to have access to essential medicines.
Studies done in Kenya have shown that public facilities experience stock-outs of basic essential
medicines for about 46 days per year.
One of the determinants of access to essential medicines is financing. Introduction of the
National Hospital Insurance Fund Civil Servants and Disciplined Services Medical Scheme
affords additional funding for a district hospital. Following this, it is expected that the stock-out
rate of essential medicines at Webuye District Hospital would reduce.
Main Objective: The study aimed to compare availability of essential medicines and funding of
essential medicines before and after implementation of the National Hospital Insurance Fund
Civil Servants and Disciplined Services Medical Scheme.
Methods: This was a retrospective longitudinal before-after study of four years; the latter two of
which the National Hospital Insurance Fund Civil Servants and Disciplined Services Medical
Scheme package was in operation. The study period was January 2010 – December 2013. Stock
control cards from the pharmacy store and accounting records were the main sources of data.
Data was extracted into various data collection forms. Data was analyzed using MS Excel, SPSS
and STATA. Results were statistically significant when p<0.05.
xiv
Results: The period after introduction of the scheme experienced a significantly higher allocation
for the medicines budget from the Facility Improvement Fund (p=0.008). Actual expenditure on
essential medicines was also higher. There was less funding from the government supplier of
medicines, KEMSA after introduction of the new scheme (p<0.0001). There was a change in
stock-out rate after introduction of the new scheme, falling from 21.75% in 2010/11 to 19.47% in
2012/13. The change was however not statistically significant (p=0.099). Regression analysis
found that an increase in the amount of Facility Improvement Fund spent on essential medicines
was a significant independent predictor of a reduction in stock-out rate, and that a higher rate of
unfilled KEMSA orders for a particular medicine predicted a higher stock-out rate.
Conclusion
Even though financing of medicines through the Facility Improvement Fund increased after
introduction of the new scheme, there was no change in the stock-out rate due to prevailing
contextual factors.
xv
CHAPTER 1: INTRODUCTION
1.1 BACKGROUND
Along with skilled and dedicated healthcare providers, medicines are the most significant means
that society possesses to prevent, alleviate, and cure disease (UN, 2005)
Essential medicines are those medicines that satisfy the priority health care needs of the
population. Essential medicines are intended to be available within the context of functioning
health systems at all times, in adequate amounts, in the appropriate dosage forms, with assured
quality and adequate information, and at a price the individual and the community can afford
(WHO, 2002).
The following criteria are used by the WHO Expert Committee on the Selection and Use of
Essential Medicine: Adequate evidence of efficacy and safety in a variety of settings, relative
cost-effectiveness, suitability of pharmacokinetic properties and formulation as single
compounds.
No health system can afford to supply all medicines that are available on the market. Lists of
essential medicines guide the procurement and supply of medicines in the public sector, schemes
that reimburse medicine costs, medicine donations, and local medicine production. The selection
of the medicines should be done with regard to evidence-based standard clinical guidelines. The
Kenya Essential Medicines List (KEML) 2010 was developed by The National Medicines &
Therapeutics Committee (NMTC) in consultation with WHO (KEML). It is planned that a new
edition of the KEML will be produced at least once every 2-3 years.
A hospital formulary is a list of medicines consisting of the most cost-effective, safe, locally
available medicines of assured quality that will satisfy the health care needs of the majority of
the patients. It is formulated by the Hospital Drugs and Therapeutic Committee using agreed
upon criteria (based on WHO criteria for selection of essential medicines).
1
Access is defined as having medicines continuously available and affordable at public or private
health facilities or medicine outlets that are within one hour’s walk from the homes of the
population (UN, 2003). Access to medicines depends on four factors as illustrated by the
framework below (Figure 1.1).
Figure 1.1: The Access Framework (Adapted from WHO Policy Perspectives on Medicines,
2004)
One of the ways hospitals can ensure rational selection and use of medicines is by coming up
with national essential medicines lists and hospital formulary lists that guide procurement and
use of medicines. Mechanisms that make medicines affordable include promoting bulk
procurement, implementing generics policies, eliminating duties, tariffs and taxes on essential
medicines and encouraging local production of essential medicines of assured quality. Reliable
supply systems can be realized by public-private-NGO partnerships in supply delivery, proper
regulatory control and exploring various purchasing schemes
In many high income countries, over 70% of pharmaceuticals are publicly funded whereas in low
and middle income countries (LMIC) public medicine expenditure does not cover the basic
needs of the majority of the population. In these countries 50-90% of the medicines are paid for
2
by patients themselves (WHO 2004b). Kenya is classified as a middle income country by the
World Bank. Health care in Kenya is relatively costly, as a result of the widespread user fees at
government health facilities together with other out-of-pocket payments at NGO and other
private health facilities (Xu et al., 2006). Households are the largest contributors of health funds
(35.9%) followed by donors (31%), and then the government (29.3%) (Luoma et al., 2010). The
total government health expenditure as a percent of total government expenditures has continued
to decline, from a high of 8.6 percent in 2001/02 to 4.6 percent in 2009/10 (MOMS & MOPHS
2011)
Currently, the public health system in Kenya relies on four main sources of financing: General
government revenues (taxes), donor funds, user fees and the National Hospital Insurance Fund
(NHIF), a government-sponsored health insurance scheme.
Total pharmaceutical expenditure accounts for 1.65 % of GDP and makes up 36.64 % of the total
health expenditure. Government expenditure on pharmaceuticals represents 9.03 % of the total
expenditure on pharmaceuticals in the Kenya (MOMS, 2010). Before the devolution of health
services, the government procured medicines through Kenya Medical Supplies Agency.
KEMSA’s 2010/2011 Government budget (not counting donor contributions) for the
procurement of essential medicines for public hospitals was US$ 19.8 million; Out of 343 items
on the Essential Drug List (EDL), KEMSA procured only about 117 selected items, based on
available funds. Many EDL medicines could not be purchased because of budgetary constraints
(UNIDO, 2010). The Ministry of Health allocated hospitals with yearly drawing rights
depending on various factors including workload, and the poverty index of the area. Hospitals
ordered from KEMSA on quarterly basis using standard order forms. The total value of the order
would not surpass the quarterly drawing rights for the hospital. The order forms contained a
limited number of essential medicines and supply of orders was characterized by a low fill rate
(The World Bank, 2009).
A large proportion of donor contributions to the health sector (78%) went to funding HIV/AIDS
related programs (Chuma and Okungu, 2011)
3
Beginning July 2013, health services in the country were devolved. Finances for essential
medicines were sent to the counties. The discretion of where to procure medicines from fell to
the counties, leaving KEMSA handling only medicines financed by donors (KEMSA, 2013).
User fees (cost-sharing), which were introduced in 1989, also pay for a portion of health services
at public facilities. NHIF reimbursements, free maternity reimbursements from the government
and cash collected from user fees are aggregated into one fund known as the Facility
Improvement Fund (FIF). This fund is used to supplement the government budget in areas such
as purchase of essential medicines, food, payment of salaries for casual workers and utility bills.
Budgeting for FIF is usually done on quarterly basis. Medicines can be purchased from local
suppliers, NGO’s such as Mission for Essential Drugs and Supplies and from the KEMSA
Supplementary Supplies Division.
Revenue from the cost-sharing system has increased exponentially over the years. However, the
revenue’s overall share of total health expenditure for Fiscal 2005/06 was just 6.4% of the
Ministry of Health’s total spending (Wamai, 2009).
The NHIF derives its mandate from the NHIF Act no. 9 of 1998 which establishes the Fund as an
autonomous state corporation. Overall coverage levels for formal and informal sector
populations have reached 4.5 million people (11% of the Kenyan population) (USAID, 2014).
The NHIF is only compulsory for the formal sector workers and until 2012, only covered part of
inpatient health care costs (UNIDO, 2010). The beneficiaries still needed to pay out-of-pocket
fees for treatment, diagnosis and pharmaceuticals (Xu et al., 2006).
There are plans to restructure NHIF in order to increase financing for health. In 2005, legislation
for a National Social Health Insurance (NSHI) was passed with the objective of covering 60 per
cent of the population by 2015 (UNIDO, 2010). The plan is to systematically enroll all of the
Kenyan population, starting with workers and civil servants, followed by the self-employed and
informal sector workers, into the new NSHI Fund.
In 2012, the NHIF introduced a Civil Servants and Disciplined Services Medical Scheme for
civil servants and disciplined services (police officers, prison officers and National Youth
Service). Civil servants and Discipline Services were allocated hospitals which were funded via
capitated payment for outpatient services and other payment policies for inpatient care. Several 4
hospitals including public hospitals were contracted to provide services towards the new scheme.
The hospitals were required to provide effective quality services and treatments of therapeutic
value to the beneficiaries according to evidence-based standards and treatment guidelines as
provided by the Ministry of Health.
Webuye District hospital is a public hospital situated in Webuye town, Bungoma East district in
Bungoma County along the Mombasa-Kampala highway. The hospital has an immediate
catchment area of 500,000 people. The hospital has a bed capacity of 217 beds and bed
occupancy of up to 150 %. On average 200-300 patients seek medical services from the casualty
department daily.
In January 2012, The Director of Medical Services via circular (Ref.MMS/ADM/1/1/16)
instructed the management of the hospital to begin offering primary health care and treatment
services to civil servants and disciplined forces who had been allocated to the hospital as per the
new NHIF scheme. The hospital also entered into a formal contract with National Hospital
Insurance Fund. The hospital received a capitation premium of Kshs 3500 per year for every
member on the scheme. This capitation was channeled through the FIF account. In return for this
capitation, the hospital agreed to provide outpatient services to all members of the health plan,
regardless of what the actual cost of these services ended up being. In addition, the hospital was
reimbursed in-patient charges at a flat rate per day depending on the job group of the member.
These rates were generally higher than for the general NHIF scheme.
The NHIF provides a list of essential drugs required to be present during accreditation as a
minimum requirement. After accreditation NHIF expected that all prescribed drugs be filled at
the hospital regardless of whether they were essential or not.
1.2 PROBLEM STATEMENT
Stock-outs of essential medicines are common in public hospitals. It is expected that once new
NHIF scheme is introduced, there would be an improvement of services due to increased
funding, and also in order to meet the expectations of the clients. It is therefore expected that
stock-outs of essential medicines will be reduced.
5
On the other hand, insurance cover is generally associated with a greater demand for services and
this may include non-essential services and goods (Moral Hazard). The hospital may therefore
stock non-essential medicines even when there are stock-outs of essential medicines. This could
lead to an increased stock-out rate of essential medicines.
All out-patients are served from the same pharmacy, regardless of whether they are on the NHIF
scheme or not. This means that when a medicine is out of stock at the hospital pharmacy, NHIF
beneficiaries have to make out-of-pocket purchases from private pharmacies.
1.3 JUSTIFICATION
Webuye District Hospital signed a contract with NHIF to provide both in-patient and out-patient
services by a capitated system. Essential services should therefore be provided adequately. It is
important that systematic studies be carried out to assess the availability of essential medicines
after implementation of the NHIF Civil Servants and Disciplined Services Medical Scheme. This
is particularly important considering that there is a possibility that the introduction of the scheme
may paradoxically compromise the availability of essential medicines.
To the best of our knowledge no studies have been done in Kenya on the effect of introduction of
the NHIF Civil Servants and Disciplined Services Medical Scheme.
Factors that may affect availability of medicines also need to be explored. These include the
KEMSA supply and amount of FIF allocated to purchase medicines.
The findings of this study will be essential in providing the management of the hospital with
information on the status of essential medicine availability before and after implementation of
NHIF and making recommendations on how to improve availability of these medicines.
1.4 RESEARCH QUESTIONS
The study aims to answer the following questions:
1. Has the funding for the budget for medicines changed since the introduction of the NHIF
Civil Servants and Disciplined Services Medical Scheme?
2. Are stock-outs of essential medicines reduced since the introduction of NHIF Civil Servants
and Disciplined Services Medical Scheme?6
3. What are the factors that affect availability of essential medicines in a hospital pharmacy?
7
1.5 NULL HYPOTHESES
1. The average amount and proportion of FIF allocated in the years 2010/2011 = years
2012/2013.
2. The percentage quantity contribution and expenditure on essential medicines for each of the
sources of essential medicines years 2010/2011= years 2012/2013
3. The frequency and duration of stock-outs of essential medicines in years 2010/2011= years
2012/2013
4. Availability of essential medicines is not influenced by any factor.
1.6 OBJECTIVES
1.6.1 Main ObjectiveThe main objective of this study was to compare the availability and funding for essential
medicines before and after introduction of National Health Insurance Fund Civil Servants and
Disciplined Services Medical Scheme at Webuye District Hospital.
1.6.2 Specific Objectives1. To compare the proportion of FIF allocated for procurement of medicines before and after
implementation of the NHIF Civil Servants and Disciplined Services Medical Scheme.
2. To measure and compare the proportional contribution of the following sources of funding to
the essential medicines budget before and after implementation of the NHIF Civil Servants
and Disciplined Services Medical Scheme: FIF, KEMSA and miscellaneous sources.
3. To determine the frequency and duration of stock-outs of essential medicines before and after
implementation of the NHIF Civil Servants and Disciplined Services Medical Scheme.
4. To explore some factors that could affect the stock-out rate of essential medicines.
8
CHAPTER 2: LITERATURE REVIEW
2.1 INTRODUCTIONMillennium Development Goal 8, Target 8.E states as follows ― “In cooperation with
pharmaceutical companies, provide access to affordable essential medicines in developing
countries” (UN, 2003)
About 30% of the world’s population, or between 1.3 and 2.1 billion people, are estimated not to
have access to the essential medicines. In India, an estimated 499–649 million people (50% to
65% of the population) do not have regular access to essential medicines. Throughout Africa, a
further 267 million people (almost half the population or 15% of the world total) also lack access
(WHO, 2004a).
A study conducted in health facilities in Kenya found that public facilities experienced stock-
outs of basic essential medicines for about 46 days per year. The public sector supply chain
was particularly prone to significant interruptions and critical stock outs, extending beyond
30 or even 90 consecutive days (MOMS & MOPHS, 2009).
The WHO created a framework for expanding access to medicines, which consists of four
components: rational selection and use of medicines, affordable prices, sustainable financing,
and reliable supply system.
One way to create access is by ensuring essential medicines are always available.
2.2 HEALTH INSURANCE
Health Insurance is a mechanism for spreading the risks of potential healthcare costs over a
group of individuals or households, with the goal of protecting the individual from a catastrophic
financial loss in the event of serious illness.
By pooling financial contributions from many people, insurance plans can cover the hospital
expenses of those experiencing catastrophic events, such as near−fatal illness or injury. Without
access to such insurance, many people are unable to obtain treatment or must incur debts to pay
hospital bills. Insurance mechanisms can also generate large volumes of revenue for health
services.
9
Many African governments in theory have been providing "health insurance" to their populations
for years in the form of free services, financed by tax. However, rising costs, limited funding,
and increasing inefficiency have greatly weakened the ability of public systems to provide
effective care and universal coverage. (Shaw, 1995)
2.2.1 Health Insurance and medicinesThere are few studies on effects of health insurance on medicine use in LMIC. Most of the
studies are observational studies whose results must be cautiously interpreted. These studies also
have different outcomes.
Results from the World Health Survey as cited by Faden et al., (2011a) suggest that insurance
hardly improves access to medicines. However, this could be because many insurance schemes
in developing countries lack a comprehensive medicines benefit or require substantial cost-
sharing for medicines (Faden et al., 2011a).
There is evidence that health insurance reduces financial barriers to access in LMIC. It has been
shown that providing health insurance can improve consumer access to and utilization of
pharmaceuticals as well as health outcomes (Faden et al., 2011b). Studies either compare the
insured to the uninsured within a population, or they compare utilization before and after
insurance was implemented. It has been shown that providing insurance was associated with an
increased use of medicines with one study finding that insurance to be most important factor for
explaining utilization of medicine (Faden et al., 2011b).
Health insurance schemes are known to have great potential to improve the cost-effective use of
medicines by incentivizing better provider prescribing, more cost-effective use by consumers,
and lower prices from pharmaceutical companies (Faden et al., 2011b).
2.2.2 Moral HazardMoral hazard refers to the change in behavior induced by insurance coverage. In other words, if
people do not experience consequences they will behave irresponsibly. This means that low out-
of-pocket cost will lead to waste. Health insurers attempt to mitigate this through the
introduction of co-payments by users. It is assumed that if users are required to make a payment
towards their health care at the point of service, more prudent utilization will be encouraged
(Shung-King, 2011).10
However, there is a concern that out-of-pocket payments, of which co-payments is one kind, may
have unintended negative consequences on the economy of households and in the context of
poorer households, may deepen their level of poverty (Shung-King, 2011).
2.3 CASE STUDIES OF INSURANCE SCHEMES AND IMPACT ON
MEDICINES
2.3.1 ChinaThe Shenzhen labor health insurance in China is a capitated social health insurance system for
migrant workers.
A study showed that the insurance had improved accessibility to essential medicines for migrant
workers (Zhu et al., 2008). Insurance indicators within two periods before and after 1st June
2006 were compared.
Percentage costs of essential medicines procured increased from 43.1% to 46.1%. This indicator
reflects the application of National Essential Medicines List as the base for purchasing of
medicines, and hence rational use of medicines. It was calculated as the costs of purchased
Essential Medicines in a period divided by the total costs of medicines purchased in that period;
costs of medicines per outpatient visit decreased from 24.94 Renminbi to 22.20 Renminbi.
Percentage costs of medicine per outpatient visit decreased, and number of outpatient visits
increased. Only the decrease in costs of medicines per outpatient visit was statistically significant
(p = 0.010).
The popularization of this insurance was recommended to other provinces in the country.
2.3.2 ThailandIn 2001, Thailand implemented the Universal Coverage Scheme (UCS), a public insurance
system. This was in order to achieve universal access to healthcare.
A study was conducted to evaluate the impact of the UCS on utilization of medicines in Thailand
for three non-communicable diseases: cancer, cardiovascular disease and diabetes. Although the
majority of increases in sales were for essential medicines, there were also post-policy increases
in sales of non-essential medicines (Garabedian et al., 2012).
11
The policy was associated with a 39% increase in antidiabetic product sales 5 years after
implementation. One year after the policy, the sale of insulin was 35% higher and at 5 years
174% higher than what would have been expected in the absence of the UCS.
There appears to have been a mixed impact on sales of cardiovascular medicines. Five years
after the policy, the sale of lipid-lowering agents was nearly double (108%) what would have
been expected in the absence of the scheme. The increase was primarily due to sales of branded
generic Simvastatin and Gemfibrozil products, which are on the National List of Essential
Medicines (NLEM), and a slight increase in sales of originator atorvastatin products, which were
not on the NLEM until 2004. For antihypertensives and cardiac medicines there was no
statistically significant change..
The results were also mixed for cancer medicines. There was no significant 1-year or 5-year
impact on the sale of antineoplastics or cytostatic hormones. However, there was an immediate
reduction in sales of immunostimulating agents to a sharp reduction in sales of interferon α-2b, a
non-NLEM medicine, around the time of UCS implementation, following recall of the product
(Garabedian et al., 2012).
2.3.3 MexicoIn a move towards universal coverage, the Sistema de Protección Social en Salud (System of
Social Protection in Health) was approved by law in 2003. Poor families could now enroll in the
Seguro Popular (People’s Insurance) which assured access to comprehensive health care.
During the reform period, regular external measurements of the availability of drugs in public
institutions were carried out. There was improved availability of drugs after the reform. In 2002,
only 55% of the prescriptions issued in Ministry of Health outpatient clinics were fully filled. By
2006, this percentage of fully filled prescriptions had increased from 55% in 2002 to 79% in
Ministry of Health outpatient clinics in general and to 89% in Ministry of Health outpatient
clinics serving Seguro Popular beneficiaries. In some states, 97% of the prescriptions issued in
outpatient clinics serving Seguro Popular beneficiaries were fully filled. In 2006, the percentage
of prescriptions issued in social security institute outpatient clinics that were fully filled was
consistently above 90, as opposed to less than 70 in 2002 (Frenk et al., 2009).
12
2.3.4 TanzaniaA survey was carried out in Tanzania to collect information on medicines coverage and health
insurance programs in the country. The National Health Insurance Scheme for public servants
covers more than 5% of the total population. It became apparent that medicines were covered by
insurance but medicines availability in the treatment facilities was an area which needed to be
addressed (Ministry Of Health And Social Services, 2008).
In another study, increasing enrolment in the government’s existing health insurance schemes
was frequently cited as one of the most promising solutions for addressing stock-out problems
(Wales et al., 2014). In the case that medicines needed by health facilities were stocked out by
Medical Stores Department, these funds could be used to supplement. At that time however, the
amount of funds available through these schemes was relatively small in comparison with the
budget for purchasing drugs from the medical stores department.
2.3.5 NigeriaIn 2005, Nigeria introduced the Formal Sector Social Health Insurance component of National
Health Insurance Scheme. Before 2005, the prescription pattern at a Nigerian military hospital
was reported as irrational with high drugs per encounter, low use of generic prescriptions and
high level of injection use (Adebayo et al., 2013). The average number of drugs per prescription
after implementation of the scheme was lower than recorded in before (2.6 vs. 3.0). It was
inferred that moral hazard may have been avoided (Adebayo et al., 2013). This could be due to
the use of capitation payments for drugs dispensed at the general outpatient clinics. There were
far fewer injections in the after than before (8.9% vs 23.9%) (Adebayo et al., 2013).
There was evidence of moral hazard shown by slight increase in antibiotic prescription more
patients encountered antibiotics in this present after than before (35.0% vs 27.8%) (Adebayo et
al., 2013).
Studies have found that in general, fee-for-service payment systems provide implicit financial
incentives to increase service quantity whereas fixed reimbursement systems, such as capitation
and fixed salaries, give doctors incentive to contain costs (Faden et al., 2011b).
13
2.3.6 GhanaIn 2005, Ghana took a step toward providing universal coverage by introducing the National
Hospital Insurance Scheme (NHIS). The beneficiaries of the scheme receive free health care,
including a drug benefit package covering a wide range of treatments.
At the end of 2008, 61% of the population was covered by the NHIS (USAID Health Systems).
Membership in the NHIS is mandatory unless individuals obtain private health insurance (less
than 1% of the population). (McIntyre et al., 2008)
In one study, the NHIS is observed to impact negatively on the ability of health facilities to
acquire medicines both in terms of quality and quantity (SEND-Ghana, 2010).This was reported
in about 92% of facilities. Reasons cited were high market prices of some drugs/medicines,
delayed claims reimbursement and the exclusion of some effective drugs from the medicines list.
Increased financing for drugs caused by the rollout of the National Health Insurance Scheme
resulted in an increase in demand for drugs as evidenced by a tripled turnover at the level of
regional medical stores in two years (Seiter, 2010). The central medical stores were however
unable to respond in time. Private wholesalers stepped in. The NHIS contracts with the both
public and private providers, including private pharmacies.
Additional financing in combination with partial privatization of the supply chain appears to
have increased access to medicines for patients in Ghana.
2.3.7 KenyaIn the year 2012, the National Hospital Insurance Fund introduced a Civil Servants and
Disciplined Services Medical Scheme. Beneficiaries of the scheme were allocated hospitals and
hospitals were funded via capitated payment for outpatient services and other payment policies
for inpatient care. The effects of this system on availability of medicines are yet to be studied.
With greater insurance coverage, the demand for medicines is naturally expected to rise although
the impact is difficult to quantify (UNIDO, 2010).
In conclusion, although local data is scarce, it is evident that insurance has varied effects on
availability of medicines. It is to be expected that there will be an increase in demand for drugs
following introduction of an insurance model. Moral hazard is an adverse phenomenon in
insurance that has the potential to reduce availability of essential medicines.
14
2.3 CONCEPTUAL FRAMEWORK
Availability of essential medicines is partly determined by financing. Sources of financing for
the essential medicines budget for the district hospital are described in the conceptual framework
below. What would happen to efficiency in utilization of funds and availability of essential
medicines if the hospital received additional funding? Other factors may also have an effect on
availability of essential medicines.
15
Process-FIF Allocation for Pharmacy Budget (Local Purchase Orders to suppliers)
Essential Medicines in hospital formulary: Analgesics, Antibiotics & Anti-fungal, Antimalarials, Cardiovascular drugs, Gastrointestinal drugs, Drugs for Metabolic Disorders, topical drugs, others
Number of days out of stock for Essential Medicines (before and after NHIF)
Efficiency in utilization of funds (funds spent per unit) before and after NHIF
Determinant of Access to Medicines Rational SelectionAffordable pricesReliable health and supply systems
Sustainable financing
-KEMSA Fill rate-Irrational selection (non-essentials) -Consumption rate -Other factors
Sources of financing for essential medicines in a district hospital-Cash paid by patients as user fees-NHIF inpatient reimbursements-Maternity reimbursementsFIF- NHIF capitation for civil servants (January 2012)
KEMSA quarterly supply of medicines- from the governmentMiscellaneous sources such as medicine donations, other facilities.
Denotes a dependent relationship; the head of the arrow points to the dependent variable.
Denotes a linkage between concepts
16
CHAPTER 3: METHODOLOGY
3.1 STUDY DESIGNA retrospective longitudinal before-after study of four years; the latter two of which the NHIF
Civil Servants and Disciplined Services Medical Scheme was in operation. The study period was
1st January 2010 – 31st December 2013. The new NHIF scheme came into operation on 1st
January 2012.
3.2 LOCATION OF STUDYThe study was conducted at Webuye District Hospital because of ease of access to past records.
This is because the hospital pharmacy is computerized. The hospital provides services to
members of the NHIF Civil Servants and Disciplined Services Medical Scheme. Webuye District
hospital is situated in Webuye town, Bungoma East district in Bungoma County along the
Mombasa-Kampala highway. The hospital is centrally placed with an immediate catchment area
of 500,000 people.
3.3 DATA COLLECTIONA list of essential medicines for evaluation had been prepared by merging the Webuye District
Hospital draft formulary list of 2010 and the KEML 2010, then selecting all medicines common
to both lists (i.e. by complete enumeration/no sampling). The resulting list had 145 essential
medicines.
Table 3.1 outlines the various types of data that were collected for each objective, and indicates
the various records from which the data was collected. The data was extracted into various data
collection forms that had been pre-tested and validated.
17
Table 3.1: Summary for data collection methods
Specific Objective Variables Indicator Method/tool Source of Data
1. Compare the proportion of
FIF allocated for procurement
of pharmaceuticals
FIF Amount allocated, Total Amount for each
quarter from 2010-2013
Difference in
Proportion
allocated for
period 2010-2011
and 2012-2013
Extracted
data into
Form 1
(Appendix I)
Quarterly
Authority to
Incur
expenditure
from accounts
department
2. To measure the proportional
contribution of the various
sources of funding to the
essential medicines budget
Quantity and Cost of each essential medicine
procured sorted by source of funding (KEMSA,
FIF, Other facility)for each quarter from 2010-
2013
Difference
between the
proportion of
medicine
procured, by
quantity and cost,
for period 2010-
2011 and 2012-
2013, sorted by
source of funding
Extracted
data into
Form 3
(Appendix I)
KEMSA orders,
Local Purchase
orders and
invoices from
suppliers, S3
cards, Pharmacy
Summary
budgets from
Supplies office
and Pharmacy
department
3.To determine the frequency
and duration of stock-outs of
Number of days out of stock for each essential
medicine for each month from 2010-2013
Total number of
days out of stock
Extracted
data into
Stock Control
Cards
18
essential medicines for period 2010-
2011 and 2012-
2013
Form 2
(Appendix I)
(electronic and
manual) from
pharmacy
department
4.To explore some factors that
could affect the stock-out rate
of essential medicines
a)KEMSA fill rate Total value of KEMSA order placed and supplied
quarterly
Value of supplies
as a proportion of
total order
Extracted
data to Form
4 (Appendix
I)
KEMSA orders
and Invoices
b)Presence on KEMSA
Standard Order Form
Presence of essential medicine on KEMSA order
form for the period 2010-2013
Yes/No Extracted
data to Form
4(Appendix
I)
KEMSA
standard Order
Forms
c)Nature of medicines procured
by FIF
Cost of Non essential drugs procured quarterly,
Total cost of medicines procured by FIF
Proportion of
expenditure on
non-essentials
Extracted
data to Form
5 (Appendix
I)
Pharmacy
Summary
Budgets
d) Consumption rate of
essential medicines
Adjusted consumption rate of essential medicine in
Daily Defined Dose, Average cost per Daily
Defined Dose
Cost of Adjusted
Consumption rate
Extract to
Form 6
(Appendix I)
Pharmacy
Stock-Control
cards
19
e) Monthly workload In-patient and out-patient workload Number of bed
days, Number of
patients
Extract to
Form7
(Appendix I)
Health Records
20
Unless listed differently in the formulary, medicines with the same formulation and active
ingredient but different strengths were considered to be the same; the quantity was adjusted to be
equivalent to the one in the formulary.
An out-of-stock situation was when the quantity of medicine remaining at a specified point in
time was zero, and was not replenished immediately. 1 day out of stock therefore means the
remaining quantity fell to zero and was replenished the next day.
Prices of medicines procured through FIF were obtained from invoices or S3 card. Prices of
medicines from KEMSA were obtained from the standard order form. Where the cost data was
not available, such as in the case of borrowing from other health facilities, KEMSA prices were
used. Where KEMSA price was not available, the current contract price was used.
3.4 DATA ANALYSIS The quantitative data was entered into Microsoft Excel. The data entered was then checked for
accuracy and completeness before being exported to STATA Software version 10.0 and SPSS
version 20 for statistical analysis.
Stock out rate was calculated in two ways
i) Percentage Monthly Stock-Out Time –for 145 medicines
= 100 %∗Total Number of stock−out days∈themonth
Total number of days∈the month X 145
ii) Stock out rate per medicine
=100 %∗Totalnumber of stock−out days∈the period
Totalnumber of days∈the period∗¿¿
*1461 days for entire period
Because the different medicines had different units, quantity received from the different sources
was converted to a percentage. The total amount received for a certain medicine was used as the
21
denominator when calculating the percentage quantity of medicine provided by KEMSA, FIF or
other facility.
Inferential and descriptive analysis was carried out. Inferential tests were selected according to
the nature of the data. Paired comparisons were done using Wilcoxon Signed Rank Test where
assumptions of the paired t-test had not been met. Friedman test was used to compare related
data of more than one group.
Time series data was tested for autocorrelation. In the event of autocorrelation Auto-regressive
Integrated Moving Average (ARIMA) modelling was used. Lagged values of the variables were
taken into account during regression of the time series.
Negative binomial regression analysis was used for count data that did not meet assumptions of
Poisson regression.
Results were considered statistically significant if the p-value is less than 0.05 or confidence
level of more than 95%.
3.5 LIMITATIONS1. There was a potential risk of missing data since the study was retrospective. Every effort was
made to obtain missing data, for example, electronic records were corroborated and/or
supplemented with manual records.
2. Anti-Retroviral drugs, anti-tuberculosis drugs and contraceptives which were not directly
under the hospital were not studied.
3. This being an observational study at one hospital, the data should be interpreted cautiously if
it is to be applied in a broader setting.
4. There were some non-essential drugs on the hospital formulary.
3.6 ASSUMPTIONIt was assumed that an out of stock in the pharmacy store meant an out of stock even at the point
of use.
22
3.7 ETHICAL CONSIDERATIONS AND APPROVALEthical approval was sought and obtained from the Kenyatta National Hospital/University of
Nairobi Ethics and Research Committee (Ref-KNH-ERC/A/83). A copy of the letter is at
Appendix II.
Approval was also obtained from the administration of Webuye District Hospital.
The information obtained from records was kept confidential and was only used for purposes of
this study.
23
CHAPTER 4: RESULTS
4.1 Allocation of FIF before and after the introduction of the NHIF civil
servants scheme
4.1.1. Quarterly allocation of funds for purchase of medicines from the FIF
(Absolute amounts)
At the beginning of every quarter, the Hospital Management Team meets and discusses various
budget proposals that have been forwarded from the departments. The FIF collections of the
previous quarter are then allocated for various purposes, including purchase of medicines,
according to the final approved budget.
Data on amount allocated for purchase of medicines from FIF was collected for the 8 quarters
before and after the implementation of the NHIF civil servants scheme.
Summary statistics for amount of money allocated for purchase of medicines are listed in Figure
4.1 below
24
Mean Minimum Maximum 50th (Median)
0.0000
2.0000
4.0000
6.0000
8.0000
10.0000
12.0000
year 2010-2011year 2012-2013
Qua
rter
ly F
IF A
lloca
tion
in m
illio
n ks
hs
Figure 4.1 Quarterly FIF allocations for purchase of medicines-summary statistics
The mean and median of the quarterly FIF allocation for purchase of medicines was greater in
the period after introduction of the new NHIF scheme (Ksh1.20 million vs. 0.73million and Ksh
1.04 million vs. 0.70million respectively). The highest quarterly FIF allocation in 2012/13 was
Ksh 2.41 million compared to Ksh 0.85 million in 2010/11. The lowest quarterly allocation in
2012/13 was Ksh 0.75 million. This is greater than the minimum quarterly allocation in 2010/11
which was Ksh 0.55 million.
25
Bivariate analysis: Comparison of amount of FIF allocate before and after new NHIF scheme
Quarterly FIF allocation for purchase of medicines in the period 2012/13 was compared to the
allocation in 2010/2011.The Wilcoxon Signed Rank Test was used because the data was paired
but did not fit the assumption of normality required for the paired t-test.
There is evidence of a significant median difference in FIF allocation for purchase of medicines
between period 2012/13 and 2010/11 (p=0.008). This suggests a significantly higher FIF
allocation for purchase of medicines in the period 2012/13 than 2010/11.
4.1.2 Proportion of total FIF allocated quarterly for purchase of medicinesThe mean quarterly amount of FIF allocated for purchase of medicines as a percentage of the
total amount available to the hospital for budgeting was calculated for each period (Table 4.1)
Table 4.1: Quarterly Percentage allocation of FIF for purchase of medicines
Mean Minimum Maximum
Year 2010 to 2011 7.55 5.14 10.66Year 2012 to 2013 9.12 6.95 12.36
The mean proportion of FIF allocated quarterly for purchase of medicines was slightly higher in
the period after introduction of the new scheme. However the paired t-test, appropriate here
because the data was paired and satisfied the assumption of normality, suggested that the
difference in these means was not statistically significant (p=0.0502).
4.2 Contributors to the Essential Medicines BudgetThree sources were found to contribute to the essential medicines budget: KEMSA, which
consisted of the supply of medicines funded by the MOH allocation/donors; FIF which entails
the purchase of medicines from local suppliers and is funded by cash collected from user fees as
well as NHIF and other government reimbursements; and other facilities, which involves direct
donations to the facility and the supply of medicines obtained from other hospitals. Of note is
26
that the period 2010-2013 experienced an average KEMSA fill rate of 42.7% (by total value) for
essential medicines, i.e. KEMSA honored 42.7% of orders for essential medicines.
4.2.1 Proportion of essential medicines procured through KEMSA, FIF and
Other FacilityTo account for the fact that different medicines are supplied in different units, the quantity of
each medicine received from each of the different sources was calculated as a proportion
(percentage) of the total quantity of that medicine received during that period. These proportions
were then averaged to yield the reported mean percentages. This was done for both periods, i.e.
2010-11 and 2012-13 (Figure 4.2).
2010/11 2012/13 2010/11 2012/13 2010/11 2012/13KEMSA FIF Other Facility
0
10
20
30
40
50
60
SOURCE
%
Figure 4.2 Mean proportion(%) of essential medicines procured and source
27
KEMSA contributed the largest average proportion of essential medicines (48.9%) in the period
before the new scheme was introduced. In the second period, FIF contributed the largest average
proportion of essential medicines (45.5%) (Figure 4.2).
Bivariate analysis: Cross-comparison of the different sources of medicines
Bivariate analysis was carried out to compare the proportions of essential medicines obtained
through the different sources of medicines. The Wilcoxon Signed Rank Tests was appropriate as
the data was paired but non-normal.
In the period 2010/11 the proportion of essential medicines from both KEMSA and FIF was
found to be significantly higher than that from other facilities i.e. KEMSA>Other Facility and
FIF> Other Facility (p=0.000). However, the study did not find a significant difference between
the proportions from KEMSA and FIF for this period (p=0.050).
In the period 2012/13, the proportion of essential medicines from the different sources was as
follows FIF>KEMSA> Other facility (p <0.017).
Bivariate analysis: Within-source comparison
For each source, Wilcoxon Signed Rank Tests were also used to compare proportions of
essential medicines procured in 2010/11 to those procured in 2012/13. The results are shown in
Table 4.2.
Table 4.2: Results from within-source comparison of proportion of essential medicine procured
Source Hypothesis p-value
KEMSA KEMSA 2012/13 < KEMSA 2010/11 0.000
FIF FIF 2012/13 > FIF 2010/11 0.000
Other Facility Other facility 2012/13 > Otherfacility2010/11 0.029
There is evidence of a significant median difference in proportion of essential medicines
received from KEMSA between the period 2012/13 and 2010/11 (p=0.000). This suggests the
28
proportion of essential medicines procured through KEMSA was significantly less after
implementation of the NHIF civil servants scheme.
Conversely, the results suggest that the proportion of essential medicines procured through FIF
was significantly greater after the implementation of the NHIF civil servants scheme (p=0.000).
A similar deduction was made for other facilities, whereby the proportion of essential medicines
received was also significantly greater after implementation of the NHIF civil servants scheme
(p=0.029).
4.2.2 Expenditure on essential medicines procured through KEMSA, FIF and
Other FacilityFor this analysis, for each of the essential medicines, the quantity of medicines received from
each source was converted to monetary terms (Kenya Shillings). Summary statistics were
calculated for the 145 medicines and mean expenditure per essential medicine determined for the
period before and after implementation of the NHIF civil servants scheme (Figure 4.3).
2010/11 2012/13 2010/11 2012/13 2010/11 2012/13KEMSA FIF Other Facility
0.0000
0.2000
0.4000
0.6000
0.8000
1.0000
1.2000
Expe
nditu
re in
'000
00 k
shs
29
Figure 4.3 Mean expenditure on essential medicines by source
As can be seen from Figure 4.3, the highest expenditure on essential medicines appears to be
incurred by KEMSA followed by FIF and while the least expenditure is by other facilities. In
2010/11 KEMSA recorded the highest expenditure per essential medicine at Ksh 111,380
followed by FIF at Ksh 29,798. In the period after implementation of the new scheme FIF
expenditure per essential medicine rose and almost equaled that of KEMSA (Kshs 48,201 vs.
Ksh 51,523). Expenditure per essential medicine for other facilities decreased in the period after
implementation of the new scheme.
Bivariate analysis: Cross comparison of different sources’ expenditure on essential medicines
The Wilcoxon Signed Rank Tests were used to compare the different sources’ expenditure on
essential medicines.
In the period 2010/11, the expenditure on essential medicines from the different sources was
found to be significantly different, i.e. KEMSA>FIF>Other Facility (p=0.000).
In the period 2012/13, FIF>Other Facility (p=0.000) and KEMSA> Other facility (p=0.000)
but the study did not find a significant difference between FIF and KEMSA expenditure on
essential medicines (p=0.338).
Bivariate analysis: Within-source comparison: Expenditure in 2010/11 and Expenditure
2012/13
For each source, Wilcoxon Signed Rank Tests were also used to compare expenditure on
essential medicines in 2010/11 to expenditure in 2012/13. The mean expenditure per essential
medicine for each source before and after implementation of the NHIF civil servants scheme was
compared (Table 4.3)
Table 4.3: Results from within-source comparison (expenditure)
Source Hypothesis P value
KEMSA KEMSA 2012/13 < KEMSA2010/11 0.000
FIF FIF 2012/13 > FIF 2010/11 0.000
Other Facility Other facility 2012/13 < Other facility 2010/11 0.122
30
There is evidence of a significant median difference in expenditure on essential medicines by
KEMSA between the period 2012/13 and 2010/11 (p=0.000). This suggests a significantly lower
expenditure on essential medicines by KEMSA after implementation of the NHIF civil servants
scheme. Conversely, the results suggest that there was a significantly higher expenditure on
essential medicines by FIF after implementation of the NHIF civil servants scheme (p=0.000).
However, there was no evidence of a significant difference in expenditure on essential medicines
by other facilities between the period 2012/13 and 2010/11 (p=0.122).
4.2.3 Efficiency Efficiency is a ratio of output vs. input (resources used). A higher efficiency indicates that for the
same output, less units of resource were used, while a lower efficiency indicates wastage in that
more units of resource were used to produce the same output. A higher efficiency in this case
means less expenditure per unit of essential medicine procured; the reverse is true.
Efficiency was compared for the two periods. For each period, efficiency was calculated for each
essential medicine as follows
Efficiency = number of unitsof essential medicine procuredtotal expenditure
Summary statistics for efficiency calculated for each essential medicine for each period are
outlined below (Table 4.4)
Table 4.4: Efficiency in number of units per Ksh - Summary statistics
N Mean 50th
(Median)
Efficiency 2010to2011 128 1.18 .11
Efficiency 2012to2013 128 1.03 .08
17 essential medicines that did not have receipts in 2010-11 and/or 2012-13 were not included in
the analysis. Analysis using the Wilcoxon Signed Rank test yielded evidence of a significantly
higher efficiency before introduction of the NHIF civil servants scheme (p<0.0001).
31
4.3 Availability of essential medicines
4.3.1 Monthly stock-out time (%)Percentage monthly stock-out time was computed for both periods as follows
Percentage Monthly Stock-Out Time = 100 %∗Total Number of stock−out days∈themonth
Total number of days∈the month X 145
Summary statistics are outlined in Table 4.5 below
Table 4.5: Monthly Stock-out time (%) - Summary statistics
N Mean Std. Deviation
Minimum Maximum Median
Year 2010to2011 24 21.75 3.54 15.12 29.08 21.45Year 2012to2013 24 19.47 6.54 11.26 38.26 19.04
The average monthly stock-out time reduced from 21.75% in 2010/11 to 19.47% in 2012/13.
This means that on average for period 2010/11, essential medicines were stocked-out 21.75% of
the time. In 2012/13, there was a decrease in stock-out rate in that essential medicines were
stocked-out 19.47% of the time.
However, Wilcoxon Signed Rank Test analysis showed that the median difference in monthly
stock-out time between the two periods was not significant (p= 0.099). This suggests that there
was no significant change in stock-out rate after introduction of the new scheme.
Figure 4.4 below displays the plot of mean percentage stock-out time versus time (in months).
32
Figure 4.4 Monthly average stock-out time (%)
There are monthly variations in the percentage stock-out time. Visual inspection of the plot
reveals a notable reduction in percentage stock-out time from June 2012 up to March 2013.
Actually this period records some of the lowest stock-out rates.
Bivariate analysis of monthly stock-out time for different months of the quarter
It appears in Figure 4.4 that there may be variation in monthly stock-out rate depending on the
position of the month in the quarter (i.e. first, second or third month of the quarter). The monthly
stock-out time for each essential medicine was therefore sorted into 3 groups- first month of
quarter, second month and third month. For each essential medicine, the total number of days out
of stock for all the first months of the quarter was computed, then converted to percentage stock-
out time according to the total number of days for the first months. The same applied to the
second and third months. Wilcoxon Signed Rank Tests were used to compare the stock-out rates
of the groups. The results are outlined in Table 4.6 below.
33
Jan-10
Mar-10
May-10
Jul-10
Sep-1
0
Nov-10
Jan-11
Mar-11
May-11
Jul-11
Sep-1
1
Nov-11
Jan-12
Mar-12
May-12
Jul-12
Sep-1
2
Nov-12
Jan-13
Mar-13
May-13
Jul-13
Sep-1
3
Nov-13
0
5
10
15
20
25
30
35
40
45%
Table 4.6: Results from comparison of stock-out rate of different months in a quarter
Ha: p-value
Mean Stock-out time month 1> Month 2 0.0002
Mean Stock-out time month 2> Month 3 0.0004
Mean Stock-out time month 1> Month 3 <0.0001
The results suggest that there were significant differences in the different months of the quarter.
Percentage stock-out time for Month 1>Month2>Month3 (p<0.017)
Class differences in stock-out rate
The Wilcoxon signed rank test was applied to compare the monthly stock-out time for each class
of essential medicines over the two periods. Only five classes had statistically significant median
differences in monthly stock-out time (Table 4.7).
Table 4.7: Significant p-values for Wilcoxon Signed Rank test for differences in monthly
stock-out time for different classes of essential medicines
Class and Ha p-value
Central Nervous System: 2012/13<2010/11 0.000
Gastrointestinal: 2012/13>2010/11 0.001
Topical Drugs :2012/13<2010/11 0.023
Anti-Malarial :2012/13< 2010/11 0.035
Thyroid drugs :2012/13<2010/11 0.001
Four out of the eighteen classes registered a significant decrease in monthly stock-out time after
introduction of the NHIF civil servants scheme. These include medicines for the central nervous
system, Thyroid drugs, Anti-Malarial drugs and Topical Drugs. Gastrointestinal medicines are
the only class that had a statistically significant increase in monthly stock-out time after
introduction of the new scheme.
34
Further analysis of the stock-out rates across the different classes of medicines revealed that, for
the entire period, anti-helminthics had the highest percentage monthly stock-out time than all
other classes (p<0.0001).
4.3.2 Stock-out rate per essential medicineThe percentage stock-out rate for each essential medicine was calculated for the entire study
period as follows.
Stock out rate per medicine=100 %∗Totalnumber of stock−out days∈the period
1461
Results for twenty medicines with highest and lowest stock-out rates are listed in table 4.8 below
35
Table 4.8: Essential medicines with highest and lowest stock-out rates in 2010-2013
20 Medicines with lowest stock-out
Rates
% stock-
out rate
20 Medicines with highest
stock-out rates
% stock
out rate
Amoxicillin 250mg caps 0.00 Chloramphenical suspension 100.00
Cotrimoxazole 400:80 tabs 0.00 Clindamycin tablets 100.00
Insulin soluble human 0.00 Fluoxetine caps 100.00
Oral Rehydration Salts 0.00 Ibuprofen suspension 100.00
Sulfadoxine/pyrimethamine tabs 0.00 Praziquantel 600mg tablets 100.00
Aspirin 300mg tabs 0.07 Salicylic acid powder 100.00
Water for injections 10ml 0.07 Lignocaine spray 100.00
Diazepam injection 0.14 Erythromycin suspension 94.93
Insulin Isophane Biphasic human 30/70 0.62 Gentamicin injection 20mg/2ml 85.56
Nystatin suspension 0.75 Haloperidol decanoate injection 78.44
Magnesium sulphate injection 0.89 Ceftriaxone Inj 250mg 73.31
Suxamethonium injection 0.96 Phenytoin Injection 64.13
Ibuprofen 200mg tablets 1.03 Ciprofloxacin ear/eye drops 59.34
Sodium bicarbonate injection 1.03
Amoxicillin-clavulanic acid
228mg/5ml 52.09
Thiopentone injection 1.03 Valproic acid 200mg 45.11
Chlorpheniramine 4mg tabs 1.16 Paracetamol suppositories 44.28
Chlorpromazine 100mg tabs 1.44 Phenobarbitone injection 41.48
Amitryptilline 25mg tabs 1.78 Dexamethasone 0.5mg tabs 39.77
Neostigmine injection 2.12 Atropine Eye Drops 1% 39.29
Paracetamol 500mg tabs 2.26 Morphine Injection 39.29
Some products were never stocked-out during the entire four years. These include antibiotics
such as amoxicillin capsules and co-trimoxazole tablets. Medicines used in theatre had very low
stock-out rates (<2.5%). These include Neostigmine Injection, Thiopentone Injection,
Suxamethonium Injection. In contrast there were some products that were stocked-out
throughout the entire period. These include antibiotics such as Clindamycin tablets,
Chloramphenicol suspension and the anti-helminthic Praziquantel. Some pediatric preparations 36
had high stock-out rates (>50%). These include Amoxicillin-Clavulanic suspension,
Erythromycin Suspension and Ibuprofen suspension. Topical preparations also had high stock-
out rates (Atropine eye drops, Ciprofloxacin eye/ear drops, Lignocaine spray, Salicylic acid
powder).
4.4 Analysis for factors that influence stock-out rateData on various factors that may influence the stock out rate was collected. This included
KEMSA fill rate, workload, FIF allocation and percentage of FIF expenditure on non essential
medicines. Two different modeling approaches were used for this analysis:
Autoregressive Integrated Moving Average based on monthly data.
Negative binomial regression based on total days out of stock for each essential medicine.
4.4.1 Monthly data- Autoregressive Integrated Moving AverageTime series analysis using an Autoregressive Integrated Moving Average (ARIMA) model was
used to model variation in monthly stock out over the study period. Monthly stock-out time was
found to be auto-correlated and at the first lag.
The monthly stock-out time data was log transformed and first differenced data was used
because the transformed data was found be non-stationary (Dicker Fuller test, p=0.0586)
The following independent variables and their first lags were tested: KEMSA expenditure, FIF
expenditure, Other Facility expenditure, % of total FIF expenditure on non-essential medicines,
outpatient workload and inpatient workload (as total bed days).
A forward stepwise approach was used and the most parsimonious model was retained.
37
Table 4.9 Significant p-values for independent variables regressed with log of monthly % stock-
out time
Independent variable Coefficient P
value
95% Confidence
Interval
FIF expenditure on essential medicines( in Kshs
100000)
-0.0337 0.025 (-0.0633,-0.0042)
One month’s lag of FIF expenditure on essential
medicines( in Kshs 100000)
-0.0280 0.022 (-0.0521,-0.0040)
The model obtained from the time series regression analysis of monthly stock-out rate showed
that the amount of money provided by FIF and its first lag (1 month) significantly influences the
monthly stock-out time. The previous month’s stock-out time also influences the stock-out time.
The money received from KEMSA and other facilities appears not to influence the monthly
stock-out time. The model was statistically significant with a probability > (chi) =0.0014.
All other factors remaining constant, a unit increase in FIF expenditure (in Kshs 100,000) for the
month will result in a decrease in monthly stock-out time by 3.31% of the original stock-out time
(95 % confidence interval of 0.42% to 6.13%).
Similarly, all other factors remaining constant, a unit increase in the previous months FIF
expenditure(in Kshs 100,000) for the month will result in a decrease in percentage monthly
stock-out time by 2.76% of the original stock-out time( confidence interval 0.40% to 5.08%)
The constant in the model was significant (p=0.010) suggesting other factors may also be
influencing monthly stock-out time.
4.4.2 Total days out of stock- Negative binomial regressionThis model was based on the individual medicines and the dependent variable was number of
days out of stock for the whole four years. Since number of days is count data, negative binomial
regression was found to be most appropriate. The data did not meet assumptions for Poisson
regression.
Independent variables explored included: number of times a medicine was ordered and not
received or was not on the KEMSA list, the number of times it was procured by FIF and cost of
adjusted consumption rate (in WHO daily defined dose).38
Table 4.10 Significant p-values for independent variables regressed with total days out of stock
Independent variable Coefficient P
value
95% confidence
interval
Number of times medicine was supplied by FIF -0.1102 0.000 (-0.1478,-0.0726)
Number of times medicine was not supplied by
KEMSA when ordered or not on KEMSA list 0.1146 0.000 (0.0863,0.1430)
All other factors remaining constant, a unit increase in number of quarters an essential medicine
was received through FIF would result in a decrease in total days out of stock by 10.43 % of the
original total days out of stock (confidence interval of 7.00% to 13.74%).
All other factors remaining constant, a unit increase in the number of times an essential medicine
was not received from KEMSA when ordered or not on KEMSA list would result in an increase
in the total days out of stock by 12.15% of the original days out of stock confidence interval
(9.01% to 15.37%).
The constant in the model was significant (p=0.000) suggesting other factors may also be
influencing the total days out of stock.
39
CHAPTER 5: DISCUSSION
Health insurance is one of several medicines financing strategies (WHO 2012). The actual
amount of money allocated for purchase of medicines increased significantly after the
implementation of the new NHIF scheme. NHIF capitation and reimbursement monies are
channeled through the FIF account. The increase was most probably due to additional funding
from NHIF. The introduction of free maternity and subsequent government reimbursements
towards the end of the year 2013 may also have contributed to this increase.
Up until implementation of devolution of health in late 2013, KEMSA was established by law as
the primary public procurement agency of essential medicines for public facilities (UNIDO,
2010). Until June 2013, KEMSA was funded by the national government and by development
partners (KEMSA, 2013). In our study, some essential medicines such as Salbutamol Inhaler,
Dexamethasone Injection and several topical medicines were never procured from KEMSA (0%
proportional quantity). This is partly because KEMSA has a standard order form that is used to
order medicines. At the beginning of 2010, the list had 116 essential medicines. Medicines may
also be ordered but are stocked out at the KEMSA warehouses hence not supplied. The average
fill rate for essential medicines on the KEMSA Standard Order Form was 42.7% meaning on
average less than half the order was usually honored. In one study, the average aggregate fill rate
for the top 100 products was 21%, suggesting chronic shortages of almost all major items (The
World Bank, 2009).
The proportional quantity contributed by FIF ranged from 0 to 100 percent. Some essential
medicines were never procured using FIF while others are procured 100% using FIF. FIF is used
to supplement KEMSA. Some of these perhaps are well provided through KEMSA while others
maybe too expensive to procure using FIF. Drugs such as Artemether Lumefantrine were never
procured using FIF, probably because of cost and also because they are to be dispensed free of
charge. Even though all the medicines are essential, some may be deemed more ‘vital’ hence will
be given priority during procurement.
40
KEMSA contributed a larger proportion of essential medicines before the new NHIF scheme. In
addition, KEMSA expenditure on essential medicines is greater before the new scheme.
Compared to the other sources, KEMSA was the largest contributor to essential medicines
expenditure in the period before introduction of the new NHIF scheme. The difference could be
attributed to the reduction in number of essential medicines included in the KEMSA Standard
Order Form with time. The reduction was aimed at increasing efficiency; on average KEMSA
prices are lower than other facilities due to economies of scale (UNIDO, 2010). At the beginning
of 2010, the list had 116 essential medicines. The number of essential medicines in the list
reduced with time. By September 2013, the list contained 71 essential medicines.
The decreasing supply of essential medicines from KEMSA could also be due to financial
constraints. In low income countries, the proportion of government spending dedicated to
pharmaceuticals has significantly reduced from 21.5% in 1990 to 16% in 2000 (WHO,
2012).WHO estimates the level of public sector funds needed to provide essential medicines in a
basic health care package at around US$ 1.5-2 per capita. The MOH allocates of US$ 1.1 per
capita which is inadequate. In addition, based on national estimates, about Ksh 7.3 billion is
required for pharmaceuticals annually, but less than 2 billion is made available (Luoma et al.,
2010). In a press statement dated 20th August 2013 by the Principal Secretary Ministry of Health,
it was acknowledged that KEMSA was only able to meet 65% of the quantified needs Public
Health Facilities due to budgetary constraints (MO H, 2013). The rest 35% was expected to
come from Development Partners, health facilities cost sharing fees, and out of pocket expenses
by patient. It was expected that once the Abuja Declaration target of 15% of the government
budget to healthcare is met, the funding gap would be eliminated.
Following devolution of health, there was a delay in supply from KEMSA in late 2013;
followed by a halt in automatic supply of medicines. Actually only 2 out of 145 essential
medicines were supplied in the last quarter of 2013. KEMSA states in its website that it was
mandated to supply up to June 2013 and following instructions from the Ministry of Health
supplied a one-off free supply that ended 30th October 2013 (KEMSA, 2013). The devolution of
health meant that the mandate to procure medicines fell to the county governments. All funds for
41
procurement of medical commodities were devolved to the Counties, and the discretion of where
to procure from fell to each particular county. Some counties would opt to continue with
KEMSA, which now was operating under a new business model (KEMSA, 2013). Some
counties opted for other private suppliers. Several counties entered into partnerships with
Mission for Essential Drugs and Supplies (MEDS). Due to the MEDS County Governments
Partnership Initiative, more than thirty counties began procuring from MEDS (MEDS
2013).Generally, public health facilities in Kenya experienced logistical problems when health
was devolved precipitately due to lack of structures and clear cut guidelines. The Parliamentary
Committee for Health received several submissions outlining the extent of shortage of drugs and
other medical supplies in public health facilities. The Committee observed that the capacity of
Counties to take up the function of procuring drugs and medicaments was weak (Parliamentary
Committee for Health).
Our findings showed that the proportion of essential medicines contributed by FIF was greater
after introduction of the new NHIF scheme at Webuye District Hospital. FIF expenditure on
essential medicines was also greater after introduction of the new scheme. Compared to the other
sources, FIF contributed the largest proportional quantity of essential medicines after
introduction of the new scheme. This is likely due to increase in amount of FIF allocated for
medicines in 2012/13 and also because receipts from KEMSA, previously the major supplier,
declined in this period. In one study, it was found that the FIF fund is increasingly being used for
procurement of medicines, to plug the gap resulting from insufficient supplies from KEMSA.
During one assessment, it emerged that between 30 and 60 percent of FIF is now utilized for
procuring pharmaceuticals and medical supplies (Luoma et al., 2010).
Our study showed that the proportion of essential medicines contributed by other facilities was
greater after introduction of the new NHIF scheme at Webuye District Hospital. This could be
attributed to increased cross-health facility exchange of essential medicines. For example,
Webuye District Hospital would exchange a commodity it has in excess for a different
commodity in X District Hospital. The increase in this activity could have been due to declining
KEMSA supplies.
42
The period before implementation of the new scheme experienced a higher efficiency in
utilization of funds probably because supply from KEMSA was higher and KEMSA is generally
cheaper due to economies of scale.
The methodology used in medicine availability studies is quite varied, making comparisons
difficult. Most studies have shown that availability of essential medicines is lower in the public
sector. In our study, overall, there was no significant change in the stock-out rate after
implementation of the new scheme. The lowest stock-out rates were recorded between August
2012 and March 2013 probably because of increase in FIF expenditure on essential medicines.
The stock-out rate seemed to increase after this period perhaps due to decreasing supply of
essential medicines from KEMSA. In contrast, reforms to the Mexico’s Seguro Popular
(People’s Insurance) resulted in improved availability of drugs. The percentage of fully filled
prescriptions increased from 55% in 2002 to above 90% in 2006. This was a scheme that
included everyone as it was aimed at universal coverage. This meant that there was more
financing made available than in the case of our study. (Frenk et al.,2009).
Stock-out rate of individual essential medicines for the entire period was highly varied. Stock-out
rates of 100% were recorded for more than one medicine, meaning some essential medicines
were not available for the entire four years. In contrast a few medicines had stock out rates as
low as 0%. Some medicines were most probably well supplied than others. It also seems that
prioritization of needs may have been in play. For example, some of the medicines with very low
stock-out rates were insulin and theatre medicines. These are used in life threatening situations.
Medicine with the highest stock out rates included some topical preparations and pediatric
formulations. These may be deemed less vital, and also unprofitable because pediatric
preparations are free to the patient. Some non-essential medicines may also have used for the
same purpose as the essential one, for example Pethidine Injection instead of Morphine Injection.
The class of medicineswith the highest stock-out rate was anti-helminthics. This could be
because there were only two medicines in this class and one of them, Praziquantel, was out of
stock during the entire period.
Various other factors were explored to explain variability in stock-out rate. The first month of the
quarter was characterized by a higher stock-out rate than the second and third month. This could
43
be due to delays in procurement especially for medicines procured through the FIF. For the FIF,
procurement of essential medicines is usually done on a quarterly basis. The earliest the authority
to incur expenditure (AIE) can be received is middle of the first month. In one study, the AIE
took an average of 38 days to be processed up to the point of arriving at the beneficiary facility.
The duration ranged from 13 to 95 days (MOH, 2009). Local Purchase Orders may take at least 1
week to prepare and supplies will take at least one week to be received at the facility. There is
still a significant gap in the financing of pharmaceuticals, despite the utilization of a diverse set
of financing mechanisms (Luoma et al., 2010). If financing was sufficient, a five month supply
of essential medicines would be procured to cater for the lead time resulting from the winding
procurement procedures.
Stock-out rate is also dependent on the previous month’s stock-out rate due to autocorrelation
which is common to time series.
The study found a significant relationship between funding and availability of medicines. In a
recent study, inadequate funding was the most strongly cited (57.9%) factor that caused
unavailability of essential medicines in public hospitals (Mwathi and Osuga, 2014). In our study,
actual FIF expenditure on essential medicines was found to have great influence on the
variability in monthly percentage stock-out time. Increase in amount of FIF spent on medicines
led to a decrease in percentage stock-out time. Paradoxically, the expenditure on essential
medicines by KEMSA appeared to have no effect on monthly percentage stock-out time even
though it was greater than the amount from FIF at the beginning of the study period. This could
be due to the fact that KEMSA lists were quite limiting. The variation in amount of money was
probably due to variation in quantity of the same supplies. This reflects lack of flexibility in
expenditure. For the FIF however, the pharmacist had discretion on what to procure. He/she
could therefore prioritize to procure medicines that were not supplied by KEMSA.
Our study also showed that stock out rate depended on the number of times an essential medicine
was purchased through FIF and the number of times it was ordered from KEMSA and not
received or was off the KEMSA list. It means therefore that when a medicine was on the
KEMSA Standard Order Form and orders for that particular medicine were honored, there was
44
decrease in the stock-out rate of that medicine. Erratic supply is therefore a factor than can lead
to stock-outs. Even though KEMSA did not seem to have an effect on monthly stock-out rate, it
had an effect on stock-out rates of individual medicines.
We concluded therefore that FIF had an effect on both variation of monthly stock-out rate and on
individual medicines stock-out rate, while KEMSA had an effect on variation in the stock-out
rate of individual medicines. From literature, inappropriate selection of medicines, irrational use
of medicines, poor inventory keeping, poor forecasting and quantification methods, poor
distribution practices are some of the other factors that are known to reduce availability of
medicines ((Mwathi and Osuga, 2014). These factors arise especially where personnel handling
medicines are non-pharmacists or are pharmacists who lack commodity management skills.
A visual inspection of Figure 4.4 indicates that there was a decrease in stock-out rate after
introduction of the new scheme, but this was not sustainable. The change could have been due to
increased funding of the medicines budget through the next major source, which is FIF. It was
short-lived probably because of the declining supply of essential medicines from KEMSA. NHIF
has low coverage. From the hospital accounting records NHIF monies for both in-patient and
outpatient accounted for at most 15% of the FIF collection. The rest 85% was paid out of pocket
by uninsured patients and from government maternity reimbursements. The money the hospital
receives from NHIF is therefore not expected to meet the needs of the general hospital
attendance. The hospital pharmacy practices equity in that all clients are treated the same, for
example, if a medicine is in stock, it is available for all. This may be seen as disadvantageous to
the NHIF beneficiaries but on moral grounds separation of services may not be advocated for.
Many countries see the initiation or promotion of one or more insurance schemes as a way to
address health financing issues. During implementation, it is important that the context and
surrounding issues are taken into account. The NHIF implemented the scheme into a system that
was experiencing major gaps in funding. The quality of care of the NHIF patients was most
probably compromised due to lack of availability of essential medicines. Issues that should be
addressed when designing a health insurance scheme include policy objectives, population
coverage, benefits to be included, organization of health services, premium calculation and
45
payment mechanism, utilization and cost-control measures and administrative arrangements
(WHO, 2012).
CONCLUSION
After introduction of the NHIF Civil Servants and Discipline Services Medical scheme there was
increased funding for medicines from the FIF but reduced financing for essential medicines from
the government at Webuye District Hospital. There was very little change in availability of
medicines. The new scheme did little to improve access to medicines for civil servants.
Contextual factors such as reduced medicines financing from the government, challenges in
devolution of health and other factors that affect availability of medicines needed to be
considered. Expanding the coverage of the scheme is recommended so as to increase financing
of essential medicines. Additional studies on availability of essential medicines after devolution
of health should be done.
RECOMMENDATIONS
1. Additional financing for essential medicines is required for Webuye District Hospital. NHIF
should be extended coverage to a larger percentage of the population. Alternatively, the
capitation rate should be increased. This will result in more funding to the hospital hence more
financing of medicines. It has been shown that increasing FIF expenditure on essential medicines
reduces the stock-out rate.
2. When implementing new insurance schemes, contextual factors that may affect quality of care
need to be taken into consideration. For example, if NHIF is to extend the outpatient benefits to 46
the general population, one consideration in implementation would be the fact that health
services have been devolved
3. KEMSA lists should be synchronized with facility formulary lists. Even though procurement
is now being done at the county level, KEMSA has an advantage due to efficiency that comes
from economies of scale. However the KEMSA fill-rate should also be improved.
4. Additional studies on determinants of stock-out rates in public hospitals in Kenya should be
done.
5. Studies on quality of medicines procured through the Facility Improvement Fund should be
done.
6. This study should be replicated in other public hospitals.
REFERENCES
Adebayo, E., Hussain, N., Ajanaku, V., (2013). Influence of Health Insurance on Rational
Use of Drugs. TAF Preventive Medicine Bulletin 12, 1.
Chuma, J., Okungu, V., (2011). Viewing the Kenyan health system through an equity
lens: implications for universal coverage. Int J Equity Health 10, 22.
Faden, L., Vialle-Valentin, C., Ross-Degnan, D., Wagner, A., (2011). Active
pharmaceutical management strategies of health insurance systems to improve
cost-effective use of medicines in low- and middle-income countries: a systematic
review of current evidence. Health Policy 100, 134–143.
Faden L, Vialle-Valentin C, Ross-Degnan D, Wagner A., (2011). The Role of Health
Insurance in the Cost-Effective Use of Medicines. WHO/HAI Project on
Medicine Prices and Availability Review Series on Pharmaceutical Pricing
Policies and Interventions Working Paper 2. Available at:
47
http://www.haiweb.org/medicineprices/24072012/HealthinsurancefinalMay2011.
pdf (accessed 22/10/2013)
Frenk, J., Gómez-Dantés, O., Knaul, F.M., (2009). The democratization of health in
Mexico: financial innovations for universal coverage. Bulletin of the World
Health Organization 87, 542–548.
Garabedian, L.F., Ross-Degnan, D., Ratanawijitrasin, S., Stephens, P., Wagner, A.K.,
(2012). Impact of universal health insurance coverage in Thailand on sales and
market share of medicines for non-communicable diseases: an interrupted time
series study. BMJ Open 2, e001686.
Global UNIDO Project, (2010). Strengthening the local production of essential generic
drugs in the least developed and developing countries- Pharmaceutical Sector
Profile: Kenya. Available at:
http://www.unido.org/fileadmin/user_media/Services/PSD/BEP/Kenya_Pharma
%20Sector%20profile_TEGLO05015_Ebook.pdf (accessed 31/10/2013)
Health Systems 20/20 Project and Research and Development Division of the Ghana
Health Service (2009) An Evaluation of the Effects of the National Health
Insurance Scheme in Ghana, Health Systems 20/20 Project, Abt Associates Inc.
USA
Kenya Medical Supplies Agency 2013. Frequently Asked Questions, (2013). Available at
http://www.kemsa.co.ke/index.php?
option=com_content&view=article&id=11&Itemid=112 (accessed 31/8/2014)
Luoma, M., Doherty, J., Muchiri, S., Barasa, T., Hofler, K., Maniscalco, L., 2010. Kenya
Health System Assessment 2010.
48
McIntyre, D., Garshong, B., Mtei, G., Meheus, F., Thiede, M., Akazili, J., Ally, M.,
Aikins, M., Mulligan, J.-A., Goudge, J., (2008). Beyond fragmentation and
towards universal coverage: insights from Ghana, South Africa and the United
Republic of Tanzania. Bulletin of the World Health Organization 86, 871–876.
Ministry Of Health and Social Services, 2008. Report on medicines coverage and Health
insurance programs survey In Tanzania. Available at
http://www.who.int/medicines/areas/coordination/
tanzania_study_insurance_coverage.pdf (accessed 22/10/2013)
Ministry of Health, 2009. Public Expenditure Tracking Survey 2008. Available at
www.ihpmr.org/wp-content/uploads/2012/10/PETS-report-210309.doc (accessed 28/11/2013)
Minitsry of Health, (2013). Press statement. Available at
http://www.kemsa.co.ke/index.php?
option=com_content&view=article&id=68:press-
statements&catid=13:news&Itemid=150 (accessed 31/8/2014)
Ministry Of Medical Services, (2010).Kenya Pharmaceutical Country Profile. Available
at:
http://www.who.int/medicines/areas/coordination/kenya_pharmaceuticalprofile_d
ecember2010.pdf (accessed 9/12/2013)
Ministry of Medical Services and Ministry of Public Health & Sanitation, (2009). Access
to Essential Medicines in Kenya -A Health Facility Survey. Available at:
http://apps.who.int/medicinedocs/documents/s18695en/s18695en.pdf
(accessed 15/09/2013)
49
Ministry of Medical Services and Ministry of Public Health & Sanitation, (2010).Kenya
Essential Medicines List 2010. Available at:
http://apps.who.int/medicinedocs/documents/s18694en/s18694en.pdf
(accessed 12/08/2013)
Ministry of Medical Services and Ministry of Public Health & Sanitation, (2011). Kenya
National Health Accounts 2009/10. Available at:
http:/apps.who.int/nha/country/ken/kenya_nha_2009-2010.pdf
(accessed 13/09/2013)
Mission for Essential Drugs and Supplies (MEDS) (2013). MEDS County Government
Partnerships Initiative 2013. Available at
http://www.meds.or.ke/Partnerships.html (accessed 31/08/2014)
Mwathi M.W, Osuga B. O., (2014). Availability of essential medicines in public
hospitals: A study of selected public hospitals in Nakuru County, Kenya.
Available at http://www.academicjournals.org/article/article1398866137_Wangu
%20and%20Osuga.pdf (accessed 16/08/2014)
Parliamentary Committee for Health. Eleventh Parliament– Second Session: Report Of
The Departmental Committee On Health On Emerging Challenges of Devolving
Health Services. Available at
http://www.parliament.go.ke/plone/national-assembly/committees/
copy_of_committee-reports/report-of-the-departmental-committee-on-health-on-
emerging-challenges-of-devolving-health-services/at_multi_download/item_files?
name=Health%20Devolution%20Report.pdf (accessed 31/08/2014)
R Paul Shaw, M.A., (1995). Financing health services through user fees and insurance.
Case studies from Sub-Saharan Africa. World Bank discussion papers.
50
Seiter, A., (2010). A practical approach to pharmaceutical policy. World Bank
Publications.
SEND – Ghana (2010) Balancing Access with Quality HealthCare: An Assessment of
the NHIS in Ghana (2004–2008) http://www.sendwestafrica.org (accessed
22/10/2013)
Shung-King, M., (2011). The Dilemmas of Co-Payment and Moral Hazard in the Context
of an NHI. HEU Working Paper. Available at : http://uct-
heu.s3.amazonaws.com/wp-content/uploads/2012/03/CO-PAYMENT-
WORKING-PAPER.pdf
(accessed 22/10/2013)
The World Bank, (2009). Public Sector Healthcare Supply Chain Strategic Network
Design For Kemsa -Driving Service Improvements through Supply Chain
Excellence. Available at:
http://siteresources.worldbank.org/INTHIVAIDS/Resources/375798-
1103037153392/SupplyChaingFinalReportKenya.pdf (accessed 13/12/2013)
UN Millennium Project, (2005). Prescription for Healthy Development: Increasing
Access to Medicines. Report of the Task Force on HIV/AIDS, Malaria, TB, and
Access to Essential Medicines, Working Group on Access to Essential Medicines.
Sterling, Va.: Earthcscan.
United Nations 2003, Indicators for Monitoring the Millennium Development Goals.
Available at: http://unstats.un.org/unsd/publication/seriesf/Seriesf_95E.pdf
(accessed 16/12/2013)
USAID; Health Finance and Governance, (2014). Case Study: Kenya National Hospital
51
Insurance Fund (NHIF) Premium Collection for the Informal Sector. Available at
https://www.hfgproject.org/wp-content/uploads/2014/06/14-0611-
NHIF_mpesa_FINAL1.pdf (accessed 14/09/2014)
Wales, J., Tobias, J., Malangalila, E., Swai, G., Wild,L, (2014). Stock-outs of essential
medicines in Tanzania: A political economy approach to analyzing problems and
identifying solutions. Available at http://twaweza.org/uploads/files/Stock-outs
%20of%20essential%20medicines%20in%20Tanzania%20-%20ODI%20&
%20TWA%20-%20final%20March2014.pdf (accessed 14/09/2014)
Wamai, R.G., (2009). The Kenya Health System–Analysis of the situation and enduring
challenges. Japan Medical Association Journal 52, 134–140.
World Health Organisation, (2002). The selection of essential medicines. WHO policy
perspectives on medicines. Available at:
http://whqlibdoc.who.int/hq/2002/WHO_EDM_2002.2.pdf (accessed 17/12/2013)
World Health Organization, (2004). The world medicines situation. Geneva: World
Health Organization. Available at
http://apps.who.int/medicinedocs/pdf/s6160e/s6160e.pdf (accessed 16/12/2013)
World Health Organisation, (2004). Equitable access to essential medicines: a framework
for collective action. WHO policy perspectives on medicines. Available at
http://whqlibdoc.who.int/hq/2004/WHO_EDM_2004.4.pdf (accessed 12/08/2013)
World Health Organization, (2006).Medicine prices surveys and proposed interventions
to improve sustainable access to affordable medicines in 6 sub-Saharan African
countries. Available at :
http://www.who.int/medicines/areas/technical_cooperation/Medpricesall8files.pdf
(accessed 13/12/2013)
52
World Health Organization, (2012). Pharmaceutical Financing Strategies. Available at
http://apps.who.int/medicinedocs/documents/s19587en/s19587en.pdf
(accessed 9/12/2013)
Xu, K., James, C., Carrin, G., Muchiri, S., (2006). An empirical model of access to
health care, health care expenditure and impoverishment in Kenya: learning from
past reforms and lessons for the future. Geneva: WHO, Department of Health
Systems Financing, Health Financing Policy, Discussion paper.
Zhu, M., Dib, H.H., Zhang, X., Tang, S., Liu, L., (2008). The influence of health
insurance towards accessing essential medicines: the experience from Shenzhen
labor health insurance. Health Policy 88, 371–380.
53
APPENDIX I: DATA COLLECTION FORMS
Form 1-FIF Allocated for purchase of medicines
2010-2011 2012-2013
Amount
allocated
(Kshs)
Total FIF
for the
quarter
% of
Total
FIF
Total
Amount
from
NHIF
Amount
allocated
(Kshs)
Total
FIF for
the
quarter
% of
Total
FIF
Total
Amount
from
NHIF
Quarter 1
Quarter 2
Quarter 3
Quarter
X
Form 2-Number of days out of stock per month
Essential Medicine STRENGTH 2010-2011 2012-2013
Number of days out of
stock in a month
(January to December)
Number of days out of
stock in a month (January
to December)
ANAESTHETIC & ALLIED
Atropine injection 1mg/ml
Bupivacaine Injection 5mg/ml
Calcium gluconate injection 1G/10ml
Ephedrine injection
Halothane inhalation 100%v/v
Ketamine injection 50mg/ml
Lignocaine adrenaline/dental catridges 2%
Lignocaine injection 20mg/ml
Neostigmine injection 2.5mg/ml
54
Naloxone Injection
Pancuronium injection 2mg/ml
Sodium bicarbonate injection 0.84G/10ml
Suxamethonium injection 50mg/ml
Thiopentone injection 500mg/vial
Xylocaine (lignocaine) spray
ANALGESICS, ANTI-INFLAMMATORY
Aspirin 300mg tabs 300mg
Diclofenac Injection 75mg/ml 75/ml
Ibuprofen 200mg tablets 200mg
Ibuprofen suspension 200mg/5ml
Morphine Injection 10mg/ml
Morphine oral solution 1mg/ml
Paracetamol 500mg tabs 500mg
Paracetamol syrup 120mg/5ml
Paracetamol suppositories 125mg
ANTIALLERGICS, ANTIASTHMA DRUGS
Aminophylline injection 250mg/10ml
Beclomethasone Inhaler 100µg/puff
Chlorpheniramine 4mg tabs 4mg
Chlorpheniramine injection 10mg/ml
Chlorpheniramine syrup 4mg/5ml
Dexamethasone 0.5mg tabs 0.5mg
Dexamethasone injection 4mg/ml
Hydrocortisone injection 100mg/vial
Prednisolone 5mg 5mg
Salbutamol 4mg tabs 4mg
Salbutamol inhaler 200µg/puff
Salbutamol nebulising soln 5mg/ml
ANTICOAGULANTS
Heparin Inj 5000iu/ml
Warfarin 5mg tabs po
ANTIDIABETICS
Glibenclamide 5mg tabs 5mg/ml
Metformin 500mg tabs 500mg
Insulin Isophane Biphasic human 30/70 1000i.u
Insulin soluble human 1000i.u
55
ANTIHELMINTHICS
Albendazole 400mg tabs 400mg
Praziquantel 600mg tablets 600mg
CARDIOVASCULAR DRUGS
Atenolol 50mg tabs 50mg
Digoxin 0.25mg tabs 0.25mg
Enalapril 5mg tabs 5mg
Frusemide 40mg tabs 40mg
Frusemide injection 20mg/2ml
Hydralazine 20mg injection 20mg
Hydrochlorothiazide 50mg tabs 50mg
Magnesium sulphate injection 5g/10ml
Methyl Dopa 250mg tabs 250mg
Propranolol 40mg tabs 40mg
CNS DRUGS
Amitryptilline 25mg tabs 25mg
Benzhexol 5mg tabs 5mg
Carbamazepine 200mg tabs 200mg
Chlorpromazine 100mg tabs 100mg
Chlorpromazine injection 50mg/2ml
Diazepam 5mg tabs 5mg
Diazepam injection 10mg/2ml
Fluphenazine decanoate injection 25mg/ml
Fluoxetine caps 20mg
Haloperidol 5mg tabs 5mg
Haloperidol decanoate injection
Phenobarbitone 30mg tabs 30mg
Phenobarbitone injection
Phenytoin 50mg tablets 50mg
Phenytoin Injection
Valproic acid 200mg
GIT ACTING DRUGS
Bisacodyl tabs 5mg 5mg
Magnesium Trisilicate-Antacids 250/125mg
Metoclopramide 10mg tablets 10mg
56
Metoclopramide injection 10ml/2ml
Omeprazole 20mg caps 20mg
Ranitidine injection 25mg/ml
IV FLUIDS
Dextrose 10%
Dextrose 5%
Dextrose 50%
Half strength darrows
Normal saline (0.9% NaCl)
Sodium Lactate (hartmann's)
ORAL ANTIBIOTICS
Amoxicillin 250mg caps 250mg
Amoxicillin dry suspension 125mg/5ml
Amoxicillin/Clavulanate 625mg tabs 625mg
Amoxicillin-clavulanic acid 228mg/5ml 228mg/5ml
Chloramphenical suspension 125mg/5ml
Chloramphenical caps 250mg
Ciprofloxacin 250mg tabs 250mg
Cotrimoxazole 400:80 tabs 480mg
Clindamycin tabs 150mg
Doxycycline 100mg caps 100mg
Erythromycin suspension 125mg/5ml
Erythromycin tabs 250mg 250mg
Flucloxacillin 250mg caps 250mg
Metronidazole 200mg tabs 200mg
Metronidazole suspension 200mg/5ml
Nitrofurantoin 100mg tabs 100mg
Tinidazole tabs 500mg
ORAL ANTIFUNGALS
Fluconazole tabs 50mg
Griseofulvin 250mg 250mg
Nystatin suspension 100,000iu/ml
EAR, EYE & SKIN PREPS
Amethocaine eye drops
Atropine Eye Drops 1%
BenzylBenzoate Emulsion
Calamine Lotion
57
Ciprofloxacin ear/eye drops
Clotrimazole cream
Gentamicin eye drops
Hydrocortisone ointment
Prednisolone eye drops
Salicylic acid powder
Silver sulphadiazine ointment
Tetracycline eye ointment
ANTIMALARIALS
Artemether/Lumefantrine 12 tabs
Artemether/Lumefantrine 18 tabs
Artemether/Lumefantrine 24 tabs
Artemether/Lumefantrine 6 tabs
Quinine 300mg tabs 300mg
Quinine injection 300mg/ml
Sulfadoxine/pyrimethamine tabs 500mg/25mg
PARENTERAL ANTIBIOTICS
Ampicillin 500mg injection 500mg
Benzathine Penicillin 2.4MU 2.4MU
Benzyl penicillin 1 MU 1MU
Benzyl penicillin 5MU 5MU
Ceftriaxone Inj 1g IG/vial
Ceftriaxone Inj 250mg 250mg
Chloramphenical injection 10mg/ml
(Flu)cloxacillin 250mgs inj 250mg
Gentamicin injection 20mg/2ml 10mg/ml
Gentamicin injection 80mg/2ml 40mg/ml
Metronidazole injection 500mg 500mg
SUPPLEMENTS
Ferrous sulphate 200mg tabs 200mg
Folic acid 5mg tabs 5mg
Oral Rehydration Salts
Potassium Chloride Injection 150mg
Zinc Tablets 20mg
THYROID&ANTITHYROID
58
Carbimazole 5mg
OTHERS
Activated Charcoal tabs
Acyclovir 400mg tablets 400mg
Clotrimazole pessaries 200mg
Dextran 70 500ml
Misoprostol tablet/pessary 200mg
Oxytocin injection
Water for injections 10ml
Form 3-Procurement of essential medicines by month
MONTH & YEAR……………………………………………….
Essential
Medicine
procured
Quantity
from
KEMSA
Quantity
from FIF
Quantity
from
other
GOK
facility
Quantity
from
other
sources
Total
Quantity
Essential
Medicine
procured
By Cost
KEMSA
By Cost
FIF
By Cost
other
GOK
facility
By Cost
from
other
sources
Total Cost
Form 4-KEMSA fill-rate and presence on KEMSA order form
CYCLE/DATE ORDER VALUE RECEIVED
VALUE
PREVIOUS
ORDER’S
FILL RATE
1
59
2
3
X
ORDER DATE………………………………….
Essential Medicine
on KEMSA list
Ordered
quantity(A)
Received
Quantity (B)
Fill ratio
(A/B)
Total Stock-
out rate
1
2
3
X
Record whether on medicine was ordered and
came, not ordered, ordered but didn’t come, not on
KEMSA list
Essential Medicine KEMSA
ORDER 1
KEMSA
ORDER 2
KEMSA
ORDER X
1
2
3
X
Form 5-Nature of drugs procured per quarter from FIF
60
MONTH………………………………………………………………..
DRUG
PROCUREDEssential?(No/Yes) Cost (KShs) On hospital draft
formulary? (No/yes)
Form 6-Consumption rate of essential drugs
Form 7-Monthly workload
61
Essential Drug UNIT
Adjusted Consumption rate
in daily defined dose(DDD)
Average cost
per DDD
1
2
3
4
Month
Outpatient
workload(numbe
r of patients)
Inpatient workload(as total
bed days)
1
2
3
4
62
APPENDIX II-Letter of Ethical Approval
63
64
